Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
"Beam Me Up, Scotty" (Beamus) AM Transmitter -- first prototype
81 views
Skip to first unread message
flipper
Jun 10, 2012, 1:39:00 PM6/10/12
to
Okay, so I like to make up cutesy names :)
I wanted to do 'something' with a 6ME8 so I tried using it for an AM
transmitter and the first prototype works rather well. Bandwidth is
'too much', less than 1 dB down at 18 KHz, but we'll worry about that
later.
This is a 'dollar days' special, 2 whole bucks worth of tubes, and L2
is a UHF converter coil scramble rewound, so that was 'free'. Power
supply is from the same converter and if the remaining issues get
worked out it'll probably end up in that cabinet as well (but where to
put the air variable?).
The schematic is rough, but should be rather self explanatory, and the
ganged tuning hasn't been fleshed out yet. I'm still using separate
caps while jiggling things around.
Orphaned web page, not yet ready for prime time, has a recording of it
playing through a table radio.
http://flipperhome.dyndns.org/Beamus.htm
I wanted to do 'something' with a 6ME8 so I tried using it for an AM
transmitter and the first prototype works rather well. Bandwidth is
'too much', less than 1 dB down at 18 KHz, but we'll worry about that
later.
This is a 'dollar days' special, 2 whole bucks worth of tubes, and L2
is a UHF converter coil scramble rewound, so that was 'free'. Power
supply is from the same converter and if the remaining issues get
worked out it'll probably end up in that cabinet as well (but where to
put the air variable?).
The schematic is rough, but should be rather self explanatory, and the
ganged tuning hasn't been fleshed out yet. I'm still using separate
caps while jiggling things around.
Orphaned web page, not yet ready for prime time, has a recording of it
playing through a table radio.
http://flipperhome.dyndns.org/Beamus.htm
Doug Bannard
Jun 10, 2012, 7:48:23 PM6/10/12
to
"flipper" <fli...@fish.net> wrote in message
news:3jl9t7hgrb5tqj0ro...@4ax.com...
looking output waveform. I'll bet that it'll sound great playing through a
vintage receiver!
Best Regards : Doug Bannard
Alex Pogossov
Jun 10, 2012, 10:50:03 PM6/10/12
to
in the vintage radio cabinets.
http://www.ebay.com.au/itm/140770446284?ssPageName=STRK:MESELX:IT&_trksid=p3984.m1555.l2649
flipper
Jun 10, 2012, 11:12:58 PM6/10/12
to
idea before the UHF converters and I even bought a supposedly non
working Philips radio to use. Turned out the radio only needed one
tube, which I had. It could also use a recap but it's in good enough
shape I didn't have the heart to scavenge it.
I might resurrect the idea for the air variable and dial indicator but
the biggest problem is I want power transformer isolation and most
don't have one (at least not the AA5s), nor 'extra room' to fit one. I
notice that Little Nipper does but it's 240 VAC.
I think my 2 bucks 2 tube Beamus scope shot looks as good as his 5
tube job ;)
flipper
Jun 11, 2012, 1:17:34 AM6/11/12
to
On Sun, 10 Jun 2012 19:48:23 -0400, "Doug Bannard" <ve3...@rogers.com>
wrote:
Thank you.
Yes, it does sound good though my Zenith 845 but by the time I got
through the cornucopia of recording problems it ended up on a circa
80's Panasonic clock radio, which is a fairly decent one with a 6x4
speaker. The single speaker helped with mic directional problems.
wrote:
Yes, it does sound good though my Zenith 845 but by the time I got
through the cornucopia of recording problems it ended up on a circa
80's Panasonic clock radio, which is a fairly decent one with a 6x4
speaker. The single speaker helped with mic directional problems.
Alex Pogossov
Jun 11, 2012, 12:16:59 AM6/11/12
to
news:ccnat7525hedeun71...@4ax.com...
Apparently, the more tubes -- the more "decent" and "professional" it might
look... But your 2-tube solution looks like gives clean modulation due to
the feedback.
However, the biggest (potential) problem of your design is frequency
instability. If you have a parasitic coupling between the aerial and the
oscillator coil, you will get the "frequency pull". If you touch the antenna
(changing the voltage on it) -- you will get a frequency shift! This pull
will also be modulated, so you will get a spurious FM. Probably acceptable
for an AM radio, but will sound crap on an SSB receiver. So your major goal
is to shield your oscillator circuit (L1, C1b and other components). And
even if you make a perfect shielding, you will not be able to completely
avoid parasitic coupling through the shaft impedance of the dual gang
variable capacitor.
look... But your 2-tube solution looks like gives clean modulation due to
the feedback.
However, the biggest (potential) problem of your design is frequency
instability. If you have a parasitic coupling between the aerial and the
oscillator coil, you will get the "frequency pull". If you touch the antenna
(changing the voltage on it) -- you will get a frequency shift! This pull
will also be modulated, so you will get a spurious FM. Probably acceptable
for an AM radio, but will sound crap on an SSB receiver. So your major goal
is to shield your oscillator circuit (L1, C1b and other components). And
even if you make a perfect shielding, you will not be able to completely
avoid parasitic coupling through the shaft impedance of the dual gang
variable capacitor.
flipper
Jun 11, 2012, 11:59:12 PM6/11/12
to
On Mon, 11 Jun 2012 14:16:59 +1000, "Alex Pogossov"
reactance tube FM transmitter. And the Twin Triode Transmitter, come
to think of it.
For not being an 'RF guy' I sure have done a lot of RF lately.
>However, the biggest (potential) problem of your design is frequency
>instability. If you have a parasitic coupling between the aerial and the
>oscillator coil, you will get the "frequency pull". If you touch the antenna
>(changing the voltage on it) -- you will get a frequency shift! This pull
>will also be modulated, so you will get a spurious FM. Probably acceptable
>for an AM radio, but will sound crap on an SSB receiver. So your major goal
>is to shield your oscillator circuit (L1, C1b and other components). And
>even if you make a perfect shielding, you will not be able to completely
>avoid parasitic coupling through the shaft impedance of the dual gang
>variable capacitor.
>
That sounds a lot like the old joke "Doc, it hurts when I poke a stick
in my eye, What should I do about that?
Don't poke a stick in your eye." ;)
Don't touch the antenna.
I do intend to shield the LO and, for cost reasons, dropping the gang
tuning idea. (Maybe I should blame it on a 'bean counter', eh?)
There's been a revision. The 70 uH was low and the scramble wind on a
UHF core not so hot, which was causing low plate swing. So I've now
gone to a 250 uH ferrite rod antenna for the plate load and am getting
80 Vpp at idle. And, good as it was, modulation depth also improved.
It's now so close to 100% you can't tell the difference unless you
blow the scale up.
Looks to me like this one is a winner but, of course, I'm a bit biased
;)
Revised schematic is posted http://flipperhome.dyndns.org/Beamus.htm
>> I think my 2 bucks 2 tube Beamus scope shot looks as good as his 5
>> tube job ;)
>
>Apparently, the more tubes -- the more "decent" and "professional" it might
>look... But your 2-tube solution looks like gives clean modulation due to
>the feedback.
Yep, NFB is the key to getting g1 linear. I did the same thing on my
>> tube job ;)
>
>Apparently, the more tubes -- the more "decent" and "professional" it might
>look... But your 2-tube solution looks like gives clean modulation due to
>the feedback.
reactance tube FM transmitter. And the Twin Triode Transmitter, come
to think of it.
For not being an 'RF guy' I sure have done a lot of RF lately.
>However, the biggest (potential) problem of your design is frequency
>instability. If you have a parasitic coupling between the aerial and the
>oscillator coil, you will get the "frequency pull". If you touch the antenna
>(changing the voltage on it) -- you will get a frequency shift! This pull
>will also be modulated, so you will get a spurious FM. Probably acceptable
>for an AM radio, but will sound crap on an SSB receiver. So your major goal
>is to shield your oscillator circuit (L1, C1b and other components). And
>even if you make a perfect shielding, you will not be able to completely
>avoid parasitic coupling through the shaft impedance of the dual gang
>variable capacitor.
>
in my eye, What should I do about that?
Don't poke a stick in your eye." ;)
Don't touch the antenna.
I do intend to shield the LO and, for cost reasons, dropping the gang
tuning idea. (Maybe I should blame it on a 'bean counter', eh?)
There's been a revision. The 70 uH was low and the scramble wind on a
UHF core not so hot, which was causing low plate swing. So I've now
gone to a 250 uH ferrite rod antenna for the plate load and am getting
80 Vpp at idle. And, good as it was, modulation depth also improved.
It's now so close to 100% you can't tell the difference unless you
blow the scale up.
Looks to me like this one is a winner but, of course, I'm a bit biased
;)
Revised schematic is posted http://flipperhome.dyndns.org/Beamus.htm
Patrick Turner
Jun 12, 2012, 5:37:15 AM6/12/12
to
On Jun 11, 12:50 pm, "Alex Pogossov" <apogo...@tpg.com.au> wrote:
> "flipper" <flip...@fish.net> wrote in message
> in the vintage radio cabinets.http://www.ebay.com.au/itm/140770446284?ssPageName=STRK:MESELX:IT&_tr...- Hide quoted text -
>
> - Show quoted text -
Hmm, I must qualify as a dumb-ass aussie because I've not tried to
make an AM transmitter in such a nice package, all kinda trick stuff.
But you know, I recall that while I studied FM tube radios so I could
service them properly in 1995, I mentioned production of an FM test
signal by means of a reactance tube FM modulator to a bunch of radio
enthusiasts at a ham gathering one evening to see if they had any
schematics or knowhow. None did, and one said "Mate, waddia wanna vary
frequency for? We jus wanna keep the frequency stable, eh..." Well,
OK, they were stuck back in 1955, most were over 70 in 1995. Now most
are dead, and I'm still lingering with audio F stuff, having
farnarkled muchly with RF home brew testorator checker thingies.
But I was about 15 when I made my own transmitter out of an existing
old radio set. I got my sister to talk into the crystal mike while
walked up the street to see how far it broadcast, using my mum's new
Japanese 9 transistor portable. After I'd gone about 1/4 mile dear
sister starts saying "This is the illegal broadcast by Patrick Turner
from No ----- street....." and then I qualified for the Olympics 400M
getting back home to shut the sister up. Females can be cheeky, and
sis was no exception.
I have an idea to make another modulator which might work merely by
having 3 cascaded variable µ pentodes, each contributing about 30% AM.
maybe it works, maybe it don't, but it'll have to wait until so many
other things more worthwhile are completed.
In Canberra we get ABC Radio National on 845kHz AM with AF BW allowed
to be 9kHz, so my AM radio has 10kHz BW using variable selectivity in
the 1st IFT by varying the distance between the coils. But one could
always use the Internet to get the podcasts at
http://www.abc.net.au/radionational/podcasts/program/
And you just subscribe. In many places RN is broadcast on FM, and now
in digital stations.
So I guess one could tune into RN on the digital and use the audio to
modulate a carrier for the AM radio to pick up, and that should sound
better than if one tries to pick up the local AM radio station pumping
out 5kW, and only 4km away.
Just don't ask me to describe how digital radio works. I looked for
radio schematics and found none, seems like secret business nobody
wants me to know, but seems like the digital band is at UHF or around
250MHz, and just what the wave forms look like I don't know.
Patrick Turner.
> "flipper" <flip...@fish.net> wrote in message
>
> - Show quoted text -
Hmm, I must qualify as a dumb-ass aussie because I've not tried to
make an AM transmitter in such a nice package, all kinda trick stuff.
But you know, I recall that while I studied FM tube radios so I could
service them properly in 1995, I mentioned production of an FM test
signal by means of a reactance tube FM modulator to a bunch of radio
enthusiasts at a ham gathering one evening to see if they had any
schematics or knowhow. None did, and one said "Mate, waddia wanna vary
frequency for? We jus wanna keep the frequency stable, eh..." Well,
OK, they were stuck back in 1955, most were over 70 in 1995. Now most
are dead, and I'm still lingering with audio F stuff, having
farnarkled muchly with RF home brew testorator checker thingies.
But I was about 15 when I made my own transmitter out of an existing
old radio set. I got my sister to talk into the crystal mike while
walked up the street to see how far it broadcast, using my mum's new
Japanese 9 transistor portable. After I'd gone about 1/4 mile dear
sister starts saying "This is the illegal broadcast by Patrick Turner
from No ----- street....." and then I qualified for the Olympics 400M
getting back home to shut the sister up. Females can be cheeky, and
sis was no exception.
I have an idea to make another modulator which might work merely by
having 3 cascaded variable µ pentodes, each contributing about 30% AM.
maybe it works, maybe it don't, but it'll have to wait until so many
other things more worthwhile are completed.
In Canberra we get ABC Radio National on 845kHz AM with AF BW allowed
to be 9kHz, so my AM radio has 10kHz BW using variable selectivity in
the 1st IFT by varying the distance between the coils. But one could
always use the Internet to get the podcasts at
http://www.abc.net.au/radionational/podcasts/program/
And you just subscribe. In many places RN is broadcast on FM, and now
in digital stations.
So I guess one could tune into RN on the digital and use the audio to
modulate a carrier for the AM radio to pick up, and that should sound
better than if one tries to pick up the local AM radio station pumping
out 5kW, and only 4km away.
Just don't ask me to describe how digital radio works. I looked for
radio schematics and found none, seems like secret business nobody
wants me to know, but seems like the digital band is at UHF or around
250MHz, and just what the wave forms look like I don't know.
Patrick Turner.
John Byrns
Jun 13, 2012, 4:48:48 PM6/13/12
to
In article <3jl9t7hgrb5tqj0ro...@4ax.com>,
Hi Flipper,
Dropping the ganged capacitor was probably a good idea, if it was a
superheterodyne capacitor there was probably no way you could have made it track
correctly. Did you ever think of using a crystal in the oscillator circuit,
that way there would be only one tuning adjustment? What is the potentiometer
in the oscillator screen circuit all about, what is the criterion for adjusting
it?
You state that 80 Vpp on the plate equates to about 35 mW, how did you calculate
that?
--
Regards,
John Byrns
Surf my web pages at, http://fmamradios.com/
Dropping the ganged capacitor was probably a good idea, if it was a
superheterodyne capacitor there was probably no way you could have made it track
correctly. Did you ever think of using a crystal in the oscillator circuit,
that way there would be only one tuning adjustment? What is the potentiometer
in the oscillator screen circuit all about, what is the criterion for adjusting
it?
You state that 80 Vpp on the plate equates to about 35 mW, how did you calculate
that?
--
Regards,
John Byrns
Surf my web pages at, http://fmamradios.com/
flipper
Jun 14, 2012, 6:38:52 AM6/14/12
to
On Wed, 13 Jun 2012 15:48:48 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
>In article <3jl9t7hgrb5tqj0ro...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> Okay, so I like to make up cutesy names :)
>>
>> I wanted to do 'something' with a 6ME8 so I tried using it for an AM
>> transmitter and the first prototype works rather well. Bandwidth is
>> 'too much', less than 1 dB down at 18 KHz, but we'll worry about that
>> later.
>>
>> This is a 'dollar days' special, 2 whole bucks worth of tubes, and L2
>> is a UHF converter coil scramble rewound, so that was 'free'. Power
>> supply is from the same converter and if the remaining issues get
>> worked out it'll probably end up in that cabinet as well (but where to
>> put the air variable?).
>>
>> The schematic is rough, but should be rather self explanatory, and the
>> ganged tuning hasn't been fleshed out yet. I'm still using separate
>> caps while jiggling things around.
>>
>> Orphaned web page, not yet ready for prime time, has a recording of it
>> playing through a table radio.
>>
>> http://flipperhome.dyndns.org/Beamus.htm
>
>Hi Flipper,
>
>Dropping the ganged capacitor was probably a good idea, if it was a
>superheterodyne capacitor there was probably no way you could have made it track
>correctly. Did you ever think of using a crystal in the oscillator circuit,
>that way there would be only one tuning adjustment?
I wanted the frequency, at least some degree, tunable.
> What is the potentiometer
>in the oscillator screen circuit all about,
That adjusts osc RF amplitude to the deflector plates.
> what is the criterion for adjusting
>it?
I'm still pondering that a bit but current procedure is to run it up
to near max mod and adjust for maximum peak to peak.
>You state that 80 Vpp on the plate equates to about 35 mW, how did you calculate
>that?
That's guesstimated EI across the plate load.
wrote:
>In article <3jl9t7hgrb5tqj0ro...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> Okay, so I like to make up cutesy names :)
>>
>> I wanted to do 'something' with a 6ME8 so I tried using it for an AM
>> transmitter and the first prototype works rather well. Bandwidth is
>> 'too much', less than 1 dB down at 18 KHz, but we'll worry about that
>> later.
>>
>> This is a 'dollar days' special, 2 whole bucks worth of tubes, and L2
>> is a UHF converter coil scramble rewound, so that was 'free'. Power
>> supply is from the same converter and if the remaining issues get
>> worked out it'll probably end up in that cabinet as well (but where to
>> put the air variable?).
>>
>> The schematic is rough, but should be rather self explanatory, and the
>> ganged tuning hasn't been fleshed out yet. I'm still using separate
>> caps while jiggling things around.
>>
>> Orphaned web page, not yet ready for prime time, has a recording of it
>> playing through a table radio.
>>
>> http://flipperhome.dyndns.org/Beamus.htm
>
>Hi Flipper,
>
>Dropping the ganged capacitor was probably a good idea, if it was a
>superheterodyne capacitor there was probably no way you could have made it track
>correctly. Did you ever think of using a crystal in the oscillator circuit,
>that way there would be only one tuning adjustment?
> What is the potentiometer
>in the oscillator screen circuit all about,
> what is the criterion for adjusting
>it?
to near max mod and adjust for maximum peak to peak.
>You state that 80 Vpp on the plate equates to about 35 mW, how did you calculate
>that?
Patrick Turner
Jun 15, 2012, 1:23:29 AM6/15/12
to
If he has 35mW and say 50Vrms anode signal at antenna output, then
0.035 = 50 x 50 / RL so RL = 71kohms. This is rather high but the tuned tank circuit probably has such a high Z at Fo.
The comments made about coils are of interest, and it should make no difference what type of core is used, because 270uH is easy to achieve with many recipes. But in the best old radios the 455kHz and MW RF coils are pie wound windings using litz wire and large Al cans to keep the Q high, ie, RF resistance low. In an AM receiver I built from scratch in 1999, I made copper cans for two RF input coils and I wound solid wire on 50mm long x 10mm dia pieces of ferrite rod normally used for ferrite rod antennas. The Q became higher than using air cored windings on say 20mm PVC pipes. In my receiver, the Q was so high the BW was less than 15kHz, giving slight cutting of sidebands and thus reducing possible audio BW, so I used TWO such coils, and a 3 gang cap, with 2 gands devoted to the the two RF coild and the other to the oscillator. To overcome the reduction of BW, I used antenna input to a tap on one coil, then a resistor from top of coil to top of next coil and then I tuned the coils so they were "stagger tuned", ie, slightly apart at the low end of the band where the LC Q is higher than at above say 1.2MHz. All that worked well, but compact fluorescent lamps and other junk creates a rectified version of incoming RF and I've had to change to a long rod ferrite antenna with a shielded coil and very short wires to keep out electrostatic portion of RF wave.
In my RF generator which I could use as transmitter, I get the same very nice AM wave on the CRO as Flipper's site shows. I have a 6BX6 for the RF output tube, and the coil is a hand wound thing on ferrite with 3 layers of not very tightly wound windings of solid wire, and with PVC tape used between layers. The self capacitance of the coil must be kept low if you wish to be able to tune the tank to as high as possible, say 1,750kHz.
Many old radios went from 550kHz to maybe 1,550kHz, a 3 fold F increase. Say you have 20pF for coil self C, then if the cap gang = 365pF max, then total max C = 385pF. Say the coil turns or ferrite position adjusted for Fo = 530kHz, then L must be 234uH. Say the C gang min C = 25pF, and you have 20pF self C
then Cmin total = 45pF which gives Fo = 1,551kHz, and not as high as anyone should accept. If C minimum could be reduced to 22.5pF, Fo = 2,194kHz, but you will never see this, and besides, one has to be able to use a trimmer cap to set the top of the F range. So the less stray C or coil self C, the better.
Keep the coil cans as big as possible, and never use iron cans.
Patrick Turner.
flipper
Jun 15, 2012, 5:49:40 AM6/15/12
to
On Thu, 14 Jun 2012 22:23:29 -0700 (PDT), Patrick Turner
<in...@turneraudio.com.au> wrote:
>On Monday, 11 June 2012 03:39:00 UTC+10, flipper wrote:
>> Okay, so I like to make up cutesy names :)
>>
>> I wanted to do 'something' with a 6ME8 so I tried using it for an AM
>> transmitter and the first prototype works rather well. Bandwidth is
>> 'too much', less than 1 dB down at 18 KHz, but we'll worry about that
>> later.
>>
>> This is a 'dollar days' special, 2 whole bucks worth of tubes, and L2
>> is a UHF converter coil scramble rewound, so that was 'free'. Power
>> supply is from the same converter and if the remaining issues get
>> worked out it'll probably end up in that cabinet as well (but where to
>> put the air variable?).
>>
>> The schematic is rough, but should be rather self explanatory, and the
>> ganged tuning hasn't been fleshed out yet. I'm still using separate
>> caps while jiggling things around.
>>
>> Orphaned web page, not yet ready for prime time, has a recording of it
>> playing through a table radio.
>>
>> http://flipperhome.dyndns.org/Beamus.htm
>
>I couldn't download Fliper's page above on his transmitter last week
Sorry about that. Every once in a while the servers is offline for
<in...@turneraudio.com.au> wrote:
>On Monday, 11 June 2012 03:39:00 UTC+10, flipper wrote:
>> Okay, so I like to make up cutesy names :)
>>
>> I wanted to do 'something' with a 6ME8 so I tried using it for an AM
>> transmitter and the first prototype works rather well. Bandwidth is
>> 'too much', less than 1 dB down at 18 KHz, but we'll worry about that
>> later.
>>
>> This is a 'dollar days' special, 2 whole bucks worth of tubes, and L2
>> is a UHF converter coil scramble rewound, so that was 'free'. Power
>> supply is from the same converter and if the remaining issues get
>> worked out it'll probably end up in that cabinet as well (but where to
>> put the air variable?).
>>
>> The schematic is rough, but should be rather self explanatory, and the
>> ganged tuning hasn't been fleshed out yet. I'm still using separate
>> caps while jiggling things around.
>>
>> Orphaned web page, not yet ready for prime time, has a recording of it
>> playing through a table radio.
>>
>> http://flipperhome.dyndns.org/Beamus.htm
>
>I couldn't download Fliper's page above on his transmitter last week
various reasons.
> but can now, and the circuit looks well.
Must say I'm rather pleased with the deep mod and excellent bandwidth
using only two 1 buck tubes.
> Possibly some improvement could be to used cathode followers to drive each beam delctor plate.
deflector plate. They're being driven single ended.
I've considered that driving the deflectors with a PP transformer
might improve RF harmonics, and maybe a CF to drive it's primary, but
I think it might be possible to simply wind a PP secondary on the osc
coil. However, I'm not convinced it's worth the added complexity.
>If he has 35mW and say 50Vrms anode signal at antenna output, then
>0.035 = 50 x 50 / RL so RL = 71kohms. This is rather high but the tuned tank circuit probably has such a high Z at Fo.
1.25mA gives the 35 mW figure and resonant impedance around 22.4 k but
I think the current guess was a tad high so impedance is probably a
bit higher.
>The comments made about coils are of interest, and it should make no difference what type of core is used, because 270uH is easy to achieve with many recipes.
uH, at least on an LCR meter, it's how it behaves at that frequency.
For example, I talked to Bob Weaver about that and he confirms
'switching power supply' ferrite cores (which I also tried) soak up RF
like, his words, "a sponge." Now, I don't know whether the UHF coil is
doing the same thing or exhibiting some other 'problem' at MW
frequencies but the volts weren't there.
Might perform better if I had some Litz wire but I'm using a ferrite
antenna for the coil now and AES has them for 3 bucks so it isn't
worth spending moolah on Litz when buying the thing 'ready made' is
just as cheap.
> But in the best old radios the 455kHz and MW RF coils are pie wound windings using litz wire and large Al cans to keep the Q high, ie, RF resistance low. In an AM receiver I built from scratch in 1999, I made copper cans for two RF input coils and I wound solid wire on 50mm long x 10mm dia pieces of ferrite rod normally used for ferrite rod antennas. The Q became higher than using air cored windings on say 20mm PVC pipes. In my receiver, the Q was so high the BW was less than 15kHz, giving slight cutting of sidebands and thus reducing possible audio BW, so I used TWO such coils, and a 3 gang cap, with 2 gands devoted to the the two RF coild and the other to the oscillator. To overcome the reduction of BW, I used antenna input to a tap on one coil, then a resistor from top of coil to top of next coil and then I tuned the coils
>so they were "stagger tuned", ie, slightly apart at the low end of the band where the LC Q is higher than at above say 1.2MHz. All that worked well, but compact fluorescent lamps and other junk creates a rectified version of incoming RF and I've had to change to a long rod ferrite antenna with a shielded coil and very short wires to keep out electrostatic portion of RF wave.
>
>In my RF generator which I could use as transmitter, I get the same very nice AM wave on the CRO as Flipper's site shows. I have a 6BX6 for the RF output tube, and the coil is a hand wound thing on ferrite with 3 layers of not very tightly wound windings of solid wire, and with PVC tape used between layers. The self capacitance of the coil must be kept low if you wish to be able to tune the tank to as high as possible, say 1,750kHz.
>Many old radios went from 550kHz to maybe 1,550kHz, a 3 fold F increase. Say you have 20pF for coil self C, then if the cap gang = 365pF max, then total max C = 385pF. Say the coil turns or ferrite position adjusted for Fo = 530kHz, then L must be 234uH. Say the C gang min C = 25pF, and you have 20pF self C
>then Cmin total = 45pF which gives Fo = 1,551kHz, and not as high as anyone should accept. If C minimum could be reduced to 22.5pF, Fo = 2,194kHz, but you will never see this, and besides, one has to be able to use a trimmer cap to set the top of the F range. So the less stray C or coil self C, the better.
radios, its a broadcaster, and being able to find a 'quiet spot' on
the dial is the only reason for tuning.
The 'problem', so to speak, isn't the local osc but the plate tank
because a 10 ft wet noodle wire antenna comes in at 30 pF-35 pF before
you even get to cap minimum, stray, and coil self capacitance.
I'm not sure yet but may go to another slug coil on the plate because
even though less inductance lowers voltage swing, hence generated
power, that may be better than cap tuning because, in that case, most
of the circulating current ends up in the tuning cap rather than the
antenna.
On the other hand, I'm getting sufficient range even with closed
plates so it may be a case of "don't fix it if it ain't broke."
Btw, I've tweaked/padded values around the existing coils and caps to
avoid the irritation of being able to 'dial' a frequency you can't
peak on the antenna tank so the tuning range is now narrowed down to
roughly 600 KHz to 1100 KHz. So, now, if you can dial it you can peak
it. Basically I put 100 pF in parallel with the osc cap and lowered
the coil to 190 uH.
I'm listening to it at about 680 KHz as I type.
John Byrns
Jun 18, 2012, 11:26:44 AM6/18/12
to
This transmitter has been bouncing around in the back of my head for the past
week because something didn't seem quite right but I couldn't put my finger on
it until this morning when I jumped out of bed.
The problem is that there is nothing wrong with the transmitter concept, except
that it doesn't take advantage of the 6ME8 in the way one might have hoped based
on your previous comments on the tube.
Applying the modulation to G1 doesn't take advantage of the beam deflection
capabilities of the tube and instead uses it in a way that a more ordinary tube,
like a dual control pentode, could serve.
Why not connect the 6ME8 cathode-grid circuit as a Hartley oscillator circuit,
as I think you have suggested in earlier posts, and connect the modulation
signal to the beam deflection plates? I suppose the downside of this scheme is
that it would require a balanced push pull RF output transformer to achieve full
modulation, and building such an RF transformer could be a non trivial project
in itself.
I forgot where you are buying your 6ME8s for a dollar apiece, could you refresh
my memory?
flipper
Jun 18, 2012, 6:37:14 PM6/18/12
to
On Mon, 18 Jun 2012 10:26:44 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
A DC pentode might work similar but I'm not sure the plate/screen
current transfer is as balanced as the 6ME8 deflection plates. I
wondered about that, though, but the final deciding factor was 'non
technical': I wanted to use a 6ME8 ;)
Might be interesting to try because I'd think it would take less RF
amplitude but, ironically, it looks like a DC pentode might pull more
B+, even without 'wasting' the 6ME8 plate two power, because it's
screen pulls so much current. It also looks like the 6ME8 might have
better plate impedance but it's hard to tell because there are no
plate curves down at the relatively low voltages it's being run on.
>Why not connect the 6ME8 cathode-grid circuit as a Hartley oscillator circuit,
>as I think you have suggested in earlier posts, and connect the modulation
>signal to the beam deflection plates? I suppose the downside of this scheme is
>that it would require a balanced push pull RF output transformer to achieve full
>modulation, and building such an RF transformer could be a non trivial project
>in itself.
That was my first thought too although you may be remembering the FM
stereo multiplex discussion, where we needed DSB-SC output, because I
think that's the only one I posted a (conceptual) schematic for.
However, unbalancing the deflectors and then 'subtracting' one plate
from the other via a PP transformer was, indeed, what I proposed to
Patrick for the '100% mod' solution back when he was tinkering with an
AM modulator for, I think, 'testing' AM radios. I did wonder, in that
discussion, what the effect of the 'extra' side bands might be. I
mean, you have, in essence, DSB with 'partially suppressed carrier'. I
may still try that some day but, yes, the PP RF transformer kind of
put me off and I'm glad I tried this one first because of what I
learned: the pot core I bought for that very idea, and would have
used, wouldn't have worked well at all.
The other 'advantage' would be self excite but everyone keeps telling
me "don't do it." Self excite 'FMs'.
It could also be done 'single ended' but then you have large audio
swing required and deflector 'non-linearity at the extremes with no
'simple' means of NFB.
So, this topology was picked because it seemed 'simpler', reduced the
number of "never done this before" things that could go Murphy's Law
wrong, and required no 'special' parts. Doubly so because I already
had them.
>I forgot where you are buying your 6ME8s for a dollar apiece, could you refresh
>my memory?
ABC vacuum tubes and ESRC1 vacuum tubes (which seem like maybe they're
the same people). Both sites have a perennial !dollar days! section,
which says qty 10 minimum but I've also been able to get a few tossed
in when buying others at 'normal' price. However, since those are
'normally' $3 (occasionally $4, as with the 6ME8) you really only have
to want '4' (maybe 3), and get 'extras' to make up 10, to come out
ahead.
vacuumtubes.net has no !dollar days! section but pretty much the same
tubes simply listed for 1 buck in their normal price lists, but I've
never tried ordering from them. I normally use ABC, mainly out of
habit.
wrote:
current transfer is as balanced as the 6ME8 deflection plates. I
wondered about that, though, but the final deciding factor was 'non
technical': I wanted to use a 6ME8 ;)
Might be interesting to try because I'd think it would take less RF
amplitude but, ironically, it looks like a DC pentode might pull more
B+, even without 'wasting' the 6ME8 plate two power, because it's
screen pulls so much current. It also looks like the 6ME8 might have
better plate impedance but it's hard to tell because there are no
plate curves down at the relatively low voltages it's being run on.
>Why not connect the 6ME8 cathode-grid circuit as a Hartley oscillator circuit,
>as I think you have suggested in earlier posts, and connect the modulation
>signal to the beam deflection plates? I suppose the downside of this scheme is
>that it would require a balanced push pull RF output transformer to achieve full
>modulation, and building such an RF transformer could be a non trivial project
>in itself.
stereo multiplex discussion, where we needed DSB-SC output, because I
think that's the only one I posted a (conceptual) schematic for.
However, unbalancing the deflectors and then 'subtracting' one plate
from the other via a PP transformer was, indeed, what I proposed to
Patrick for the '100% mod' solution back when he was tinkering with an
AM modulator for, I think, 'testing' AM radios. I did wonder, in that
discussion, what the effect of the 'extra' side bands might be. I
mean, you have, in essence, DSB with 'partially suppressed carrier'. I
may still try that some day but, yes, the PP RF transformer kind of
put me off and I'm glad I tried this one first because of what I
learned: the pot core I bought for that very idea, and would have
used, wouldn't have worked well at all.
The other 'advantage' would be self excite but everyone keeps telling
me "don't do it." Self excite 'FMs'.
It could also be done 'single ended' but then you have large audio
swing required and deflector 'non-linearity at the extremes with no
'simple' means of NFB.
So, this topology was picked because it seemed 'simpler', reduced the
number of "never done this before" things that could go Murphy's Law
wrong, and required no 'special' parts. Doubly so because I already
had them.
>I forgot where you are buying your 6ME8s for a dollar apiece, could you refresh
>my memory?
the same people). Both sites have a perennial !dollar days! section,
which says qty 10 minimum but I've also been able to get a few tossed
in when buying others at 'normal' price. However, since those are
'normally' $3 (occasionally $4, as with the 6ME8) you really only have
to want '4' (maybe 3), and get 'extras' to make up 10, to come out
ahead.
vacuumtubes.net has no !dollar days! section but pretty much the same
tubes simply listed for 1 buck in their normal price lists, but I've
never tried ordering from them. I normally use ABC, mainly out of
habit.
Patrick Turner
Jun 18, 2012, 7:40:45 PM6/18/12
to
On 19 June, 01:26, John Byrns <byr...@sbcglobal.net> wrote:
> In article <3jl9t7hgrb5tqj0ro2r4gf8nlajofnd...@4ax.com>,
A balanced RF tranny that is tuned requires coil with more turns on
the same core to get twice the inductance, fairly easy if you start
with a ferrite rod.
I'd probably say the best wire for low capacitance of the coil is
solid telephone hook up wire, or strands taken from a cat5 cable.
You then need two tuning gangs, 20-365pF are OK. The cap frame and
moving plates are at 0V and if you want to avoid B+ across the tuning
caps, then cap frame is bolted to chassis and 0.1uF caps from coil to
fixed plates. Maybe put 2M2 from fised plates to 0V to bias them down.
This allows you to have a 3 gang cap and use one gang for the
oscillator coil.
The old HP606 I have has this sort of set up. It used 6B4 to cathode
modulate a pair of 6CL6. Oscillator is PP type using maybe 12AT7, I
forget, but its designed for many ranges from 300kHz to 65MHz, all
with well calibrated dial. The whole thing has very high electro-
mechanical integrity; must have cost a huge amount in 1955. Its got
NFB around the modulator to make the AM more linear. But my home brew
SE pentode modulator also works just as well for 2 bands up to 1.7MHz
so I can test the BC band or any IF channel. I don't need the more
complex HP PP circuit. I can't see any problem having NFB around ANY
form of AM modulator because whatever triode is used to as a modulator
can be one of a pair of triodes in an LTP with TWO input ports, and
one port is for the detected AF NFB signal, and the other is for AF
input, so the modulator amplifies the difference between input and FB
signal and applies a correction signal at the output.
Such an LTP for an SE modulator isn't really a pair, because you only
need one anode's output with the other anode grounded via electro cap.
But the two available output could be used for a PP modulator.
BTW John, be careful jumping out of bed. Lotsa guys come to grief that
way :-).
Patrick Turner.
> In article <3jl9t7hgrb5tqj0ro2r4gf8nlajofnd...@4ax.com>,
the same core to get twice the inductance, fairly easy if you start
with a ferrite rod.
I'd probably say the best wire for low capacitance of the coil is
solid telephone hook up wire, or strands taken from a cat5 cable.
You then need two tuning gangs, 20-365pF are OK. The cap frame and
moving plates are at 0V and if you want to avoid B+ across the tuning
caps, then cap frame is bolted to chassis and 0.1uF caps from coil to
fixed plates. Maybe put 2M2 from fised plates to 0V to bias them down.
This allows you to have a 3 gang cap and use one gang for the
oscillator coil.
The old HP606 I have has this sort of set up. It used 6B4 to cathode
modulate a pair of 6CL6. Oscillator is PP type using maybe 12AT7, I
forget, but its designed for many ranges from 300kHz to 65MHz, all
with well calibrated dial. The whole thing has very high electro-
mechanical integrity; must have cost a huge amount in 1955. Its got
NFB around the modulator to make the AM more linear. But my home brew
SE pentode modulator also works just as well for 2 bands up to 1.7MHz
so I can test the BC band or any IF channel. I don't need the more
complex HP PP circuit. I can't see any problem having NFB around ANY
form of AM modulator because whatever triode is used to as a modulator
can be one of a pair of triodes in an LTP with TWO input ports, and
one port is for the detected AF NFB signal, and the other is for AF
input, so the modulator amplifies the difference between input and FB
signal and applies a correction signal at the output.
Such an LTP for an SE modulator isn't really a pair, because you only
need one anode's output with the other anode grounded via electro cap.
But the two available output could be used for a PP modulator.
>
> I forgot where you are buying your 6ME8s for a dollar apiece, could you refresh
> my memory?
What other beam deflector tubes are usable?
> I forgot where you are buying your 6ME8s for a dollar apiece, could you refresh
> my memory?
BTW John, be careful jumping out of bed. Lotsa guys come to grief that
way :-).
Patrick Turner.
John Byrns
Jun 24, 2012, 4:59:48 PM6/24/12
to
In article <7u4vt7h09i6pieb9i...@4ax.com>,
Would "FMing" actually occur with a "Self excited" oscillator built with a beam
deflection tube, or would the grid-cathode oscillator circuit be isolated from
modulation effects that would cause "FMing", by the construction of the beam
deflection tube? Since the sum of the plate currents for both plates remains
constant with modulation applied to the beam deflection electrodes, I would
think that the cathode-grid oscillator circuit wouldn't even see the modulation
applied to the deflection electrodes. G3, the accelerating electrode probably
provides further isolation between the oscillator and modulation effects.
The 6SA7/6BE6/6SC6 family of heptodes would also seem to provide isolation
between a cathode-grid "self excited" oscillator circuit and modulation applied
to G3 as the total cathode current appears to be largely independent of the
voltage on G3, minimizing "FMing". If this weren't the case, the local
oscillator frequency, in an AM receiver using one of these tubes, would vary
with changes in the AGC voltage due to fading.
Both of these schemes differ from the common phono oscillator circuit in that
they require two tuned circuits, one for the oscillator and a second for the
plate circuit feeding the antenna, the plate current can be completely cut off,
neglecting leakage and so forth, without affecting the oscillator circuit to any
great extent. The single tuned circuit approach of the common phono oscillator
circuit would have a greater sensitivity to "FMing", and the oscillator dies
completely when the plate current is cut off.
deflection tube, or would the grid-cathode oscillator circuit be isolated from
modulation effects that would cause "FMing", by the construction of the beam
deflection tube? Since the sum of the plate currents for both plates remains
constant with modulation applied to the beam deflection electrodes, I would
think that the cathode-grid oscillator circuit wouldn't even see the modulation
applied to the deflection electrodes. G3, the accelerating electrode probably
provides further isolation between the oscillator and modulation effects.
The 6SA7/6BE6/6SC6 family of heptodes would also seem to provide isolation
between a cathode-grid "self excited" oscillator circuit and modulation applied
to G3 as the total cathode current appears to be largely independent of the
voltage on G3, minimizing "FMing". If this weren't the case, the local
oscillator frequency, in an AM receiver using one of these tubes, would vary
with changes in the AGC voltage due to fading.
Both of these schemes differ from the common phono oscillator circuit in that
they require two tuned circuits, one for the oscillator and a second for the
plate circuit feeding the antenna, the plate current can be completely cut off,
neglecting leakage and so forth, without affecting the oscillator circuit to any
great extent. The single tuned circuit approach of the common phono oscillator
circuit would have a greater sensitivity to "FMing", and the oscillator dies
completely when the plate current is cut off.
flipper
Jun 24, 2012, 8:01:43 PM6/24/12
to
On Sun, 24 Jun 2012 15:59:48 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
Well, your logic is pretty much what mine was, plus the same 'constant
current' theory applying to dual control pentodes as well, but over at
radiomuseum (plus others) I read an exhaustive thread where the poster
said "will NEVER try THAT again." I can't find the link off hand.
He did try to theorize some kind of miller interaction but was vague
enough that I couldn't quite follow the logic. He also dealt with your
idea 'that can't be' or else a typical radio wouldn't work right. His
argument was the only reason it did work was the small RF levels and
LO carrier being suppressed.
Anyhow, everyone tells me don't do it but I might try anyway ;)
As for the 6ME8, I thought, like you, cathode current should be
'constant' since the 'beam' is just being 'deflected' to the two
plates but was surprised to find gobs of RF on the cathode. I guess
the modest deflector bias, which is changing with RF, does affect it.
Maybe it wouldn't if I drove both.
Btw, my instinct was to 'clean it up' but that made things worse so I
finally decided to not try 'fixing' what ain't broke.
I'm also going to try the DC pentode because, like you said, it seems
to me that ought to work about the same.
wrote:
current' theory applying to dual control pentodes as well, but over at
radiomuseum (plus others) I read an exhaustive thread where the poster
said "will NEVER try THAT again." I can't find the link off hand.
He did try to theorize some kind of miller interaction but was vague
enough that I couldn't quite follow the logic. He also dealt with your
idea 'that can't be' or else a typical radio wouldn't work right. His
argument was the only reason it did work was the small RF levels and
LO carrier being suppressed.
Anyhow, everyone tells me don't do it but I might try anyway ;)
As for the 6ME8, I thought, like you, cathode current should be
'constant' since the 'beam' is just being 'deflected' to the two
plates but was surprised to find gobs of RF on the cathode. I guess
the modest deflector bias, which is changing with RF, does affect it.
Maybe it wouldn't if I drove both.
Btw, my instinct was to 'clean it up' but that made things worse so I
finally decided to not try 'fixing' what ain't broke.
I'm also going to try the DC pentode because, like you said, it seems
to me that ought to work about the same.
flipper
Jun 24, 2012, 8:15:45 PM6/24/12
to
http://www.radiomuseum.org/forum/granco_granco_am_modulatortransmitter.html
Down at the bottom, "The sharp-cutoff heptode solution" by Jacob
Roschy. His second circuit is "The cathode oscillator attempt." "Since
this resulted simultaneously with the desired amplitude modulation
into a strong undesired frequency modulation, I abandoned this
experiment and scrapped this circuit very soon, I will never try this
any more !"
Alex Pogossov
Jun 25, 2012, 5:36:07 AM6/25/12
to
"John Byrns" <byr...@sbcglobal.net> wrote in message
news:byrnsj-7C2B68....@news.giganews.com...
> The 6SA7/6BE6/6SC6 family of heptodes would also seem to provide isolation
> between a cathode-grid "self excited" oscillator circuit and modulation
> applied
> to G3 as the total cathode current appears to be largely independent of
> the
> voltage on G3, minimizing "FMing". If this weren't the case, the local
> oscillator frequency, in an AM receiver using one of these tubes, would
> vary
> with changes in the AGC voltage due to fading.
G3.
With 6BE6 it is almost impossible to cut it off completely, so 100%
modulation os impossible, and deep modulatio will be distorted, even with
feedback.
By the way, in this case NFB shall be applied from the plate of a heptode,
not from its cathode.
It is better to use a separate oscillator and geef it to G3 of a heptode,
while feeding audio to G1. In this case any heptode can be used. NFB can be
taken from the cathode since plate current is *sort of* proportional to
cathode current. Do not forget to decouple G2+G4 to cathode (!), not to GND
and use a large electrolytic for passing AF as well, not RF only. But still
it is better to take NFB from the plate (I mean AF component, not RF).
Even in a best heptode, even with a separate oscillator you will have
residual FM -- due to space charge and stray capacitances. A small FM of say
100...1000Hz is acceptable for listening on an AM radio, but will not be
listenable on a synchrodyne or an SSB receiver.
flipper
Jun 25, 2012, 5:49:32 PM6/25/12
to
On Mon, 25 Jun 2012 19:36:07 +1000, "Alex Pogossov"
<apog...@tpg.com.au> wrote:
>
>"John Byrns" <byr...@sbcglobal.net> wrote in message
>news:byrnsj-7C2B68....@news.giganews.com...
>
>> The 6SA7/6BE6/6SC6 family of heptodes would also seem to provide isolation
>> between a cathode-grid "self excited" oscillator circuit and modulation
>> applied
>> to G3 as the total cathode current appears to be largely independent of
>> the
>> voltage on G3, minimizing "FMing". If this weren't the case, the local
>> oscillator frequency, in an AM receiver using one of these tubes, would
>> vary
>> with changes in the AGC voltage due to fading.
>
>Of these tubes only 6CS6 (not 6SC6) is suitable as it has sharp cut-off on
>G3.
>With 6BE6 it is almost impossible to cut it off completely, so 100%
>modulation os impossible, and deep modulatio will be distorted, even with
>feedback.
>
>By the way, in this case NFB shall be applied from the plate of a heptode,
>not from its cathode.
Yes, audio to G3 is the most common implementation with people looking
<apog...@tpg.com.au> wrote:
>
>"John Byrns" <byr...@sbcglobal.net> wrote in message
>news:byrnsj-7C2B68....@news.giganews.com...
>
>> The 6SA7/6BE6/6SC6 family of heptodes would also seem to provide isolation
>> between a cathode-grid "self excited" oscillator circuit and modulation
>> applied
>> to G3 as the total cathode current appears to be largely independent of
>> the
>> voltage on G3, minimizing "FMing". If this weren't the case, the local
>> oscillator frequency, in an AM receiver using one of these tubes, would
>> vary
>> with changes in the AGC voltage due to fading.
>
>Of these tubes only 6CS6 (not 6SC6) is suitable as it has sharp cut-off on
>G3.
>With 6BE6 it is almost impossible to cut it off completely, so 100%
>modulation os impossible, and deep modulatio will be distorted, even with
>feedback.
>
>By the way, in this case NFB shall be applied from the plate of a heptode,
>not from its cathode.
to the G3 curve for linearity, which is why dual control pentodes seem
to be the preferred choice. For one, as you mentioned, the 6CS6 is one
of the few sharp heptodes, the ECH84 being the only other one I can
think of off hand, and there's more DC pentode choices. Second, the
dual control pentode G3 'linear' region remains relatively constant
over bias while the 6CS6 cutoff region shifts with screen volts and
bias.. That makes the DC pentode easier to bias.
There's a ton of Dual Control Pentode AM broadcaster schematics online
and I've breadboarded a couple of them. In fact, my LO was originally
developed for a 6GY6 version using a 1 MHz brick osc.
>It is better to use a separate oscillator and geef it to G3 of a heptode,
>while feeding audio to G1. In this case any heptode can be used. NFB can be
>taken from the cathode since plate current is *sort of* proportional to
>cathode current. Do not forget to decouple G2+G4 to cathode (!), not to GND
>and use a large electrolytic for passing AF as well, not RF only. But still
>it is better to take NFB from the plate (I mean AF component, not RF).
alternate to my 'Beamus' 6ME8 modulator. It does seem like it ought to
work similarly.
I managed to get a dual control pentode (6HZ6) model working with
Circuitmaker, although I'm not sure how 'good' it is. At any rate, it
seems to not like going over 80%, or so, mod. Can't drive it to near
cutoff. Another thing, input impedance to G3 seems incredibly low, at
least at RF, and, if the simulation is even remotely accurate, there's
no way my LO can drive it.
Both those 'problem's may be the model, so I'll probably try building
it anyway, but if the simulation is valid then the 6ME8 works quite a
bit better.
>Even in a best heptode, even with a separate oscillator you will have
>residual FM -- due to space charge and stray capacitances. A small FM of say
>100...1000Hz is acceptable for listening on an AM radio, but will not be
>listenable on a synchrodyne or an SSB receiver.
Modulating grid capacitance, which is coupled to the LO tank?
Alex Pogossov
Jun 26, 2012, 8:09:21 AM6/26/12
to
news:jvkhu7t0q8atdfn2n...@4ax.com...
plate voltage.
Dual control pentodes have one undesirable disadvantage. Their plate
impedance is not high and more importantly, it is not constant. It is high
at G3 cutoff (obviously!) and high at G3=0...+5V (where it works as a
conventional pentode without much current returning to G2). At the very
middle of the "linear" G3 zone the plate impedance is low, as in a triode
where the cathode is the space charge rather than a red hot physical
cathode. If such modulator runs into a LC tank, its Q factor will be
modulated. Peaks and troughs will be emphasized, carrier level -- reduced.
Hence -- the second harmonic of the envelope.
>
> There's a ton of Dual Control Pentode AM broadcaster schematics online
> and I've breadboarded a couple of them. In fact, my LO was originally
> developed for a 6GY6 version using a 1 MHz brick osc.
>
>>It is better to use a separate oscillator and geef it to G3 of a heptode,
>>while feeding audio to G1. In this case any heptode can be used. NFB can
>>be
>>taken from the cathode since plate current is *sort of* proportional to
>>cathode current. Do not forget to decouple G2+G4 to cathode (!), not to
>>GND
>>and use a large electrolytic for passing AF as well, not RF only. But
>>still
>>it is better to take NFB from the plate (I mean AF component, not RF).
>
> This, using a dual control pentode, is what John was suggesting as an
> alternate to my 'Beamus' 6ME8 modulator. It does seem like it ought to
> work similarly.
>
> I managed to get a dual control pentode (6HZ6) model working with
> Circuitmaker, although I'm not sure how 'good' it is. At any rate, it
> seems to not like going over 80%, or so, mod. Can't drive it to near
> cutoff. Another thing, input impedance to G3 seems incredibly low, at
> least at RF, and, if the simulation is even remotely accurate, there's
> no way my LO can drive it.
>
> Both those 'problem's may be the model, so I'll probably try building
> it anyway, but if the simulation is valid then the 6ME8 works quite a
> bit better.
Yes, 6ME8 seems a good choice, though this tube is "exotic" -- not found in
evereyone's junk box, unlike say 6BE6.
>
>>Even in a best heptode, even with a separate oscillator you will have
>>residual FM -- due to space charge and stray capacitances. A small FM of
>>say
>>100...1000Hz is acceptable for listening on an AM radio, but will not be
>>listenable on a synchrodyne or an SSB receiver.
>
> I'm curious how space charge makes it across to a separate LO.
> Modulating grid capacitance, which is coupled to the LO tank?
Exactly. Likewise, in 6ME8 density of the beam (cathode current) affects the
space charge between the deflection plates which modulated capacutance
between the deflection plates. I do not know the magnitude of this effect
compared to a heptode. In a heptode typically space chage "equals" 0.3pF.
As far as the parasitic FM is concerned, do two simple experiments.
1. On a quality comms receiver in SSB mode listen to the carrier (set beat
frequency to a few hundred hertz) of your transmitter.
Without applying any audio, halve the cathode resistor of 6ME8 (touch it
with another resistor in parallel), effectively simulating +100% modulation
excursion. Ideally, pitch of the beat tone shall not change. If it
changes... you have FM.
2. On a quality comms receiver in SSB mode listen to the carrier (set beat
frequency to a few hundred hertz) of your transmitter.
Without applying any audio, play with the antenna tuning capacitor of your
transmitter. Ideally, pitch shall not change, only the volume as you tune to
and detune from the resonance peak. If it changes...
If you do these tests, could you "report" the results here. Then we can
discuss what causes FM in each of these cases.
Regards,
Alex
> On Mon, 25 Jun 2012 19:36:07 +1000, "Alex Pogossov"
> <apog...@tpg.com.au> wrote:
> Yes, audio to G3 is the most common implementation with people looking
> to the G3 curve for linearity, which is why dual control pentodes seem
> to be the preferred choice. For one, as you mentioned, the 6CS6 is one
> of the few sharp heptodes, the ECH84 being the only other one I can
> think of off hand, and there's more DC pentode choices. Second, the
> dual control pentode G3 'linear' region remains relatively constant
> over bias while the 6CS6 cutoff region shifts with screen volts and
> bias.. That makes the DC pentode easier to bias.
Likewise in a dual control pentode the "linear G3 region" shifts with the
> <apog...@tpg.com.au> wrote:
> Yes, audio to G3 is the most common implementation with people looking
> to the G3 curve for linearity, which is why dual control pentodes seem
> to be the preferred choice. For one, as you mentioned, the 6CS6 is one
> of the few sharp heptodes, the ECH84 being the only other one I can
> think of off hand, and there's more DC pentode choices. Second, the
> dual control pentode G3 'linear' region remains relatively constant
> over bias while the 6CS6 cutoff region shifts with screen volts and
> bias.. That makes the DC pentode easier to bias.
plate voltage.
Dual control pentodes have one undesirable disadvantage. Their plate
impedance is not high and more importantly, it is not constant. It is high
at G3 cutoff (obviously!) and high at G3=0...+5V (where it works as a
conventional pentode without much current returning to G2). At the very
middle of the "linear" G3 zone the plate impedance is low, as in a triode
where the cathode is the space charge rather than a red hot physical
cathode. If such modulator runs into a LC tank, its Q factor will be
modulated. Peaks and troughs will be emphasized, carrier level -- reduced.
Hence -- the second harmonic of the envelope.
>
> There's a ton of Dual Control Pentode AM broadcaster schematics online
> and I've breadboarded a couple of them. In fact, my LO was originally
> developed for a 6GY6 version using a 1 MHz brick osc.
>
>>It is better to use a separate oscillator and geef it to G3 of a heptode,
>>while feeding audio to G1. In this case any heptode can be used. NFB can
>>be
>>taken from the cathode since plate current is *sort of* proportional to
>>cathode current. Do not forget to decouple G2+G4 to cathode (!), not to
>>GND
>>and use a large electrolytic for passing AF as well, not RF only. But
>>still
>>it is better to take NFB from the plate (I mean AF component, not RF).
>
> This, using a dual control pentode, is what John was suggesting as an
> alternate to my 'Beamus' 6ME8 modulator. It does seem like it ought to
> work similarly.
>
> I managed to get a dual control pentode (6HZ6) model working with
> Circuitmaker, although I'm not sure how 'good' it is. At any rate, it
> seems to not like going over 80%, or so, mod. Can't drive it to near
> cutoff. Another thing, input impedance to G3 seems incredibly low, at
> least at RF, and, if the simulation is even remotely accurate, there's
> no way my LO can drive it.
>
> Both those 'problem's may be the model, so I'll probably try building
> it anyway, but if the simulation is valid then the 6ME8 works quite a
> bit better.
evereyone's junk box, unlike say 6BE6.
>
>>Even in a best heptode, even with a separate oscillator you will have
>>residual FM -- due to space charge and stray capacitances. A small FM of
>>say
>>100...1000Hz is acceptable for listening on an AM radio, but will not be
>>listenable on a synchrodyne or an SSB receiver.
>
> I'm curious how space charge makes it across to a separate LO.
> Modulating grid capacitance, which is coupled to the LO tank?
space charge between the deflection plates which modulated capacutance
between the deflection plates. I do not know the magnitude of this effect
compared to a heptode. In a heptode typically space chage "equals" 0.3pF.
As far as the parasitic FM is concerned, do two simple experiments.
1. On a quality comms receiver in SSB mode listen to the carrier (set beat
frequency to a few hundred hertz) of your transmitter.
Without applying any audio, halve the cathode resistor of 6ME8 (touch it
with another resistor in parallel), effectively simulating +100% modulation
excursion. Ideally, pitch of the beat tone shall not change. If it
changes... you have FM.
2. On a quality comms receiver in SSB mode listen to the carrier (set beat
frequency to a few hundred hertz) of your transmitter.
Without applying any audio, play with the antenna tuning capacitor of your
transmitter. Ideally, pitch shall not change, only the volume as you tune to
and detune from the resonance peak. If it changes...
If you do these tests, could you "report" the results here. Then we can
discuss what causes FM in each of these cases.
Regards,
Alex
flipper
Jun 26, 2012, 3:17:16 PM6/26/12
to
On Tue, 26 Jun 2012 22:09:21 +1000, "Alex Pogossov"
Look at the Tung-Sol (because it has the combined curves) datasheet
for the 6AS6 and the G.E. datasheet for the 6CS6. On the 6AS6 all the
plate curves tend to converge in the same cutoff region but it's
dramatically different for the 6CS6. Now, the 'linear' region of the
6AS6 does shift some horizontally but it's nothing like the 6CS6.
>Dual control pentodes have one undesirable disadvantage. Their plate
>impedance is not high and more importantly, it is not constant. It is high
>at G3 cutoff (obviously!) and high at G3=0...+5V (where it works as a
>conventional pentode without much current returning to G2). At the very
>middle of the "linear" G3 zone the plate impedance is low, as in a triode
>where the cathode is the space charge rather than a red hot physical
>cathode. If such modulator runs into a LC tank, its Q factor will be
>modulated. Peaks and troughs will be emphasized, carrier level -- reduced.
>Hence -- the second harmonic of the envelope.
Yeah, I noticed the plate impedance curves and the change is rather
dramatic.
Are you suggesting a heptode would be a better 'alternate choice', for
the 6ME8 audio to G1 topology, than the dual control pentode?
Well, they're only a buck online, which is cheaper than the 6BE6.
>>>Even in a best heptode, even with a separate oscillator you will have
>>>residual FM -- due to space charge and stray capacitances. A small FM of
>>>say
>>>100...1000Hz is acceptable for listening on an AM radio, but will not be
>>>listenable on a synchrodyne or an SSB receiver.
>>
>> I'm curious how space charge makes it across to a separate LO.
>> Modulating grid capacitance, which is coupled to the LO tank?
>
>Exactly. Likewise, in 6ME8 density of the beam (cathode current) affects the
>space charge between the deflection plates which modulated capacutance
>between the deflection plates. I do not know the magnitude of this effect
>compared to a heptode. In a heptode typically space chage "equals" 0.3pF.
Wow, my guess wasn't bad for 'not an RF guy' ;)
>As far as the parasitic FM is concerned, do two simple experiments.
>
>1. On a quality comms receiver in SSB mode listen to the carrier (set beat
>frequency to a few hundred hertz) of your transmitter.
>Without applying any audio, halve the cathode resistor of 6ME8 (touch it
>with another resistor in parallel), effectively simulating +100% modulation
>excursion. Ideally, pitch of the beat tone shall not change. If it
>changes... you have FM.
>
>2. On a quality comms receiver in SSB mode listen to the carrier (set beat
>frequency to a few hundred hertz) of your transmitter.
>Without applying any audio, play with the antenna tuning capacitor of your
>transmitter. Ideally, pitch shall not change, only the volume as you tune to
>and detune from the resonance peak. If it changes...
>
>If you do these tests, could you "report" the results here. Then we can
>discuss what causes FM in each of these cases.
I don't have a SSB receiver.
<apog...@tpg.com.au> wrote:
>
>"flipper" <fli...@fish.net> wrote in message
>news:jvkhu7t0q8atdfn2n...@4ax.com...
>> On Mon, 25 Jun 2012 19:36:07 +1000, "Alex Pogossov"
>> <apog...@tpg.com.au> wrote:
>
>> Yes, audio to G3 is the most common implementation with people looking
>> to the G3 curve for linearity, which is why dual control pentodes seem
>> to be the preferred choice. For one, as you mentioned, the 6CS6 is one
>> of the few sharp heptodes, the ECH84 being the only other one I can
>> think of off hand, and there's more DC pentode choices. Second, the
>> dual control pentode G3 'linear' region remains relatively constant
>> over bias while the 6CS6 cutoff region shifts with screen volts and
>> bias.. That makes the DC pentode easier to bias.
>
>Likewise in a dual control pentode the "linear G3 region" shifts with the
>plate voltage.
I wasn't clear. I meant on the horizontal axis.
>
>"flipper" <fli...@fish.net> wrote in message
>news:jvkhu7t0q8atdfn2n...@4ax.com...
>> On Mon, 25 Jun 2012 19:36:07 +1000, "Alex Pogossov"
>> <apog...@tpg.com.au> wrote:
>
>> Yes, audio to G3 is the most common implementation with people looking
>> to the G3 curve for linearity, which is why dual control pentodes seem
>> to be the preferred choice. For one, as you mentioned, the 6CS6 is one
>> of the few sharp heptodes, the ECH84 being the only other one I can
>> think of off hand, and there's more DC pentode choices. Second, the
>> dual control pentode G3 'linear' region remains relatively constant
>> over bias while the 6CS6 cutoff region shifts with screen volts and
>> bias.. That makes the DC pentode easier to bias.
>
>Likewise in a dual control pentode the "linear G3 region" shifts with the
>plate voltage.
Look at the Tung-Sol (because it has the combined curves) datasheet
for the 6AS6 and the G.E. datasheet for the 6CS6. On the 6AS6 all the
plate curves tend to converge in the same cutoff region but it's
dramatically different for the 6CS6. Now, the 'linear' region of the
6AS6 does shift some horizontally but it's nothing like the 6CS6.
>Dual control pentodes have one undesirable disadvantage. Their plate
>impedance is not high and more importantly, it is not constant. It is high
>at G3 cutoff (obviously!) and high at G3=0...+5V (where it works as a
>conventional pentode without much current returning to G2). At the very
>middle of the "linear" G3 zone the plate impedance is low, as in a triode
>where the cathode is the space charge rather than a red hot physical
>cathode. If such modulator runs into a LC tank, its Q factor will be
>modulated. Peaks and troughs will be emphasized, carrier level -- reduced.
>Hence -- the second harmonic of the envelope.
dramatic.
Are you suggesting a heptode would be a better 'alternate choice', for
the 6ME8 audio to G1 topology, than the dual control pentode?
>>>Even in a best heptode, even with a separate oscillator you will have
>>>residual FM -- due to space charge and stray capacitances. A small FM of
>>>say
>>>100...1000Hz is acceptable for listening on an AM radio, but will not be
>>>listenable on a synchrodyne or an SSB receiver.
>>
>> I'm curious how space charge makes it across to a separate LO.
>> Modulating grid capacitance, which is coupled to the LO tank?
>
>Exactly. Likewise, in 6ME8 density of the beam (cathode current) affects the
>space charge between the deflection plates which modulated capacutance
>between the deflection plates. I do not know the magnitude of this effect
>compared to a heptode. In a heptode typically space chage "equals" 0.3pF.
>As far as the parasitic FM is concerned, do two simple experiments.
>
>1. On a quality comms receiver in SSB mode listen to the carrier (set beat
>frequency to a few hundred hertz) of your transmitter.
>Without applying any audio, halve the cathode resistor of 6ME8 (touch it
>with another resistor in parallel), effectively simulating +100% modulation
>excursion. Ideally, pitch of the beat tone shall not change. If it
>changes... you have FM.
>
>2. On a quality comms receiver in SSB mode listen to the carrier (set beat
>frequency to a few hundred hertz) of your transmitter.
>Without applying any audio, play with the antenna tuning capacitor of your
>transmitter. Ideally, pitch shall not change, only the volume as you tune to
>and detune from the resonance peak. If it changes...
>
>If you do these tests, could you "report" the results here. Then we can
>discuss what causes FM in each of these cases.
John Byrns
Jun 26, 2012, 4:36:48 PM6/26/12
to
In article <g5bfu79udllqk65f4...@4ax.com>,
Actually that wasn't what I was trying to suggest, that was either a "typo" on
my part or I was having a "senior moment". What I was actually talking about
was a circuit using a heptode such as a 6BE6, 6CS6, or 6BY6, I'm not sure how
the term "dual control pentode" managed to escape my keyboard.
An interesting point about your 6ME8 transmitter is that it appears to be a
vacuum state analog to the solid state designs using a three transistor "cell",
which seem to be the designs of choice with many old radio fans. In this
circuit two of the transistors are connected as a long tailed differential pair
with RF drive applied to the base of one transistor and with the modulated RF
output taken from one of the collectors. The third transistor operates as a
current source in the emitter "tail" circuit of the first two transistors, with
audio applied to its base to modulate the RF output of the differential pair.
There are also a bunch of dual plate tubes that were used in electric organs,
that I have been meaning to lookup for a couple of years to see if there are any
that might be useful in these circuits, so far I haven't gotten around to it.
> I found the page.
>
> http://www.radiomuseum.org/forum/granco_granco_am_modulatortransmitter.html
>
> Down at the bottom, "The sharp-cutoff heptode solution" by Jacob
> Roschy. His second circuit is "The cathode oscillator attempt." "Since
> this resulted simultaneously with the desired amplitude modulation
> into a strong undesired frequency modulation, I abandoned this
> experiment and scrapped this circuit very soon, I will never try this
> any more !"
That's an interesting page, but back to that in a moment.
Re unintentional FMing of AM carriers, that seems to have been a traditional
part of early AM broadcasting, although the radios back then may have had a much
wider bandwidth than today's radios, so the FMing may not have mattered much.
Consider the FMing that was likely present in the following circuit taken from
an old notebook that belonged to a gentleman doing research on building a radio
station back in the day, which I bought on eBay.
http://fmamradios.com/Old_Notebook/KDKA_Transmitter_Schematic.jpg
And here is a photograph of what is probably a later but similar transmitter.
Unfortunately this photo doesn't include the power supply and modulation reactor.
http://www.hammondmuseumofradio.org/kdka/early-kdka-exp.jpg
While this is obviously not a photograph of the exact transmitter described by
the schematic, I suspect that it is similar, possibly with improvements to
eliminate FMing, but probably not to improve modulation capability.
The Radiomuseum web page really took me back as I owned a Motorola FM converter
using the 12-volt tubes back in the 1962-66 time frame. The Motorola was
probably a little more upscale than the Granco discussed on the Radiomuseum
page, IIRC it had an extra IF stage and a nicer case than the Granco, however it
used a "phono oscillator" circuit similar to the Granco's to drive the existing
AM radio. At the time I was impressed with how good the "transmitter" circuit
in the Motorola FM converter sounded, and the audio level was not especially
lower than on an AM station.
I had been wondering how much of the FMing with the self-excited heptodes is due
to coupling within the tube, and how much was due to layout and shielding
issues. Joe seems to have gone into that to some extent, although a deeper
exploration of the subject would be more satisfying to me. Jacob's use of the
ECH84 is interesting in that it presumably eliminates any direct "electron
cloud" coupling issues, but the oscillator and modulated amplifier elements are
still in close proximity so that there could still be inter-electrode coupling
issues. None of Jacob's circuits include a buffer stage between the oscillator
and modulated amplifier stage to completely eliminate the last vestiges of
"electron cloud" coupling affecting the oscillator frequency.
Another issue that Jacob addressed that I have been wondering about is what he
calls "G3 Blurring", which he reduces by using a cathode follower to drive G3.
It occurs to me that a refinement of this approach would be to use a
direct-coupled cathode follower to allow driving G3 positive, although that
would probably require an additional power supply, increasing complexity.
I was not previously aware of the 6BY6 heptode mentioned on this page, it
appears to have a considerably more linear G3 voltage vs. plate current curve
than the 6CS6 most people on this side of the pond seem to talk about.
Something else that I learned from this page, that should have been obvious to
me before, is that the two tuned circuit heptode circuit that I suggested
provides a degree of isolation between varying antenna capacity and the
oscillator frequency.
One thing that really strikes me about pages like this Radiomuseum web page is
that the authors rarely state what all their design goals are. If I were
building a small transmitter using four tube elements as Jacob did, I would
probably go with a plate modulation approach, eliminating several problems
presented by the grid modulation approach Jacob uses. The attractiveness of the
heptodes to me are the possibilities for building a single tube transmitter
my part or I was having a "senior moment". What I was actually talking about
was a circuit using a heptode such as a 6BE6, 6CS6, or 6BY6, I'm not sure how
the term "dual control pentode" managed to escape my keyboard.
An interesting point about your 6ME8 transmitter is that it appears to be a
vacuum state analog to the solid state designs using a three transistor "cell",
which seem to be the designs of choice with many old radio fans. In this
circuit two of the transistors are connected as a long tailed differential pair
with RF drive applied to the base of one transistor and with the modulated RF
output taken from one of the collectors. The third transistor operates as a
current source in the emitter "tail" circuit of the first two transistors, with
audio applied to its base to modulate the RF output of the differential pair.
There are also a bunch of dual plate tubes that were used in electric organs,
that I have been meaning to lookup for a couple of years to see if there are any
that might be useful in these circuits, so far I haven't gotten around to it.
> I found the page.
>
> http://www.radiomuseum.org/forum/granco_granco_am_modulatortransmitter.html
>
> Down at the bottom, "The sharp-cutoff heptode solution" by Jacob
> Roschy. His second circuit is "The cathode oscillator attempt." "Since
> this resulted simultaneously with the desired amplitude modulation
> into a strong undesired frequency modulation, I abandoned this
> experiment and scrapped this circuit very soon, I will never try this
> any more !"
Re unintentional FMing of AM carriers, that seems to have been a traditional
part of early AM broadcasting, although the radios back then may have had a much
wider bandwidth than today's radios, so the FMing may not have mattered much.
Consider the FMing that was likely present in the following circuit taken from
an old notebook that belonged to a gentleman doing research on building a radio
station back in the day, which I bought on eBay.
http://fmamradios.com/Old_Notebook/KDKA_Transmitter_Schematic.jpg
And here is a photograph of what is probably a later but similar transmitter.
Unfortunately this photo doesn't include the power supply and modulation reactor.
http://www.hammondmuseumofradio.org/kdka/early-kdka-exp.jpg
While this is obviously not a photograph of the exact transmitter described by
the schematic, I suspect that it is similar, possibly with improvements to
eliminate FMing, but probably not to improve modulation capability.
The Radiomuseum web page really took me back as I owned a Motorola FM converter
using the 12-volt tubes back in the 1962-66 time frame. The Motorola was
probably a little more upscale than the Granco discussed on the Radiomuseum
page, IIRC it had an extra IF stage and a nicer case than the Granco, however it
used a "phono oscillator" circuit similar to the Granco's to drive the existing
AM radio. At the time I was impressed with how good the "transmitter" circuit
in the Motorola FM converter sounded, and the audio level was not especially
lower than on an AM station.
I had been wondering how much of the FMing with the self-excited heptodes is due
to coupling within the tube, and how much was due to layout and shielding
issues. Joe seems to have gone into that to some extent, although a deeper
exploration of the subject would be more satisfying to me. Jacob's use of the
ECH84 is interesting in that it presumably eliminates any direct "electron
cloud" coupling issues, but the oscillator and modulated amplifier elements are
still in close proximity so that there could still be inter-electrode coupling
issues. None of Jacob's circuits include a buffer stage between the oscillator
and modulated amplifier stage to completely eliminate the last vestiges of
"electron cloud" coupling affecting the oscillator frequency.
Another issue that Jacob addressed that I have been wondering about is what he
calls "G3 Blurring", which he reduces by using a cathode follower to drive G3.
It occurs to me that a refinement of this approach would be to use a
direct-coupled cathode follower to allow driving G3 positive, although that
would probably require an additional power supply, increasing complexity.
I was not previously aware of the 6BY6 heptode mentioned on this page, it
appears to have a considerably more linear G3 voltage vs. plate current curve
than the 6CS6 most people on this side of the pond seem to talk about.
Something else that I learned from this page, that should have been obvious to
me before, is that the two tuned circuit heptode circuit that I suggested
provides a degree of isolation between varying antenna capacity and the
oscillator frequency.
One thing that really strikes me about pages like this Radiomuseum web page is
that the authors rarely state what all their design goals are. If I were
building a small transmitter using four tube elements as Jacob did, I would
probably go with a plate modulation approach, eliminating several problems
presented by the grid modulation approach Jacob uses. The attractiveness of the
heptodes to me are the possibilities for building a single tube transmitter
John Byrns
Jun 26, 2012, 4:38:20 PM6/26/12
to
In article <jvkhu7t0q8atdfn2n...@4ax.com>,
Actually that wasn't what I was trying to suggest, that was either a "typo" on
my part or I was having a "senior moment". What I was actually talking about
was a circuit using a heptode such as a 6BE6, 6CS6, or 6BY6, I'm not sure how
the term "dual control pentode" managed to escape my keyboard.
my part or I was having a "senior moment". What I was actually talking about
was a circuit using a heptode such as a 6BE6, 6CS6, or 6BY6, I'm not sure how
the term "dual control pentode" managed to escape my keyboard.
flipper
Jun 26, 2012, 10:15:27 PM6/26/12
to
On Tue, 26 Jun 2012 15:36:48 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
<snip.
>> >I'm also going to try the DC pentode because, like you said, it seems
>> >to me that ought to work about the same.
>
>Actually that wasn't what I was trying to suggest, that was either a "typo" on
>my part or I was having a "senior moment". What I was actually talking about
>was a circuit using a heptode such as a 6BE6, 6CS6, or 6BY6, I'm not sure how
>the term "dual control pentode" managed to escape my keyboard.
You're quite right. You said the "6SA7/6BE6/6SC6 family of heptodes"
and it was I who 'converted' it to dual control heptodes because
'total cathode current appears to be largely independent of the
voltage on G3' on those too.
>An interesting point about your 6ME8 transmitter is that it appears to be a
>vacuum state analog to the solid state designs using a three transistor "cell",
>which seem to be the designs of choice with many old radio fans. In this
>circuit two of the transistors are connected as a long tailed differential pair
>with RF drive applied to the base of one transistor and with the modulated RF
>output taken from one of the collectors. The third transistor operates as a
>current source in the emitter "tail" circuit of the first two transistors, with
>audio applied to its base to modulate the RF output of the differential pair.
Yes, I know the topology and was 'that close' to building one but am
not sure which was the cart and which the horse here. I originally
started thinking of the 6ME8 back when Patrick was tinkering with
Gilbert Cells because I noticed a 'similarity' between it and the 3
transistor 'cell', of which the Gilbert has two on top a current
source.
Anyway, when working on this transmitter I tried 'emulating' something
akin to the beam deflection tube because I don't have a model for one
and constructed a 'long tail pair' using a pentode for the tail, which
is also akin to the 3 transistor 'cell'. That did point out one
difference, though. The deflection plates do not depend on cathodes
(emitters) driving the 'other side' to get the 'see saw' effect.
>There are also a bunch of dual plate tubes that were used in electric organs,
>that I have been meaning to lookup for a couple of years to see if there are any
>that might be useful in these circuits, so far I haven't gotten around to it.
The one I'm aware of is the 12FQ8 but have never been able to figure
out how it was intended to function.
Yeah, and he also doesn't provide any measurements of the 'FMing' in
either. Just that it's apparently 'bad' in one and the other circuit
works well, whatever that means.
>Another issue that Jacob addressed that I have been wondering about is what he
>calls "G3 Blurring", which he reduces by using a cathode follower to drive G3.
>It occurs to me that a refinement of this approach would be to use a
>direct-coupled cathode follower to allow driving G3 positive, although that
>would probably require an additional power supply, increasing complexity.
>
>I was not previously aware of the 6BY6 heptode mentioned on this page, it
>appears to have a considerably more linear G3 voltage vs. plate current curve
>than the 6CS6 most people on this side of the pond seem to talk about.
I've got both and they're both '1 buck' tubes. I think people kind of
like the 6CS6 because of the unusually high G3 gm, so it can
potentially be driven without a preamp.
>Something else that I learned from this page, that should have been obvious to
>me before, is that the two tuned circuit heptode circuit that I suggested
>provides a degree of isolation between varying antenna capacity and the
>oscillator frequency.
>
>One thing that really strikes me about pages like this Radiomuseum web page is
>that the authors rarely state what all their design goals are. If I were
>building a small transmitter using four tube elements as Jacob did, I would
>probably go with a plate modulation approach, eliminating several problems
>presented by the grid modulation approach Jacob uses. The attractiveness of the
>heptodes to me are the possibilities for building a single tube transmitter
Well, the 6CS6 mat be the choice, then, because of that high G3 gm, if
the FMing could be solved. Joe Sousa, who has another interesting
article on what he calls his 'hi-fi' transmitter, seems to think it
could be neutralized out.
Btw, his 'hi-fi' transmitter is the same idea I was proposing to
Patrick with the 6ME8 but he's using a dual control pentode to do it.
http://www.radiomuseum.org/forum/hi_fi_am_transmitter.html
One tube job, like your goal. Btw, he speaks of neutralization in that
article too (3- Frequency Stability) but decides his circuit doesn't
really need it.
wrote:
<snip.
>> >I'm also going to try the DC pentode because, like you said, it seems
>> >to me that ought to work about the same.
>
>Actually that wasn't what I was trying to suggest, that was either a "typo" on
>my part or I was having a "senior moment". What I was actually talking about
>was a circuit using a heptode such as a 6BE6, 6CS6, or 6BY6, I'm not sure how
>the term "dual control pentode" managed to escape my keyboard.
and it was I who 'converted' it to dual control heptodes because
'total cathode current appears to be largely independent of the
voltage on G3' on those too.
>An interesting point about your 6ME8 transmitter is that it appears to be a
>vacuum state analog to the solid state designs using a three transistor "cell",
>which seem to be the designs of choice with many old radio fans. In this
>circuit two of the transistors are connected as a long tailed differential pair
>with RF drive applied to the base of one transistor and with the modulated RF
>output taken from one of the collectors. The third transistor operates as a
>current source in the emitter "tail" circuit of the first two transistors, with
>audio applied to its base to modulate the RF output of the differential pair.
not sure which was the cart and which the horse here. I originally
started thinking of the 6ME8 back when Patrick was tinkering with
Gilbert Cells because I noticed a 'similarity' between it and the 3
transistor 'cell', of which the Gilbert has two on top a current
source.
Anyway, when working on this transmitter I tried 'emulating' something
akin to the beam deflection tube because I don't have a model for one
and constructed a 'long tail pair' using a pentode for the tail, which
is also akin to the 3 transistor 'cell'. That did point out one
difference, though. The deflection plates do not depend on cathodes
(emitters) driving the 'other side' to get the 'see saw' effect.
>There are also a bunch of dual plate tubes that were used in electric organs,
>that I have been meaning to lookup for a couple of years to see if there are any
>that might be useful in these circuits, so far I haven't gotten around to it.
out how it was intended to function.
either. Just that it's apparently 'bad' in one and the other circuit
works well, whatever that means.
>Another issue that Jacob addressed that I have been wondering about is what he
>calls "G3 Blurring", which he reduces by using a cathode follower to drive G3.
>It occurs to me that a refinement of this approach would be to use a
>direct-coupled cathode follower to allow driving G3 positive, although that
>would probably require an additional power supply, increasing complexity.
>
>I was not previously aware of the 6BY6 heptode mentioned on this page, it
>appears to have a considerably more linear G3 voltage vs. plate current curve
>than the 6CS6 most people on this side of the pond seem to talk about.
like the 6CS6 because of the unusually high G3 gm, so it can
potentially be driven without a preamp.
>Something else that I learned from this page, that should have been obvious to
>me before, is that the two tuned circuit heptode circuit that I suggested
>provides a degree of isolation between varying antenna capacity and the
>oscillator frequency.
>
>One thing that really strikes me about pages like this Radiomuseum web page is
>that the authors rarely state what all their design goals are. If I were
>building a small transmitter using four tube elements as Jacob did, I would
>probably go with a plate modulation approach, eliminating several problems
>presented by the grid modulation approach Jacob uses. The attractiveness of the
>heptodes to me are the possibilities for building a single tube transmitter
the FMing could be solved. Joe Sousa, who has another interesting
article on what he calls his 'hi-fi' transmitter, seems to think it
could be neutralized out.
Btw, his 'hi-fi' transmitter is the same idea I was proposing to
Patrick with the 6ME8 but he's using a dual control pentode to do it.
http://www.radiomuseum.org/forum/hi_fi_am_transmitter.html
One tube job, like your goal. Btw, he speaks of neutralization in that
article too (3- Frequency Stability) but decides his circuit doesn't
really need it.
flipper
Jun 26, 2012, 10:17:59 PM6/26/12
to
On Tue, 26 Jun 2012 15:38:20 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
Sorry. I'm the one who 'converted' it to dual control pentode.
wrote:
John Byrns
Jun 26, 2012, 11:29:35 PM6/26/12
to
In article <q6rku7989he2i4oi8...@4ax.com>,
Are you trying to play with my mind? I was the first to use the term "dual
control pentode in this thread when I made the statement "Applying the
converted it!
control pentode in this thread when I made the statement "Applying the
modulation to G1 doesn't take advantage of the beam deflection capabilities of
the tube and instead uses it in a way that a more ordinary tube, like a dual
control pentode, could serve." when I meant to say heptode. I'm the one that
the tube and instead uses it in a way that a more ordinary tube, like a dual
converted it!
flipper
Jun 27, 2012, 12:44:41 AM6/27/12
to
On Tue, 26 Jun 2012 22:29:35 -0500, John Byrns <byr...@sbcglobal.net>
LOL. Okay, sorry again ;)
For this application I tend to think of them as almost the same thing
but am beginning to realize that the 'extra' G4 screen is probably
what gives a heptode the higher plate impedance. Is that why, despite
the 'accidental' mentioning of dual control pentode, you seem to be
adamant about heptodes, or is there another reason?
For this application I tend to think of them as almost the same thing
but am beginning to realize that the 'extra' G4 screen is probably
what gives a heptode the higher plate impedance. Is that why, despite
the 'accidental' mentioning of dual control pentode, you seem to be
adamant about heptodes, or is there another reason?
Alex Pogossov
Jun 27, 2012, 2:21:53 AM6/27/12
to
news:p40ku71bm4ted609u...@4ax.com...
Then do another trick instead.
Tune any AM radio into a strong local AM station within tuning range of your
transmitter. Then tune your transmitter to a close frequency so that you
hear beat frequency on the radio. It is better to have the signal from the
transmitter stronger than this station so that you do not hear the audio but
only the beat whining (with some sideband chatter of course). Then proceed
with the above described tests. (A carrier of an AM station is even more
stable than any communications receiver.)
Regards,
Alex
> On Tue, 26 Jun 2012 22:09:21 +1000, "Alex Pogossov"
> <apog...@tpg.com.au> wrote:
>>As far as the parasitic FM is concerned, do two simple experiments.
>>
>>1. On a quality comms receiver in SSB mode listen to the carrier (set beat
>>frequency to a few hundred hertz) of your transmitter.
>>Without applying any audio, halve the cathode resistor of 6ME8 (touch it
>>with another resistor in parallel), effectively simulating +100%
>>modulation
>>excursion. Ideally, pitch of the beat tone shall not change. If it
>>changes... you have FM.
>>
>>2. On a quality comms receiver in SSB mode listen to the carrier (set beat
>>frequency to a few hundred hertz) of your transmitter.
>>Without applying any audio, play with the antenna tuning capacitor of your
>>transmitter. Ideally, pitch shall not change, only the volume as you tune
>>to
>>and detune from the resonance peak. If it changes...
>>
>>If you do these tests, could you "report" the results here. Then we can
>>discuss what causes FM in each of these cases.
>
> I don't have a SSB receiver.
>
Oops, surely you are not an RF guy.
> <apog...@tpg.com.au> wrote:
>>As far as the parasitic FM is concerned, do two simple experiments.
>>
>>1. On a quality comms receiver in SSB mode listen to the carrier (set beat
>>frequency to a few hundred hertz) of your transmitter.
>>Without applying any audio, halve the cathode resistor of 6ME8 (touch it
>>with another resistor in parallel), effectively simulating +100%
>>modulation
>>excursion. Ideally, pitch of the beat tone shall not change. If it
>>changes... you have FM.
>>
>>2. On a quality comms receiver in SSB mode listen to the carrier (set beat
>>frequency to a few hundred hertz) of your transmitter.
>>Without applying any audio, play with the antenna tuning capacitor of your
>>transmitter. Ideally, pitch shall not change, only the volume as you tune
>>to
>>and detune from the resonance peak. If it changes...
>>
>>If you do these tests, could you "report" the results here. Then we can
>>discuss what causes FM in each of these cases.
>
> I don't have a SSB receiver.
>
Then do another trick instead.
Tune any AM radio into a strong local AM station within tuning range of your
transmitter. Then tune your transmitter to a close frequency so that you
hear beat frequency on the radio. It is better to have the signal from the
transmitter stronger than this station so that you do not hear the audio but
only the beat whining (with some sideband chatter of course). Then proceed
with the above described tests. (A carrier of an AM station is even more
stable than any communications receiver.)
Regards,
Alex
flipper
Jun 27, 2012, 5:35:50 PM6/27/12
to
On Wed, 27 Jun 2012 16:21:53 +1000, "Alex Pogossov"
Wouldn't my frequency counter be simpler? hehe
With carrier at 640 KHz frequency deviation from idle to twice idle is
roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
.64022 MHz
With carrier at 640 KHz frequency deviation from idle to twice idle is
roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
.64022 MHz
Patrick Turner
Jun 27, 2012, 8:30:35 PM6/27/12
to
On Jun 27, 12:15 pm, flipper <flip...@fish.net> wrote:
> On Tue, 26 Jun 2012 15:36:48 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >In article <g5bfu79udllqk65f4e7i3lvptmcg6jh...@4ax.com>,
> On Tue, 26 Jun 2012 15:36:48 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
about a Gilbert Cell. I couldn't spend another 3 months perfecting
what I did with 3 transistors because I always have other things to
do. Using a gilbert cell chip probably would have worked better then
me trying to fiddle with discretes. But could a Gilbert cell be made
using 3 triodes?
Another tube with strong g3 action is the 7 pin 6DT6 commonly used for
quadrature FM detection. Another one I was intersted in was "gated
beam pentode" 6BN6, which some said never worked well as 6DT6. Maybe
these have apps at BC band F.
Patrick Turner.
flipper
Jun 27, 2012, 11:03:31 PM6/27/12
to
takes more than 3 transistors.
> Using a gilbert cell chip probably would have worked better then
>me trying to fiddle with discretes. But could a Gilbert cell be made
>using 3 triodes?
For the '3 transistor' cell I think pentodes might be more appropriate
because the bipolar collector current vs collector-emitter voltage
looks just like a pentode. Although, I'm not sure what the effect of
screen current mixed in might be. Or a single 6ME8 because that's
essentially how it works with the deflector plates acting as the two
upper transistors and G1 the bottom.
That's why I used a 6ME8: it's 'conceptually pure' with equivalent
'ins and outs'. As used, however, one plate is simply tied to B+ so,
if screen current and plate current 'see saw' in a heptode, which the
curves indicate is the case, then it might work similarly. I.E. what
goes into the 'unused' second plate externally connected to B+ is
'internally attached' screen current in the heptode. At least it seems
to be, in theory, but I didn't know what 'subtleties' might crop up
and, especially when doing something 'new', like to cut down the
unknowns as much as possible.
>Another tube with strong g3 action is the 7 pin 6DT6 commonly used for
>quadrature FM detection. Another one I was intersted in was "gated
>beam pentode" 6BN6, which some said never worked well as 6DT6. Maybe
>these have apps at BC band F.
>Patrick Turner.
discriminator made for FM demod and has a 'limiter' built in, which is
sort of the antithesis of AM. Of course, one doesn't have to use it
'as intended' but it's beyond me.
John Byrns
Jun 27, 2012, 11:06:46 PM6/27/12
to
In article <4f3lu7d1v6egkpmfa...@4ax.com>,
The reason I am "adamant about heptodes", if that is what you want to call it,
is not because of the plate resistance, it is simply because I have no
familiarity at all with DC pentodes and their characteristics.
I don't even know any DC Pentode type numbers and am not sophisticated enough to
know that the plate resistance of a DC pentode is lower than a heptode.
Assuming the plate resistance of a DC Pentode is lower than a heptode, is that a
bad thing?
I am not "adamant about heptodes", in fact I don't particularly care for grid
modulation, preferring plate modulation instead, however it is hard to conceive
of how to build a plate modulated transmitter using only one single section
tube, leaving the heptode, or possibly the DC Pentode, as the only easy choice
for a single tube transmitter. I do have a design for a single tube transmitter
using your 6ME8 that meets my design brief, however it would require me to build
my own well balanced push-pull RF antenna coupling transformer.
is not because of the plate resistance, it is simply because I have no
familiarity at all with DC pentodes and their characteristics.
I don't even know any DC Pentode type numbers and am not sophisticated enough to
know that the plate resistance of a DC pentode is lower than a heptode.
Assuming the plate resistance of a DC Pentode is lower than a heptode, is that a
bad thing?
I am not "adamant about heptodes", in fact I don't particularly care for grid
modulation, preferring plate modulation instead, however it is hard to conceive
of how to build a plate modulated transmitter using only one single section
tube, leaving the heptode, or possibly the DC Pentode, as the only easy choice
for a single tube transmitter. I do have a design for a single tube transmitter
using your 6ME8 that meets my design brief, however it would require me to build
my own well balanced push-pull RF antenna coupling transformer.
flipper
Jun 27, 2012, 11:59:46 PM6/27/12
to
On Wed, 27 Jun 2012 22:06:46 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
>In article <4f3lu7d1v6egkpmfa...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>>
>> For this application I tend to think of them as almost the same thing
>> but am beginning to realize that the 'extra' G4 screen is probably
>> what gives a heptode the higher plate impedance. Is that why, despite
>> the 'accidental' mentioning of dual control pentode, you seem to be
>> adamant about heptodes, or is there another reason?
>
>The reason I am "adamant about heptodes", if that is what you want to call it,
>is not because of the plate resistance, it is simply because I have no
>familiarity at all with DC pentodes and their characteristics.
Oh, okay. Well, just saying you're not familiar with them was enough
;)
>I don't even know any DC Pentode type numbers and am not sophisticated enough to
>know that the plate resistance of a DC pentode is lower than a heptode.
>Assuming the plate resistance of a DC Pentode is lower than a heptode, is that a
>bad thing?
Well, it affects plate tank Q.
6AS6, 6DT6, 6GY6, and 6888 are some of the more 'popular' ones for
broadcasters.
>I am not "adamant about heptodes", in fact I don't particularly care for grid
>modulation, preferring plate modulation instead, however it is hard to conceive
>of how to build a plate modulated transmitter using only one single section
>tube, leaving the heptode, or possibly the DC Pentode, as the only easy choice
>for a single tube transmitter. I do have a design for a single tube transmitter
>using your 6ME8 that meets my design brief, however it would require me to build
>my own well balanced push-pull RF antenna coupling transformer.
Yeah, a PP transformer is in some of my designs too but I'm skittish
about trying to wind one.
wrote:
>In article <4f3lu7d1v6egkpmfa...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>>
>> For this application I tend to think of them as almost the same thing
>> but am beginning to realize that the 'extra' G4 screen is probably
>> what gives a heptode the higher plate impedance. Is that why, despite
>> the 'accidental' mentioning of dual control pentode, you seem to be
>> adamant about heptodes, or is there another reason?
>
>The reason I am "adamant about heptodes", if that is what you want to call it,
>is not because of the plate resistance, it is simply because I have no
>familiarity at all with DC pentodes and their characteristics.
;)
>I don't even know any DC Pentode type numbers and am not sophisticated enough to
>know that the plate resistance of a DC pentode is lower than a heptode.
>Assuming the plate resistance of a DC Pentode is lower than a heptode, is that a
>bad thing?
6AS6, 6DT6, 6GY6, and 6888 are some of the more 'popular' ones for
broadcasters.
>I am not "adamant about heptodes", in fact I don't particularly care for grid
>modulation, preferring plate modulation instead, however it is hard to conceive
>of how to build a plate modulated transmitter using only one single section
>tube, leaving the heptode, or possibly the DC Pentode, as the only easy choice
>for a single tube transmitter. I do have a design for a single tube transmitter
>using your 6ME8 that meets my design brief, however it would require me to build
>my own well balanced push-pull RF antenna coupling transformer.
about trying to wind one.
Alex Pogossov
Jun 28, 2012, 6:31:40 AM6/28/12
to
news:djumu79p9rea3ptlh...@4ax.com...
> .64022 MHz.
It is not bad. I expected worse. Probably 6ME8 has less space charge
coupling than a regular heptode.
Still 70Hz is not acceptable for listening on a synchronous or SSB detector.
In these cases parasitic FM index shall not exceed about 0.25, so with the
lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
But who would deliberately listening to your transmitter on an SSB radio? No
one.
So, well done anyway!
It is not bad. I expected worse. Probably 6ME8 has less space charge
coupling than a regular heptode.
Still 70Hz is not acceptable for listening on a synchronous or SSB detector.
In these cases parasitic FM index shall not exceed about 0.25, so with the
lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
But who would deliberately listening to your transmitter on an SSB radio? No
one.
So, well done anyway!
John Byrns
Jun 28, 2012, 6:47:29 PM6/28/12
to
In article <hpknu799ef2up6tk1...@4ax.com>,
avoid sideband clipping and provide High Fidelity audio.
> 6AS6, 6DT6, 6GY6, and 6888 are some of the more 'popular' ones for
> broadcasters.
>
> >I am not "adamant about heptodes", in fact I don't particularly care for
> >grid
> >modulation, preferring plate modulation instead, however it is hard to
> >conceive
> >of how to build a plate modulated transmitter using only one single section
> >tube, leaving the heptode, or possibly the DC Pentode, as the only easy
> >choice
> >for a single tube transmitter. I do have a design for a single tube
> >transmitter
> >using your 6ME8 that meets my design brief, however it would require me to
> >build
> >my own well balanced push-pull RF antenna coupling transformer.
>
> Yeah, a PP transformer is in some of my designs too but I'm skittish
> about trying to wind one.
After a little out of the box thinking, it occurs to me that it may be possible
to make lemonade out of these lemons. It appears that by intentionally
unbalancing the transformer you can generate AM without the need for biasing the
deflectors to produce a carrier. Tight coupling between the two sections of the
winding would probably still be required to insure that no weird phase shifts
are produced which would prevent achieving 100% negative modulation. Hopefully
using a toroidal transformer core would provide adequate coupling between the
two plates?
flipper <fli...@fish.net> wrote:
> On Wed, 27 Jun 2012 22:06:46 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >In article <4f3lu7d1v6egkpmfa...@4ax.com>,
> > flipper <fli...@fish.net> wrote:
> >
> >>
> >> For this application I tend to think of them as almost the same thing
> >> but am beginning to realize that the 'extra' G4 screen is probably
> >> what gives a heptode the higher plate impedance. Is that why, despite
> >> the 'accidental' mentioning of dual control pentode, you seem to be
> >> adamant about heptodes, or is there another reason?
> >
> >The reason I am "adamant about heptodes", if that is what you want to call
> >it,
> >is not because of the plate resistance, it is simply because I have no
> >familiarity at all with DC pentodes and their characteristics.
>
> Oh, okay. Well, just saying you're not familiar with them was enough
> ;)
>
> >I don't even know any DC Pentode type numbers and am not sophisticated
> >enough to
> >know that the plate resistance of a DC pentode is lower than a heptode.
> >Assuming the plate resistance of a DC Pentode is lower than a heptode, is
> >that a
> >bad thing?
>
> Well, it affects plate tank Q.
But that is a good thing, as we need to keep the plate tank Q within reason to
> On Wed, 27 Jun 2012 22:06:46 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >In article <4f3lu7d1v6egkpmfa...@4ax.com>,
> > flipper <fli...@fish.net> wrote:
> >
> >>
> >> For this application I tend to think of them as almost the same thing
> >> but am beginning to realize that the 'extra' G4 screen is probably
> >> what gives a heptode the higher plate impedance. Is that why, despite
> >> the 'accidental' mentioning of dual control pentode, you seem to be
> >> adamant about heptodes, or is there another reason?
> >
> >The reason I am "adamant about heptodes", if that is what you want to call
> >it,
> >is not because of the plate resistance, it is simply because I have no
> >familiarity at all with DC pentodes and their characteristics.
>
> Oh, okay. Well, just saying you're not familiar with them was enough
> ;)
>
> >I don't even know any DC Pentode type numbers and am not sophisticated
> >enough to
> >know that the plate resistance of a DC pentode is lower than a heptode.
> >Assuming the plate resistance of a DC Pentode is lower than a heptode, is
> >that a
> >bad thing?
>
> Well, it affects plate tank Q.
avoid sideband clipping and provide High Fidelity audio.
> 6AS6, 6DT6, 6GY6, and 6888 are some of the more 'popular' ones for
> broadcasters.
>
> >I am not "adamant about heptodes", in fact I don't particularly care for
> >grid
> >modulation, preferring plate modulation instead, however it is hard to
> >conceive
> >of how to build a plate modulated transmitter using only one single section
> >tube, leaving the heptode, or possibly the DC Pentode, as the only easy
> >choice
> >for a single tube transmitter. I do have a design for a single tube
> >transmitter
> >using your 6ME8 that meets my design brief, however it would require me to
> >build
> >my own well balanced push-pull RF antenna coupling transformer.
>
> Yeah, a PP transformer is in some of my designs too but I'm skittish
> about trying to wind one.
to make lemonade out of these lemons. It appears that by intentionally
unbalancing the transformer you can generate AM without the need for biasing the
deflectors to produce a carrier. Tight coupling between the two sections of the
winding would probably still be required to insure that no weird phase shifts
are produced which would prevent achieving 100% negative modulation. Hopefully
using a toroidal transformer core would provide adequate coupling between the
two plates?
flipper
Jun 28, 2012, 10:21:10 PM6/28/12
to
On Thu, 28 Jun 2012 17:47:29 -0500, John Byrns <byr...@sbcglobal.net>
I don't know if that's necessarily a 'good thing' or not but you seem
to presuming 110k, or whatever the particular DC pentode comes out as,
results in 'the optimum value'.
One can always lower Q but you can't do better than the device limits
so, if all else were equal (which is seldom the case), my gut instinct
would be to 'err' on the side of 'too much' because that's readily
correctable.
>> 6AS6, 6DT6, 6GY6, and 6888 are some of the more 'popular' ones for
>> broadcasters.
>>
>> >I am not "adamant about heptodes", in fact I don't particularly care for
>> >grid
>> >modulation, preferring plate modulation instead, however it is hard to
>> >conceive
>> >of how to build a plate modulated transmitter using only one single section
>> >tube, leaving the heptode, or possibly the DC Pentode, as the only easy
>> >choice
>> >for a single tube transmitter. I do have a design for a single tube
>> >transmitter
>> >using your 6ME8 that meets my design brief, however it would require me to
>> >build
>> >my own well balanced push-pull RF antenna coupling transformer.
>>
>> Yeah, a PP transformer is in some of my designs too but I'm skittish
>> about trying to wind one.
>
>After a little out of the box thinking, it occurs to me that it may be possible
>to make lemonade out of these lemons. It appears that by intentionally
>unbalancing the transformer you can generate AM without the need for biasing the
>deflectors to produce a carrier. Tight coupling between the two sections of the
>winding would probably still be required to insure that no weird phase shifts
>are produced which would prevent achieving 100% negative modulation. Hopefully
>using a toroidal transformer core would provide adequate coupling between the
>two plates?
I've debated myself into recursive circles trying to decide which
'imbalance' is the least problematic but, at the moment, I'm on the
'balanced transformer', offset deflectors, side of the circle
precisely because of the potential phase shift issue you mentioned. I
also kind of like the idea that deflector offset is adjustable.
Btw, that previously posted 'hi-hi' link is the same thing using a
dual control pentode. He does use an imbalanced transformer but I'm
not sure how that compares because, in a DC pentode, the plate and
screen are not 'equal' like the dual plates on a 6ME8. I also kind of
wonder about screen modulation but it seems to work, maybe because
amplitudes are so small.
to presuming 110k, or whatever the particular DC pentode comes out as,
results in 'the optimum value'.
One can always lower Q but you can't do better than the device limits
so, if all else were equal (which is seldom the case), my gut instinct
would be to 'err' on the side of 'too much' because that's readily
correctable.
>> 6AS6, 6DT6, 6GY6, and 6888 are some of the more 'popular' ones for
>> broadcasters.
>>
>> >I am not "adamant about heptodes", in fact I don't particularly care for
>> >grid
>> >modulation, preferring plate modulation instead, however it is hard to
>> >conceive
>> >of how to build a plate modulated transmitter using only one single section
>> >tube, leaving the heptode, or possibly the DC Pentode, as the only easy
>> >choice
>> >for a single tube transmitter. I do have a design for a single tube
>> >transmitter
>> >using your 6ME8 that meets my design brief, however it would require me to
>> >build
>> >my own well balanced push-pull RF antenna coupling transformer.
>>
>> Yeah, a PP transformer is in some of my designs too but I'm skittish
>> about trying to wind one.
>
>After a little out of the box thinking, it occurs to me that it may be possible
>to make lemonade out of these lemons. It appears that by intentionally
>unbalancing the transformer you can generate AM without the need for biasing the
>deflectors to produce a carrier. Tight coupling between the two sections of the
>winding would probably still be required to insure that no weird phase shifts
>are produced which would prevent achieving 100% negative modulation. Hopefully
>using a toroidal transformer core would provide adequate coupling between the
>two plates?
'imbalance' is the least problematic but, at the moment, I'm on the
'balanced transformer', offset deflectors, side of the circle
precisely because of the potential phase shift issue you mentioned. I
also kind of like the idea that deflector offset is adjustable.
Btw, that previously posted 'hi-hi' link is the same thing using a
dual control pentode. He does use an imbalanced transformer but I'm
not sure how that compares because, in a DC pentode, the plate and
screen are not 'equal' like the dual plates on a 6ME8. I also kind of
wonder about screen modulation but it seems to work, maybe because
amplitudes are so small.
Patrick Turner
Jun 29, 2012, 6:59:51 AM6/29/12
to
Its awhile since I mucked around with a Gilner, or Turbert Cell or
whatever I tried to use as an am modulator with some simularity to
Gilbert, which does have more than 3 transistors, 6 as you say.
But looking at my notes and the collection of Gilbert cell schematics
I have, what I used was a 1/2 Gilbert cell with an LTP driving
balanced RF OPT, one side of LTP to 0V, other side to RF carrier F,
and tail is transistor collector with AF applied to base. Very simple,
but not as good as using tubes as I had already. Maybe Gilbert himself
used 6 triodes to make his original Cell, so 3 x 12AT7 might work
fine, although with two LTPs using one 12AT7, Ik is twice Ia in one
tube, so the cathode drive might best be a 6DJ8 or a pentode. I've
never had time to findoutabout and do my own complete research and
development. But one reason behind good sounding stereo FM is due to
Gilbert cells. Paravicini tried to con the world with a design for an
"all tubed MPX decoder" over 10 years ago, but it was just a chip with
Gibert within, and buffer tubes on output. The guy didn't do a real
tubed MPX decoder with tubed Gilbert cell. Manufacturers avoided doing
it like the plague because the cost of 3 twin triodes was far higher
than the usual crappy diode matrix.
I recall Fisher used 6BN6. I never bothered to build anything using
the tube. But 6DT6 was used in very many TV sets to get the audio from
the FM audio signal and chief benefit was the large amount of audio
output compared to using Foster Seely discriminator or ratio detector.
The 6DT6 gave quite low enough THD, and its use could eliminate the
use of a preamp tube so hence it "paid for itself" The g3 has fairly
high gm, and in FM demod, the circuit connected to g3 is an oscillator
at Fm frequency, very easy to get running reliably. Anyway, maybe the
tube could work well as an Am modulator.
I know you are hooked on using just one tube, and someone said stuff
about FM being created and maybe that's true, because the audio
changes the effective C across a coil that determines the F so the
tuning F changes. I have a separate triode oscillator, then CF buffer,
and audio cannot get near anything oscillating at RF. I have a pentode
RF amp tube, and fairly large Rk bypassed for RF, but not for AF, so
when AF is applied to g1 with the RF from resistance divider from CF
RF source, the tube current is changed fairly linearly because there
is local current FB at AF, so wave form wasn't so terribly bad as say
50% mod. But adding more tubes to get NFB applied to audio stage much
reduced THD in wave form, and got me closer to 100% mod. I found many
old radios make a real mess when they try to detect audio when
modulation exceeds 50%, and of course back in good old days AM was not
fully modulated. But when you look at radio station waves, there is
high % mod, and of course compression maybe. And many superhet radio
sets can't produce a clean 100% mod IF wave when there is a 100% mod
RF wave input. I'm not sure how much NFB you have but certainly it
does work to make AM better you have non linear tubes or feeble tubes
to deal with. I've not tried to use plate modulation which ppl say is
best, one reason being that the modulator adds power to the output
signal, and % mod is more linear to AF input to modulator.
There must be many ways to make AM.
I googled "tube gilbert cell" and other like things but no tubed
Gilbert Cell came onto my screen when I clicked 'images' ....so all
trial and error needed.
Patrick T.
Patrick Turner
Jun 29, 2012, 7:16:31 AM6/29/12
to
On Jun 28, 1:59 pm, flipper <flip...@fish.net> wrote:
> On Wed, 27 Jun 2012 22:06:46 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >In article <4f3lu7d1v6egkpmfapqsb3b2kgimmkb...@4ax.com>,
> On Wed, 27 Jun 2012 22:06:46 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
double tuning gang with frame and moving plates at 0V with 0.01uF caps
from fixed plated to coil ends.
the thing is to allow the tubes to drive a nice high ohm RLa-a load,
and have a secondary wound over the middle part of the winding, ie,
the earthy part, and have this winding about 1/5 of the total P turns,
so you get a 25:1 impedance match so antena C has hardly any effect on
coil tuning. if the RLa-a in parallel with tube Ra was say 25k, then
Zout from sec is 1k0, quite low enough to get over effects of C across
OP terminals etc. probably an air cored coil is OK using 40mm PVC pipe
former with cat 5 strands of solid wire for coil. But maybe best if
you drill many holes as possible in PVC to reduce diaelectric losses.
Layout and keeping coil away from other metal and keeping leads tidy
assist in overall Q which need not be huge. Another way to get low Z
output is to have a tap on one side of PP coil, at say 20% of turns so
you can use 0.01 cap to take out SE signal. No sec winding, and you
still get benefits of the Z transformation and stray output C from
antenna won't worry the coil.
Patrick Turner.
flipper
Jun 29, 2012, 7:18:36 PM6/29/12
to
On Thu, 28 Jun 2012 20:31:40 +1000, "Alex Pogossov"
> Probably 6ME8 has less space charge
>coupling than a regular heptode.
I suspect so and it's one of the things I think are 'different'
because of the beam deflection.
>Still 70Hz is not acceptable for listening on a synchronous or SSB detector.
>In these cases parasitic FM index shall not exceed about 0.25, so with the
>lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
>
>But who would deliberately listening to your transmitter on an SSB radio? No
>one.
Right. What I'd rather know is how much deviation is 'acceptable' for
the BCB.
>So, well done anyway!
Thanks. I'm getting the impression it might not be worth even trying a
heptode or DC pentode.
<apog...@tpg.com.au> wrote:
>
>"flipper" <fli...@fish.net> wrote in message
>news:djumu79p9rea3ptlh...@4ax.com...
>> On Wed, 27 Jun 2012 16:21:53 +1000, "Alex Pogossov"
>> <apog...@tpg.com.au> wrote:
>>>Oops, surely you are not an RF guy.
>>>Then do another trick instead.
>>>Tune any AM radio into a strong local AM station within tuning range of
>>>your
>>>transmitter. Then tune your transmitter to a close frequency so that you
>>>hear beat frequency on the radio. It is better to have the signal from the
>>>transmitter stronger than this station so that you do not hear the audio
>>>but
>>>only the beat whining (with some sideband chatter of course). Then proceed
>>>with the above described tests. (A carrier of an AM station is even more
>>>stable than any communications receiver.)
>>>
>>>Regards,
>>>Alex
>>
>>
>> Wouldn't my frequency counter be simpler? hehe
>>
>> With carrier at 640 KHz frequency deviation from idle to twice idle is
>> roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
>> .64022 MHz.
>
>It is not bad. I expected worse.
Were you thinking maybe a CF would be needed?
>
>"flipper" <fli...@fish.net> wrote in message
>news:djumu79p9rea3ptlh...@4ax.com...
>> On Wed, 27 Jun 2012 16:21:53 +1000, "Alex Pogossov"
>> <apog...@tpg.com.au> wrote:
>>>Oops, surely you are not an RF guy.
>>>Then do another trick instead.
>>>Tune any AM radio into a strong local AM station within tuning range of
>>>your
>>>transmitter. Then tune your transmitter to a close frequency so that you
>>>hear beat frequency on the radio. It is better to have the signal from the
>>>transmitter stronger than this station so that you do not hear the audio
>>>but
>>>only the beat whining (with some sideband chatter of course). Then proceed
>>>with the above described tests. (A carrier of an AM station is even more
>>>stable than any communications receiver.)
>>>
>>>Regards,
>>>Alex
>>
>>
>> Wouldn't my frequency counter be simpler? hehe
>>
>> With carrier at 640 KHz frequency deviation from idle to twice idle is
>> roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
>> .64022 MHz.
>
>It is not bad. I expected worse.
> Probably 6ME8 has less space charge
>coupling than a regular heptode.
because of the beam deflection.
>Still 70Hz is not acceptable for listening on a synchronous or SSB detector.
>In these cases parasitic FM index shall not exceed about 0.25, so with the
>lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
>
>But who would deliberately listening to your transmitter on an SSB radio? No
>one.
the BCB.
>So, well done anyway!
Thanks. I'm getting the impression it might not be worth even trying a
heptode or DC pentode.
flipper
Jun 29, 2012, 8:20:57 PM6/29/12
to
work fine, unless you were holding one side to 0V without bipolar
supplies because there has to be 'something' across the bottom
transistor.
The basic topology you describe is the one I 'almost built':
http://flipperhome.dyndns.org/AM%20Transmitter.htm
And, as previously mentioned, it is the solid stage version of Beamus.
I.E., compared to the 6ME8, the bottom transistor is G1 and the top
two are the deflection plates.
> Maybe Gilbert himself
>used 6 triodes to make his original Cell,
> so 3 x 12AT7 might work
>fine, although with two LTPs using one 12AT7, Ik is twice Ia in one
>tube, so the cathode drive might best be a 6DJ8 or a pentode. I've
>never had time to findoutabout and do my own complete research and
>development. But one reason behind good sounding stereo FM is due to
>Gilbert cells. Paravicini tried to con the world with a design for an
>"all tubed MPX decoder" over 10 years ago, but it was just a chip with
>Gibert within, and buffer tubes on output. The guy didn't do a real
>tubed MPX decoder with tubed Gilbert cell. Manufacturers avoided doing
>it like the plague because the cost of 3 twin triodes was far higher
>than the usual crappy diode matrix.
>
>I recall Fisher used 6BN6. I never bothered to build anything using
>the tube. But 6DT6 was used in very many TV sets to get the audio from
>the FM audio signal and chief benefit was the large amount of audio
>output compared to using Foster Seely discriminator or ratio detector.
>The 6DT6 gave quite low enough THD, and its use could eliminate the
>use of a preamp tube so hence it "paid for itself" The g3 has fairly
>high gm, and in FM demod, the circuit connected to g3 is an oscillator
>at Fm frequency, very easy to get running reliably. Anyway, maybe the
>tube could work well as an Am modulator.
http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
'conventional' audio to G3 approach.
The 6BN6 is a whole different thing: a gated beam discriminator and
limiter with the express purpose of rejecting AM.
>I know you are hooked on using just one tube,
experimenting with using one tube. That was 'the point' of it: to see
what one could do with one tube.
But everything I've done since that experiment involves more than one
tube.
What would be a more accurate characterization is I like the challenge
of getting the most'est from the least'est.
> and someone said stuff
>about FM being created and maybe that's true, because the audio
>changes the effective C across a coil that determines the F so the
>tuning F changes. I have a separate triode oscillator,
> then CF buffer,
>and audio cannot get near anything oscillating at RF. I have a pentode
>RF amp tube, and fairly large Rk bypassed for RF, but not for AF, so
>when AF is applied to g1 with the RF from resistance divider from CF
>RF source, the tube current is changed fairly linearly because there
>is local current FB at AF, so wave form wasn't so terribly bad as say
>50% mod. But adding more tubes to get NFB applied to audio stage much
>reduced THD in wave form, and got me closer to 100% mod. I found many
>old radios make a real mess when they try to detect audio when
>modulation exceeds 50%, and of course back in good old days AM was not
>fully modulated. But when you look at radio station waves, there is
>high % mod, and of course compression maybe. And many superhet radio
>sets can't produce a clean 100% mod IF wave when there is a 100% mod
>RF wave input. I'm not sure how much NFB you have but certainly it
>does work to make AM better you have non linear tubes or feeble tubes
>to deal with.
between 15 dB and 20 dB.
Alex Pogossov
Jun 30, 2012, 1:45:37 AM6/30/12
to
news:qhbsu7dtce6cv7uj5...@4ax.com...
for some FM.
>
>> Probably 6ME8 has less space charge
>>coupling than a regular heptode.
>
> I suspect so and it's one of the things I think are 'different'
> because of the beam deflection.
In the deflection tube the electron beam is far away (relatively) from the
deflection plates, not brushing them even when the deflection plates sit at
positive bias. In a heptode the space charge is right in the vicinity of the
G3, enveloping it and nearly touching it. Therefore the deflection tube
space charge effects are smaller.
But... the deflection tube requires +/-30V for full beam cutoff/switchover,
while a heptode probably would cutoff at -5V. Six times less. Now if you
arrange a tap on the oscillator running at 30V at 1/6-th, and feed it to a
heptode G3, the effect will be 6*6=36 times smaller. So you end up with the
same FM normalised to an available oscillator voltage. A hetode will give
low FM if coupled to a tap of the oscillator tank.
Because of huge deflection drive requirement, such tubes never gained
popularity as generic mixers.
>
>>Still 70Hz is not acceptable for listening on a synchronous or SSB
>>detector.
>>In these cases parasitic FM index shall not exceed about 0.25, so with the
>>lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
>>
>>But who would deliberately listening to your transmitter on an SSB radio?
>>No
>>one.
>
> Right. What I'd rather know is how much deviation is 'acceptable' for
> the BCB.
Up to 1kHz, I would guess.
> On Thu, 28 Jun 2012 20:31:40 +1000, "Alex Pogossov"
>>> With carrier at 640 KHz frequency deviation from idle to twice idle is
>>> roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
>>> .64022 MHz.
>>
>>It is not bad. I expected worse.
>
> Were you thinking maybe a CF would be needed?
In a quality equipment -- yes, but not in your case when you knowingly allow
>>> roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
>>> .64022 MHz.
>>
>>It is not bad. I expected worse.
>
> Were you thinking maybe a CF would be needed?
for some FM.
>
>> Probably 6ME8 has less space charge
>>coupling than a regular heptode.
>
> I suspect so and it's one of the things I think are 'different'
> because of the beam deflection.
deflection plates, not brushing them even when the deflection plates sit at
positive bias. In a heptode the space charge is right in the vicinity of the
G3, enveloping it and nearly touching it. Therefore the deflection tube
space charge effects are smaller.
But... the deflection tube requires +/-30V for full beam cutoff/switchover,
while a heptode probably would cutoff at -5V. Six times less. Now if you
arrange a tap on the oscillator running at 30V at 1/6-th, and feed it to a
heptode G3, the effect will be 6*6=36 times smaller. So you end up with the
same FM normalised to an available oscillator voltage. A hetode will give
low FM if coupled to a tap of the oscillator tank.
Because of huge deflection drive requirement, such tubes never gained
popularity as generic mixers.
>
>>Still 70Hz is not acceptable for listening on a synchronous or SSB
>>detector.
>>In these cases parasitic FM index shall not exceed about 0.25, so with the
>>lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
>>
>>But who would deliberately listening to your transmitter on an SSB radio?
>>No
>>one.
>
> Right. What I'd rather know is how much deviation is 'acceptable' for
> the BCB.
Patrick Turner
Jun 30, 2012, 2:49:00 AM6/30/12
to
> >Its awhile since I mucked around with a Gilner, or Turbert Cell or
> >whatever I tried to use as an am modulator with some simularity to
> >Gilbert, which does have more than 3 transistors, 6 as you say.
> >But looking at my notes and the collection of Gilbert cell schematics
> >I have, what I used was a 1/2 Gilbert cell with an LTP driving
> >balanced RF OPT, one side of LTP to 0V, other side to RF carrier F,
> >and tail is transistor collector with AF applied to base. Very simple,
> >but not as good as using tubes as I had already.
>
> Well, I think the balanced RF OPT was overkill but the 'LTP' should
> work fine, unless you were holding one side to 0V without bipolar
> supplies because there has to be 'something' across the bottom
> transistor.
>
> The basic topology you describe is the one I 'almost built':http://flipperhome.dyndns.org/AM%20Transmitter.htm
> And, as previously mentioned, it is the solid stage version of Beamus.
> I.E., compared to the 6ME8, the bottom transistor is G1 and the top
> two are the deflection plates.
>
> > Maybe Gilbert himself
> >used 6 triodes to make his original Cell,
>
> Barrie's original cell was bipolar.
I suspected it was, nothing tubed on Google, but whatever you can do
> >whatever I tried to use as an am modulator with some simularity to
> >Gilbert, which does have more than 3 transistors, 6 as you say.
> >But looking at my notes and the collection of Gilbert cell schematics
> >I have, what I used was a 1/2 Gilbert cell with an LTP driving
> >balanced RF OPT, one side of LTP to 0V, other side to RF carrier F,
> >and tail is transistor collector with AF applied to base. Very simple,
> >but not as good as using tubes as I had already.
>
> Well, I think the balanced RF OPT was overkill but the 'LTP' should
> work fine, unless you were holding one side to 0V without bipolar
> supplies because there has to be 'something' across the bottom
> transistor.
>
> The basic topology you describe is the one I 'almost built':http://flipperhome.dyndns.org/AM%20Transmitter.htm
> And, as previously mentioned, it is the solid stage version of Beamus.
> I.E., compared to the 6ME8, the bottom transistor is G1 and the top
> two are the deflection plates.
>
> > Maybe Gilbert himself
> >used 6 triodes to make his original Cell,
>
> Barrie's original cell was bipolar.
with bjts, you can also do with tubes, well, nearly whatever anyway,
until you need a pnp tube, which can't exist, and tubes are all npn,
and I mean figuratively, not literally..
>
> The 6DT6, yes, because it's simply a dual control pentode. Here's one http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
> 'conventional' audio to G3 approach.
Gee, that looks really nice and simple! Makes me wanna build one, I
> The 6DT6, yes, because it's simply a dual control pentode. Here's one http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
> 'conventional' audio to G3 approach.
might have a few 6DT6 laying around.
>
> The 6BN6 is a whole different thing: a gated beam discriminator and
> limiter with the express purpose of rejecting AM.
>
> >I know you are hooked on using just one tube,
>
> Well, no. I was only 'hooked' on using one tube back when I was
> experimenting with using one tube. That was 'the point' of it: to see
> what one could do with one tube.
wear so many hats and does so many things, and which hardly anyone
knows anything about.
>
> But everything I've done since that experiment involves more than one
> tube.
>
> What would be a more accurate characterization is I like the challenge
> of getting the most'est from the least'est.
OK, but to me the 6ME6 is a "leastest with mostest" type of tube.
> But everything I've done since that experiment involves more than one
> tube.
>
> What would be a more accurate characterization is I like the challenge
> of getting the most'est from the least'est.
>
> > and someone said stuff
> >about FM being created and maybe that's true, because the audio
> >changes the effective C across a coil that determines the F so the
> >tuning F changes. I have a separate triode oscillator,
>
> So does Beamus. Well, except I used a pentode osc.
>
>
>
>
>
> > then CF buffer,
> >and audio cannot get near anything oscillating at RF. I have a pentode
> >RF amp tube, and fairly large Rk bypassed for RF, but not for AF, so
> >when AF is applied to g1 with the RF from resistance divider from CF
> >RF source, the tube current is changed fairly linearly because there
> >is local current FB at AF, so wave form wasn't so terribly bad as say
> >50% mod. But adding more tubes to get NFB applied to audio stage much
> >reduced THD in wave form, and got me closer to 100% mod. I found many
> >old radios make a real mess when they try to detect audio when
> >modulation exceeds 50%, and of course back in good old days AM was not
> >fully modulated. But when you look at radio station waves, there is
> >high % mod, and of course compression maybe. And many superhet radio
> >sets can't produce a clean 100% mod IF wave when there is a 100% mod
> >RF wave input. I'm not sure how much NFB you have but certainly it
> >does work to make AM better you have non linear tubes or feeble tubes
> >to deal with.
>
> I'm not sure what 'feeble' is supposed to mean but I estimate NFB at
> between 15 dB and 20 dB.
Feeble means low Ia and low gm, and thast usually means low "figure of
> > and someone said stuff
> >about FM being created and maybe that's true, because the audio
> >changes the effective C across a coil that determines the F so the
> >tuning F changes. I have a separate triode oscillator,
>
> So does Beamus. Well, except I used a pentode osc.
>
>
>
>
>
> > then CF buffer,
> >and audio cannot get near anything oscillating at RF. I have a pentode
> >RF amp tube, and fairly large Rk bypassed for RF, but not for AF, so
> >when AF is applied to g1 with the RF from resistance divider from CF
> >RF source, the tube current is changed fairly linearly because there
> >is local current FB at AF, so wave form wasn't so terribly bad as say
> >50% mod. But adding more tubes to get NFB applied to audio stage much
> >reduced THD in wave form, and got me closer to 100% mod. I found many
> >old radios make a real mess when they try to detect audio when
> >modulation exceeds 50%, and of course back in good old days AM was not
> >fully modulated. But when you look at radio station waves, there is
> >high % mod, and of course compression maybe. And many superhet radio
> >sets can't produce a clean 100% mod IF wave when there is a 100% mod
> >RF wave input. I'm not sure how much NFB you have but certainly it
> >does work to make AM better you have non linear tubes or feeble tubes
> >to deal with.
>
> I'm not sure what 'feeble' is supposed to mean but I estimate NFB at
> between 15 dB and 20 dB.
merit". For example, EF86 is a gutless wonder of a tube which ws used
in thousands of amps in 1955 where one wanted gain between 80 and 150
and with only one nine pin socket and 0.7mA of anode current. It
replaced the use of a 12AU7 with 2 halves cascaded to get similar
gain, but with more total Ia and more R&C parts. I'd rather have the
12AU7. Look at Quad-II with EF86. Bloody awful! To get the gain needed
Quad used 180k RLa resistors and following bias R for KT66 were 680k
which often began to develop a high positive biasing voltage when
tubes age a bit. One shouldn't have Rg more than 120k max, ever, but
that weighs down the gain of feeble EF86. Later, in 1995, Chinese who
bought Quad used Andy Grove's revised Quad-II-Forty circuit with 6SH7
in pentode, almost identical in operation to early Quad-II. 6SH7 is an
octal tube with potentially far higher gm than EF86, or the 6SJ and
6SJ7 family of pentodes. Why Grove went with the original Quad-II
schematic is a mystery when it would have been much better to use 6SH7
in triode followed by 6SN7 as an LTP with CCS tail is beyond me. I
have repaired and re-wired a few Quad-II-Forty, because once you take
a careful look beyond the nice looking Chinese paint, its a POS amp,
and they have 6SH7 operating with Ia way below where its figure of
merit would be high, and with 470k biasing for the KT88 instead of the
KT66. Instead of gutless wonder pentodes with Ia at 1mA, I like more
linear triodes with 4 or 5 mA at least.
Enough about feeble tubes. In RF work, I like to see healthy Ia
because capacitances anywhere tend to have low Z paths at RF.
15-20dB FB is good, enough to do something worth doing, assuming the
gear makes maybe 7% Dn at 90% mod without NFB.
But I found that with NFB applied, the modulation bandwidth with
undistorted wave at say 95% mod was reduced, and to be expected of
course. In other words, modulation can be 95% OK with AF from say 15Hz
to 5kHz, with envelope wave having THD < 1%, and enough to test most
AM radios, but above 5kHz, the envelope THD increases. OK though at
50% mod, good from 15Hz to 20kHz, less than maybe 0.5%. Its hard to
measure because one needs a linear detector, But then with dual trace
CRO one can compare input AF with recovered AF at detector output of
receiver, and if that detector is my double triode type with two CF
triodes then any visible distortion is likely to be in receiver mixer
and IF stage, not in sig gene, not in detector.
The AM transmissions here from radio stations are pretty good, and
probably with lower THD + IMD than in 1955, so recievers are the weak
link. My own AM set has cascode vari-mu twin triode input, 6AN7 mixer,
then 6BX6 IF amp with unbypassed Rk for some local current FB and no
AGC voltage is applied to it. 6BX6 or EF80 is a sharp cut off pentode
with strong gm normally used for 4.5Mhz TV IF stages. Its operating
with fixed bias. AGC voltage is generated by the detector CF stage,
and it works on mixer just a bit, and on RF input tube. I have
variable spacing between IFT1 like Hellicrafter used on their
communications radios with 3 IFTs to vary the Q for the IF band. But I
vary between flat top pass band and twin peak, so with two IFTs, and
IFT2 with normal 9kHz pass band, I get 18kHz pass band and 9kHz of AF
bandwidth. Some HF emphasis in R&C network boosts the AF BW to 10kHz.
I hate to seem immodest, but my AM radio outperforms ALL other AM
radios I have ever placed beside it to compare them, including Quad's
AM tuner which I think is next best. My point is that OK, one makes a
nice trick thing like a decent AM "mini-transmitter" and you can put
on a CD in the garage and fool ppl into believing there's this nice
new radio station without adds and some BS artist, but what about the
radio receiver perfomance?
Most AM radios are shit boxes, fair dunkum. The worst are the 3
transistor types added to an AM/FM receiver-tuner, 1.5kHz of audio BW
if you are lucky and very high THD IMD. Tubed types vary between
passable and "it needs total re-engineering". Passable means 4kHz BW,
typical THD/IMD 4%. Good means 7kHz BW, THD/IMD < 2%, and Excellent
means 10kHz and Dn < 1%.
There's no use going for more BW because AM transmissions are limited
to less than 10kHz, usually. Only 4.5kHz, where stations are close
together.
Patrick Turner.
John Byrns
Jun 30, 2012, 12:37:49 PM6/30/12
to
In article <fda19320-b0ca-40b9...@h10g2000pbi.googlegroups.com>,
behind the Turner tuner? At first glance QUAD's AM design doesn't seem much
different than dozens of other AM tuner designs, exactly what is it that gives
it the edge over the other pretenders?
Patrick Turner <in...@turneraudio.com.au> wrote:
> The AM transmissions here from radio stations are pretty good, and
> probably with lower THD + IMD than in 1955, so recievers are the weak
> link. My own AM set has cascode vari-mu twin triode input, 6AN7 mixer,
> then 6BX6 IF amp with unbypassed Rk for some local current FB and no
> AGC voltage is applied to it. 6BX6 or EF80 is a sharp cut off pentode
> with strong gm normally used for 4.5Mhz TV IF stages. Its operating
> with fixed bias. AGC voltage is generated by the detector CF stage,
> and it works on mixer just a bit, and on RF input tube. I have
> variable spacing between IFT1 like Hellicrafter used on their
> communications radios with 3 IFTs to vary the Q for the IF band. But I
> vary between flat top pass band and twin peak, so with two IFTs, and
> IFT2 with normal 9kHz pass band, I get 18kHz pass band and 9kHz of AF
> bandwidth. Some HF emphasis in R&C network boosts the AF BW to 10kHz.
> I hate to seem immodest, but my AM radio outperforms ALL other AM
> radios I have ever placed beside it to compare them, including Quad's
> AM tuner which I think is next best.
What is it about QUAD'a AM tuner that places it in the much coveted second place
> The AM transmissions here from radio stations are pretty good, and
> probably with lower THD + IMD than in 1955, so recievers are the weak
> link. My own AM set has cascode vari-mu twin triode input, 6AN7 mixer,
> then 6BX6 IF amp with unbypassed Rk for some local current FB and no
> AGC voltage is applied to it. 6BX6 or EF80 is a sharp cut off pentode
> with strong gm normally used for 4.5Mhz TV IF stages. Its operating
> with fixed bias. AGC voltage is generated by the detector CF stage,
> and it works on mixer just a bit, and on RF input tube. I have
> variable spacing between IFT1 like Hellicrafter used on their
> communications radios with 3 IFTs to vary the Q for the IF band. But I
> vary between flat top pass band and twin peak, so with two IFTs, and
> IFT2 with normal 9kHz pass band, I get 18kHz pass band and 9kHz of AF
> bandwidth. Some HF emphasis in R&C network boosts the AF BW to 10kHz.
> I hate to seem immodest, but my AM radio outperforms ALL other AM
> radios I have ever placed beside it to compare them, including Quad's
> AM tuner which I think is next best.
behind the Turner tuner? At first glance QUAD's AM design doesn't seem much
different than dozens of other AM tuner designs, exactly what is it that gives
it the edge over the other pretenders?
Patrick Turner
Jun 30, 2012, 9:26:04 PM6/30/12
to
On Jul 1, 2:37 am, John Byrns <byr...@sbcglobal.net> wrote:
> In article <fda19320-b0ca-40b9-89a2-547158659...@h10g2000pbi.googlegroups.com>,
Well, Quad managed to get the selectivity variation just right with
switched tertiary on IFT 1. Most other radio makers steadfastly
avoided such delights and fed the buying public with half the audio BW
of Quad. Quad also got it right with traditional diode detector AF
detector, not too much Dn. But nearly all these old AM radios and
tuners need a ferrite rod antenna with short leads to pick up the
magnetic wave more than the electrostatic wave which comes to most old
sets to a high Z coil loosely magneticly coupled to tuned RF input
coil ahed of an RF amp or mixer.
Today's household wull be filled with fluroresant lamps each with tiny
SMPS plus all the other SMPS crap, plus linear PSU crap which all
manages to modulate station signals which arrive at the house with
100Hz wave, so the receivers pick up some radio stations which have
100Hz added modulation, and this hum can't be avoided when tuning.
So in a good AM tuner for best fidelity local pick up, you need a
ferrite rod antenna, a switched IFT 1 for wider AF BW, and so Quad had
some way to go to equal what I have. But in 1999 when I made my radio
from spare junk-box parts there was little radio interference and I
had two cascaded RF input coils with stagger tuning at the low end of
the band for better AF bw on stations which I like. But I had to
change later to a ferrite rod because of noise. The Quad also has
beautifully made tuning controls and various bands, where just have
one band. Quad's detector could have been better, but they'd have had
to use something similar to my cathode followers with ge diodes etc.
If someone were to spend a few days altering a Quad AM tuner to make
it better, I'm sure they could.
Because AM radio stations have crystal controlled oscillators set for
very specific frequencies, it would be nice to add a digital press
button frequency generator in receiver with setable F to tune instead
of having any oscillator. But you'd still have to tune the RF coil, so
as they are will do. I must put a tuning meter and better dial into my
set, and I have a suitable old FM tuner module which sounds well which
I could install in my set, because I often switch from AM to kitchen
to get the least boring program or best music. Then all AM and FM can
be in the one box which really needs its timber front altered and a
new longer slide type dial put in. Another fix-up job to attend to
during my retirement beginning on August 1 in a month.
Its all not really needed now because we have Digital Audio
Broadcasting, and a digital radio is cheap because all such gadgetty
stuff is made in Asia for 20c each. But speakers and audio amps in
little digital radios are awful, so one can focus on tube audio amps
and decent speakers. I might switch to stereo for the kitchen, so
speakers would be small, but like my old VAF bookshelfs for my TV set,
bokkshelfs can be good enough. The 1995 Metz TV I was given had The
World's Worst audio system which I removed entirely. Audio comes to
old solid state pioneer amp and VAF speakers from the back of the set-
top box. Good enough for TV sound which is often poor to begin with.
Maybe I get a 105cm LCD screen soon, but still have to do better with
sound. Like a lot of things, I don't have the time or inclination to
stay put mentally in 1955, or 1999, when other things have come along.
Well done AM is cabable of very good reception, but it all got dumbed
down with many stations put too close to allow wider AF. I've often
thought of using IF at 2MHz, and coils would be easier, and probably 3
IFTs would be needed. Even with Q = 100, pass band is 20kHz, so 10kHz
of audio.
In 1999 I tried first to make a DG Tucker style tubed synchrodyne, but
that proved much more difficult while offering poor performance
compared to a good superhet. The coils needed were beyound my time
allowances. There were no data sheets or how-to-do anywhere in 1999,
and I didn't go online till 2000.
The Gilbert Cell IC chips can make a nice synchrodyne, ie, direct
conversion receiver. I just NEVER get time for all this going back to
finish 1955 stuff which was interrupted by having to grow up, face the
world, get a job, raise kids, etc, whatever. Just because one cannot
take up where one left off at 1955 does not mean there isn't good
stuff to do now. Hell, I spend 3 days a week on a bicycle, and I don't
miss the soldrin' iron then. I'd rather have a resting HR of 48, keep
my docs happy, not take any pills, and enjoy the great outdoors and
speed and zing of cycling around than being couped up in the shed.
As for DAB, don't ask me about it because I haven't a clue exactly
what the transmitted wave looks like, have no clue how the transmitted
wave is constructed, and have no idea what happens to it when is
picked up by somebody's set. The boffins want us to buy, plug in, play
it, and not understand it, let alone add tubey stuff anywhere to make
it go better, or to build a digital radio with tubes. Probably that
would be as bad as tryna make a CD player using tubes only. You'd end
up with a thousand tubes, and because random failure rates are high
for tubes, the darn thing would fail regularly, besides, if each tube
consumes 2 Watts, then you have a room heater - not wanted in summer.
Tubed AM/FM tuners are not always reliable, like old tubed TV sets.
Patrick Turner.
> --
> Regards,
>
> John Byrns
>
> Surf my web pages at, http://fmamradios.com/- Hide quoted text -
>
> - Show quoted text -
> In article <fda19320-b0ca-40b9-89a2-547158659...@h10g2000pbi.googlegroups.com>,
switched tertiary on IFT 1. Most other radio makers steadfastly
avoided such delights and fed the buying public with half the audio BW
of Quad. Quad also got it right with traditional diode detector AF
detector, not too much Dn. But nearly all these old AM radios and
tuners need a ferrite rod antenna with short leads to pick up the
magnetic wave more than the electrostatic wave which comes to most old
sets to a high Z coil loosely magneticly coupled to tuned RF input
coil ahed of an RF amp or mixer.
Today's household wull be filled with fluroresant lamps each with tiny
SMPS plus all the other SMPS crap, plus linear PSU crap which all
manages to modulate station signals which arrive at the house with
100Hz wave, so the receivers pick up some radio stations which have
100Hz added modulation, and this hum can't be avoided when tuning.
So in a good AM tuner for best fidelity local pick up, you need a
ferrite rod antenna, a switched IFT 1 for wider AF BW, and so Quad had
some way to go to equal what I have. But in 1999 when I made my radio
from spare junk-box parts there was little radio interference and I
had two cascaded RF input coils with stagger tuning at the low end of
the band for better AF bw on stations which I like. But I had to
change later to a ferrite rod because of noise. The Quad also has
beautifully made tuning controls and various bands, where just have
one band. Quad's detector could have been better, but they'd have had
to use something similar to my cathode followers with ge diodes etc.
If someone were to spend a few days altering a Quad AM tuner to make
it better, I'm sure they could.
Because AM radio stations have crystal controlled oscillators set for
very specific frequencies, it would be nice to add a digital press
button frequency generator in receiver with setable F to tune instead
of having any oscillator. But you'd still have to tune the RF coil, so
as they are will do. I must put a tuning meter and better dial into my
set, and I have a suitable old FM tuner module which sounds well which
I could install in my set, because I often switch from AM to kitchen
to get the least boring program or best music. Then all AM and FM can
be in the one box which really needs its timber front altered and a
new longer slide type dial put in. Another fix-up job to attend to
during my retirement beginning on August 1 in a month.
Its all not really needed now because we have Digital Audio
Broadcasting, and a digital radio is cheap because all such gadgetty
stuff is made in Asia for 20c each. But speakers and audio amps in
little digital radios are awful, so one can focus on tube audio amps
and decent speakers. I might switch to stereo for the kitchen, so
speakers would be small, but like my old VAF bookshelfs for my TV set,
bokkshelfs can be good enough. The 1995 Metz TV I was given had The
World's Worst audio system which I removed entirely. Audio comes to
old solid state pioneer amp and VAF speakers from the back of the set-
top box. Good enough for TV sound which is often poor to begin with.
Maybe I get a 105cm LCD screen soon, but still have to do better with
sound. Like a lot of things, I don't have the time or inclination to
stay put mentally in 1955, or 1999, when other things have come along.
Well done AM is cabable of very good reception, but it all got dumbed
down with many stations put too close to allow wider AF. I've often
thought of using IF at 2MHz, and coils would be easier, and probably 3
IFTs would be needed. Even with Q = 100, pass band is 20kHz, so 10kHz
of audio.
In 1999 I tried first to make a DG Tucker style tubed synchrodyne, but
that proved much more difficult while offering poor performance
compared to a good superhet. The coils needed were beyound my time
allowances. There were no data sheets or how-to-do anywhere in 1999,
and I didn't go online till 2000.
The Gilbert Cell IC chips can make a nice synchrodyne, ie, direct
conversion receiver. I just NEVER get time for all this going back to
finish 1955 stuff which was interrupted by having to grow up, face the
world, get a job, raise kids, etc, whatever. Just because one cannot
take up where one left off at 1955 does not mean there isn't good
stuff to do now. Hell, I spend 3 days a week on a bicycle, and I don't
miss the soldrin' iron then. I'd rather have a resting HR of 48, keep
my docs happy, not take any pills, and enjoy the great outdoors and
speed and zing of cycling around than being couped up in the shed.
As for DAB, don't ask me about it because I haven't a clue exactly
what the transmitted wave looks like, have no clue how the transmitted
wave is constructed, and have no idea what happens to it when is
picked up by somebody's set. The boffins want us to buy, plug in, play
it, and not understand it, let alone add tubey stuff anywhere to make
it go better, or to build a digital radio with tubes. Probably that
would be as bad as tryna make a CD player using tubes only. You'd end
up with a thousand tubes, and because random failure rates are high
for tubes, the darn thing would fail regularly, besides, if each tube
consumes 2 Watts, then you have a room heater - not wanted in summer.
Tubed AM/FM tuners are not always reliable, like old tubed TV sets.
Patrick Turner.
> --
> Regards,
>
> John Byrns
>
> Surf my web pages at, http://fmamradios.com/- Hide quoted text -
>
> - Show quoted text -
John Byrns
Jul 1, 2012, 10:36:54 AM7/1/12
to
In article <72702337-ebe9-4afb...@nw7g2000pbb.googlegroups.com>,
circuit, in the day this same type of circuit was virtually universal in Hi-Fi
AM tuners! Is your point simply that QUAD's bandwidth choices were better than
the choices made by other manufacturers, if so this choice would seem to be
largely a matter of individual taste and reception conditions?
The detector issue is probably a little more complicated, some manufacturers
clearly screwed up, while others got it right like QUAD. I don't know what
might have accounted for this, but I suspect that it may be partly the result of
AM tuner specs being measured at 30% modulation. As a result IIRC signal
generators of the day only provided low distortion up to the 30% modulation
figure and generator distortion increased radically between there and 100%
modulation, if the generator would even do anywhere near 100% modulation. As a
result distortion measurements above the required 30% modulation figure were
likely as much a measure of the signal generator's distortion as they were of
the detectors distortion, so the designers may not have been able to tell if
their detector designs were working as they expected.
Patrick Turner <in...@turneraudio.com.au> wrote:
> On Jul 1, 2:37 am, John Byrns <byr...@sbcglobal.net> wrote:
> >
> > What is it about QUAD'a AM tuner that places it in the much coveted second
> > place
> > behind the Turner tuner? At first glance QUAD's AM design doesn't seem
> > much
> > different than dozens of other AM tuner designs, exactly what is it that
> > gives
> > it the edge over the other pretenders?
>
> Well, Quad managed to get the selectivity variation just right with
> switched tertiary on IFT 1. Most other radio makers steadfastly
> avoided such delights and fed the buying public with half the audio BW
> of Quad. Quad also got it right with traditional diode detector AF
> detector, not too much Dn.
It's not clear to me what was unique/better about QUAD's "selectivity variation"
> On Jul 1, 2:37 am, John Byrns <byr...@sbcglobal.net> wrote:
> >
> > What is it about QUAD'a AM tuner that places it in the much coveted second
> > place
> > behind the Turner tuner? At first glance QUAD's AM design doesn't seem
> > much
> > different than dozens of other AM tuner designs, exactly what is it that
> > gives
> > it the edge over the other pretenders?
>
> Well, Quad managed to get the selectivity variation just right with
> switched tertiary on IFT 1. Most other radio makers steadfastly
> avoided such delights and fed the buying public with half the audio BW
> of Quad. Quad also got it right with traditional diode detector AF
> detector, not too much Dn.
circuit, in the day this same type of circuit was virtually universal in Hi-Fi
AM tuners! Is your point simply that QUAD's bandwidth choices were better than
the choices made by other manufacturers, if so this choice would seem to be
largely a matter of individual taste and reception conditions?
The detector issue is probably a little more complicated, some manufacturers
clearly screwed up, while others got it right like QUAD. I don't know what
might have accounted for this, but I suspect that it may be partly the result of
AM tuner specs being measured at 30% modulation. As a result IIRC signal
generators of the day only provided low distortion up to the 30% modulation
figure and generator distortion increased radically between there and 100%
modulation, if the generator would even do anywhere near 100% modulation. As a
result distortion measurements above the required 30% modulation figure were
likely as much a measure of the signal generator's distortion as they were of
the detectors distortion, so the designers may not have been able to tell if
their detector designs were working as they expected.
John L Stewart
Jul 1, 2012, 11:24:50 AM7/1/12
to
Patrick Turner;959148 Wrote:
> On Jul 1, 2:37*am, John Byrns <byr...@sbcglobal.net> wrote:-
> > In article
> <fda19320-b0ca-40b9-89a2-547158659...@h10g2000pbi.googlegroups..com>,
> > *Patrick Turner <i...@turneraudio.com.au> wrote:
> >
> >
> >
> >
> >-
> > > The AM transmissions here from radio stations are pretty good, and
> > > probably with lower THD + IMD than in 1955, so recievers are the
> weak
> > > link. My own AM set has cascode vari-mu twin triode input, 6AN7
> mixer,
> > > then 6BX6 IF amp with unbypassed Rk for some local current FB and
> no
> > > AGC voltage is applied to it. 6BX6 or EF80 is a sharp cut off
> pentode
> > > with strong gm normally used for 4.5Mhz TV IF stages. Its operating
> > > with fixed bias. AGC voltage is generated by the detector CF stage,
> > > and it works on mixer just a bit, and on RF input tube. I have
> > > variable spacing between IFT1 like Hellicrafter used on their
> > > communications radios with 3 IFTs to vary the Q for the IF band. But
> I
> > > vary between flat top pass band and twin peak, so with two IFTs,
> and
> > > IFT2 with normal 9kHz pass band, I get 18kHz pass band and 9kHz of
> AF
> > > bandwidth. Some HF emphasis in R&C network boosts the AF BW to
> 10kHz.
> > > I hate to seem immodest, but my AM radio outperforms ALL other AM
> > > radios I have ever placed beside it to compare them, including
> Quad's
> > > AM tuner which I think is next best.-
> > > probably with lower THD + IMD than in 1955, so recievers are the
> weak
> > > link. My own AM set has cascode vari-mu twin triode input, 6AN7
> mixer,
> > > then 6BX6 IF amp with unbypassed Rk for some local current FB and
> no
> > > AGC voltage is applied to it. 6BX6 or EF80 is a sharp cut off
> pentode
> > > with strong gm normally used for 4.5Mhz TV IF stages. Its operating
> > > with fixed bias. AGC voltage is generated by the detector CF stage,
> > > and it works on mixer just a bit, and on RF input tube. I have
> > > variable spacing between IFT1 like Hellicrafter used on their
> > > communications radios with 3 IFTs to vary the Q for the IF band. But
> I
> > > vary between flat top pass band and twin peak, so with two IFTs,
> and
> > > IFT2 with normal 9kHz pass band, I get 18kHz pass band and 9kHz of
> AF
> > > bandwidth. Some HF emphasis in R&C network boosts the AF BW to
> 10kHz.
> > > I hate to seem immodest, but my AM radio outperforms ALL other AM
> > > radios I have ever placed beside it to compare them, including
> Quad's
> >
> > What is it about QUAD'a AM tuner that places it in the much coveted
> second place
> > behind the Turner tuner? *At first glance QUAD's AM design doesn't
> > What is it about QUAD'a AM tuner that places it in the much coveted
> second place
> seem much
> > different than dozens of other AM tuner designs, exactly what is it
> that gives
> > it the edge over the other pretenders?-
> > different than dozens of other AM tuner designs, exactly what is it
> that gives
> >
> > - Show quoted text --
The EICO HFT-94 has both IF1 & IF2 with selective BW. They claim overall
selectivity of 7 & 14 KHz. Audio spec is 20-9000 Hz WB & 20-5000 Hz NB
at 3 db down. Also using a CK885 Ge diode as the detector followed by a
10KHz whistle filter, then a FB pair with a 12AU7 to audio output.
The RF selectivity is good since it has an RF amp ahead of the
convertor.
I built two of these from the kits around 1960. One still sounds
excellant & runs most days while I'm in the workshop. The other is on
the shelf for now.
Hey Patrick, is your DAB (Digital Audio) by Ubiquity as here on the
regular FM band or is the transmission on the 'L' Band. Just curious. I
sometimes think I hear interference on one of the locals broadcasting
both analogue & digital. The tuner in that case is a Sansui TU-717
Google IBOC (In Band On Channel) for info on the Ibiquity System. One of
the companies I sold for, Aeroflex has a good AN on the system.
Cheers to all, John
+-------------------------------------------------------------------+
|Filename: 150px-Alfred_E__Neumann.jpe |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=290|
+-------------------------------------------------------------------+
--
John L Stewart
flipper
Jul 1, 2012, 10:37:25 PM7/1/12
to
On Sat, 30 Jun 2012 15:45:37 +1000, "Alex Pogossov"
<apog...@tpg.com.au> wrote:
>
>"flipper" <fli...@fish.net> wrote in message
>news:qhbsu7dtce6cv7uj5...@4ax.com...
>> On Thu, 28 Jun 2012 20:31:40 +1000, "Alex Pogossov"
>
>>>> With carrier at 640 KHz frequency deviation from idle to twice idle is
>>>> roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
>>>> .64022 MHz.
>>>
>>>It is not bad. I expected worse.
>>
>> Were you thinking maybe a CF would be needed?
>
>In a quality equipment -- yes, but not in your case when you knowingly allow
>for some FM.
Ahem. "Quality" is a subjective thing. Maybe change that to
<apog...@tpg.com.au> wrote:
>
>"flipper" <fli...@fish.net> wrote in message
>news:qhbsu7dtce6cv7uj5...@4ax.com...
>> On Thu, 28 Jun 2012 20:31:40 +1000, "Alex Pogossov"
>
>>>> With carrier at 640 KHz frequency deviation from idle to twice idle is
>>>> roughly 70 Hz. Actual measurement was .64015 MHz (resolution limit) to
>>>> .64022 MHz.
>>>
>>>It is not bad. I expected worse.
>>
>> Were you thinking maybe a CF would be needed?
>
>In a quality equipment -- yes, but not in your case when you knowingly allow
>for some FM.
"Professional Broadcast" equipment because I think its pretty dern
good 'quality' for 2 one buck tubes ;)
>>> Probably 6ME8 has less space charge
>>>coupling than a regular heptode.
>>
>> I suspect so and it's one of the things I think are 'different'
>> because of the beam deflection.
>
>In the deflection tube the electron beam is far away (relatively) from the
>deflection plates, not brushing them even when the deflection plates sit at
>positive bias. In a heptode the space charge is right in the vicinity of the
>G3, enveloping it and nearly touching it. Therefore the deflection tube
>space charge effects are smaller.
deflection plates.
>But... the deflection tube requires +/-30V for full beam cutoff/switchover,
>while a heptode probably would cutoff at -5V. Six times less. Now if you
>arrange a tap on the oscillator running at 30V at 1/6-th, and feed it to a
>heptode G3, the effect will be 6*6=36 times smaller. So you end up with the
>same FM normalised to an available oscillator voltage. A hetode will give
>low FM if coupled to a tap of the oscillator tank.
vicinity" heptode's space charge coupling is. I presume you're
implying it should be '6 times' as much based on the drive required
but are were sure that holds true when the mechanism, beam deflection,
is different? I mean, "same FM?"
Frankly, I have no idea how one calculates 'effective capacitance'
chance from space charge coupling.
Still, the point is taken. The existing grid takeoff was strictly for
amplitude and if I were to try using a heptode I'd likely be using the
coil cathode tap. Which, btw, if 5V would be enough for a heptode,
that is exactly what's at the coil tap right now.
>Because of huge deflection drive requirement, such tubes never gained
>popularity as generic mixers.
>
>>
>>>Still 70Hz is not acceptable for listening on a synchronous or SSB
>>>detector.
>>>In these cases parasitic FM index shall not exceed about 0.25, so with the
>>>lowest audio frequency of 50Hz, FM deviation shall not exceed 15Hz.
>>>
>>>But who would deliberately listening to your transmitter on an SSB radio?
>>>No
>>>one.
>>
>> Right. What I'd rather know is how much deviation is 'acceptable' for
>> the BCB.
>
>Up to 1kHz, I would guess.
flipper
Jul 1, 2012, 11:19:30 PM7/1/12
to
Gilbert Cell than you did with the discrete BJTs.
Except there's not much reason to do so since that's what beam
deflection tubes are for.
>> The 6DT6, yes, because it's simply a dual control pentode. Here's one http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
>> 'conventional' audio to G3 approach.
>
>Gee, that looks really nice and simple! Makes me wanna build one, I
>might have a few 6DT6 laying around.
and I think most are just tweaked derivations of Norm's 6888 version.
Linearity depends on the G3 curve.
If you like that one you might also get a kick out of this one
(completely different approach by Robert Weaver )
http://www.radiomuseum.org/forum/a_one_tube_controlled_carrier_am_transmitter.html
He gets by with self excite because it's crystal controlled.
>> The 6BN6 is a whole different thing: a gated beam discriminator and
>> limiter with the express purpose of rejecting AM.
>>
>> >I know you are hooked on using just one tube,
>>
>> Well, no. I was only 'hooked' on using one tube back when I was
>> experimenting with using one tube. That was 'the point' of it: to see
>> what one could do with one tube.
>
>And now you know, and not bad considering you chose a tube which can
>wear so many hats and does so many things, and which hardly anyone
>knows anything about.
The 'one tube' attempt was 3 years ago using a 6SC6 'phono oscillator'
topology.
>> But everything I've done since that experiment involves more than one
>> tube.
>>
>> What would be a more accurate characterization is I like the challenge
>> of getting the most'est from the least'est.
>
>OK, but to me the 6ME6 is a "leastest with mostest" type of tube.
then, because its gm is 4400.
>Enough about feeble tubes. In RF work, I like to see healthy Ia
>because capacitances anywhere tend to have low Z paths at RF.
of the tube.
>15-20dB FB is good, enough to do something worth doing, assuming the
>gear makes maybe 7% Dn at 90% mod without NFB.
>
>But I found that with NFB applied, the modulation bandwidth with
>undistorted wave at say 95% mod was reduced, and to be expected of
>course.
I'd only 'expect' that if the audio loop were bandwidth or slew rate
limited and neither are the case here.
flipper
Jul 1, 2012, 11:37:07 PM7/1/12
to
core so I don't have to screw with the misery of winding a gazillion
turns on a toroid
> You can have just one winding with CT to B+, and
>double tuning gang with frame and moving plates at 0V with 0.01uF caps
>from fixed plated to coil ends.
Dual gang tuning the plates looks simple enough till you then have to
add antenna tuning off the secondary.
>the thing is to allow the tubes to drive a nice high ohm RLa-a load,
>and have a secondary wound over the middle part of the winding, ie,
>the earthy part, and have this winding about 1/5 of the total P turns,
>so you get a 25:1 impedance match so antena C has hardly any effect on
>coil tuning. if the RLa-a in parallel with tube Ra was say 25k, then
>Zout from sec is 1k0, quite low enough to get over effects of C across
>OP terminals etc. probably an air cored coil is OK using 40mm PVC pipe
>former with cat 5 strands of solid wire for coil. But maybe best if
>you drill many holes as possible in PVC to reduce diaelectric losses.
>Layout and keeping coil away from other metal and keeping leads tidy
>assist in overall Q which need not be huge. Another way to get low Z
>output is to have a tap on one side of PP coil, at say 20% of turns so
>you can use 0.01 cap to take out SE signal. No sec winding, and you
>still get benefits of the Z transformation and stray output C from
>antenna won't worry the coil.
>Patrick Turner.
and space winding isn't all that easy. Maybe use Litz with insulation
providing a 'natural' spacing.
At any rate, what seemed 'simple' now has a raft of questions about
just how to accomplish it whereas with single ended I had a couple of
ready made parts that could be tried without adding risk.
With the 100 mW limit I certainly don't need 'more power' so the only
reason I can think of to use a balanced PP in the existing design
would be if the 'balance' significantly improved something else to
make up for the added complexity.
John L Stewart
Jul 1, 2012, 8:55:12 PM7/1/12
to
This link will take you to the App Note on IBOC I refered to earlier-
http://tinyurl.com/86zy4be
The Spec A in the article was developed by Will'Tek in Germany. In the
US it was sometimes sold under the Boonton name.
That part of the company was sold off to Aeroflex about two years ago.
A very nice portable SA to work with. I managed to sell quite a few.
Here is a link to iBiquity for the curious-
http://www.ibiquity.com/i/pdfs/Conversion_Requirements.pdf
Cheers, John
+-------------------------------------------------------------------+
John Byrns
Jul 4, 2012, 7:10:40 PM7/4/12
to
In article <jd22v7d8ori552bd3...@4ax.com>,
interesting radiomuseum articles I have come across. While this transmitter
isn't exactly my cup of tea, it is amazing what the author did with so little,
back to that later!
The use of screen modulation is a departure for these small transmitters as most
seem to use G3 modulation of either a heptode or a Dual Control Pentode, with a
few using Plate modulation as in the series modulation approach, and now the
beam deflection approach used in your unique "Beam Me Up Scotty" transmitter.
Of course screen modulation is nothing unique having been used in real AM
broadcast transmitters, especially those built by Continental, like the 316B 10
kW transmitters installed by CBS as backup transmitters at their O&O stations
around 1960 when they went to remote control for their transmitters.
The manual (23MB), with schematic, for a screen modulated Continental 1 kW MW-AM
broadcast transmitter can be found here:
http://louise.hallikainen.org/BH/uploads/ContinentalElectronics3141kWAMTx.pdf
There are at least two types of dynamic carrier control, the BBC has written
some good research reports on the subject. One type of DCC, as used in Weaver's
transmitter, reduces the amplitude of the carrier when the modulation level is
low. The second type does essentially the opposite and reduces the carrier
amplitude when the modulation level is high. I believe the second form of DCC,
as opposed to the form used by Weaver, affords greater operating efficiency and
economy as well as reducing the peak power that the transmitter must produce.
The second approach also offers a subjectively lower subjective noise level IIRC.
It is fascinating how Weaver achieves DCC essentially by reducing the bias on
the modulator tube causing grid rectification of the audio and consequent change
in the carrier amplitude. This actually saves a few components that would
otherwise be required to provide normal modulator bias.
While the BBC used DCC for economic reasons, to reduce power and equipment
costs, Weaver's main reason for using DCC seems to be as a poor man's audio
processor. It isn't clear exactly how much compression Weaver's scheme
provides, perhaps as little as 4 dB? It would have been helpful if Weaver had
included trapezoidal oscilloscope patterns for his 300mV p-p and 2.2V p-p test
cases using tone modulation. This would allow judging the actual audio
compression effect, which is produced by the receiver's AGC system.
It would have also have been nice if Weaver had tried some of the pentodes he
mentioned in place of the tetrode he used, so we could have a better idea if it
is true that tetrodes provide more linear screen modulation than pentodes do, as
Weaver reports from the technical literature. References to the relevant
technical literature would also help here, I am particularly curious if this
effect applies to these small receiving tubes, or if it is mainly applicable to
larger transmitting tubes?
Previously I have only read the radiomuseum mini transmitter pages that I have
found through Google, or that people have linked to. Upon looking through their
project index I was surprised, and greatly pleased, to find that Weaver also has
a design for a three tube C-Quam AM Stereo transmitter, although I haven't had a
chance to read the rather lengthy article yet. I was acquainted with the
inventor of the C-Quam AM Stereo system so I have a special interest in this
transmitter design. In the past I have given some thought to designs for tube
based C-Quam receivers. I spent a few minutes Googling a couple of aspects of
AM stereo and was sadly surprised to see that Leonard Kahn, the designer of a
competing AM Stereo System, the Compatible Sideband System, passed away about a
month ago.
--
Regards,
John Byrns
Surf my web pages at, http://fmamradios.com/
flipper <fli...@fish.net> wrote:
> On Fri, 29 Jun 2012 23:49:00 -0700 (PDT), Patrick Turner
> <in...@turneraudio.com.au> wrote:
>
> >> The 6DT6, yes, because it's simply a dual control pentode. Here's one
> >> http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
> >> 'conventional' audio to G3 approach.
> >
> >Gee, that looks really nice and simple! Makes me wanna build one, I
> >might have a few 6DT6 laying around.
>
> There's a version of it for just about every dual control pentode made
> and I think most are just tweaked derivations of Norm's 6888 version.
> Linearity depends on the G3 curve.
>
> If you like that one you might also get a kick out of this one
> (completely different approach by Robert Weaver )
> http://www.radiomuseum.org/forum/a_one_tube_controlled_carrier_am_transmitter.
> html
>
> He gets by with self excite because it's crystal controlled.
I hadn't seen this radiomueseum article before, it is one of the most
> On Fri, 29 Jun 2012 23:49:00 -0700 (PDT), Patrick Turner
> <in...@turneraudio.com.au> wrote:
>
> >> The 6DT6, yes, because it's simply a dual control pentode. Here's one
> >> http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
> >> 'conventional' audio to G3 approach.
> >
> >Gee, that looks really nice and simple! Makes me wanna build one, I
> >might have a few 6DT6 laying around.
>
> There's a version of it for just about every dual control pentode made
> and I think most are just tweaked derivations of Norm's 6888 version.
> Linearity depends on the G3 curve.
>
> If you like that one you might also get a kick out of this one
> (completely different approach by Robert Weaver )
> http://www.radiomuseum.org/forum/a_one_tube_controlled_carrier_am_transmitter.
> html
>
> He gets by with self excite because it's crystal controlled.
interesting radiomuseum articles I have come across. While this transmitter
isn't exactly my cup of tea, it is amazing what the author did with so little,
back to that later!
The use of screen modulation is a departure for these small transmitters as most
seem to use G3 modulation of either a heptode or a Dual Control Pentode, with a
few using Plate modulation as in the series modulation approach, and now the
beam deflection approach used in your unique "Beam Me Up Scotty" transmitter.
Of course screen modulation is nothing unique having been used in real AM
broadcast transmitters, especially those built by Continental, like the 316B 10
kW transmitters installed by CBS as backup transmitters at their O&O stations
around 1960 when they went to remote control for their transmitters.
The manual (23MB), with schematic, for a screen modulated Continental 1 kW MW-AM
broadcast transmitter can be found here:
http://louise.hallikainen.org/BH/uploads/ContinentalElectronics3141kWAMTx.pdf
There are at least two types of dynamic carrier control, the BBC has written
some good research reports on the subject. One type of DCC, as used in Weaver's
transmitter, reduces the amplitude of the carrier when the modulation level is
low. The second type does essentially the opposite and reduces the carrier
amplitude when the modulation level is high. I believe the second form of DCC,
as opposed to the form used by Weaver, affords greater operating efficiency and
economy as well as reducing the peak power that the transmitter must produce.
The second approach also offers a subjectively lower subjective noise level IIRC.
It is fascinating how Weaver achieves DCC essentially by reducing the bias on
the modulator tube causing grid rectification of the audio and consequent change
in the carrier amplitude. This actually saves a few components that would
otherwise be required to provide normal modulator bias.
While the BBC used DCC for economic reasons, to reduce power and equipment
costs, Weaver's main reason for using DCC seems to be as a poor man's audio
processor. It isn't clear exactly how much compression Weaver's scheme
provides, perhaps as little as 4 dB? It would have been helpful if Weaver had
included trapezoidal oscilloscope patterns for his 300mV p-p and 2.2V p-p test
cases using tone modulation. This would allow judging the actual audio
compression effect, which is produced by the receiver's AGC system.
It would have also have been nice if Weaver had tried some of the pentodes he
mentioned in place of the tetrode he used, so we could have a better idea if it
is true that tetrodes provide more linear screen modulation than pentodes do, as
Weaver reports from the technical literature. References to the relevant
technical literature would also help here, I am particularly curious if this
effect applies to these small receiving tubes, or if it is mainly applicable to
larger transmitting tubes?
Previously I have only read the radiomuseum mini transmitter pages that I have
found through Google, or that people have linked to. Upon looking through their
project index I was surprised, and greatly pleased, to find that Weaver also has
a design for a three tube C-Quam AM Stereo transmitter, although I haven't had a
chance to read the rather lengthy article yet. I was acquainted with the
inventor of the C-Quam AM Stereo system so I have a special interest in this
transmitter design. In the past I have given some thought to designs for tube
based C-Quam receivers. I spent a few minutes Googling a couple of aspects of
AM stereo and was sadly surprised to see that Leonard Kahn, the designer of a
competing AM Stereo System, the Compatible Sideband System, passed away about a
month ago.
--
Regards,
John Byrns
Surf my web pages at, http://fmamradios.com/
flipper
Jul 5, 2012, 1:02:23 AM7/5/12
to
On Wed, 04 Jul 2012 18:10:40 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
>In article <jd22v7d8ori552bd3...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Fri, 29 Jun 2012 23:49:00 -0700 (PDT), Patrick Turner
>> <in...@turneraudio.com.au> wrote:
>>
>> >> The 6DT6, yes, because it's simply a dual control pentode. Here's one
>> >> http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
>> >> 'conventional' audio to G3 approach.
>> >
>> >Gee, that looks really nice and simple! Makes me wanna build one, I
>> >might have a few 6DT6 laying around.
>>
>> There's a version of it for just about every dual control pentode made
>> and I think most are just tweaked derivations of Norm's 6888 version.
>> Linearity depends on the G3 curve.
>>
>> If you like that one you might also get a kick out of this one
>> (completely different approach by Robert Weaver )
>> http://www.radiomuseum.org/forum/a_one_tube_controlled_carrier_am_transmitter.
>> html
>>
>> He gets by with self excite because it's crystal controlled.
>
>
>I hadn't seen this radiomueseum article before, it is one of the most
>interesting radiomuseum articles I have come across. While this transmitter
>isn't exactly my cup of tea, it is amazing what the author did with so little,
>back to that later!
Yes, Robert Weaver is a font of cleverness but if that one amazes you
then his single tube 6ME8 double reflex superhet will blow your mind.
http://electronbunker.ca/OneTubeSuper.html
Talk about making a tube work for it's keep.
Btw, this is the original carrier controlled transmitter he mentions
in the radiomuseum article. http://electronbunker.ca/OneTubeXMTR.html
>The use of screen modulation is a departure for these small transmitters as most
>seem to use G3 modulation of either a heptode or a Dual Control Pentode, with a
>few using Plate modulation as in the series modulation approach, and now the
>beam deflection approach used in your unique "Beam Me Up Scotty" transmitter.
>Of course screen modulation is nothing unique having been used in real AM
>broadcast transmitters, especially those built by Continental, like the 316B 10
>kW transmitters installed by CBS as backup transmitters at their O&O stations
>around 1960 when they went to remote control for their transmitters.
>
>The manual (23MB), with schematic, for a screen modulated Continental 1 kW MW-AM
>broadcast transmitter can be found here:
>http://louise.hallikainen.org/BH/uploads/ContinentalElectronics3141kWAMTx.pdf
>
>There are at least two types of dynamic carrier control, the BBC has written
>some good research reports on the subject. One type of DCC, as used in Weaver's
>transmitter, reduces the amplitude of the carrier when the modulation level is
>low. The second type does essentially the opposite and reduces the carrier
>amplitude when the modulation level is high. I believe the second form of DCC,
>as opposed to the form used by Weaver, affords greater operating efficiency and
>economy as well as reducing the peak power that the transmitter must produce.
>The second approach also offers a subjectively lower subjective noise level IIRC.
I'm not sure of the history but think the modern trend is a
combination of both, lowering carrier 'in the middle' and raising it
on the ends. An example:
http://people.wallawalla.edu/~rob.frohne/Dynamic%20Carrier%20Control/DCC1.html
As I think of it, though, I don't see how the 'second form' you
mention would increase efficiency much since most audio is not at
'high mod' levels.
>It is fascinating how Weaver achieves DCC essentially by reducing the bias on
>the modulator tube causing grid rectification of the audio and consequent change
>in the carrier amplitude. This actually saves a few components that would
>otherwise be required to provide normal modulator bias.
>
>While the BBC used DCC for economic reasons, to reduce power and equipment
>costs, Weaver's main reason for using DCC seems to be as a poor man's audio
>processor. It isn't clear exactly how much compression Weaver's scheme
>provides, perhaps as little as 4 dB? It would have been helpful if Weaver had
>included trapezoidal oscilloscope patterns for his 300mV p-p and 2.2V p-p test
>cases using tone modulation. This would allow judging the actual audio
>compression effect, which is produced by the receiver's AGC system.
Yes, his purpose was as 'compressor', despite it appearing to be the
opposite. Receiver AGC does the other half of the work.
I'm not sure if he had a numerical target in mind or simply tweaked
for best 'sound', which I imagine might leave an issue about the AGC
characteristics of the particular receiver.
>It would have also have been nice if Weaver had tried some of the pentodes he
>mentioned in place of the tetrode he used, so we could have a better idea if it
>is true that tetrodes provide more linear screen modulation than pentodes do, as
>Weaver reports from the technical literature. References to the relevant
>technical literature would also help here, I am particularly curious if this
>effect applies to these small receiving tubes, or if it is mainly applicable to
>larger transmitting tubes?
He also posts on ARF and there's a thread there where he did try some
of those, including a 6V6.
http://www.antiqueradios.com/forums/viewtopic.php?f=12&t=170605&start=40
There was also some discussion about whether the 6CQ8 is really a
tetrode or a 'beam' pentode.
>Previously I have only read the radiomuseum mini transmitter pages that I have
>found through Google, or that people have linked to. Upon looking through their
>project index I was surprised, and greatly pleased, to find that Weaver also has
>a design for a three tube C-Quam AM Stereo transmitter, although I haven't had a
>chance to read the rather lengthy article yet. I was acquainted with the
>inventor of the C-Quam AM Stereo system so I have a special interest in this
>transmitter design. In the past I have given some thought to designs for tube
>based C-Quam receivers. I spent a few minutes Googling a couple of aspects of
>AM stereo and was sadly surprised to see that Leonard Kahn, the designer of a
>competing AM Stereo System, the Compatible Sideband System, passed away about a
>month ago.
Yes, he posted that one on ARF too. Oh, I see in the radiomuseum
article he brings up the 'beam pentode' tetrode matter that was
discussed on ARF.
Btw, his cabinet work is every bit as good as his tube designs.
wrote:
>In article <jd22v7d8ori552bd3...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Fri, 29 Jun 2012 23:49:00 -0700 (PDT), Patrick Turner
>> <in...@turneraudio.com.au> wrote:
>>
>> >> The 6DT6, yes, because it's simply a dual control pentode. Here's one
>> >> http://amradio.freeiz.com/transmitter/6bk76dt6.jpg using the
>> >> 'conventional' audio to G3 approach.
>> >
>> >Gee, that looks really nice and simple! Makes me wanna build one, I
>> >might have a few 6DT6 laying around.
>>
>> There's a version of it for just about every dual control pentode made
>> and I think most are just tweaked derivations of Norm's 6888 version.
>> Linearity depends on the G3 curve.
>>
>> If you like that one you might also get a kick out of this one
>> (completely different approach by Robert Weaver )
>> http://www.radiomuseum.org/forum/a_one_tube_controlled_carrier_am_transmitter.
>> html
>>
>> He gets by with self excite because it's crystal controlled.
>
>
>I hadn't seen this radiomueseum article before, it is one of the most
>interesting radiomuseum articles I have come across. While this transmitter
>isn't exactly my cup of tea, it is amazing what the author did with so little,
>back to that later!
then his single tube 6ME8 double reflex superhet will blow your mind.
http://electronbunker.ca/OneTubeSuper.html
Talk about making a tube work for it's keep.
Btw, this is the original carrier controlled transmitter he mentions
in the radiomuseum article. http://electronbunker.ca/OneTubeXMTR.html
>The use of screen modulation is a departure for these small transmitters as most
>seem to use G3 modulation of either a heptode or a Dual Control Pentode, with a
>few using Plate modulation as in the series modulation approach, and now the
>beam deflection approach used in your unique "Beam Me Up Scotty" transmitter.
>Of course screen modulation is nothing unique having been used in real AM
>broadcast transmitters, especially those built by Continental, like the 316B 10
>kW transmitters installed by CBS as backup transmitters at their O&O stations
>around 1960 when they went to remote control for their transmitters.
>
>The manual (23MB), with schematic, for a screen modulated Continental 1 kW MW-AM
>broadcast transmitter can be found here:
>http://louise.hallikainen.org/BH/uploads/ContinentalElectronics3141kWAMTx.pdf
>
>There are at least two types of dynamic carrier control, the BBC has written
>some good research reports on the subject. One type of DCC, as used in Weaver's
>transmitter, reduces the amplitude of the carrier when the modulation level is
>low. The second type does essentially the opposite and reduces the carrier
>amplitude when the modulation level is high. I believe the second form of DCC,
>as opposed to the form used by Weaver, affords greater operating efficiency and
>economy as well as reducing the peak power that the transmitter must produce.
>The second approach also offers a subjectively lower subjective noise level IIRC.
combination of both, lowering carrier 'in the middle' and raising it
on the ends. An example:
http://people.wallawalla.edu/~rob.frohne/Dynamic%20Carrier%20Control/DCC1.html
As I think of it, though, I don't see how the 'second form' you
mention would increase efficiency much since most audio is not at
'high mod' levels.
>It is fascinating how Weaver achieves DCC essentially by reducing the bias on
>the modulator tube causing grid rectification of the audio and consequent change
>in the carrier amplitude. This actually saves a few components that would
>otherwise be required to provide normal modulator bias.
>
>While the BBC used DCC for economic reasons, to reduce power and equipment
>costs, Weaver's main reason for using DCC seems to be as a poor man's audio
>processor. It isn't clear exactly how much compression Weaver's scheme
>provides, perhaps as little as 4 dB? It would have been helpful if Weaver had
>included trapezoidal oscilloscope patterns for his 300mV p-p and 2.2V p-p test
>cases using tone modulation. This would allow judging the actual audio
>compression effect, which is produced by the receiver's AGC system.
opposite. Receiver AGC does the other half of the work.
I'm not sure if he had a numerical target in mind or simply tweaked
for best 'sound', which I imagine might leave an issue about the AGC
characteristics of the particular receiver.
>It would have also have been nice if Weaver had tried some of the pentodes he
>mentioned in place of the tetrode he used, so we could have a better idea if it
>is true that tetrodes provide more linear screen modulation than pentodes do, as
>Weaver reports from the technical literature. References to the relevant
>technical literature would also help here, I am particularly curious if this
>effect applies to these small receiving tubes, or if it is mainly applicable to
>larger transmitting tubes?
of those, including a 6V6.
http://www.antiqueradios.com/forums/viewtopic.php?f=12&t=170605&start=40
There was also some discussion about whether the 6CQ8 is really a
tetrode or a 'beam' pentode.
>Previously I have only read the radiomuseum mini transmitter pages that I have
>found through Google, or that people have linked to. Upon looking through their
>project index I was surprised, and greatly pleased, to find that Weaver also has
>a design for a three tube C-Quam AM Stereo transmitter, although I haven't had a
>chance to read the rather lengthy article yet. I was acquainted with the
>inventor of the C-Quam AM Stereo system so I have a special interest in this
>transmitter design. In the past I have given some thought to designs for tube
>based C-Quam receivers. I spent a few minutes Googling a couple of aspects of
>AM stereo and was sadly surprised to see that Leonard Kahn, the designer of a
>competing AM Stereo System, the Compatible Sideband System, passed away about a
>month ago.
article he brings up the 'beam pentode' tetrode matter that was
discussed on ARF.
Btw, his cabinet work is every bit as good as his tube designs.
flipper
Jul 5, 2012, 5:42:29 AM7/5/12
to
On Tue, 26 Jun 2012 22:29:35 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
>
dirty heptode/DC pentode (same pinout so easy to try both)
'substitute' for the 6ME8 to test the theory they should behave
similarly.
Abstract summary is they didn't.
On first power up a 90% mod waveform looked a lot like a mountain with
a huge deep crater where the peak should be and after twiddling around
a bit it occurred to me, uh oh, G3 is relative to cathode, and that's
going up and down with audio, so either the RF amplitude, or 'center
point', needed to swing from full on to cutoff changes. That doesn't
happen with the 6ME8 because the deflection plates are relative to
each other and any change on the cathode is either blocked by the
screen or goes equally to both.
I tried connecting the G3 grid leak to cathode and the result looked
'almost' right but I think it was still distorted. So maybe there's a
way around it but they sure aren't a simple 'plop in' kind of thing.
wrote:
>
>Are you trying to play with my mind? I was the first to use the term "dual
>control pentode in this thread when I made the statement "Applying the
>modulation to G1 doesn't take advantage of the beam deflection capabilities of
>the tube and instead uses it in a way that a more ordinary tube, like a dual
>control pentode, could serve." when I meant to say heptode. I'm the one that
>converted it!
Setting side who's mind is going... hehe. I threw together a quick and
>control pentode in this thread when I made the statement "Applying the
>modulation to G1 doesn't take advantage of the beam deflection capabilities of
>the tube and instead uses it in a way that a more ordinary tube, like a dual
>control pentode, could serve." when I meant to say heptode. I'm the one that
>converted it!
dirty heptode/DC pentode (same pinout so easy to try both)
'substitute' for the 6ME8 to test the theory they should behave
similarly.
Abstract summary is they didn't.
On first power up a 90% mod waveform looked a lot like a mountain with
a huge deep crater where the peak should be and after twiddling around
a bit it occurred to me, uh oh, G3 is relative to cathode, and that's
going up and down with audio, so either the RF amplitude, or 'center
point', needed to swing from full on to cutoff changes. That doesn't
happen with the 6ME8 because the deflection plates are relative to
each other and any change on the cathode is either blocked by the
screen or goes equally to both.
I tried connecting the G3 grid leak to cathode and the result looked
'almost' right but I think it was still distorted. So maybe there's a
way around it but they sure aren't a simple 'plop in' kind of thing.
John Byrns
Jul 5, 2012, 10:50:23 PM7/5/12
to
In article <ei3av79cq9u583rue...@4ax.com>,
It's too late to blow my mind with that, you already sent us there last year so
my mind is already blown, can't blow it twice.
> Btw, this is the original carrier controlled transmitter he mentions
> in the radiomuseum article. http://electronbunker.ca/OneTubeXMTR.html
Yes, I had found that myself after I saw his mention of it.
Yes, I have seen that one, there is a BBC paper here:
http://downloads.bbc.co.uk/rd/pubs/reports/1994-01.pdf
This paper has a bibliography which lists a number of earlier papers, including
several BBC papers from about 10 years earlier which is where I think I first
learned of these techniques. I have mostly forgotten how this stuff works and
will have to dig out some of the references and read them through.
> As I think of it, though, I don't see how the 'second form' you
> mention would increase efficiency much since most audio is not at
> 'high mod' levels.
I don't know about that, I remember back in the early 1960s an associate of mine
worked at a station where he proudly pointed out that the modulation monitor
meter rarely dropped off the 100% mark. Maybe the second scheme I described was
just a harebrained scheme I thought up years ago and never really existed
anywhere else. However I would think it could work for stations with very dense
continuous modulation like a rock music station. With this second scheme you
could potentially remove the peak tube from a Doherty style transmitter, saving
some filament power, and perhaps change the pulley ratio on the cooling blower
reducing the power consumed by the blower. Again I will have to read through
some of the references and refresh my memory on all this.
> >It is fascinating how Weaver achieves DCC essentially by reducing the bias
> >on
> >the modulator tube causing grid rectification of the audio and consequent
> >change
> >in the carrier amplitude. This actually saves a few components that would
> >otherwise be required to provide normal modulator bias.
> >
> >While the BBC used DCC for economic reasons, to reduce power and equipment
> >costs, Weaver's main reason for using DCC seems to be as a poor man's audio
> >processor. It isn't clear exactly how much compression Weaver's scheme
> >provides, perhaps as little as 4 dB? It would have been helpful if Weaver
> >had
> >included trapezoidal oscilloscope patterns for his 300mV p-p and 2.2V p-p
> >test
> >cases using tone modulation. This would allow judging the actual audio
> >compression effect, which is produced by the receiver's AGC system.
>
> Yes, his purpose was as 'compressor', despite it appearing to be the
> opposite. Receiver AGC does the other half of the work.
>
> I'm not sure if he had a numerical target in mind or simply tweaked
> for best 'sound', which I imagine might leave an issue about the AGC
> characteristics of the particular receiver.
I got the impression that he was constrained by the circuit an tweaked it for
the best performance he could get out of it. It looks like there might be a few
things he could do to further improve the performance, but the ones that come to
my mind all complicate the circuit. A direct coupled cathode follower between
the modulator and the modulated screen grid might extend the performance range.
IIRC one of the BBC papers, maybe several, delves into the AGC compatibility
issues.
> >It would have also have been nice if Weaver had tried some of the pentodes
> >he
> >mentioned in place of the tetrode he used, so we could have a better idea if
> >it
> >is true that tetrodes provide more linear screen modulation than pentodes
> >do, as
> >Weaver reports from the technical literature. References to the relevant
> >technical literature would also help here, I am particularly curious if this
> >effect applies to these small receiving tubes, or if it is mainly applicable
> >to
> >larger transmitting tubes?
>
> He also posts on ARF and there's a thread there where he did try some
> of those, including a 6V6.
>
> http://www.antiqueradios.com/forums/viewtopic.php?f=12&t=170605&start=40
>
> There was also some discussion about whether the 6CQ8 is really a
> tetrode or a 'beam' pentode.
I saw that on one of his radiomuseum pages.
> >Previously I have only read the radiomuseum mini transmitter pages that I
> >have
> >found through Google, or that people have linked to. Upon looking through
> >their
> >project index I was surprised, and greatly pleased, to find that Weaver also
> >has
> >a design for a three tube C-Quam AM Stereo transmitter, although I haven't
> >had a
> >chance to read the rather lengthy article yet. I was acquainted with the
> >inventor of the C-Quam AM Stereo system so I have a special interest in this
> >transmitter design. In the past I have given some thought to designs for
> >tube
> >based C-Quam receivers. I spent a few minutes Googling a couple of aspects
> >of
> >AM stereo and was sadly surprised to see that Leonard Kahn, the designer of
> >a
> >competing AM Stereo System, the Compatible Sideband System, passed away
> >about a
> >month ago.
>
> Yes, he posted that one on ARF too.
Not sure what you are talking about here, what did he post on "ARF too"?
> Oh, I see in the radiomuseum
> article he brings up the 'beam pentode' tetrode matter that was
> discussed on ARF.
Did he say anything more about it on "ARF"?
> Btw, his cabinet work is every bit as good as his tube designs.
It looks that way.
my mind is already blown, can't blow it twice.
> Btw, this is the original carrier controlled transmitter he mentions
> in the radiomuseum article. http://electronbunker.ca/OneTubeXMTR.html
http://downloads.bbc.co.uk/rd/pubs/reports/1994-01.pdf
This paper has a bibliography which lists a number of earlier papers, including
several BBC papers from about 10 years earlier which is where I think I first
learned of these techniques. I have mostly forgotten how this stuff works and
will have to dig out some of the references and read them through.
> As I think of it, though, I don't see how the 'second form' you
> mention would increase efficiency much since most audio is not at
> 'high mod' levels.
worked at a station where he proudly pointed out that the modulation monitor
meter rarely dropped off the 100% mark. Maybe the second scheme I described was
just a harebrained scheme I thought up years ago and never really existed
anywhere else. However I would think it could work for stations with very dense
continuous modulation like a rock music station. With this second scheme you
could potentially remove the peak tube from a Doherty style transmitter, saving
some filament power, and perhaps change the pulley ratio on the cooling blower
reducing the power consumed by the blower. Again I will have to read through
some of the references and refresh my memory on all this.
> >It is fascinating how Weaver achieves DCC essentially by reducing the bias
> >on
> >the modulator tube causing grid rectification of the audio and consequent
> >change
> >in the carrier amplitude. This actually saves a few components that would
> >otherwise be required to provide normal modulator bias.
> >
> >While the BBC used DCC for economic reasons, to reduce power and equipment
> >costs, Weaver's main reason for using DCC seems to be as a poor man's audio
> >processor. It isn't clear exactly how much compression Weaver's scheme
> >provides, perhaps as little as 4 dB? It would have been helpful if Weaver
> >had
> >included trapezoidal oscilloscope patterns for his 300mV p-p and 2.2V p-p
> >test
> >cases using tone modulation. This would allow judging the actual audio
> >compression effect, which is produced by the receiver's AGC system.
>
> Yes, his purpose was as 'compressor', despite it appearing to be the
> opposite. Receiver AGC does the other half of the work.
>
> I'm not sure if he had a numerical target in mind or simply tweaked
> for best 'sound', which I imagine might leave an issue about the AGC
> characteristics of the particular receiver.
the best performance he could get out of it. It looks like there might be a few
things he could do to further improve the performance, but the ones that come to
my mind all complicate the circuit. A direct coupled cathode follower between
the modulator and the modulated screen grid might extend the performance range.
IIRC one of the BBC papers, maybe several, delves into the AGC compatibility
issues.
> >It would have also have been nice if Weaver had tried some of the pentodes
> >he
> >mentioned in place of the tetrode he used, so we could have a better idea if
> >it
> >is true that tetrodes provide more linear screen modulation than pentodes
> >do, as
> >Weaver reports from the technical literature. References to the relevant
> >technical literature would also help here, I am particularly curious if this
> >effect applies to these small receiving tubes, or if it is mainly applicable
> >to
> >larger transmitting tubes?
>
> He also posts on ARF and there's a thread there where he did try some
> of those, including a 6V6.
>
> http://www.antiqueradios.com/forums/viewtopic.php?f=12&t=170605&start=40
>
> There was also some discussion about whether the 6CQ8 is really a
> tetrode or a 'beam' pentode.
> >Previously I have only read the radiomuseum mini transmitter pages that I
> >have
> >found through Google, or that people have linked to. Upon looking through
> >their
> >project index I was surprised, and greatly pleased, to find that Weaver also
> >has
> >a design for a three tube C-Quam AM Stereo transmitter, although I haven't
> >had a
> >chance to read the rather lengthy article yet. I was acquainted with the
> >inventor of the C-Quam AM Stereo system so I have a special interest in this
> >transmitter design. In the past I have given some thought to designs for
> >tube
> >based C-Quam receivers. I spent a few minutes Googling a couple of aspects
> >of
> >AM stereo and was sadly surprised to see that Leonard Kahn, the designer of
> >a
> >competing AM Stereo System, the Compatible Sideband System, passed away
> >about a
> >month ago.
>
> Yes, he posted that one on ARF too.
> Oh, I see in the radiomuseum
> article he brings up the 'beam pentode' tetrode matter that was
> discussed on ARF.
> Btw, his cabinet work is every bit as good as his tube designs.
John Byrns
Jul 5, 2012, 10:52:00 PM7/5/12
to
In article <remav7pebvlvouhcd...@4ax.com>,
you are saying the 6ME8 behaved differently than the heptode and that the
heptode behaved differently than the dual control pentode?
> On first power up a 90% mod waveform looked a lot like a mountain with
> a huge deep crater where the peak should be and after twiddling around
> a bit it occurred to me, uh oh, G3 is relative to cathode, and that's
> going up and down with audio, so either the RF amplitude, or 'center
> point', needed to swing from full on to cutoff changes.
By the context you are clearly talking about either the heptode or the dual
control pentode here, did both exhibit this mountain crater effect? Why is
there audio on the cathode? I assume we are talking about a self excited
circuit so that there would be a lot of RF on the cathode, but audio?
Is the plate loading set correctly? Incorrect plate loading would cause an
effect similar to what you are describing.
> That doesn't
> happen with the 6ME8 because the deflection plates are relative to
> each other and any change on the cathode is either blocked by the
> screen or goes equally to both.
>
> I tried connecting the G3 grid leak to cathode and the result looked
> 'almost' right but I think it was still distorted. So maybe there's a
> way around it but they sure aren't a simple 'plop in' kind of thing.
I don't follow your reference to "a simple 'plop in' kind of thing", what do you
mean?
flipper <fli...@fish.net> wrote:
> On Tue, 26 Jun 2012 22:29:35 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >
> >Are you trying to play with my mind? I was the first to use the term "dual
> >control pentode in this thread when I made the statement "Applying the
> >modulation to G1 doesn't take advantage of the beam deflection capabilities
> >of
> >the tube and instead uses it in a way that a more ordinary tube, like a dual
> >control pentode, could serve." when I meant to say heptode. I'm the one
> >that
> >converted it!
>
> Setting side who's mind is going... hehe. I threw together a quick and
> dirty heptode/DC pentode (same pinout so easy to try both)
> 'substitute' for the 6ME8 to test the theory they should behave
> similarly.
>
> Abstract summary is they didn't.
Sounds reasonable, I would be surprised if they did behave the same. I take it
> On Tue, 26 Jun 2012 22:29:35 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >
> >Are you trying to play with my mind? I was the first to use the term "dual
> >control pentode in this thread when I made the statement "Applying the
> >modulation to G1 doesn't take advantage of the beam deflection capabilities
> >of
> >the tube and instead uses it in a way that a more ordinary tube, like a dual
> >control pentode, could serve." when I meant to say heptode. I'm the one
> >that
> >converted it!
>
> Setting side who's mind is going... hehe. I threw together a quick and
> dirty heptode/DC pentode (same pinout so easy to try both)
> 'substitute' for the 6ME8 to test the theory they should behave
> similarly.
>
> Abstract summary is they didn't.
you are saying the 6ME8 behaved differently than the heptode and that the
heptode behaved differently than the dual control pentode?
> On first power up a 90% mod waveform looked a lot like a mountain with
> a huge deep crater where the peak should be and after twiddling around
> a bit it occurred to me, uh oh, G3 is relative to cathode, and that's
> going up and down with audio, so either the RF amplitude, or 'center
> point', needed to swing from full on to cutoff changes.
control pentode here, did both exhibit this mountain crater effect? Why is
there audio on the cathode? I assume we are talking about a self excited
circuit so that there would be a lot of RF on the cathode, but audio?
Is the plate loading set correctly? Incorrect plate loading would cause an
effect similar to what you are describing.
> That doesn't
> happen with the 6ME8 because the deflection plates are relative to
> each other and any change on the cathode is either blocked by the
> screen or goes equally to both.
>
> I tried connecting the G3 grid leak to cathode and the result looked
> 'almost' right but I think it was still distorted. So maybe there's a
> way around it but they sure aren't a simple 'plop in' kind of thing.
mean?
flipper
Jul 5, 2012, 11:35:10 PM7/5/12
to
On Thu, 05 Jul 2012 21:52:00 -0500, John Byrns <byr...@sbcglobal.net>
>> On first power up a 90% mod waveform looked a lot like a mountain with
>> a huge deep crater where the peak should be and after twiddling around
>> a bit it occurred to me, uh oh, G3 is relative to cathode, and that's
>> going up and down with audio, so either the RF amplitude, or 'center
>> point', needed to swing from full on to cutoff changes.
>
>By the context you are clearly talking about either the heptode or the dual
>control pentode here, did both exhibit this mountain crater effect?
Yes.
>Why is
>there audio on the cathode? I assume we are talking about a self excited
>circuit so that there would be a lot of RF on the cathode, but audio?
No, I'm not talking about audio to G3 with RF to G1, self excite or
not. That's the dime a dozen, pick your tube and there's a ready made
schematic already there to download, way of doing it.
Maybe I misunderstood your "like a (heptode) dual control pentode
could serve" but I took it as suggesting audio to G1 with RF to G3,
'like done with the 6ME8', should produce results 'like' the 6ME8. As
in 'why use a 6ME8 when a plain ole heptode would do the same thing'?
And that's the context all my discussions on them has been in.
So I disconnected the 6ME8, ran RF to heptode/DC pentode G3 with audio
to G1 which, of course, puts audio on the cathode through Rk which, in
turn, is the preamp NFB... just as with the 6ME8. I.E. a 'plop in'
replacement to see if it 'works like the 6ME8' worked.
It didn't.
>Is the plate loading set correctly? Incorrect plate loading would cause an
>effect similar to what you are describing.
Plate load was fine, and peaked. The problem is G3 being relative to
the cathode, which is going up and down with audio.
>> That doesn't
>> happen with the 6ME8 because the deflection plates are relative to
>> each other and any change on the cathode is either blocked by the
>> screen or goes equally to both.
>>
>> I tried connecting the G3 grid leak to cathode and the result looked
>> 'almost' right but I think it was still distorted. So maybe there's a
>> way around it but they sure aren't a simple 'plop in' kind of thing.
>
>I don't follow your reference to "a simple 'plop in' kind of thing", what do you
>mean?
See above.
wrote:
>In article <remav7pebvlvouhcd...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Tue, 26 Jun 2012 22:29:35 -0500, John Byrns <byr...@sbcglobal.net>
>> wrote:
>>
>> >
>> >Are you trying to play with my mind? I was the first to use the term "dual
>> >control pentode in this thread when I made the statement "Applying the
>> >modulation to G1 doesn't take advantage of the beam deflection capabilities
>> >of
>> >the tube and instead uses it in a way that a more ordinary tube, like a dual
>> >control pentode, could serve." when I meant to say heptode. I'm the one
>> >that
>> >converted it!
>>
>> Setting side who's mind is going... hehe. I threw together a quick and
>> dirty heptode/DC pentode (same pinout so easy to try both)
>> 'substitute' for the 6ME8 to test the theory they should behave
>> similarly.
>>
>> Abstract summary is they didn't.
>
>Sounds reasonable, I would be surprised if they did behave the same. I take it
>you are saying the 6ME8 behaved differently than the heptode and that the
>heptode behaved differently than the dual control pentode?
See below.
>In article <remav7pebvlvouhcd...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Tue, 26 Jun 2012 22:29:35 -0500, John Byrns <byr...@sbcglobal.net>
>> wrote:
>>
>> >
>> >Are you trying to play with my mind? I was the first to use the term "dual
>> >control pentode in this thread when I made the statement "Applying the
>> >modulation to G1 doesn't take advantage of the beam deflection capabilities
>> >of
>> >the tube and instead uses it in a way that a more ordinary tube, like a dual
>> >control pentode, could serve." when I meant to say heptode. I'm the one
>> >that
>> >converted it!
>>
>> Setting side who's mind is going... hehe. I threw together a quick and
>> dirty heptode/DC pentode (same pinout so easy to try both)
>> 'substitute' for the 6ME8 to test the theory they should behave
>> similarly.
>>
>> Abstract summary is they didn't.
>
>Sounds reasonable, I would be surprised if they did behave the same. I take it
>you are saying the 6ME8 behaved differently than the heptode and that the
>heptode behaved differently than the dual control pentode?
>> On first power up a 90% mod waveform looked a lot like a mountain with
>> a huge deep crater where the peak should be and after twiddling around
>> a bit it occurred to me, uh oh, G3 is relative to cathode, and that's
>> going up and down with audio, so either the RF amplitude, or 'center
>> point', needed to swing from full on to cutoff changes.
>
>By the context you are clearly talking about either the heptode or the dual
>control pentode here, did both exhibit this mountain crater effect?
>Why is
>there audio on the cathode? I assume we are talking about a self excited
>circuit so that there would be a lot of RF on the cathode, but audio?
not. That's the dime a dozen, pick your tube and there's a ready made
schematic already there to download, way of doing it.
Maybe I misunderstood your "like a (heptode) dual control pentode
could serve" but I took it as suggesting audio to G1 with RF to G3,
'like done with the 6ME8', should produce results 'like' the 6ME8. As
in 'why use a 6ME8 when a plain ole heptode would do the same thing'?
And that's the context all my discussions on them has been in.
So I disconnected the 6ME8, ran RF to heptode/DC pentode G3 with audio
to G1 which, of course, puts audio on the cathode through Rk which, in
turn, is the preamp NFB... just as with the 6ME8. I.E. a 'plop in'
replacement to see if it 'works like the 6ME8' worked.
It didn't.
>Is the plate loading set correctly? Incorrect plate loading would cause an
>effect similar to what you are describing.
the cathode, which is going up and down with audio.
>> That doesn't
>> happen with the 6ME8 because the deflection plates are relative to
>> each other and any change on the cathode is either blocked by the
>> screen or goes equally to both.
>>
>> I tried connecting the G3 grid leak to cathode and the result looked
>> 'almost' right but I think it was still distorted. So maybe there's a
>> way around it but they sure aren't a simple 'plop in' kind of thing.
>
>I don't follow your reference to "a simple 'plop in' kind of thing", what do you
>mean?
John Byrns
Jul 5, 2012, 11:50:49 PM7/5/12
to
In article <byrnsj-93F190....@news.giganews.com>,
I did some digging around on the interweb based on the bibliography in the the
BBC paper I linked to above. I found the following two BBC papers:
http://downloads.bbc.co.uk/rd/pubs/reports/1985-13.pdf
http://downloads.bbc.co.uk/rd/pubs/reports/1988-15.pdf
I think these are the papers I read way back when, I haven't yet reviewed them
again, but I think they both discuss what I called the second DCC scheme because
"Fig. 2" in each paper illustrates the scheme as I conceive of it, I will have
to see what the papers have to say beyond that.
I wonder if there is a poor man's approach to implementing this DCC scheme
analogous to Weaver's implementation of the first DCC scheme?
BBC paper I linked to above. I found the following two BBC papers:
http://downloads.bbc.co.uk/rd/pubs/reports/1985-13.pdf
http://downloads.bbc.co.uk/rd/pubs/reports/1988-15.pdf
I think these are the papers I read way back when, I haven't yet reviewed them
again, but I think they both discuss what I called the second DCC scheme because
"Fig. 2" in each paper illustrates the scheme as I conceive of it, I will have
to see what the papers have to say beyond that.
I wonder if there is a poor man's approach to implementing this DCC scheme
analogous to Weaver's implementation of the first DCC scheme?
flipper
Jul 6, 2012, 12:15:10 AM7/6/12
to
On Thu, 05 Jul 2012 21:50:23 -0500, John Byrns <byr...@sbcglobal.net>
Hehe. We must have similar google settings as I saw that one too.
Interesting but in the 'results' they never speak of achieving the
'goals', just perception analysis.
>
>http://downloads.bbc.co.uk/rd/pubs/reports/1994-01.pdf
>
>This paper has a bibliography which lists a number of earlier papers, including
>several BBC papers from about 10 years earlier which is where I think I first
>learned of these techniques. I have mostly forgotten how this stuff works and
>will have to dig out some of the references and read them through.
>
>> As I think of it, though, I don't see how the 'second form' you
>> mention would increase efficiency much since most audio is not at
>> 'high mod' levels.
>
>I don't know about that, I remember back in the early 1960s an associate of mine
>worked at a station where he proudly pointed out that the modulation monitor
>meter rarely dropped off the 100% mark.
Hmm. Well, in that case, wouldn't it equate to simply having a lower
power transmitter since the 'compression' would be essentially 'always
on' constant?
> Maybe the second scheme I described was
>just a harebrained scheme I thought up years ago and never really existed
>anywhere else. However I would think it could work for stations with very dense
>continuous modulation like a rock music station. With this second scheme you
>could potentially remove the peak tube from a Doherty style transmitter, saving
>some filament power, and perhaps change the pulley ratio on the cooling blower
>reducing the power consumed by the blower. Again I will have to read through
>some of the references and refresh my memory on all this.
I think it depends on what the 'goal' is and, in the final analysis,
probably not quite as 'simple' as it seems.
To me DAM makes the most sense, if the idea is to conserve power.
Possibly but the beauty of the thing is it's utter simplicity and that
he doesn't 'add complexity' but gets a new function almost by just
'rearranging' things. Not quite, of course, because his tetrode isn't
a dual control pentode, like in a 'conventional' broadcaster, but the
number of devices is the same.
>IIRC one of the BBC papers, maybe several, delves into the AGC compatibility
>issues.
Yeah, but they're being considerably more 'fancy'.
His 3 tube C-QUAM AM Stereo Transmitter.
>> Oh, I see in the radiomuseum
>> article he brings up the 'beam pentode' tetrode matter that was
>> discussed on ARF.
>
>Did he say anything more about it on "ARF"?
I posted a link to it, up above.
Interesting but in the 'results' they never speak of achieving the
'goals', just perception analysis.
>
>http://downloads.bbc.co.uk/rd/pubs/reports/1994-01.pdf
>
>This paper has a bibliography which lists a number of earlier papers, including
>several BBC papers from about 10 years earlier which is where I think I first
>learned of these techniques. I have mostly forgotten how this stuff works and
>will have to dig out some of the references and read them through.
>
>> As I think of it, though, I don't see how the 'second form' you
>> mention would increase efficiency much since most audio is not at
>> 'high mod' levels.
>
>I don't know about that, I remember back in the early 1960s an associate of mine
>worked at a station where he proudly pointed out that the modulation monitor
>meter rarely dropped off the 100% mark.
power transmitter since the 'compression' would be essentially 'always
on' constant?
> Maybe the second scheme I described was
>just a harebrained scheme I thought up years ago and never really existed
>anywhere else. However I would think it could work for stations with very dense
>continuous modulation like a rock music station. With this second scheme you
>could potentially remove the peak tube from a Doherty style transmitter, saving
>some filament power, and perhaps change the pulley ratio on the cooling blower
>reducing the power consumed by the blower. Again I will have to read through
>some of the references and refresh my memory on all this.
probably not quite as 'simple' as it seems.
To me DAM makes the most sense, if the idea is to conserve power.
he doesn't 'add complexity' but gets a new function almost by just
'rearranging' things. Not quite, of course, because his tetrode isn't
a dual control pentode, like in a 'conventional' broadcaster, but the
number of devices is the same.
>IIRC one of the BBC papers, maybe several, delves into the AGC compatibility
>issues.
>> Oh, I see in the radiomuseum
>> article he brings up the 'beam pentode' tetrode matter that was
>> discussed on ARF.
>
>Did he say anything more about it on "ARF"?
John Byrns
Jul 6, 2012, 5:16:26 PM7/6/12
to
In article <8rlcv7lg9am0qrlle...@4ax.com>,
thought about the fact that the cathode in your circuit isn't grounded for audio
frequencies.
Speaking of not being sure what circuit was being discussed, when you were
making those "FMing" measurements for Alex Pogossov I was never clear on which
circuit you were measuring? Was it the full "Beam Me Up Scotty" transmitter
with separate oscillator or something else?
> Maybe I misunderstood your "like a (heptode) dual control pentode
> could serve" but I took it as suggesting audio to G1 with RF to G3,
> 'like done with the 6ME8', should produce results 'like' the 6ME8. As
> in 'why use a 6ME8 when a plain ole heptode would do the same thing'?
> And that's the context all my discussions on them has been in.
As near as I can tell you correctly understood my original question.
> So I disconnected the 6ME8, ran RF to heptode/DC pentode G3 with audio
> to G1 which, of course, puts audio on the cathode through Rk which, in
> turn, is the preamp NFB... just as with the 6ME8. I.E. a 'plop in'
> replacement to see if it 'works like the 6ME8' worked.
>
> It didn't.
First let me check to see that I correctly understand where the "craters" you
are observing actually are on the transmitted waveform, as I am interpreting
what you said they are on the positive modulation peaks, is that correct?
I still think incorrect plate loading is the cause of this problem, although if
there weren't audio on the cathode you would just get plain old clipping rather
than craters. The G3 drivelevel may also have some impact on this. While I
think you understood my original question I didn't mean to imply that one of the
other tubes could be "plopped" directly in place of the 6ME8 without tweaking
the circuit parameters to suit the new tube, especially the plate loading and
the G3 drive level.
Another issue that may be compounding the problem is that the G1 control is
presumably like a single ended class A audio amplifier which produces a fair
amount of second harmonic distortion. IIRC, and I hope I don't have this
backwards, the tube output on negative G1 excursions is compressed relative to
positive G1 excursions. If I have this the right way around that would imply
that if you adjust for 90% negative modulation, then the positive modulation
peaks may be considerably greater than 90% and may be limited by "bumping into
the cathode". If this is the case, correct plate loading should fix the problem.
> >Is the plate loading set correctly? Incorrect plate loading would cause an
> >effect similar to what you are describing.
>
> Plate load was fine, and peaked.
I understand the "peaked" part, but how are you determining that the "plate load
was fine"? Correct plate loading is important on these grid modulated
transmitters, unlike plate modulated transmitters which are more forgiving of
incorrect plate loading. I have not seen anyone, on radiomuseum, or elsewhere,
delve into the loading issues with these small transmitters. I thought Robert
Weaver was going to get into it in his C-Quam article where he calculates the
impedance of a short monopole antenna, but he balked and didn't discuss the
loading issue.
> The problem is G3 being relative to
> the cathode, which is going up and down with audio.
I understand what you are saying about the cathode going up and down with the
audio, I'm not sure I understand what G3 has to do with the problem though,
other than possibly the G3 drive level?
> >> That doesn't
> >> happen with the 6ME8 because the deflection plates are relative to
> >> each other and any change on the cathode is either blocked by the
> >> screen or goes equally to both.
> >> screen or goes equally to both.
The question here seems to be did you just get lucky with the loading for the
6ME8, or is the plate somehow blocked from the cathode as you speculate? In any
case any transmitter is going to eventually clip in some way on positive peaks.
> >> I tried connecting the G3 grid leak to cathode and the result looked
> >> 'almost' right but I think it was still distorted. So maybe there's a
> >> way around it but they sure aren't a simple 'plop in' kind of thing.
That's an interesting result, I wonder if the change was due to the consequent
change in bias on G3?
> >I don't follow your reference to "a simple 'plop in' kind of thing", what do
> >you mean?
>
> See above.
I wouldn't expect it to be as simple as "a simple 'plop in' kind of thing", I
would expect that the circuit parameters would need tweaking for the different
tube.
See above.
flipper <fli...@fish.net> wrote:
> On Thu, 05 Jul 2012 21:52:00 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >Why is
> >there audio on the cathode? I assume we are talking about a self excited
> >circuit so that there would be a lot of RF on the cathode, but audio?
>
> No, I'm not talking about audio to G3 with RF to G1, self excite or
> not. That's the dime a dozen, pick your tube and there's a ready made
> schematic already there to download, way of doing it.
Yeah, I forgot exactly what circuit we were talking about, plus I never really
> On Thu, 05 Jul 2012 21:52:00 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >Why is
> >there audio on the cathode? I assume we are talking about a self excited
> >circuit so that there would be a lot of RF on the cathode, but audio?
>
> No, I'm not talking about audio to G3 with RF to G1, self excite or
> not. That's the dime a dozen, pick your tube and there's a ready made
> schematic already there to download, way of doing it.
thought about the fact that the cathode in your circuit isn't grounded for audio
frequencies.
Speaking of not being sure what circuit was being discussed, when you were
making those "FMing" measurements for Alex Pogossov I was never clear on which
circuit you were measuring? Was it the full "Beam Me Up Scotty" transmitter
with separate oscillator or something else?
> Maybe I misunderstood your "like a (heptode) dual control pentode
> could serve" but I took it as suggesting audio to G1 with RF to G3,
> 'like done with the 6ME8', should produce results 'like' the 6ME8. As
> in 'why use a 6ME8 when a plain ole heptode would do the same thing'?
> And that's the context all my discussions on them has been in.
> So I disconnected the 6ME8, ran RF to heptode/DC pentode G3 with audio
> to G1 which, of course, puts audio on the cathode through Rk which, in
> turn, is the preamp NFB... just as with the 6ME8. I.E. a 'plop in'
> replacement to see if it 'works like the 6ME8' worked.
>
> It didn't.
are observing actually are on the transmitted waveform, as I am interpreting
what you said they are on the positive modulation peaks, is that correct?
I still think incorrect plate loading is the cause of this problem, although if
there weren't audio on the cathode you would just get plain old clipping rather
than craters. The G3 drivelevel may also have some impact on this. While I
think you understood my original question I didn't mean to imply that one of the
other tubes could be "plopped" directly in place of the 6ME8 without tweaking
the circuit parameters to suit the new tube, especially the plate loading and
the G3 drive level.
Another issue that may be compounding the problem is that the G1 control is
presumably like a single ended class A audio amplifier which produces a fair
amount of second harmonic distortion. IIRC, and I hope I don't have this
backwards, the tube output on negative G1 excursions is compressed relative to
positive G1 excursions. If I have this the right way around that would imply
that if you adjust for 90% negative modulation, then the positive modulation
peaks may be considerably greater than 90% and may be limited by "bumping into
the cathode". If this is the case, correct plate loading should fix the problem.
> >Is the plate loading set correctly? Incorrect plate loading would cause an
> >effect similar to what you are describing.
>
> Plate load was fine, and peaked.
was fine"? Correct plate loading is important on these grid modulated
transmitters, unlike plate modulated transmitters which are more forgiving of
incorrect plate loading. I have not seen anyone, on radiomuseum, or elsewhere,
delve into the loading issues with these small transmitters. I thought Robert
Weaver was going to get into it in his C-Quam article where he calculates the
impedance of a short monopole antenna, but he balked and didn't discuss the
loading issue.
> The problem is G3 being relative to
> the cathode, which is going up and down with audio.
audio, I'm not sure I understand what G3 has to do with the problem though,
other than possibly the G3 drive level?
> >> That doesn't
> >> happen with the 6ME8 because the deflection plates are relative to
> >> each other and any change on the cathode is either blocked by the
> >> screen or goes equally to both.
> >> screen or goes equally to both.
6ME8, or is the plate somehow blocked from the cathode as you speculate? In any
case any transmitter is going to eventually clip in some way on positive peaks.
> >> I tried connecting the G3 grid leak to cathode and the result looked
> >> 'almost' right but I think it was still distorted. So maybe there's a
> >> way around it but they sure aren't a simple 'plop in' kind of thing.
change in bias on G3?
> >I don't follow your reference to "a simple 'plop in' kind of thing", what do
> >you mean?
>
> See above.
would expect that the circuit parameters would need tweaking for the different
tube.
See above.
John Byrns
Jul 6, 2012, 7:23:04 PM7/6/12
to
In article <8sncv7hd9pvv76f9i...@4ax.com>,
Did you see the links in my update post, particularly this one?
http://downloads.bbc.co.uk/rd/pubs/reports/1988-15.pdf
There is considerable discussion of the results in terms of energy savings and
operational experience with actual BBC transmitters for "AMC" which is my
second, and preferred, method. They even mention removing the transmitters peak
tube when using 6 dB of compression, this paper must be where I originally got
my information.
> >http://downloads.bbc.co.uk/rd/pubs/reports/1994-01.pdf
> >
> >This paper has a bibliography which lists a number of earlier papers,
> >including
> >several BBC papers from about 10 years earlier which is where I think I
> >first
> >learned of these techniques. I have mostly forgotten how this stuff works
> >and
> >will have to dig out some of the references and read them through.
> >
> >> As I think of it, though, I don't see how the 'second form' you
> >> mention would increase efficiency much since most audio is not at
> >> 'high mod' levels.
> >
> >I don't know about that, I remember back in the early 1960s an associate of
> >mine
> >worked at a station where he proudly pointed out that the modulation monitor
> >meter rarely dropped off the 100% mark.
>
> Hmm. Well, in that case, wouldn't it equate to simply having a lower
> power transmitter since the 'compression' would be essentially 'always
> on' constant?
I suppose so, except that the AMC scheme on a higher powered transmitter has a
better subjective signal to noise ratio because the carrier power is increased
during quiet periods, causing the receiver's AGC to reduce the gain and hence
the noise.
> > Maybe the second scheme I described was
> >just a harebrained scheme I thought up years ago and never really existed
> >anywhere else. However I would think it could work for stations with very
> >dense
> >continuous modulation like a rock music station. With this second scheme
> >you
> >could potentially remove the peak tube from a Doherty style transmitter,
> >saving
> >some filament power, and perhaps change the pulley ratio on the cooling
> >blower
> >reducing the power consumed by the blower. Again I will have to read
> >through
> >some of the references and refresh my memory on all this.
I seem to have remembered all this correctly from the BBC research paper rather
than it having been a harebrained idea of my own, or maybe the BBC engineers are
also harebrained.
> I think it depends on what the 'goal' is and, in the final analysis,
> probably not quite as 'simple' as it seems.
>
> To me DAM makes the most sense, if the idea is to conserve power.
The BBC appears to have chosen to implement AMC, and to have achieved a
considerable energy savings, even with only 3 dB of compression, while my gold
standard is 6 dB which should provide even greater energy savings, although in
my quick scan I didn't notice that the BBC provided any energy savings data for
that mode.
Yes, I agree that the beauty of Weaver's "DAM" transmitter is its extreme
simplicity with no added complexity. I have come up with two simple poor man's
schemes for implementing my second form of DCC, or "AMC". One is a simple
modification of your "Beam Me Up Scotty" transmitter that actually saves a few
parts. My second approach, which I prefer, is a slight modification of Weaver's
transmitter that requires a few extra Rs & Cs, plus a diode.
Where Weaver's poor man's DAM scheme causes compression of the received audio,
my poor man's AMC scheme would cause expansion of the received audio. Of course
in the real rich man's systems appropriate expansion or compression is applied
to the audio at the transmitter to eliminate the compression or expansion
effects of the poor man's versions.
http://downloads.bbc.co.uk/rd/pubs/reports/1988-15.pdf
There is considerable discussion of the results in terms of energy savings and
operational experience with actual BBC transmitters for "AMC" which is my
second, and preferred, method. They even mention removing the transmitters peak
tube when using 6 dB of compression, this paper must be where I originally got
my information.
> >http://downloads.bbc.co.uk/rd/pubs/reports/1994-01.pdf
> >
> >This paper has a bibliography which lists a number of earlier papers,
> >including
> >several BBC papers from about 10 years earlier which is where I think I
> >first
> >learned of these techniques. I have mostly forgotten how this stuff works
> >and
> >will have to dig out some of the references and read them through.
> >
> >> As I think of it, though, I don't see how the 'second form' you
> >> mention would increase efficiency much since most audio is not at
> >> 'high mod' levels.
> >
> >I don't know about that, I remember back in the early 1960s an associate of
> >mine
> >worked at a station where he proudly pointed out that the modulation monitor
> >meter rarely dropped off the 100% mark.
>
> Hmm. Well, in that case, wouldn't it equate to simply having a lower
> power transmitter since the 'compression' would be essentially 'always
> on' constant?
better subjective signal to noise ratio because the carrier power is increased
during quiet periods, causing the receiver's AGC to reduce the gain and hence
the noise.
> > Maybe the second scheme I described was
> >just a harebrained scheme I thought up years ago and never really existed
> >anywhere else. However I would think it could work for stations with very
> >dense
> >continuous modulation like a rock music station. With this second scheme
> >you
> >could potentially remove the peak tube from a Doherty style transmitter,
> >saving
> >some filament power, and perhaps change the pulley ratio on the cooling
> >blower
> >reducing the power consumed by the blower. Again I will have to read
> >through
> >some of the references and refresh my memory on all this.
than it having been a harebrained idea of my own, or maybe the BBC engineers are
also harebrained.
> I think it depends on what the 'goal' is and, in the final analysis,
> probably not quite as 'simple' as it seems.
>
> To me DAM makes the most sense, if the idea is to conserve power.
considerable energy savings, even with only 3 dB of compression, while my gold
standard is 6 dB which should provide even greater energy savings, although in
my quick scan I didn't notice that the BBC provided any energy savings data for
that mode.
simplicity with no added complexity. I have come up with two simple poor man's
schemes for implementing my second form of DCC, or "AMC". One is a simple
modification of your "Beam Me Up Scotty" transmitter that actually saves a few
parts. My second approach, which I prefer, is a slight modification of Weaver's
transmitter that requires a few extra Rs & Cs, plus a diode.
Where Weaver's poor man's DAM scheme causes compression of the received audio,
my poor man's AMC scheme would cause expansion of the received audio. Of course
in the real rich man's systems appropriate expansion or compression is applied
to the audio at the transmitter to eliminate the compression or expansion
effects of the poor man's versions.
John L Stewart
Jul 6, 2012, 4:40:08 PM7/6/12
to
Another system that managed to save some transmitter power & some BW as
well saw some popularity in the late 80s. It is ACSB (Amplitude
Compandered Sideband). It did not take off well commercially here as its
proponents hoped. I've still got the Aerotron stuff in my file.
ACSB is basically SSB with a 3.1 KHz pilot tone included, so demod is
somewhat simpler.
It was directed at the 2-way com market (PMR, Public Mobile Radio) as a
replacement for NBFM where channels occupied 25 KHz & the carrier runs
continuously. ACSB could be fit into a 5 KHz channel, allowing lots more
utilization in a given spectrum.
The local rep pushing the system called looking for suitable test
equipment while I was at R&S. It looked like we could do it with the R&S
CMT or CMTA if the SSB option was loaded.
Seems like ACSB may still be in use by some amature radio operators. See
this link- http://forums.qrz.com/archive/index.php/t-265754.html
Today APCO 25 digital systems C4FM (12.5 KHz BW) & CQPSK (6.25 KHz BW)
pretty well solve the BW & power problems. The Aeroflex 3920 can do
both.
Cheers to all, John
+-------------------------------------------------------------------+
|Download: http://www.audiobanter.com/attachment.php?attachmentid=292|
flipper
Jul 6, 2012, 11:24:52 PM7/6/12
to
On Fri, 06 Jul 2012 16:16:26 -0500, John Byrns <byr...@sbcglobal.net>
>> Maybe I misunderstood your "like a (heptode) dual control pentode
>> could serve" but I took it as suggesting audio to G1 with RF to G3,
>> 'like done with the 6ME8', should produce results 'like' the 6ME8. As
>> in 'why use a 6ME8 when a plain ole heptode would do the same thing'?
>> And that's the context all my discussions on them has been in.
>
>As near as I can tell you correctly understood my original question.
>
>> So I disconnected the 6ME8, ran RF to heptode/DC pentode G3 with audio
>> to G1 which, of course, puts audio on the cathode through Rk which, in
>> turn, is the preamp NFB... just as with the 6ME8. I.E. a 'plop in'
>> replacement to see if it 'works like the 6ME8' worked.
>>
>> It didn't.
>
>First let me check to see that I correctly understand where the "craters" you
>are observing actually are on the transmitted waveform, as I am interpreting
>what you said they are on the positive modulation peaks, is that correct?
Yep.
>I still think incorrect plate loading is the cause of this problem,
What kind of "incorrect plate loading" do you mean?
> although if
>there weren't audio on the cathode you would just get plain old clipping rather
>than craters. The G3 drivelevel may also have some impact on this.
Well, as I said, "after twiddling around a bit." I tried RF drive
levels everywhere from 1V to 15V.
> While I
>think you understood my original question I didn't mean to imply that one of the
>other tubes could be "plopped" directly in place of the 6ME8 without tweaking
>the circuit parameters to suit the new tube, especially the plate loading and
>the G3 drive level.
I didn't expect to 'do nothing' either and by 'plop in' I simply mean
the same basic topology.
>Another issue that may be compounding the problem is that the G1 control is
>presumably like a single ended class A audio amplifier which produces a fair
>amount of second harmonic distortion. IIRC, and I hope I don't have this
>backwards, the tube output on negative G1 excursions is compressed relative to
>positive G1 excursions. If I have this the right way around that would imply
>that if you adjust for 90% negative modulation, then the positive modulation
>peaks may be considerably greater than 90% and may be limited by "bumping into
>the cathode". If this is the case, correct plate loading should fix the problem.
RF amplitude was no where close to 'bumping into the cathode." I wish
it were because that could be easily fixed and would mean I was
getting maximum power generated.
>> >Is the plate loading set correctly? Incorrect plate loading would cause an
>> >effect similar to what you are describing.
>>
>> Plate load was fine, and peaked.
>
>I understand the "peaked" part, but how are you determining that the "plate load
>was fine"? Correct plate loading is important on these grid modulated
>transmitters, unlike plate modulated transmitters which are more forgiving of
>incorrect plate loading. I have not seen anyone, on radiomuseum, or elsewhere,
>delve into the loading issues with these small transmitters. I thought Robert
>Weaver was going to get into it in his C-Quam article where he calculates the
>impedance of a short monopole antenna, but he balked and didn't discuss the
>loading issue.
I don't think it comes up because all of these things end up with
relatively low plate loads.
>> The problem is G3 being relative to
>> the cathode, which is going up and down with audio.
>
>I understand what you are saying about the cathode going up and down with the
>audio, I'm not sure I understand what G3 has to do with the problem though,
>other than possibly the G3 drive level?
G3 is relative to the cathode. Let's say there's 1 mA total cathode
current going through the 5600 Ohm Rk. That puts the cathode at 5.6 V
and, since the G3 grid leak is normally connected to ground, means G3
bias is -5.6 V. Now stick audio into G1. Cathode voltage can swing
from 0 to 11.2 V (assuming no 2'nd harmonic, which will hopefully be
low because we're using NFB to correct it), which means G3 bias is now
changing with the audio from 0 to -11.2 V.
Now, 11.2 V is probably enough to drive all of them to cutoff,
ironically at the point that should be 'max power' (peak cathode
current), so without even getting around to applying RF, and pondering
'drive levels', it seems to me we're already 'in trouble'.
In short, we apply audio to G1, which puts audio on the cathode,
which, in turn, impresses the same audio onto G3. Except, oops, we
don't want audio on G3 and there's no way to filter it off the cathode
because that's necessary for NFB into the preamp driving G1.
You can see why it 'craters' the peaks. Cathode audio drives G3 bias
dramatically negative at the peaks, cutting off plate current.
I feel fairly confident with that analysis because it not only
explains the observed behavior but my 'testable hypothesis' of moving
the G3 grid leak to cathode produced results consistent with
predictions. (All very "scientific method," eh? ;) )
>> >> That doesn't
>> >> happen with the 6ME8 because the deflection plates are relative to
>> >> each other and any change on the cathode is either blocked by the
>> >> screen or goes equally to both.
>> >> screen or goes equally to both.
>
>The question here seems to be did you just get lucky with the loading for the
>6ME8,
It depends on how you define 'lucky'. None of these tube broadcasters
using simply a plate tank have high enough load impedance to get very
large plate swings so generated power is only a fraction of plate
power.
I suppose it's 'lucky' you never have to worry about slamming into the
cathode but it's not so 'lucky' if you're hoping for any kind of
efficiency.
> or is the plate somehow blocked from the cathode as you speculate? In any
>case any transmitter is going to eventually clip in some way on positive peaks.
>
>> >> I tried connecting the G3 grid leak to cathode and the result looked
>> >> 'almost' right but I think it was still distorted. So maybe there's a
>> >> way around it but they sure aren't a simple 'plop in' kind of thing.
>
>That's an interesting result, I wonder if the change was due to the consequent
>change in bias on G3?
Interesting way of putting it as it's due to G3 bias 'not changing'
any more. I.E. assuming impedances work out, which I didn't go to the
effort of trying to refine, G3 then goes 'up and down' in sync with
the cathode going 'up and down' so, as far as G3 is concerned, it's
'static' relative to the cathode. Well, before we apply RF, that is.
That introduces all sorts of impedance and bias current conundrums,
like G3 leakage currents into Rk, and I keep wondering if the screen
should be bypassed to the top of Rk rather than ground, but then
there's screen currents into Rk. Anyway, it became clear it wasn't a
case of simply translating the existing 6ME8 circuit into a heptode/DC
pentode version.
It does make me wonder if, despite the 6ME8 deflection plates seeming
to be being more or less independent of cathode voltage, if doing the
same thing might improve distortion slightly because they're possibly
not '100%' independent. I mean, they need a bias voltage, which also
might be 'changing' because cathode has audio on it, but just not as
much relative to the 35V bias. I'm not sure because it isn't obvious
to me whether that bias is relative to cathode or the screen but
Murphy's Law says it's whichever is worst to deal with, which would be
the cathode. I may try the same idea on Beamus but it's so clean now
I'm not sure I could tell by just looking at the scope, unless it
dramatically screws things up.
>> >I don't follow your reference to "a simple 'plop in' kind of thing", what do
>> >you mean?
>>
>> See above.
>
>I wouldn't expect it to be as simple as "a simple 'plop in' kind of thing", I
>would expect that the circuit parameters would need tweaking for the different
>tube.
It isn't a matter of 'tweaking' because G3 is intimately tied to
cathode voltage.
wrote:
>In article <8rlcv7lg9am0qrlle...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Thu, 05 Jul 2012 21:52:00 -0500, John Byrns <byr...@sbcglobal.net>
>> wrote:
>>
>> >Why is
>> >there audio on the cathode? I assume we are talking about a self excited
>> >circuit so that there would be a lot of RF on the cathode, but audio?
>>
>> No, I'm not talking about audio to G3 with RF to G1, self excite or
>> not. That's the dime a dozen, pick your tube and there's a ready made
>> schematic already there to download, way of doing it.
>
>Yeah, I forgot exactly what circuit we were talking about, plus I never really
>thought about the fact that the cathode in your circuit isn't grounded for audio
>frequencies.
>
>Speaking of not being sure what circuit was being discussed, when you were
>making those "FMing" measurements for Alex Pogossov I was never clear on which
>circuit you were measuring? Was it the full "Beam Me Up Scotty" transmitter
>with separate oscillator or something else?
Beamus
>In article <8rlcv7lg9am0qrlle...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Thu, 05 Jul 2012 21:52:00 -0500, John Byrns <byr...@sbcglobal.net>
>> wrote:
>>
>> >Why is
>> >there audio on the cathode? I assume we are talking about a self excited
>> >circuit so that there would be a lot of RF on the cathode, but audio?
>>
>> No, I'm not talking about audio to G3 with RF to G1, self excite or
>> not. That's the dime a dozen, pick your tube and there's a ready made
>> schematic already there to download, way of doing it.
>
>Yeah, I forgot exactly what circuit we were talking about, plus I never really
>thought about the fact that the cathode in your circuit isn't grounded for audio
>frequencies.
>
>Speaking of not being sure what circuit was being discussed, when you were
>making those "FMing" measurements for Alex Pogossov I was never clear on which
>circuit you were measuring? Was it the full "Beam Me Up Scotty" transmitter
>with separate oscillator or something else?
>> Maybe I misunderstood your "like a (heptode) dual control pentode
>> could serve" but I took it as suggesting audio to G1 with RF to G3,
>> 'like done with the 6ME8', should produce results 'like' the 6ME8. As
>> in 'why use a 6ME8 when a plain ole heptode would do the same thing'?
>> And that's the context all my discussions on them has been in.
>
>As near as I can tell you correctly understood my original question.
>
>> So I disconnected the 6ME8, ran RF to heptode/DC pentode G3 with audio
>> to G1 which, of course, puts audio on the cathode through Rk which, in
>> turn, is the preamp NFB... just as with the 6ME8. I.E. a 'plop in'
>> replacement to see if it 'works like the 6ME8' worked.
>>
>> It didn't.
>
>First let me check to see that I correctly understand where the "craters" you
>are observing actually are on the transmitted waveform, as I am interpreting
>what you said they are on the positive modulation peaks, is that correct?
>I still think incorrect plate loading is the cause of this problem,
> although if
>there weren't audio on the cathode you would just get plain old clipping rather
>than craters. The G3 drivelevel may also have some impact on this.
levels everywhere from 1V to 15V.
> While I
>think you understood my original question I didn't mean to imply that one of the
>other tubes could be "plopped" directly in place of the 6ME8 without tweaking
>the circuit parameters to suit the new tube, especially the plate loading and
>the G3 drive level.
the same basic topology.
>Another issue that may be compounding the problem is that the G1 control is
>presumably like a single ended class A audio amplifier which produces a fair
>amount of second harmonic distortion. IIRC, and I hope I don't have this
>backwards, the tube output on negative G1 excursions is compressed relative to
>positive G1 excursions. If I have this the right way around that would imply
>that if you adjust for 90% negative modulation, then the positive modulation
>peaks may be considerably greater than 90% and may be limited by "bumping into
>the cathode". If this is the case, correct plate loading should fix the problem.
it were because that could be easily fixed and would mean I was
getting maximum power generated.
>> >Is the plate loading set correctly? Incorrect plate loading would cause an
>> >effect similar to what you are describing.
>>
>> Plate load was fine, and peaked.
>
>I understand the "peaked" part, but how are you determining that the "plate load
>was fine"? Correct plate loading is important on these grid modulated
>transmitters, unlike plate modulated transmitters which are more forgiving of
>incorrect plate loading. I have not seen anyone, on radiomuseum, or elsewhere,
>delve into the loading issues with these small transmitters. I thought Robert
>Weaver was going to get into it in his C-Quam article where he calculates the
>impedance of a short monopole antenna, but he balked and didn't discuss the
>loading issue.
relatively low plate loads.
>> The problem is G3 being relative to
>> the cathode, which is going up and down with audio.
>
>I understand what you are saying about the cathode going up and down with the
>audio, I'm not sure I understand what G3 has to do with the problem though,
>other than possibly the G3 drive level?
current going through the 5600 Ohm Rk. That puts the cathode at 5.6 V
and, since the G3 grid leak is normally connected to ground, means G3
bias is -5.6 V. Now stick audio into G1. Cathode voltage can swing
from 0 to 11.2 V (assuming no 2'nd harmonic, which will hopefully be
low because we're using NFB to correct it), which means G3 bias is now
changing with the audio from 0 to -11.2 V.
Now, 11.2 V is probably enough to drive all of them to cutoff,
ironically at the point that should be 'max power' (peak cathode
current), so without even getting around to applying RF, and pondering
'drive levels', it seems to me we're already 'in trouble'.
In short, we apply audio to G1, which puts audio on the cathode,
which, in turn, impresses the same audio onto G3. Except, oops, we
don't want audio on G3 and there's no way to filter it off the cathode
because that's necessary for NFB into the preamp driving G1.
You can see why it 'craters' the peaks. Cathode audio drives G3 bias
dramatically negative at the peaks, cutting off plate current.
I feel fairly confident with that analysis because it not only
explains the observed behavior but my 'testable hypothesis' of moving
the G3 grid leak to cathode produced results consistent with
predictions. (All very "scientific method," eh? ;) )
>> >> That doesn't
>> >> happen with the 6ME8 because the deflection plates are relative to
>> >> each other and any change on the cathode is either blocked by the
>> >> screen or goes equally to both.
>> >> screen or goes equally to both.
>
>The question here seems to be did you just get lucky with the loading for the
>6ME8,
using simply a plate tank have high enough load impedance to get very
large plate swings so generated power is only a fraction of plate
power.
I suppose it's 'lucky' you never have to worry about slamming into the
cathode but it's not so 'lucky' if you're hoping for any kind of
efficiency.
> or is the plate somehow blocked from the cathode as you speculate? In any
>case any transmitter is going to eventually clip in some way on positive peaks.
>
>> >> I tried connecting the G3 grid leak to cathode and the result looked
>> >> 'almost' right but I think it was still distorted. So maybe there's a
>> >> way around it but they sure aren't a simple 'plop in' kind of thing.
>
>That's an interesting result, I wonder if the change was due to the consequent
>change in bias on G3?
any more. I.E. assuming impedances work out, which I didn't go to the
effort of trying to refine, G3 then goes 'up and down' in sync with
the cathode going 'up and down' so, as far as G3 is concerned, it's
'static' relative to the cathode. Well, before we apply RF, that is.
That introduces all sorts of impedance and bias current conundrums,
like G3 leakage currents into Rk, and I keep wondering if the screen
should be bypassed to the top of Rk rather than ground, but then
there's screen currents into Rk. Anyway, it became clear it wasn't a
case of simply translating the existing 6ME8 circuit into a heptode/DC
pentode version.
It does make me wonder if, despite the 6ME8 deflection plates seeming
to be being more or less independent of cathode voltage, if doing the
same thing might improve distortion slightly because they're possibly
not '100%' independent. I mean, they need a bias voltage, which also
might be 'changing' because cathode has audio on it, but just not as
much relative to the 35V bias. I'm not sure because it isn't obvious
to me whether that bias is relative to cathode or the screen but
Murphy's Law says it's whichever is worst to deal with, which would be
the cathode. I may try the same idea on Beamus but it's so clean now
I'm not sure I could tell by just looking at the scope, unless it
dramatically screws things up.
>> >I don't follow your reference to "a simple 'plop in' kind of thing", what do
>> >you mean?
>>
>> See above.
>
>I wouldn't expect it to be as simple as "a simple 'plop in' kind of thing", I
>would expect that the circuit parameters would need tweaking for the different
>tube.
cathode voltage.
flipper
Jul 7, 2012, 12:45:52 AM7/7/12
to
On Fri, 06 Jul 2012 18:23:04 -0500, John Byrns <byr...@sbcglobal.net>
Yes, but it hadn't been posted yet when I commented on the other.
Yeah, and when I first saw it I thought improved audio was 'the point'
of it.
>> > Maybe the second scheme I described was
>> >just a harebrained scheme I thought up years ago and never really existed
>> >anywhere else. However I would think it could work for stations with very
>> >dense
>> >continuous modulation like a rock music station. With this second scheme
>> >you
>> >could potentially remove the peak tube from a Doherty style transmitter,
>> >saving
>> >some filament power, and perhaps change the pulley ratio on the cooling
>> >blower
>> >reducing the power consumed by the blower. Again I will have to read
>> >through
>> >some of the references and refresh my memory on all this.
>
>I seem to have remembered all this correctly from the BBC research paper rather
>than it having been a harebrained idea of my own, or maybe the BBC engineers are
>also harebrained.
>
>> I think it depends on what the 'goal' is and, in the final analysis,
>> probably not quite as 'simple' as it seems.
>>
>> To me DAM makes the most sense, if the idea is to conserve power.
>
>The BBC appears to have chosen to implement AMC, and to have achieved a
>considerable energy savings, even with only 3 dB of compression, while my gold
>standard is 6 dB which should provide even greater energy savings, although in
>my quick scan I didn't notice that the BBC provided any energy savings data for
>that mode.
Well, I have changed my opinion since then ;)
Actually, I was thinking of a modified DAM but the data does suggest
AMC is 'better'.
I think we should also note that 'affecting' the audio was Weaver's
*goal* because these little broadcasters have no 'compression' like
'the big boys' routinely apply before we get to considering 'power
savings'.
> I have come up with two simple poor man's
>schemes for implementing my second form of DCC, or "AMC".
Are we trying to save power on a home broadcaster?
> One is a simple
>modification of your "Beam Me Up Scotty" transmitter that actually saves a few
>parts.
I'd love to hear it. Are you thinking maybe a grid clamp on G1?
Although, I don't see that saving parts.
> My second approach, which I prefer, is a slight modification of Weaver's
>transmitter that requires a few extra Rs & Cs, plus a diode.
>
>Where Weaver's poor man's DAM scheme causes compression of the received audio,
>my poor man's AMC scheme would cause expansion of the received audio.
Yeah, and I'm not sure what the audible result would be because, as
Weaver points out, part of the 'problem' with home broadcasters is the
lack of compression causes them to sound 'weaker' than a 'big boy'
broadcast since, to stay below serious clipping, the average program
level has to be lower.
The missing element is AGC action so I don't know if a "poor man's
AMC" would 'enhance' dynamic range or just make it sound even weaker.
> Of course
>in the real rich man's systems appropriate expansion or compression is applied
>to the audio at the transmitter to eliminate the compression or expansion
>effects of the poor man's versions.
Yes, but that was the point of Weaver's design: actual audible
'compression' reducing the dynamic range so the end result sounds
'fuller' and 'louder' like 'the big boys' do with direct audio
compression of the source material. In his case you don't want to
'compensate' for it because he's not trying to do a "poor man's DAM"
but a "poor man's audio compressor." Which makes it doubly clever in
using DCC to accomplish a different job.
If I understand correctly AMC must be 'further compressing'
(relatively speaking) the already compressed audio because, after
receiver AGC 'compensates' for it, you still want the audio compressed
like it would be from a 'normal' transmitter.
of it.
>> > Maybe the second scheme I described was
>> >just a harebrained scheme I thought up years ago and never really existed
>> >anywhere else. However I would think it could work for stations with very
>> >dense
>> >continuous modulation like a rock music station. With this second scheme
>> >you
>> >could potentially remove the peak tube from a Doherty style transmitter,
>> >saving
>> >some filament power, and perhaps change the pulley ratio on the cooling
>> >blower
>> >reducing the power consumed by the blower. Again I will have to read
>> >through
>> >some of the references and refresh my memory on all this.
>
>I seem to have remembered all this correctly from the BBC research paper rather
>than it having been a harebrained idea of my own, or maybe the BBC engineers are
>also harebrained.
>
>> I think it depends on what the 'goal' is and, in the final analysis,
>> probably not quite as 'simple' as it seems.
>>
>> To me DAM makes the most sense, if the idea is to conserve power.
>
>The BBC appears to have chosen to implement AMC, and to have achieved a
>considerable energy savings, even with only 3 dB of compression, while my gold
>standard is 6 dB which should provide even greater energy savings, although in
>my quick scan I didn't notice that the BBC provided any energy savings data for
>that mode.
Actually, I was thinking of a modified DAM but the data does suggest
AMC is 'better'.
*goal* because these little broadcasters have no 'compression' like
'the big boys' routinely apply before we get to considering 'power
savings'.
> I have come up with two simple poor man's
>schemes for implementing my second form of DCC, or "AMC".
> One is a simple
>modification of your "Beam Me Up Scotty" transmitter that actually saves a few
>parts.
Although, I don't see that saving parts.
> My second approach, which I prefer, is a slight modification of Weaver's
>transmitter that requires a few extra Rs & Cs, plus a diode.
>
>Where Weaver's poor man's DAM scheme causes compression of the received audio,
>my poor man's AMC scheme would cause expansion of the received audio.
Weaver points out, part of the 'problem' with home broadcasters is the
lack of compression causes them to sound 'weaker' than a 'big boy'
broadcast since, to stay below serious clipping, the average program
level has to be lower.
The missing element is AGC action so I don't know if a "poor man's
AMC" would 'enhance' dynamic range or just make it sound even weaker.
> Of course
>in the real rich man's systems appropriate expansion or compression is applied
>to the audio at the transmitter to eliminate the compression or expansion
>effects of the poor man's versions.
'compression' reducing the dynamic range so the end result sounds
'fuller' and 'louder' like 'the big boys' do with direct audio
compression of the source material. In his case you don't want to
'compensate' for it because he's not trying to do a "poor man's DAM"
but a "poor man's audio compressor." Which makes it doubly clever in
using DCC to accomplish a different job.
If I understand correctly AMC must be 'further compressing'
(relatively speaking) the already compressed audio because, after
receiver AGC 'compensates' for it, you still want the audio compressed
like it would be from a 'normal' transmitter.
John Byrns
Jul 7, 2012, 11:10:25 PM7/7/12
to
In article <3l3fv7590f4bkph9u...@4ax.com>,
tube plate is too high.
> >Another issue that may be compounding the problem is that the G1 control is
> >presumably like a single ended class A audio amplifier which produces a fair
> >amount of second harmonic distortion. IIRC, and I hope I don't have this
> >backwards, the tube output on negative G1 excursions is compressed relative
> >to
> >positive G1 excursions. If I have this the right way around that would
> >imply
> >that if you adjust for 90% negative modulation, then the positive modulation
> >peaks may be considerably greater than 90% and may be limited by "bumping
> >into
> >the cathode". If this is the case, correct plate loading should fix the
> >problem.
>
> RF amplitude was no where close to 'bumping into the cathode." I wish
> it were because that could be easily fixed and would mean I was
> getting maximum power generated.
"Bumping into the cathode" was a bad choice of words on my part and doesn't
correctly reflect what I was trying to say, however I will leave it at that
since I am accepting your analysis further down.
> >> >Is the plate loading set correctly? Incorrect plate loading would cause
> >> >an
> >> >effect similar to what you are describing.
> >>
> >> Plate load was fine, and peaked.
> >
> >I understand the "peaked" part, but how are you determining that the "plate
> >load
> >was fine"? Correct plate loading is important on these grid modulated
> >transmitters, unlike plate modulated transmitters which are more forgiving
> >of
> >incorrect plate loading. I have not seen anyone, on radiomuseum, or
> >elsewhere,
> >delve into the loading issues with these small transmitters. I thought
> >Robert
> >Weaver was going to get into it in his C-Quam article where he calculates
> >the
> >impedance of a short monopole antenna, but he balked and didn't discuss the
> >loading issue.
>
> I don't think it comes up because all of these things end up with
> relatively low plate loads.
If it is the case that "all of these things end up with relatively low plate
loads", why do so many of these circuits have to add a resistor across the plate
tank circuit?
If it is the case that "all of these things end up with relatively low plate
loads", why do so many of these circuits have to add a resistor across the plate
tank circuit?
flipper <fli...@fish.net> wrote:
> On Fri, 06 Jul 2012 16:16:26 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >First let me check to see that I correctly understand where the "craters"
> >you
> >are observing actually are on the transmitted waveform, as I am interpreting
> >what you said they are on the positive modulation peaks, is that correct?
>
> Yep.
>
> >I still think incorrect plate loading is the cause of this problem,
>
> What kind of "incorrect plate loading" do you mean?
I mean that the load resistance reflected from the antenna/ground system to the
> On Fri, 06 Jul 2012 16:16:26 -0500, John Byrns <byr...@sbcglobal.net>
> wrote:
>
> >First let me check to see that I correctly understand where the "craters"
> >you
> >are observing actually are on the transmitted waveform, as I am interpreting
> >what you said they are on the positive modulation peaks, is that correct?
>
> Yep.
>
> >I still think incorrect plate loading is the cause of this problem,
>
> What kind of "incorrect plate loading" do you mean?
tube plate is too high.
> >Another issue that may be compounding the problem is that the G1 control is
> >presumably like a single ended class A audio amplifier which produces a fair
> >amount of second harmonic distortion. IIRC, and I hope I don't have this
> >backwards, the tube output on negative G1 excursions is compressed relative
> >to
> >positive G1 excursions. If I have this the right way around that would
> >imply
> >that if you adjust for 90% negative modulation, then the positive modulation
> >peaks may be considerably greater than 90% and may be limited by "bumping
> >into
> >the cathode". If this is the case, correct plate loading should fix the
> >problem.
>
> RF amplitude was no where close to 'bumping into the cathode." I wish
> it were because that could be easily fixed and would mean I was
> getting maximum power generated.
correctly reflect what I was trying to say, however I will leave it at that
since I am accepting your analysis further down.
> >> >Is the plate loading set correctly? Incorrect plate loading would cause
> >> >an
> >> >effect similar to what you are describing.
> >>
> >> Plate load was fine, and peaked.
> >
> >I understand the "peaked" part, but how are you determining that the "plate
> >load
> >was fine"? Correct plate loading is important on these grid modulated
> >transmitters, unlike plate modulated transmitters which are more forgiving
> >of
> >incorrect plate loading. I have not seen anyone, on radiomuseum, or
> >elsewhere,
> >delve into the loading issues with these small transmitters. I thought
> >Robert
> >Weaver was going to get into it in his C-Quam article where he calculates
> >the
> >impedance of a short monopole antenna, but he balked and didn't discuss the
> >loading issue.
>
> I don't think it comes up because all of these things end up with
> relatively low plate loads.
loads", why do so many of these circuits have to add a resistor across the plate
tank circuit?
loads", why do so many of these circuits have to add a resistor across the plate
tank circuit?
John Byrns
Jul 7, 2012, 11:17:00 PM7/7/12
to
In article <jmcfv755fekvrn5k3...@4ax.com>,
Yes, the article made it clear that his goal was the compression the scheme
provided, although it is not clear from the data he presented how much
compression he actually achieves. Perhaps a simple compressor circuit could do
better at the expense of an additional tube? I suppose an extra tube would
defile the beauty of it though.
> > I have come up with two simple poor man's
> >schemes for implementing my second form of DCC, or "AMC".
>
> Are we trying to save power on a home broadcaster?
No, but since I had originally mentioned two forms of DCC I felt compelled to
offer an idea for applying the second form to a home broadcaster even if doing
so might be counter productive. The simple one tube compressor mentioned above
might ameliorate the problem though, although it would make more sense to simply
apply the simple one tube compressor to a standard home broadcaster.
> > One is a simple
> >modification of your "Beam Me Up Scotty" transmitter that actually saves a
> >few
> >parts.
>
> I'd love to hear it. Are you thinking maybe a grid clamp on G1?
> Although, I don't see that saving parts.
Yes, I was thinking of the grid clamp idea like weaver uses, except applied to
the 6ME8 G1 where it would have the opposite effect of Weaver's grid clamp on
the audio preamp tube. This assumes that the 6ME8 operating conditions could be
suitably tweaked, perhaps G2 voltage, to compensate for the changed G1
conditions. The parts saving would be the elimination of the 6ME8 cathode
resistor and feedback network.
> > My second approach, which I prefer, is a slight modification of Weaver's
> >transmitter that requires a few extra Rs & Cs, plus a diode.
> >
> >Where Weaver's poor man's DAM scheme causes compression of the received
> >audio,
> >my poor man's AMC scheme would cause expansion of the received audio.
>
> Yeah, and I'm not sure what the audible result would be because, as
> Weaver points out, part of the 'problem' with home broadcasters is the
> lack of compression causes them to sound 'weaker' than a 'big boy'
> broadcast since, to stay below serious clipping, the average program
> level has to be lower.
>
> The missing element is AGC action so I don't know if a "poor man's
> AMC" would 'enhance' dynamic range or just make it sound even weaker.
Yes, I agree, as I said it was just an exercise to go along with my mention of
the second form of DCC. I think just adding a simple one tube compressor to the
standard home transmitter is the way to go.
> > Of course
> >in the real rich man's systems appropriate expansion or compression is
> >applied
> >to the audio at the transmitter to eliminate the compression or expansion
> >effects of the poor man's versions.
>
> Yes, but that was the point of Weaver's design: actual audible
> 'compression' reducing the dynamic range so the end result sounds
> 'fuller' and 'louder' like 'the big boys' do with direct audio
> compression of the source material. In his case you don't want to
> 'compensate' for it because he's not trying to do a "poor man's DAM"
> but a "poor man's audio compressor." Which makes it doubly clever in
> using DCC to accomplish a different job.
Very True.
> If I understand correctly AMC must be 'further compressing'
> (relatively speaking) the already compressed audio because, after
> receiver AGC 'compensates' for it, you still want the audio compressed
> like it would be from a 'normal' transmitter.
Yes, that is my understanding also, assuming I am parsing your words correctly.
provided, although it is not clear from the data he presented how much
compression he actually achieves. Perhaps a simple compressor circuit could do
better at the expense of an additional tube? I suppose an extra tube would
defile the beauty of it though.
> > I have come up with two simple poor man's
> >schemes for implementing my second form of DCC, or "AMC".
>
> Are we trying to save power on a home broadcaster?
offer an idea for applying the second form to a home broadcaster even if doing
so might be counter productive. The simple one tube compressor mentioned above
might ameliorate the problem though, although it would make more sense to simply
apply the simple one tube compressor to a standard home broadcaster.
> > One is a simple
> >modification of your "Beam Me Up Scotty" transmitter that actually saves a
> >few
> >parts.
>
> I'd love to hear it. Are you thinking maybe a grid clamp on G1?
> Although, I don't see that saving parts.
the 6ME8 G1 where it would have the opposite effect of Weaver's grid clamp on
the audio preamp tube. This assumes that the 6ME8 operating conditions could be
suitably tweaked, perhaps G2 voltage, to compensate for the changed G1
conditions. The parts saving would be the elimination of the 6ME8 cathode
resistor and feedback network.
> > My second approach, which I prefer, is a slight modification of Weaver's
> >transmitter that requires a few extra Rs & Cs, plus a diode.
> >
> >Where Weaver's poor man's DAM scheme causes compression of the received
> >audio,
> >my poor man's AMC scheme would cause expansion of the received audio.
>
> Yeah, and I'm not sure what the audible result would be because, as
> Weaver points out, part of the 'problem' with home broadcasters is the
> lack of compression causes them to sound 'weaker' than a 'big boy'
> broadcast since, to stay below serious clipping, the average program
> level has to be lower.
>
> The missing element is AGC action so I don't know if a "poor man's
> AMC" would 'enhance' dynamic range or just make it sound even weaker.
the second form of DCC. I think just adding a simple one tube compressor to the
standard home transmitter is the way to go.
> > Of course
> >in the real rich man's systems appropriate expansion or compression is
> >applied
> >to the audio at the transmitter to eliminate the compression or expansion
> >effects of the poor man's versions.
>
> Yes, but that was the point of Weaver's design: actual audible
> 'compression' reducing the dynamic range so the end result sounds
> 'fuller' and 'louder' like 'the big boys' do with direct audio
> compression of the source material. In his case you don't want to
> 'compensate' for it because he's not trying to do a "poor man's DAM"
> but a "poor man's audio compressor." Which makes it doubly clever in
> using DCC to accomplish a different job.
> If I understand correctly AMC must be 'further compressing'
> (relatively speaking) the already compressed audio because, after
> receiver AGC 'compensates' for it, you still want the audio compressed
> like it would be from a 'normal' transmitter.
flipper
Jul 8, 2012, 12:36:12 AM7/8/12
to
On Sat, 07 Jul 2012 22:10:25 -0500, John Byrns <byr...@sbcglobal.net>
Beamus, meaning with the 2.2k Rk, only did 90 Vp at 100% mod, leaving
65 V to cathode, since B+ came in lower at the higher current draw. On
the '100 mW' Beamus B+ is upwards to 190 V and the lower plate current
doesn't produce as much swing, of course.
The heptode/DC pentode version was doing essentially the same thing
because the load and plate current was the same, by intent.
At the moment I'm not quite sure how much because, somehow, I've
screwed the dern thing up, I mean the reassembled '100 mW' Beamus,
and, after 8 hours of rechecking every connection and component, have
not a blooming clue what's wrong with it.
The only reason I mentioned it is, since I have no idea what nor when
things went wrong, no measurements done since 'converting' it to the
heptode, and back, are reliable.
Odds are I didn't something 'stupid' when restoring it back to the
6ME8 but I'll be damned if I can find it.
>> >Another issue that may be compounding the problem is that the G1 control is
>> >presumably like a single ended class A audio amplifier which produces a fair
>> >amount of second harmonic distortion. IIRC, and I hope I don't have this
>> >backwards, the tube output on negative G1 excursions is compressed relative
>> >to
>> >positive G1 excursions. If I have this the right way around that would
>> >imply
>> >that if you adjust for 90% negative modulation, then the positive modulation
>> >peaks may be considerably greater than 90% and may be limited by "bumping
>> >into
>> >the cathode". If this is the case, correct plate loading should fix the
>> >problem.
>>
>> RF amplitude was no where close to 'bumping into the cathode." I wish
>> it were because that could be easily fixed and would mean I was
>> getting maximum power generated.
>
>"Bumping into the cathode" was a bad choice of words on my part and doesn't
>correctly reflect what I was trying to say, however I will leave it at that
>since I am accepting your analysis further down.
Yeah, I understand and wasn't going to quibble about minimum plate
voltage either. Near as I can recall maximum swing at 100% mod was on
the order of 50V peak so there was gobs of B+ left.
>> >> >Is the plate loading set correctly? Incorrect plate loading would cause
>> >> >an
>> >> >effect similar to what you are describing.
>> >>
>> >> Plate load was fine, and peaked.
>> >
>> >I understand the "peaked" part, but how are you determining that the "plate
>> >load
>> >was fine"? Correct plate loading is important on these grid modulated
>> >transmitters, unlike plate modulated transmitters which are more forgiving
>> >of
>> >incorrect plate loading. I have not seen anyone, on radiomuseum, or
>> >elsewhere,
>> >delve into the loading issues with these small transmitters. I thought
>> >Robert
>> >Weaver was going to get into it in his C-Quam article where he calculates
>> >the
>> >impedance of a short monopole antenna, but he balked and didn't discuss the
>> >loading issue.
>>
>> I don't think it comes up because all of these things end up with
>> relatively low plate loads.
>
>If it is the case that "all of these things end up with relatively low plate
>loads", why do so many of these circuits have to add a resistor across the plate
>tank circuit?
I'm not aware of any plain ole tank loaded dual control pentode
broadcasters that do, including Norm's original 'power monster' 6888
version.
Same answer: I don't know of any. I imagine that might be due to the
dual control pentode's relative low plate impedance limiting Q and
perhaps it could be more of an issue with heptodes.
wrote:
>In article <3l3fv7590f4bkph9u...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Fri, 06 Jul 2012 16:16:26 -0500, John Byrns <byr...@sbcglobal.net>
>> wrote:
>>
>> >First let me check to see that I correctly understand where the "craters"
>> >you
>> >are observing actually are on the transmitted waveform, as I am interpreting
>> >what you said they are on the positive modulation peaks, is that correct?
>>
>> Yep.
>>
>> >I still think incorrect plate loading is the cause of this problem,
>>
>> What kind of "incorrect plate loading" do you mean?
>
>I mean that the load resistance reflected from the antenna/ground system to the
>tube plate is too high.
Okay. Well, that's what I figured you meant but even the 'high power'
>In article <3l3fv7590f4bkph9u...@4ax.com>,
> flipper <fli...@fish.net> wrote:
>
>> On Fri, 06 Jul 2012 16:16:26 -0500, John Byrns <byr...@sbcglobal.net>
>> wrote:
>>
>> >First let me check to see that I correctly understand where the "craters"
>> >you
>> >are observing actually are on the transmitted waveform, as I am interpreting
>> >what you said they are on the positive modulation peaks, is that correct?
>>
>> Yep.
>>
>> >I still think incorrect plate loading is the cause of this problem,
>>
>> What kind of "incorrect plate loading" do you mean?
>
>I mean that the load resistance reflected from the antenna/ground system to the
>tube plate is too high.
Beamus, meaning with the 2.2k Rk, only did 90 Vp at 100% mod, leaving
65 V to cathode, since B+ came in lower at the higher current draw. On
the '100 mW' Beamus B+ is upwards to 190 V and the lower plate current
doesn't produce as much swing, of course.
The heptode/DC pentode version was doing essentially the same thing
because the load and plate current was the same, by intent.
At the moment I'm not quite sure how much because, somehow, I've
screwed the dern thing up, I mean the reassembled '100 mW' Beamus,
and, after 8 hours of rechecking every connection and component, have
not a blooming clue what's wrong with it.
The only reason I mentioned it is, since I have no idea what nor when
things went wrong, no measurements done since 'converting' it to the
heptode, and back, are reliable.
Odds are I didn't something 'stupid' when restoring it back to the
6ME8 but I'll be damned if I can find it.
>> >Another issue that may be compounding the problem is that the G1 control is
>> >presumably like a single ended class A audio amplifier which produces a fair
>> >amount of second harmonic distortion. IIRC, and I hope I don't have this
>> >backwards, the tube output on negative G1 excursions is compressed relative
>> >to
>> >positive G1 excursions. If I have this the right way around that would
>> >imply
>> >that if you adjust for 90% negative modulation, then the positive modulation
>> >peaks may be considerably greater than 90% and may be limited by "bumping
>> >into
>> >the cathode". If this is the case, correct plate loading should fix the
>> >problem.
>>
>> RF amplitude was no where close to 'bumping into the cathode." I wish
>> it were because that could be easily fixed and would mean I was
>> getting maximum power generated.
>
>"Bumping into the cathode" was a bad choice of words on my part and doesn't
>correctly reflect what I was trying to say, however I will leave it at that
>since I am accepting your analysis further down.
voltage either. Near as I can recall maximum swing at 100% mod was on
the order of 50V peak so there was gobs of B+ left.
>> >> >Is the plate loading set correctly? Incorrect plate loading would cause
>> >> >an
>> >> >effect similar to what you are describing.
>> >>
>> >> Plate load was fine, and peaked.
>> >
>> >I understand the "peaked" part, but how are you determining that the "plate
>> >load
>> >was fine"? Correct plate loading is important on these grid modulated
>> >transmitters, unlike plate modulated transmitters which are more forgiving
>> >of
>> >incorrect plate loading. I have not seen anyone, on radiomuseum, or
>> >elsewhere,
>> >delve into the loading issues with these small transmitters. I thought
>> >Robert
>> >Weaver was going to get into it in his C-Quam article where he calculates
>> >the
>> >impedance of a short monopole antenna, but he balked and didn't discuss the
>> >loading issue.
>>
>> I don't think it comes up because all of these things end up with
>> relatively low plate loads.
>
>If it is the case that "all of these things end up with relatively low plate
>loads", why do so many of these circuits have to add a resistor across the plate
>tank circuit?
broadcasters that do, including Norm's original 'power monster' 6888
version.
dual control pentode's relative low plate impedance limiting Q and
perhaps it could be more of an issue with heptodes.
John Byrns
Jul 8, 2012, 12:47:38 AM7/8/12
to
In article <bl0iv758cvt3018nb...@4ax.com>,
Ops, Sorry, I made an editing mistake and pasted in the wrong text making for
the repeat of the same text from above. The last bit was supposed to be:
"OK, since you are confident in your analysis I will accept it. Given that is
the problem wouldn't replacing the "G3 grid leak" resistor with an RFC between
G3 and the cathode fix the problem?"
the repeat of the same text from above. The last bit was supposed to be:
"OK, since you are confident in your analysis I will accept it. Given that is
the problem wouldn't replacing the "G3 grid leak" resistor with an RFC between
G3 and the cathode fix the problem?"
flipper
Jul 8, 2012, 12:55:04 AM7/8/12
to
On Sat, 07 Jul 2012 22:17:00 -0500, John Byrns <byr...@sbcglobal.net>
I don't remember in which of his posts I read it but modulation was
effectively '200%' so that would be 6 dB, if AGC fully compensated.
>> > I have come up with two simple poor man's
>> >schemes for implementing my second form of DCC, or "AMC".
>>
>> Are we trying to save power on a home broadcaster?
>
>No, but since I had originally mentioned two forms of DCC I felt compelled to
>offer an idea for applying the second form to a home broadcaster even if doing
>so might be counter productive. The simple one tube compressor mentioned above
>might ameliorate the problem though, although it would make more sense to simply
>apply the simple one tube compressor to a standard home broadcaster.
Yes. I'm normally all for 'simple' and 'clever' solutions but, in this
case, think I'd prefer a 'separate' compressor. For one, since I
normally use a PC for audio I can use much more sophisticated software
compression.
I though it sounded great when first tried but, you know, I think I
prefer the 'weaker' uncompressed playback. At least I have a choice
when they're separate.
>> > One is a simple
>> >modification of your "Beam Me Up Scotty" transmitter that actually saves a
>> >few
>> >parts.
>>
>> I'd love to hear it. Are you thinking maybe a grid clamp on G1?
>> Although, I don't see that saving parts.
>
>Yes, I was thinking of the grid clamp idea like weaver uses, except applied to
>the 6ME8 G1 where it would have the opposite effect of Weaver's grid clamp on
>the audio preamp tube. This assumes that the 6ME8 operating conditions could be
>suitably tweaked, perhaps G2 voltage, to compensate for the changed G1
>conditions. The parts saving would be the elimination of the 6ME8 cathode
>resistor and feedback network.
Pulling out feedback is a problem because linearity sucks without it.
>> > My second approach, which I prefer, is a slight modification of Weaver's
>> >transmitter that requires a few extra Rs & Cs, plus a diode.
>> >
>> >Where Weaver's poor man's DAM scheme causes compression of the received
>> >audio,
>> >my poor man's AMC scheme would cause expansion of the received audio.
>>
>> Yeah, and I'm not sure what the audible result would be because, as
>> Weaver points out, part of the 'problem' with home broadcasters is the
>> lack of compression causes them to sound 'weaker' than a 'big boy'
>> broadcast since, to stay below serious clipping, the average program
>> level has to be lower.
>>
>> The missing element is AGC action so I don't know if a "poor man's
>> AMC" would 'enhance' dynamic range or just make it sound even weaker.
>
>Yes, I agree, as I said it was just an exercise to go along with my mention of
>the second form of DCC. I think just adding a simple one tube compressor to the
>standard home transmitter is the way to go.
Me too. Or the software approach, although I suppose it's kind of
weird to be all hot on 'tube' transmitters and then stick a DSP in the
chain.
>> > Of course
>> >in the real rich man's systems appropriate expansion or compression is
>> >applied
>> >to the audio at the transmitter to eliminate the compression or expansion
>> >effects of the poor man's versions.
>>
>> Yes, but that was the point of Weaver's design: actual audible
>> 'compression' reducing the dynamic range so the end result sounds
>> 'fuller' and 'louder' like 'the big boys' do with direct audio
>> compression of the source material. In his case you don't want to
>> 'compensate' for it because he's not trying to do a "poor man's DAM"
>> but a "poor man's audio compressor." Which makes it doubly clever in
>> using DCC to accomplish a different job.
>
>Very True.
>
>> If I understand correctly AMC must be 'further compressing'
>> (relatively speaking) the already compressed audio because, after
>> receiver AGC 'compensates' for it, you still want the audio compressed
>> like it would be from a 'normal' transmitter.
>
>Yes, that is my understanding also, assuming I am parsing your words correctly.
Its really kind of amazing just how mangled AM transmissions are.
effectively '200%' so that would be 6 dB, if AGC fully compensated.
>> > I have come up with two simple poor man's
>> >schemes for implementing my second form of DCC, or "AMC".
>>
>> Are we trying to save power on a home broadcaster?
>
>No, but since I had originally mentioned two forms of DCC I felt compelled to
>offer an idea for applying the second form to a home broadcaster even if doing
>so might be counter productive. The simple one tube compressor mentioned above
>might ameliorate the problem though, although it would make more sense to simply
>apply the simple one tube compressor to a standard home broadcaster.
case, think I'd prefer a 'separate' compressor. For one, since I
normally use a PC for audio I can use much more sophisticated software
compression.
I though it sounded great when first tried but, you know, I think I
prefer the 'weaker' uncompressed playback. At least I have a choice
when they're separate.
>> > One is a simple
>> >modification of your "Beam Me Up Scotty" transmitter that actually saves a
>> >few
>> >parts.
>>
>> I'd love to hear it. Are you thinking maybe a grid clamp on G1?
>> Although, I don't see that saving parts.
>
>Yes, I was thinking of the grid clamp idea like weaver uses, except applied to
>the 6ME8 G1 where it would have the opposite effect of Weaver's grid clamp on
>the audio preamp tube. This assumes that the 6ME8 operating conditions could be
>suitably tweaked, perhaps G2 voltage, to compensate for the changed G1
>conditions. The parts saving would be the elimination of the 6ME8 cathode
>resistor and feedback network.
>> > My second approach, which I prefer, is a slight modification of Weaver's
>> >transmitter that requires a few extra Rs & Cs, plus a diode.
>> >
>> >Where Weaver's poor man's DAM scheme causes compression of the received
>> >audio,
>> >my poor man's AMC scheme would cause expansion of the received audio.
>>
>> Yeah, and I'm not sure what the audible result would be because, as
>> Weaver points out, part of the 'problem' with home broadcasters is the
>> lack of compression causes them to sound 'weaker' than a 'big boy'
>> broadcast since, to stay below serious clipping, the average program
>> level has to be lower.
>>
>> The missing element is AGC action so I don't know if a "poor man's
>> AMC" would 'enhance' dynamic range or just make it sound even weaker.
>
>Yes, I agree, as I said it was just an exercise to go along with my mention of
>the second form of DCC. I think just adding a simple one tube compressor to the
>standard home transmitter is the way to go.
weird to be all hot on 'tube' transmitters and then stick a DSP in the
chain.
>> > Of course
>> >in the real rich man's systems appropriate expansion or compression is
>> >applied
>> >to the audio at the transmitter to eliminate the compression or expansion
>> >effects of the poor man's versions.
>>
>> Yes, but that was the point of Weaver's design: actual audible
>> 'compression' reducing the dynamic range so the end result sounds
>> 'fuller' and 'louder' like 'the big boys' do with direct audio
>> compression of the source material. In his case you don't want to
>> 'compensate' for it because he's not trying to do a "poor man's DAM"
>> but a "poor man's audio compressor." Which makes it doubly clever in
>> using DCC to accomplish a different job.
>
>Very True.
>
>> If I understand correctly AMC must be 'further compressing'
>> (relatively speaking) the already compressed audio because, after
>> receiver AGC 'compensates' for it, you still want the audio compressed
>> like it would be from a 'normal' transmitter.
>
>Yes, that is my understanding also, assuming I am parsing your words correctly.
flipper
Jul 8, 2012, 1:52:28 AM7/8/12
to
On Sat, 07 Jul 2012 23:47:38 -0500, John Byrns <byr...@sbcglobal.net>
Oh, okay. Well, things are a bit confusing since I've now got the 'its
broke' problem, so I can't be 100% sure about its behavior when I
tried putting the G3 grid leak to cathode, but I'm not sure what the
suggested RFC is intended to 'fix'. With G3 grid leak to ground the
problem is audio on G3 and an RFC isn't going to 'help' there because
it's not in the G3-cathode path, which is 'through the tube'. I don't
think the 'its broke' problem has any impact on this aspect because
the analysis 'is clear' whether you even build it or not.
With G3 grid leak on the cathode I didn't note an 'RF problem' but
can't say much else because I don't know if things were 'right'. I
mean, in theory it had a chance of working, and the results were
'close', so whatever is 'broke' may have affected the result.
However, I *think* the heptode/dc pentode version was 'correct' and
that the 'mistake' came in restoring the 6ME8 but it could have also
happened when I tried putting the 6ME8 deflector bias to cathode, like
I indicated I might try. I don't know which since I didn't 'retest'
Beamus after removing the heptode/dc pentode, because 'we know Beamus
works' <rolling eyes>. So then I tried the Beamus deflector to cathode
test and the deflector behavior looked as expected with lovely RF, en
block, moving up and down with the audio. But there was a rather large
'floor' to negative mod, below which it would not go. I'm not quite
sure why but suspect stored charge on the bias bypass cap, which is
needed to 'feed through' full audio amplitude to the bias point. Also,
deflector bias behaved oddly. In the 'normal' version there is a point
of 'peak output amplitude' between 35V and 40V but with the revised
circuit it just kept on growing and growing as bias was increased, all
the way to 100 V, as if they were acting as a screen. Anyway, I didn't
see a 'quick fix' and, so, restored it back to 'normal', or so I
thought, but now there's compression of positive mod.
All bias voltages look right, RF amplitude is right, I've swapped
tubes, and even re-ohmed every resistor in the circuit. Everything
looks perfectly fine, except for not working right.
I'm thinking of taking a break to see if the alternate reality
twilight zone black hole that's apparently engulfed my workroom will
drift off to harass someone else.
broke' problem, so I can't be 100% sure about its behavior when I
tried putting the G3 grid leak to cathode, but I'm not sure what the
suggested RFC is intended to 'fix'. With G3 grid leak to ground the
problem is audio on G3 and an RFC isn't going to 'help' there because
it's not in the G3-cathode path, which is 'through the tube'. I don't
think the 'its broke' problem has any impact on this aspect because
the analysis 'is clear' whether you even build it or not.
With G3 grid leak on the cathode I didn't note an 'RF problem' but
can't say much else because I don't know if things were 'right'. I
mean, in theory it had a chance of working, and the results were
'close', so whatever is 'broke' may have affected the result.
However, I *think* the heptode/dc pentode version was 'correct' and
that the 'mistake' came in restoring the 6ME8 but it could have also
happened when I tried putting the 6ME8 deflector bias to cathode, like
I indicated I might try. I don't know which since I didn't 'retest'
Beamus after removing the heptode/dc pentode, because 'we know Beamus
works' <rolling eyes>. So then I tried the Beamus deflector to cathode
test and the deflector behavior looked as expected with lovely RF, en
block, moving up and down with the audio. But there was a rather large
'floor' to negative mod, below which it would not go. I'm not quite
sure why but suspect stored charge on the bias bypass cap, which is
needed to 'feed through' full audio amplitude to the bias point. Also,
deflector bias behaved oddly. In the 'normal' version there is a point
of 'peak output amplitude' between 35V and 40V but with the revised
circuit it just kept on growing and growing as bias was increased, all
the way to 100 V, as if they were acting as a screen. Anyway, I didn't
see a 'quick fix' and, so, restored it back to 'normal', or so I
thought, but now there's compression of positive mod.
All bias voltages look right, RF amplitude is right, I've swapped
tubes, and even re-ohmed every resistor in the circuit. Everything
looks perfectly fine, except for not working right.
I'm thinking of taking a break to see if the alternate reality
twilight zone black hole that's apparently engulfed my workroom will
drift off to harass someone else.
Patrick Turner
Jul 8, 2012, 6:51:40 AM7/8/12
to
Fipper mentioned something everyone wanting hifi from AM should
know :-
> Its really kind of amazing just how mangled AM transmissions are.
are definately more natural true to life sounding than others. We have
a couple of them here which make owning an AM radio worthwhile.
I don't have a huge amount of time to occupy my brain about the 1,001
issues surrounding fascinating use of dual deflection plate pentodes.
If I ever get around to using one, I'll just try something that's
already been done and measure/observe/think/tweak as one does while
wanting to farnarkle around in one's shed because there's FAE2doo.
BUT, I suddenly remembered the type number of a couple of such tubes
given to me about 7 years ago just before when a dear radio ham friend
of 72 died. The tube is the 7360, and if you look it up in TDSL you
find a pdf with a circuit for it.
http://www.tubebooks.org/tubedata/HB-3/Receiving-Type_Industrial_Tubes/7360.PDF
Patrick Turner.
John Byrns
Jul 9, 2012, 6:41:55 PM7/9/12
to
In article <io3iv7pt93rcfo1ln...@4ax.com>,
Robert Weaver said in the "Update" to his "A One-Tube Controlled Carrier AM
Transmitter":
"From this trace I made the following measurements:"
"Upward modulation = (7.0 - 2.2)/2.2 * 100 = 218%"
"Downward modulation = (2.2 - 0.2)/2.2 * 100 = 91%"
"The average of the upward and downward modulation is then 154%"
Is this where you got the "200%" figure, or was it elsewhere?
I don't think you can infer, from the modulation percentage alone, the degree of
compression produced by Weaver's scheme in the receiver. Weaver's scheme is
linear with respect to the audio applied to the transmitter; it is the change in
carrier level that produces the compression in the receiver. Using the scope
picture from which Weaver calculated the modulation percentages above, one can
estimate the carrier level at full modulation is about 3.5 divisions, while the
unmodulated carrier is 2.2 divisions in the picture. If the receiver AGC is
fully effective then this would equate to compression of 4 dB.
While Weaver's scheme is clever, it's a good thing it doesn't add any complexity
to his transmitter, as a maximum of only 4 dB compression isn't much, and since
the actual compression occurs in the receiver it doesn't do anything for the
received signal to noise ratio either.
A simple one tube compressor could provide considerably more than 4 dB of
compression and would actually improve the received signal to noise ratio.
> >No, but since I had originally mentioned two forms of DCC I felt compelled
> >to
> >offer an idea for applying the second form to a home broadcaster even if
> >doing
> >so might be counter productive. The simple one tube compressor mentioned
> >above
> >might ameliorate the problem though, although it would make more sense to
> >simply
> >apply the simple one tube compressor to a standard home broadcaster.
>
> Yes. I'm normally all for 'simple' and 'clever' solutions but, in this
> case, think I'd prefer a 'separate' compressor. For one, since I
> normally use a PC for audio I can use much more sophisticated software
> compression.
>
> I though it sounded great when first tried but, you know, I think I
> prefer the 'weaker' uncompressed playback. At least I have a choice
> when they're separate.
Weaver's transmitter could provide a "choice" with simple modification. By
placing the parallel combination of a resistor, capacitor, and switch in the
cathode circuit of the triode modulator section, one would have Weaver's DCC
operation when the switch is closed, grounding the cathode, and normal constant
carrier operation when the switch is open changing the modulator bias to
eliminate the grid clipping.
Transmitter":
"From this trace I made the following measurements:"
"Upward modulation = (7.0 - 2.2)/2.2 * 100 = 218%"
"Downward modulation = (2.2 - 0.2)/2.2 * 100 = 91%"
"The average of the upward and downward modulation is then 154%"
Is this where you got the "200%" figure, or was it elsewhere?
I don't think you can infer, from the modulation percentage alone, the degree of
compression produced by Weaver's scheme in the receiver. Weaver's scheme is
linear with respect to the audio applied to the transmitter; it is the change in
carrier level that produces the compression in the receiver. Using the scope
picture from which Weaver calculated the modulation percentages above, one can
estimate the carrier level at full modulation is about 3.5 divisions, while the
unmodulated carrier is 2.2 divisions in the picture. If the receiver AGC is
fully effective then this would equate to compression of 4 dB.
While Weaver's scheme is clever, it's a good thing it doesn't add any complexity
to his transmitter, as a maximum of only 4 dB compression isn't much, and since
the actual compression occurs in the receiver it doesn't do anything for the
received signal to noise ratio either.
A simple one tube compressor could provide considerably more than 4 dB of
compression and would actually improve the received signal to noise ratio.
> >No, but since I had originally mentioned two forms of DCC I felt compelled
> >to
> >offer an idea for applying the second form to a home broadcaster even if
> >doing
> >so might be counter productive. The simple one tube compressor mentioned
> >above
> >might ameliorate the problem though, although it would make more sense to
> >simply
> >apply the simple one tube compressor to a standard home broadcaster.
>
> Yes. I'm normally all for 'simple' and 'clever' solutions but, in this
> case, think I'd prefer a 'separate' compressor. For one, since I
> normally use a PC for audio I can use much more sophisticated software
> compression.
>
> I though it sounded great when first tried but, you know, I think I
> prefer the 'weaker' uncompressed playback. At least I have a choice
> when they're separate.
placing the parallel combination of a resistor, capacitor, and switch in the
cathode circuit of the triode modulator section, one would have Weaver's DCC
operation when the switch is closed, grounding the cathode, and normal constant
carrier operation when the switch is open changing the modulator bias to
eliminate the grid clipping.
flipper
Jul 9, 2012, 8:05:51 PM7/9/12
to
On Sun, 8 Jul 2012 03:51:40 -0700 (PDT), Patrick Turner
<in...@turneraudio.com.au> wrote:
>Fipper mentioned something everyone wanting hifi from AM should
>know :-
Anyone wanting 'hi-fi' from existing AM is in trouble long before
<in...@turneraudio.com.au> wrote:
>Fipper mentioned something everyone wanting hifi from AM should
>know :-
issues of compression come up because the S/N ratio and allocated
bandwidth are simply insufficient to achieve any reasonable definition
of the term.
You can, however, manage results better than the 'typical' AM radio,
sometimes referred to as 'wideband AM' or some other euphemism, but to
say 'hi-fi' is wishful thinking.
>> Its really kind of amazing just how mangled AM transmissions are.
>
>Indeed. So many stations are not so good to listen to. But then some
>are definately more natural true to life sounding than others. We have
>a couple of them here which make owning an AM radio worthwhile.
compression. FM is compressed and, unfortunately, so are most 'rock
and roll' CDs. Apparently digital radio is as well.
Having said that, there are 'pro' arguments to compressing, such as
the relatively low S/N ratio of AM, ambient noise level in automobiles
and various 'public' venues, etc. but 'hi-hi' isn't one of them.
What I was hinting at, though, without being explicit, is that AMC
layers 'more distortion' on top of existing distortion and the claimed
result is 'not much worse'. Well, you can 'not much worse' yourself
into garbage if your comparison is always against the last 'not much
worse' degradation over the previous 'not much worse'. For example, AM
audio bandwidth in the US used to be 15 KHz but was reduced in 1989 to
10.2 KHz which, I imagine, was subjectively 'not much worse', at least
to whatever they deemed to be the 'average listener', but it
effectively killed any notion of 'hi-fi', or semi 'hi-fi', AM so we're
now comparing 'not much worse' against "why bother trying?"
Note, though, that means there's actually a useful purpose, beyond
simply choosing your own program material, to a 'home broadcaster'
because, under U.S. Part 15 rules, you're pretty much free to transmit
whatever you like as long as its confined to your own domain and does
not interfere with licensed broadcasters.
So, in effect, you're free to broadcast something that has a fighting
chance of resembling 'hi-hi' AM, including uncompressed if one so
desires, to your old, much wider bandwidth, AM radios.
I had that in mind with Beamus and you'll notice it measured flat to
18 KHz, so it is capable of as much 'fi' as AM can produce.
>I don't have a huge amount of time to occupy my brain about the 1,001
>issues surrounding fascinating use of dual deflection plate pentodes.
>If I ever get around to using one, I'll just try something that's
>already been done and measure/observe/think/tweak as one does while
>wanting to farnarkle around in one's shed because there's FAE2doo.
>
>BUT, I suddenly remembered the type number of a couple of such tubes
>given to me about 7 years ago just before when a dear radio ham friend
>of 72 died. The tube is the 7360, and if you look it up in TDSL you
>find a pdf with a circuit for it.
>
>http://www.tubebooks.org/tubedata/HB-3/Receiving-Type_Industrial_Tubes/7360.PDF
>
>Patrick Turner.
want some more schematics to ponder copying then try this one.
http://jlandrigan.com/files/Receivers/SSB%20Exciter%20Circuits%20Using%20the%207360.pdf
Of potential interest, it has a product detector at the bottom.
You might want to consider selling those on Ebay and tinkering with
less expensive tubes because the 7360 is in short supply these days
and commands prices in the 30 buck range. That, of course, is why I'm
using the $1 6ME8.
Patrick Turner
Jul 9, 2012, 8:32:36 PM7/9/12
to
On Jul 10, 10:05 am, flipper <flip...@fish.net> wrote:
> On Sun, 8 Jul 2012 03:51:40 -0700 (PDT), Patrick Turner
>
> On Sun, 8 Jul 2012 03:51:40 -0700 (PDT), Patrick Turner
>
broadcasters wanting more stations.
But Oz is a big place, so 2CY in Canberra can broadcast with 9kHz of
audio BW without being heard where some other station is on the same
F, or be heard where some other station is only 9kHz away, because
side bands extend out to +/-9 kHz each side of the carrier F.
> Note, though, that means there's actually a useful purpose, beyond
> simply choosing your own program material, to a 'home broadcaster'
> because, under U.S. Part 15 rules, you're pretty much free to transmit
> whatever you like as long as its confined to your own domain and does
> not interfere with licensed broadcasters.
>
> So, in effect, you're free to broadcast something that has a fighting
> chance of resembling 'hi-hi' AM, including uncompressed if one so
> desires, to your old, much wider bandwidth, AM radios.
>
> I had that in mind with Beamus and you'll notice it measured flat to
> 18 KHz, so it is capable of as much 'fi' as AM can produce.
>
> >I don't have a huge amount of time to occupy my brain about the 1,001
> >issues surrounding fascinating use of dual deflection plate pentodes.
> >If I ever get around to using one, I'll just try something that's
> >already been done and measure/observe/think/tweak as one does while
> >wanting to farnarkle around in one's shed because there's FAE2doo.
>
> >BUT, I suddenly remembered the type number of a couple of such tubes
> >given to me about 7 years ago just before when a dear radio ham friend
> >of 72 died. The tube is the 7360, and if you look it up in TDSL you
> >find a pdf with a circuit for it.
>
> >http://www.tubebooks.org/tubedata/HB-3/Receiving-Type_Industrial_Tube...
> >I don't have a huge amount of time to occupy my brain about the 1,001
> >issues surrounding fascinating use of dual deflection plate pentodes.
> >If I ever get around to using one, I'll just try something that's
> >already been done and measure/observe/think/tweak as one does while
> >wanting to farnarkle around in one's shed because there's FAE2doo.
>
> >BUT, I suddenly remembered the type number of a couple of such tubes
> >given to me about 7 years ago just before when a dear radio ham friend
> >of 72 died. The tube is the 7360, and if you look it up in TDSL you
> >find a pdf with a circuit for it.
>
>
> >Patrick Turner.
>
> Thank you but I already have it in my 'beam tubes' folder. But if you
> want some more schematics to ponder copying then try this one.
>
> http://jlandrigan.com/files/Receivers/SSB%20Exciter%20Circuits%20Usin...
> >Patrick Turner.
>
> Thank you but I already have it in my 'beam tubes' folder. But if you
> want some more schematics to ponder copying then try this one.
>
>
> Of potential interest, it has a product detector at the bottom.
>
> You might want to consider selling those on Ebay and tinkering with
> less expensive tubes because the 7360 is in short supply these days
> and commands prices in the 30 buck range. That, of course, is why I'm
> using the $1 6ME8.
Although poor as a church mouse, I could afford a 7360 if I wanted
> Of potential interest, it has a product detector at the bottom.
>
> You might want to consider selling those on Ebay and tinkering with
> less expensive tubes because the 7360 is in short supply these days
> and commands prices in the 30 buck range. That, of course, is why I'm
> using the $1 6ME8.
one.
I wonder what use they could be in a direct conversion receiver or in
an MPX stereo decoder get the L-R signal.
Probably my couple of 7360 are old pulls, not sure, but maybe some NOS
here in Oz somewhere.
I still have work to finish on audio gear before I go "upstairs" -
ie, to work on RF. I also like riding a bike a bit these days.
Patrick Turner.
flipper
Jul 9, 2012, 9:06:49 PM7/9/12
to
On Mon, 09 Jul 2012 17:41:55 -0500, John Byrns <byr...@sbcglobal.net>
Looks like I remembered the 218% and forgot the rest.
>I don't think you can infer, from the modulation percentage alone, the degree of
>compression produced by Weaver's scheme in the receiver. Weaver's scheme is
>linear with respect to the audio applied to the transmitter; it is the change in
>carrier level that produces the compression in the receiver. Using the scope
>picture from which Weaver calculated the modulation percentages above, one can
>estimate the carrier level at full modulation is about 3.5 divisions, while the
>unmodulated carrier is 2.2 divisions in the picture. If the receiver AGC is
>fully effective then this would equate to compression of 4 dB.
Well, that's pretty close to taking the modulation percentage.
4 dB doesn't sound like much but its apparently enough that people who
listen to it are 'impressed'. Maybe he gets more 'effective'
compression with the 'soft limiting' allowing more mod than the 4 dB
implies.
If I remember correctly, when playing with software compression more
than 6 dB was noticeably irritating, or tiring, to me.
>While Weaver's scheme is clever, it's a good thing it doesn't add any complexity
>to his transmitter, as a maximum of only 4 dB compression isn't much, and since
>the actual compression occurs in the receiver it doesn't do anything for the
>received signal to noise ratio either.
I don't think that kind of carrier control *can* improve the signal to
noise ratio.
>A simple one tube compressor could provide considerably more than 4 dB of
>compression and would actually improve the received signal to noise ratio.
That would be because it's compressing the source audio going into the
carrier but the 'noise reduction' is, again, a 'receiver function'. I
mean, since compression moves the average audio level upward the
volume control can be turned downward, which turns down the noise.
>> >No, but since I had originally mentioned two forms of DCC I felt compelled
>> >to
>> >offer an idea for applying the second form to a home broadcaster even if
>> >doing
>> >so might be counter productive. The simple one tube compressor mentioned
>> >above
>> >might ameliorate the problem though, although it would make more sense to
>> >simply
>> >apply the simple one tube compressor to a standard home broadcaster.
>>
>> Yes. I'm normally all for 'simple' and 'clever' solutions but, in this
>> case, think I'd prefer a 'separate' compressor. For one, since I
>> normally use a PC for audio I can use much more sophisticated software
>> compression.
>>
>> I though it sounded great when first tried but, you know, I think I
>> prefer the 'weaker' uncompressed playback. At least I have a choice
>> when they're separate.
>
>Weaver's transmitter could provide a "choice" with simple modification. By
>placing the parallel combination of a resistor, capacitor, and switch in the
>cathode circuit of the triode modulator section, one would have Weaver's DCC
>operation when the switch is closed, grounding the cathode, and normal constant
>carrier operation when the switch is open changing the modulator bias to
>eliminate the grid clipping.
Yeah, I thought of that but then the issue of two tubes operating open
loop comes up. With compression active it's 'a necessary sacrifice'
but without it I'd like to put some effort into lowering distortion.
On the other hand, lowering transmitter distortion only to fed
distorted audio into it seems a bit odd as well.
>I don't think you can infer, from the modulation percentage alone, the degree of
>compression produced by Weaver's scheme in the receiver. Weaver's scheme is
>linear with respect to the audio applied to the transmitter; it is the change in
>carrier level that produces the compression in the receiver. Using the scope
>picture from which Weaver calculated the modulation percentages above, one can
>estimate the carrier level at full modulation is about 3.5 divisions, while the
>unmodulated carrier is 2.2 divisions in the picture. If the receiver AGC is
>fully effective then this would equate to compression of 4 dB.
4 dB doesn't sound like much but its apparently enough that people who
listen to it are 'impressed'. Maybe he gets more 'effective'
compression with the 'soft limiting' allowing more mod than the 4 dB
implies.
If I remember correctly, when playing with software compression more
than 6 dB was noticeably irritating, or tiring, to me.
>While Weaver's scheme is clever, it's a good thing it doesn't add any complexity
>to his transmitter, as a maximum of only 4 dB compression isn't much, and since
>the actual compression occurs in the receiver it doesn't do anything for the
>received signal to noise ratio either.
noise ratio.
>A simple one tube compressor could provide considerably more than 4 dB of
>compression and would actually improve the received signal to noise ratio.
carrier but the 'noise reduction' is, again, a 'receiver function'. I
mean, since compression moves the average audio level upward the
volume control can be turned downward, which turns down the noise.
>> >No, but since I had originally mentioned two forms of DCC I felt compelled
>> >to
>> >offer an idea for applying the second form to a home broadcaster even if
>> >doing
>> >so might be counter productive. The simple one tube compressor mentioned
>> >above
>> >might ameliorate the problem though, although it would make more sense to
>> >simply
>> >apply the simple one tube compressor to a standard home broadcaster.
>>
>> Yes. I'm normally all for 'simple' and 'clever' solutions but, in this
>> case, think I'd prefer a 'separate' compressor. For one, since I
>> normally use a PC for audio I can use much more sophisticated software
>> compression.
>>
>> I though it sounded great when first tried but, you know, I think I
>> prefer the 'weaker' uncompressed playback. At least I have a choice
>> when they're separate.
>
>Weaver's transmitter could provide a "choice" with simple modification. By
>placing the parallel combination of a resistor, capacitor, and switch in the
>cathode circuit of the triode modulator section, one would have Weaver's DCC
>operation when the switch is closed, grounding the cathode, and normal constant
>carrier operation when the switch is open changing the modulator bias to
>eliminate the grid clipping.
loop comes up. With compression active it's 'a necessary sacrifice'
but without it I'd like to put some effort into lowering distortion.
On the other hand, lowering transmitter distortion only to fed
distorted audio into it seems a bit odd as well.
flipper
Jul 10, 2012, 1:28:10 AM7/10/12
to
On Sun, 08 Jul 2012 00:52:28 -0500, flipper <fli...@fish.net> wrote:
<snip>
>
>All bias voltages look right, RF amplitude is right, I've swapped
>tubes, and even re-ohmed every resistor in the circuit. Everything
>looks perfectly fine, except for not working right.
>
>I'm thinking of taking a break to see if the alternate reality
>twilight zone black hole that's apparently engulfed my workroom will
>drift off to harass someone else.
>
Ya know, if I didn't have scope pictures I'd be wondering if it ever
worked right.
Turns out it works at 1150 KHz, or thereabouts, but there is a
HUUUUUUUUge difference in drive level needed for 100% mod from the low
to high end of the frequency range. It's almost as if, for some
inexplicable reason, preamp gain goes to nil when RF frequency is
adjusted to the low end of the band.
That makes no sense at all but at least it's a clue, and that's more
than I had before.
<snip>
>
>All bias voltages look right, RF amplitude is right, I've swapped
>tubes, and even re-ohmed every resistor in the circuit. Everything
>looks perfectly fine, except for not working right.
>
>I'm thinking of taking a break to see if the alternate reality
>twilight zone black hole that's apparently engulfed my workroom will
>drift off to harass someone else.
>
worked right.
Turns out it works at 1150 KHz, or thereabouts, but there is a
HUUUUUUUUge difference in drive level needed for 100% mod from the low
to high end of the frequency range. It's almost as if, for some
inexplicable reason, preamp gain goes to nil when RF frequency is
adjusted to the low end of the band.
That makes no sense at all but at least it's a clue, and that's more
than I had before.
flipper
Jul 10, 2012, 3:25:21 AM7/10/12
to
Beamus began as a 'since I've got an oscillator' project and that came
about from looking to design an amplitude stabilized OSC for one of
those dual control pentode transmitters.
The original design was a pentode, like now, but had a triode sensing
RF peak and, from that, controlling screen volts to maintain constant
OSC amplitude. Well, I also found that, when loaded (deflection plate
100k), OSC amplitude remained virtually constant over the whole 365 pF
tuning range even without the triode and, so, I didn't use it with
Beamus.
But, *surprise*, when I readjusted the OSC coil from 250 uH to 190 uH,
and padded the 365 pF cap with 100 pF to narrow the tuning range, OSC
amplitude now varies all over the blooming place, say 3 to 1 (maybe
more), with lowest amplitude at low frequency, which is why it works
at the top end but goes to hell at the low. There's simply not enough
RF amplitude to drive the deflectors so it can't 'peak power' the
output.
I didn't notice that, at first, because it was running on the top end
of the dial but, later, when 'experimenting' I switched to the low end
to find a quiet spot.
Boy, that much amplitude variation sure took me by surprise.
John Byrns
Jul 10, 2012, 9:38:52 PM7/10/12
to
In article <15jnv7hcnbjt8eso8...@4ax.com>,
I thought "Beamus" had an RF drive control to set the correct deflector drive
voltage? Am I confused, why didn't this control insure the correct drivevoltage
across the band? Maybe "Beamus" should have a multifunction meter that could be
switched to measure RF drive voltage to the deflectors, carrier output level,
and the 6ME8 cathode current, to facilitate setting up on different frequencies.
voltage? Am I confused, why didn't this control insure the correct drivevoltage
across the band? Maybe "Beamus" should have a multifunction meter that could be
switched to measure RF drive voltage to the deflectors, carrier output level,
and the 6ME8 cathode current, to facilitate setting up on different frequencies.
flipper
Jul 11, 2012, 12:37:26 AM7/11/12
to
On Tue, 10 Jul 2012 20:38:52 -0500, John Byrns <byr...@sbcglobal.net>
wrote:
It does have an 'RF drive' adjustment but that was intended to be an
'initial setup' thing and I consider it 'bad enough' you have to
re-tune the antenna on a frequency change.
> Maybe "Beamus" should have a multifunction meter that could be
>switched to measure RF drive voltage to the deflectors, carrier output level,
>and the 6ME8 cathode current, to facilitate setting up on different frequencies.
I'd rather it not be so complicated but even that is moot because the
amplitude variation is so huge that the 'RF drive' pot can't come even
close to compensating for it.
Btw, there's also a secondary effect in that RF OSC current drain is
significant so when amplitude drops B+ rises, which affects deflector
and screen bias. Its a beautiful example of Murphy's Law because as RF
amplitude drops, making it unable to drive 'peak power', B+ rise
increases the power level it needs to drive, because screen volts
increase. That made the LED 'antenna tuning indicator' give the
impression the low end had 'more output power' than the upper end,
because it did.
Those 'problems' will go away again when I solve the RF amplitude
issue, though.
Actually, it could be easily solved with different air variables (in
fact, I was thinking of redoing the antenna matching network anyway in
an attempt to reduce cap circulating currents), because the OSC works
perfectly fine with 250 uH (or 220 uH, I think), but now I'm curious
why lowering inductance made such a dramatic difference. I don't know
a heck of a lot about oscillators but suspect the lower inductance
doesn't have enough 'kick' to sustain both the tuning cap circulating
current and deflector bias resistor (load) current so maybe increasing
those 100ks to 220k might be enough to solve it. I also have another
idea, which is to balanced drive the deflectors off the OSC coil
'primary' but I'm not sure there's enough amplitude there. Anyway, the
problem is solvable, it's just a matter of finding a 'simple', cost
effective, one.
Having said that, B+ rise makes me wonder if I should consider
stabilizing it because that makes the two adjustments to some degree
interactive but maybe the range isn't significant enough to matter
when things are working 'right'. And maybe, adjust deflectors, adjust
RF amplitude, and then go back to 'fine tune' the deflector voltage
isn't enough of an 'extra step' to warrant adding complexity.
wrote:
'initial setup' thing and I consider it 'bad enough' you have to
re-tune the antenna on a frequency change.
> Maybe "Beamus" should have a multifunction meter that could be
>switched to measure RF drive voltage to the deflectors, carrier output level,
>and the 6ME8 cathode current, to facilitate setting up on different frequencies.
amplitude variation is so huge that the 'RF drive' pot can't come even
close to compensating for it.
Btw, there's also a secondary effect in that RF OSC current drain is
significant so when amplitude drops B+ rises, which affects deflector
and screen bias. Its a beautiful example of Murphy's Law because as RF
amplitude drops, making it unable to drive 'peak power', B+ rise
increases the power level it needs to drive, because screen volts
increase. That made the LED 'antenna tuning indicator' give the
impression the low end had 'more output power' than the upper end,
because it did.
Those 'problems' will go away again when I solve the RF amplitude
issue, though.
Actually, it could be easily solved with different air variables (in
fact, I was thinking of redoing the antenna matching network anyway in
an attempt to reduce cap circulating currents), because the OSC works
perfectly fine with 250 uH (or 220 uH, I think), but now I'm curious
why lowering inductance made such a dramatic difference. I don't know
a heck of a lot about oscillators but suspect the lower inductance
doesn't have enough 'kick' to sustain both the tuning cap circulating
current and deflector bias resistor (load) current so maybe increasing
those 100ks to 220k might be enough to solve it. I also have another
idea, which is to balanced drive the deflectors off the OSC coil
'primary' but I'm not sure there's enough amplitude there. Anyway, the
problem is solvable, it's just a matter of finding a 'simple', cost
effective, one.
Having said that, B+ rise makes me wonder if I should consider
stabilizing it because that makes the two adjustments to some degree
interactive but maybe the range isn't significant enough to matter
when things are working 'right'. And maybe, adjust deflectors, adjust
RF amplitude, and then go back to 'fine tune' the deflector voltage
isn't enough of an 'extra step' to warrant adding complexity.
flipper
Jul 13, 2012, 6:22:13 PM7/13/12
to
>Actually, it could be easily solved with different air variables (in
>fact, I was thinking of redoing the antenna matching network anyway in
>an attempt to reduce cap circulating currents), because the OSC works
>perfectly fine with 250 uH (or 220 uH, I think), but now I'm curious
>why lowering inductance made such a dramatic difference. I don't know
>a heck of a lot about oscillators but suspect the lower inductance
>doesn't have enough 'kick' to sustain both the tuning cap circulating
>current and deflector bias resistor (load) current so maybe increasing
>those 100ks to 220k might be enough to solve it. I also have another
>idea, which is to balanced drive the deflectors off the OSC coil
>'primary' but I'm not sure there's enough amplitude there. Anyway, the
>problem is solvable, it's just a matter of finding a 'simple', cost
>effective, one.
I haven't had a chance to do extensive testing but the 'simple'
>fact, I was thinking of redoing the antenna matching network anyway in
>an attempt to reduce cap circulating currents), because the OSC works
>perfectly fine with 250 uH (or 220 uH, I think), but now I'm curious
>why lowering inductance made such a dramatic difference. I don't know
>a heck of a lot about oscillators but suspect the lower inductance
>doesn't have enough 'kick' to sustain both the tuning cap circulating
>current and deflector bias resistor (load) current so maybe increasing
>those 100ks to 220k might be enough to solve it. I also have another
>idea, which is to balanced drive the deflectors off the OSC coil
>'primary' but I'm not sure there's enough amplitude there. Anyway, the
>problem is solvable, it's just a matter of finding a 'simple', cost
>effective, one.
solutions of changing deflector bias resistors and/or increasing
cathode current didn't have a significant effect. And, for some
reason, even with 250 uH RF amplitude now varies more over the 365pF
range than I originally measured. Also, strangely, the 'best'
deflector bias seems to have become 50 V, vs 35 V, like Weaver
suggested it should, and it looks like RF peak should 'match'. What I
haven't rechecked is that I swapped in a new 6ME8 when trying to find
out "what's wrong" and perhaps the 'optimum' bias point varies by
tube.
At any rate, it's the low end where amplitude drops off and, on top of
that, 50 Vp is near RF OSC maximum so there's not much room to
compensate.
I think the simplest solution is to change air variables, since I was
narrowing the range anyway, and a smaller cap is much less expensive.
For example, radiodaze has a 170 pF air variable for $3.00 vs the
'traditional' $15.00 for a 365 pF. Use the same for both OSC and
antenna tank and then trim the OSC range with the coil and add a cap
trimmer if we need to snug the OSC range slightly under the tank's
tuning range. That removes the need for a 100 pF padder to narrow the
range (making for more efficient use of the air variable) and it's the
'large capacitance' that seems to suffer OSC amplitude.
Considering antenna capacitance that should allow an almost 2 to 1
tuning range, which should be more than sufficient since the purpose
is only to find a 'quiet' spot on the dial (otherwise we'd use a
crystal) and, for antenna efficiency (lousy though it is), the upper
end is preferred.anyway.
I tell you one thing, though, when it's tuned up right she sure sounds
sweet (for AM). Actually, I can't tell 'by sound' when RF OSC
amplitude is low because I think that 'distortion' primarily shows up
at max mod (near clipping) nor is it terribly obvious on the scope
unless you're looking for it, which is why I didn't notice it for a
while.
Coincidentally, I discovered that one of my RF noise problems, in
addition to this location being an RF nightmare, was one of my PCs.
The dern thing splatters RF all over the place even when turned off, I
have to pull the plug to get it off the air, so I guess the power
supply is going out. Fortunately it's not the server.
John L Stewart
Jul 14, 2012, 5:45:27 AM7/14/12
to
Coincidentally, I discovered that one of my RF noise problems, in
addition to this location being an RF nightmare, was one of my PCs.
The dern thing splatters RF all over the place even when turned off, I
have to pull the plug to get it off the air, so I guess the power
supply is going out. Fortunately it's not the server.
of off. Not unusual at all. I found similar problems while testing some of
my experimental AF amps in the past.
Cheers, John
+-------------------------------------------------------------------+
|Filename: avatar50384_3_gif_pagespeed_ce_dAT16iGY8O.gif |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=293|
Don Pearce
Jul 14, 2012, 10:06:54 AM7/14/12
to
On Sat, 14 Jul 2012 09:45:27 +0000, John L Stewart
<John.L.Stew...@audiobanter.com> wrote:
>
>Coincidentally, I discovered that one of my RF noise problems, in
>addition to this location being an RF nightmare, was one of my PCs.
>The dern thing splatters RF all over the place even when turned off, I
>have to pull the plug to get it off the air, so I guess the power
>supply is going out. Fortunately it's not the server.
>
>It is the switch mode PS that generates lots of RFI, wether the PC is on
>of off. Not unusual at all. I found similar problems while testing some of
>my experimental AF amps in the past.
>
>Cheers, John
>
I frequently have to carry out RF approvals work on products, and this
<John.L.Stew...@audiobanter.com> wrote:
>
>Coincidentally, I discovered that one of my RF noise problems, in
>addition to this location being an RF nightmare, was one of my PCs.
>The dern thing splatters RF all over the place even when turned off, I
>have to pull the plug to get it off the air, so I guess the power
>supply is going out. Fortunately it's not the server.
>
>It is the switch mode PS that generates lots of RFI, wether the PC is on
>of off. Not unusual at all. I found similar problems while testing some of
>my experimental AF amps in the past.
>
>Cheers, John
>
is the situation I have found. PC power supplies get designed, for the
most part, quite well, following the chip manufacturer's application
note. The power supply is then given its EMC approval.
After that, knowing that the thing will never be looked at again, half
the suppression components are removed from the PCB. Take the lid off
yours and see how many bits are not fitted. The best thing you can do
is go into the after-market and find a decent replacement.
d
flipper
Jul 15, 2012, 2:34:10 AM7/15/12
to
>Coincidentally, I discovered that one of my RF noise problems, in
>addition to this location being an RF nightmare, was one of my PCs.
>The dern thing splatters RF all over the place even when turned off, I
>have to pull the plug to get it off the air, so I guess the power
>supply is going out. Fortunately it's not the server.
>
>It is the switch mode PS that generates lots of RFI, wether the PC is on
>of off. Not unusual at all. I found similar problems while testing some of
>my experimental AF amps in the past.
>
>Cheers, John
Thanks. Yes, I know its the switching power supply. That's why I
>addition to this location being an RF nightmare, was one of my PCs.
>The dern thing splatters RF all over the place even when turned off, I
>have to pull the plug to get it off the air, so I guess the power
>supply is going out. Fortunately it's not the server.
>
>It is the switch mode PS that generates lots of RFI, wether the PC is on
>of off. Not unusual at all. I found similar problems while testing some of
>my experimental AF amps in the past.
>
>Cheers, John
speculated it was going out.
I've got a dozen PCs here and that's the only one doing it but I don't
really know when it 'started' because its reserved for testing things
and doesn't always have even the power cord connected since that's
'shared', along the monitor, keyboard, mouse, et all, with another one
I also use for testing (one is Windows 7 and the other Windows XP).
I mentioned it because I imagine there might still be people who think
'off' really means 'off' but that hasn't been the case ever since the
'soft' power button.
It took me a while to pin that particular noise to the PC because
noise here is sometimes so bad I can barely receive even the strongest
local stations and I know that's nothing in the house because I've
tripped the main breaker and had the power company out. They seemed
'surprised' there was so much noise, and confirmed its coming in on
their lines, but then informed me the power company is only obligated
to provide 'power' and not 'noise free' power.
flipper
Jul 16, 2012, 12:57:35 AM7/16/12
to
On Sun, 15 Jul 2012 13:22:58 +0000, John L Stewart
<John.L.Stew...@audiobanter.com> wrote:
>
>flipper;959374 Wrote:
>> On Sat, 14 Jul 2012 09:45:27 +0000, John L Stewart
>> <John.L.Stew...@audiobanter.com> wrote:
>> -
>beautiful comb spectrum that can really get in the way as we try to
>chase down some kinds of problems.
>
>Good Luck, John
Yes, good point, and I do have a few CRT monitors left but they're
just occasionally powered up to do maintenance on a few machines and
are off 95% of the time. The rest are LCD but then the switching power
supply 'soft power' switch issue potentially crops up again, not to
mention the plethora of 'low power' switchers running everything from
routers to cell phone chargers.
Fact is, RF is never safe in this 'digital world' ;)
Amusingly enough, the first 'humongous noise' problem I had, way back
in the 'olden days', was a defective doorbell transformer.
That's why, for the 'acid test', I killed the main breaker because,
funny as it may sound, it's not a trivial task these days to track
down everything electrical/electronic. It's worse than an Easter egg
hunt and just when you think you've got them all, oh yeah, there's a
clock in the oven <rolling eyes> (or a doorbell transformer). It's
just a heck of a lot simpler to flip the mains. Of course, if that
does kill the noise it's in the house so then you can flip off all the
branches, bring the mains up, and reconnect branch by branch to
isolate in which section the culprit lives.
That doesn't do me any good when its coming in from the power company
but, fortunately, whoever it is 'out there' that's causing the noise
doesn't seem to be active of late.
<John.L.Stew...@audiobanter.com> wrote:
>
>flipper;959374 Wrote:
>> On Sat, 14 Jul 2012 09:45:27 +0000, John L Stewart
>> <John.L.Stew...@audiobanter.com> wrote:
>> >
>> >Coincidentally, I discovered that one of my RF noise problems, in
>> >addition to this location being an RF nightmare, was one of my PCs.
>> >The dern thing splatters RF all over the place even when turned off, I
>> >have to pull the plug to get it off the air, so I guess the power
>> >supply is going out. Fortunately it's not the server.
>> >
>> >It is the switch mode PS that generates lots of RFI, wether the PC is
>> on
>> >of off. Not unusual at all. I found similar problems while testing some
>> of
>> >my experimental AF amps in the past.
>> >
>> >Cheers, John-
>> >Coincidentally, I discovered that one of my RF noise problems, in
>> >addition to this location being an RF nightmare, was one of my PCs.
>> >The dern thing splatters RF all over the place even when turned off, I
>> >have to pull the plug to get it off the air, so I guess the power
>> >supply is going out. Fortunately it's not the server.
>> >
>> >It is the switch mode PS that generates lots of RFI, wether the PC is
>> on
>> >of off. Not unusual at all. I found similar problems while testing some
>> of
>> >my experimental AF amps in the past.
>> >
>>
>> Thanks. Yes, I know its the switching power supply. That's why I
>> speculated it was going out.
>>
>> I've got a dozen PCs here and that's the only one doing it but I don't
>> really know when it 'started' because its reserved for testing things
>> and doesn't always have even the power cord connected since that's
>> 'shared', along the monitor, keyboard, mouse, et all, with another one
>> I also use for testing (one is Windows 7 and the other Windows XP).
>>
>> I mentioned it because I imagine there might still be people who think
>> 'off' really means 'off' but that hasn't been the case ever since the
>> 'soft' power button.
>>
>> It took me a while to pin that particular noise to the PC because
>> noise here is sometimes so bad I can barely receive even the strongest
>> local stations and I know that's nothing in the house because I've
>> tripped the main breaker and had the power company out. They seemed
>> 'surprised' there was so much noise, and confirmed its coming in on
>> their lines, but then informed me the power company is only obligated
>> to provide 'power' and not 'noise free' power.
>
>Another common & obvious nusance is the CRT type display. Generates a
>> Thanks. Yes, I know its the switching power supply. That's why I
>> speculated it was going out.
>>
>> I've got a dozen PCs here and that's the only one doing it but I don't
>> really know when it 'started' because its reserved for testing things
>> and doesn't always have even the power cord connected since that's
>> 'shared', along the monitor, keyboard, mouse, et all, with another one
>> I also use for testing (one is Windows 7 and the other Windows XP).
>>
>> I mentioned it because I imagine there might still be people who think
>> 'off' really means 'off' but that hasn't been the case ever since the
>> 'soft' power button.
>>
>> It took me a while to pin that particular noise to the PC because
>> noise here is sometimes so bad I can barely receive even the strongest
>> local stations and I know that's nothing in the house because I've
>> tripped the main breaker and had the power company out. They seemed
>> 'surprised' there was so much noise, and confirmed its coming in on
>> their lines, but then informed me the power company is only obligated
>> to provide 'power' and not 'noise free' power.
>
>beautiful comb spectrum that can really get in the way as we try to
>chase down some kinds of problems.
>
>Good Luck, John
Yes, good point, and I do have a few CRT monitors left but they're
just occasionally powered up to do maintenance on a few machines and
are off 95% of the time. The rest are LCD but then the switching power
supply 'soft power' switch issue potentially crops up again, not to
mention the plethora of 'low power' switchers running everything from
routers to cell phone chargers.
Fact is, RF is never safe in this 'digital world' ;)
Amusingly enough, the first 'humongous noise' problem I had, way back
in the 'olden days', was a defective doorbell transformer.
That's why, for the 'acid test', I killed the main breaker because,
funny as it may sound, it's not a trivial task these days to track
down everything electrical/electronic. It's worse than an Easter egg
hunt and just when you think you've got them all, oh yeah, there's a
clock in the oven <rolling eyes> (or a doorbell transformer). It's
just a heck of a lot simpler to flip the mains. Of course, if that
does kill the noise it's in the house so then you can flip off all the
branches, bring the mains up, and reconnect branch by branch to
isolate in which section the culprit lives.
That doesn't do me any good when its coming in from the power company
but, fortunately, whoever it is 'out there' that's causing the noise
doesn't seem to be active of late.
Patrick Turner
Jul 16, 2012, 10:07:21 AM7/16/12
to
>
> - Show quoted text -
Gee, and I thought I had bad problems with RFI from 101 gadgets.
Mains here are filthy with noise, but probably most is RF pick up
because the mains street wiring acts as a vast antenna network. The
source impedance of the noise is low, so just putting 0.01uF across
active and neutral may not do much. RF noise comes in via common mode
and differential mode, and one way to stop mains noise is to put
bigger value caps across active to neutral, and between both active
neutral to an earth stake or water pipe, and still you'd find noise.
Earth paths have impedance and copper pipes buried in the ground under
a house probably offer a good earth but one that still has impedance.
So when a radio is also well earthed to the same water pipes, its all
floating on the same earth signal as the mains, but the impedance of
the earthing may not be linear, and could be diode like, and so you
get rectified RF waves applied to the receiver and hum in the audio
signal. I have aluminium foil insulation under my roof tiles, and I
put it there before knowing it could seriously affect radio reception.
It acts as a large capacitor plate that isn't earthed, and even if I
did earth it, the wire used acts like a coil and there's a resonant CL
circuit. And maybe slight aluminium oxide also is a bit diode like. To
try to fully understand what causes noise in terms of the equivalent
LCR model for the mains, earth and insulation would be pointless - too
complex. I should live in a house without foil insulation and with no
known noisy gadgets, be like the Amish, no electricity.
I have tried putting my AM set inside a steel box with careful mains
input filtering but then the ferrite rod doesn't work very well. Maybe
an aluminium box would be better, or use bronze fly wire.
So, my favorite AM station has hum if I use a bit-of-wire antenna. But
if I use a ferrite rod the noise is 30dB less, and tolerable, although
some days hum will just switch on when something unknown begins to
cause it.
Probably my best way of getting good AM would be to live 15km away on
the far side of the radio stations and in a farm house with hardly any
other people around, and few ppl between the farm house and stations.
But I can't afford just move. I could maybe rent a farm house and rent
out my house to someone who paid more than farm house rent, but all
too much bother for not a huge problem.
I don't have an oven with electronic clock, don't have a phone
answering gadget, and don't have so many things. I did once have a
Shimasu phone answer gadget, and with a linear supply, and It made
terrible noise on AM, and nothing I did would silence it until I just
unplugged it, cut its cable, and stuck it the rubbish bin. If people
really wanted to phone me they'd better try me later when I am likely
to be home. I was getting fed up with returning a call to them and I'd
have to leave a message on their answer machine. So for last 10 years,
no answer machine, because all those calls at 22c each add up.
Patrick Turner.
0 new messages