DHT hum
len...@my-dejanews.com
I just finished a Svetlana 811-10 Single Ended with a
design from their website.
I am using AC on the 811's heaters although all designs I
have seen use DC.
There is fairly audible hum. I've tried hum bucking resistors which
seem to make it worse.
Any ideas ?
Thanks,
Lennard
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
PhCATuan
You got to use DC for the heater to get rid of the hum. There is no other
alternative.
TubeGarden
len...@my-dejanews.com
Yeah, it is being used.
In article <19981122082833...@ng-fb1.aol.com>,
-----------== Posted via Deja News, The Discussion Network ==----------
John Byrns
Hi Al & All,
A few thoughts with respect to AC heated filaments, and DHT hum come to
mind. First the disclaimer, I am a IHT kind of guy, and don't really know
anything about DHT's, so take what I say with a grain of salt.
1. DC heating can always be used, but it is reported to result in sonic
compromises, plus it may lead to shorter filament life.
2. It would seem like PP operation of DHT's would cancel most of the AC
heated filament hum. But of course then you loose any sonic benefits of
SE operation.
3. Due to the half wave rectification effect of the filament operation,
DHT hum does not occur at the 60 Hz line frequency, but rather at 120 Hz,
and other higher harmonics.
4. It seems to me that it should be possible to cancel the major hum
component at 120 Hz by running two DHT's in PSE, with the 60 Hz heating
current for the two filaments supplied 90 degrees out of phase. This
would result in the 120 Hz hum components to be 180 degrees out of phase,
and cancel.
Running the filaments in a PSE amp 90 degrees out of phase will require an
extra filament transformer, as well as other phase shifting components,
but should be worth it in terms of preserving the sonic benefits of AC
heating. If you are lucky enough to have a 3 phase AC supply available in
your listening room, then it should be a simple matter to arrange the two
DHT filament transformers in a Scott-T type of connection to generate the
required 2 phase heating currents. For those of us without 3 phase power
connections in our listening rooms, we will have to take other measures.
A suitable value capacitor in series with the AC feed to the second
filament transformer, and a shunt inductor across the primary could be
configured as a 1/4 wave length line section to provide the required 90
degree phase shift. The shunt inductor could even be eliminated, if a
cooperative transformer winder could be found, to wind the second filament
transformer with the exact required value of primary inductance.
With this scheme, the 120 Hz component of DHT hum could be eliminated in a
PSE amplifier with only the addition of a second filament heating
transformer (of custom design) and a series capacitor. This would
eliminate the need to resort to DC heating and the resultant sonic
compromises.
Any thoughts on this idea?
Regards,
John Byrns
TubeGarden
Christopher A. Beck
Take a look at my website. I have a very similar amp, using a SV572-10,
with AC on the heaters. I use a 50 ohm, 3 watt wirewound pot to cancel
hum. The hum pot is right near the 572, and the center tap on the
filament winding is not grounded. The 50 ohm pot creates what I would
call a "virtual" center tap, and allows you to null out the hum.
I have almost zero noise. What is there is so slight to be
insignificant to me. You will never get it dead quiet with AC, but I
refuse to put nasty diodes and electrolytic caps in the filament
circuit. Just enjoy the music.
Go to: http://www.execpc.com/~n9zes
Go look at the schematic on my amplifier. Scroll down a little to see
the power supply. It just works. And the amps really sound sweet.
Chris Beck
remove .nojunk to reply via e-mail
len...@my-dejanews.com wrote:
>
> I just finished a Svetlana 811-10 Single Ended with a
> design from their website.
> I am using AC on the 811's heaters although all designs I
> have seen use DC.
>
> There is fairly audible hum. I've tried hum bucking resistors which
> seem to make it worse.
>
> Any ideas ?
>
> Thanks,
>
> Lennard
>
Chris Hornbeck
wrote:
>3. Due to the half wave rectification effect of the filament operation,
>DHT hum does not occur at the 60 Hz line frequency, but rather at 120 Hz,
>and other higher harmonics.
First I've heard of anything remotely like this. Could you comment
further? Thanks,
Chris Hornbeck, Guyville
"Second star to the right,
Then straight on 'til morning."
rco...@ibm.net
Please contact Steve Deckert at www.decware.com.
The essence of his Single ended triodes is the SV83's and a 6n1p, not the
same thing but he regularly deals with this issue.
Specifically, he quieted my amps for me to near inaudible humm, triode
characteristic.
Tell him you spoke to me.
Thanks,
In article <737s58$qr0$1...@nnrp1.dejanews.com>, len...@my-dejanews.com says...
Ccusano
my 300B amps, and I can zero out the hum quickly and easily.
Chris
Gabe Velez
battery operated antique radios and the small 1 volt filament portable
tube radios. So...
> 1. DC heating can always be used, but it is reported to result in
> sonic
> compromises, plus it may lead to shorter filament life.
Direct heated Triodes are made for DC, if memory serves. My antique
radios are still using original 1920's tubes. I don't htink that AC or
DC makes much of a difference in longevity, though. Just voltage level
and current. Unless the filament is made from a different kind of
material for either current (I vaguely seem to recall that this may be
the case).
> 2. It would seem like PP operation of DHT's would cancel most of the
> AC
> heated filament hum. But of course then you loose any sonic benefits
> of
> SE operation.
This is true.
> 3. Due to the half wave rectification effect of the filament
> operation,
> DHT hum does not occur at the 60 Hz line frequency, but rather at 120
> Hz,
> and other higher harmonics.
I don't think so here. Half of a 60 hertz wave is the same as 60
hertz of full wave. In full wave rectifiers the only reason the cycles
become 120 hertz is because the negative side of the 60 hertz is now
brought positive in between the positive waves of the 60 cycles making
it 120. But, according to your logic, the rectification becomes only
half wave. 60 hertz. But, I could be wrong.
> 4. It seems to me that it should be possible to cancel the major hum
> component at 120 Hz by running two DHT's in PSE, with the 60 Hz
> heating
> current for the two filaments supplied 90 degrees out of phase. This
> would result in the 120 Hz hum components to be 180 degrees out of
> phase,
> and cancel.
Sounds interesting, and I think it may work, but it is all 60 hertz.
> Running the filaments in a PSE amp 90 degrees out of phase will
> require an
> extra filament transformer, as well as other phase shifting
> components,
> but should be worth it in terms of preserving the sonic benefits of AC
>
> heating.
What sonic benefits are there? This is a curiosity.
> If you are lucky enough to have a 3 phase AC supply available in
> your listening room, then it should be a simple matter to arrange the
> two
> DHT filament transformers in a Scott-T type of connection to generate
> the
> required 2 phase heating currents. For those of us without 3 phase
> power
> connections in our listening rooms, we will have to take other
> measures.
> A suitable value capacitor in series with the AC feed to the second
> filament transformer, and a shunt inductor across the primary could be
>
> configured as a 1/4 wave length line section to provide the required
> 90
> degree phase shift. The shunt inductor could even be eliminated, if a
>
> cooperative transformer winder could be found, to wind the second
> filament
> transformer with the exact required value of primary inductance.
Wouldn't a bridge rectifier for a buck and a couple of 4700
microfarad caps at 35 volts for $3.50 apiece be a bit cheaper and
easier?
> With this scheme, the 120 Hz component of DHT hum could be eliminated
> in a
> PSE amplifier with only the addition of a second filament heating
> transformer (of custom design) and a series capacitor. This would
> eliminate the need to resort to DC heating and the resultant sonic
> compromises.
> Any thoughts on this idea?
Yeah, this transformer could cost about $150 (custom made? Maybe
more), if it is feasible. But, the $8.00 alternative that everyone else
uses... K.I.S.S.
Have Fun!
Gabe
http://www.trigon.com
http://members.tripod.com/~gabevee
http://www.neutronicstechcorp.com
Gabe Velez
> Dear Sir,
>
> Please contact Steve Deckert at www.decware.com.
>
> The essence of his Single ended triodes is the SV83's and a 6n1p, not
> the
> same thing but he regularly deals with this issue.
>
> Specifically, he quieted my amps for me to near inaudible hum, triode
> characteristic.
>
> Tell him you spoke to me.
>
Steve's amps use 1)Indirectly heated cathodes and 2) pentodes that
are connected in triode mode. Not quite the same. Your hum problem might
have been for a different reason.
This isn't to take anything away from the man. I think he is a
genius. I have one of his Death Box speakers and I must say that I have
never heard such a subectively flat and powerful bass response from such
a cheap setup. So drop him a line, by all means!
John Byrns
wrote:
> Hey, John, I have a bit of DHT experience, due to my working with
> battery operated antique radios and the small 1 volt filament portable
> tube radios. So...
>
> > 1. DC heating can always be used, but it is reported to result in
> > sonic
> > compromises, plus it may lead to shorter filament life.
>
> Direct heated Triodes are made for DC, if memory serves. My antique
> radios are still using original 1920's tubes. I don't htink that AC or
> DC makes much of a difference in longevity, though. Just voltage level
> and current. Unless the filament is made from a different kind of
> material for either current (I vaguely seem to recall that this may be
> the case).
I don't believe that all Direct Heated Triodes are made for DC heating.
Some large directly heated transmitting tubes are designed for AC
heating. I think there are even a few designed for a polyphase supply to
the filaments, to reduce the hum. A lot of people claim DC heating
reduces longevity because of the higher current density at the positive
end of the filament, I don't know if this is true or not.
> > 2. It would seem like PP operation of DHT's would cancel most of the
> > AC
> > heated filament hum. But of course then you loose any sonic benefits
> > of
> > SE operation.
>
> This is true.
>
> > 3. Due to the half wave rectification effect of the filament
> > operation,
> > DHT hum does not occur at the 60 Hz line frequency, but rather at 120
> > Hz,
> > and other higher harmonics.
>
> I don't think so here. Half of a 60 hertz wave is the same as 60
> hertz of full wave. In full wave rectifiers the only reason the cycles
> become 120 hertz is because the negative side of the 60 hertz is now
> brought positive in between the positive waves of the 60 cycles making
> it 120. But, according to your logic, the rectification becomes only
> half wave. 60 hertz. But, I could be wrong.
Sorry, that was a typo, I meant to say "full wave rectification effect".
> > 4. It seems to me that it should be possible to cancel the major hum
> > component at 120 Hz by running two DHT's in PSE, with the 60 Hz
> > heating
> > current for the two filaments supplied 90 degrees out of phase. This
> > would result in the 120 Hz hum components to be 180 degrees out of
> > phase,
> > and cancel.
>
> Sounds interesting, and I think it may work, but it is all 60 hertz.
It does work, but it wasn't really my idea. This scheme was used in the
high powered RF amps of some old AM broadcast transmitters, which used PSE
RF finals, it wasn't necessary in the PP modulator, because of the
cancellation effect provided by the PP operation. It occurred to me that
it might be a clever idea to apply this scheme to DHT PSE amps. Of course
those transmitters had a 3-phase supply available, so it was a simple
matter to generate the 2-phase filament currents with a Scott-T, so they
didn't need the phasing circuit I suggested. There is no 60 Hz hum in an
AC heated DHT, it is 120 Hz, and various other harmonics.
> > Running the filaments in a PSE amp 90 degrees out of phase will
> > require an
> > extra filament transformer, as well as other phase shifting
> > components,
> > but should be worth it in terms of preserving the sonic benefits of AC
> >
> > heating.
>
> What sonic benefits are there? This is a curiosity.
I don't know if there are sonic benefits to AC heating or not, but there
are a lot of people that think there is, so I am interested in what the
reason might be. That's one of the reasons I was casting about for ways
to suppress the 120 Hz hum of an AC heated DTH, to remove that as a
variable in the equation.
> > If you are lucky enough to have a 3 phase AC supply available in
> > your listening room, then it should be a simple matter to arrange the
> > two
> > DHT filament transformers in a Scott-T type of connection to generate
> > the
> > required 2 phase heating currents. For those of us without 3 phase
> > power
> > connections in our listening rooms, we will have to take other
> > measures.
> > A suitable value capacitor in series with the AC feed to the second
> > filament transformer, and a shunt inductor across the primary could be
> >
> > configured as a 1/4 wave length line section to provide the required
> > 90
> > degree phase shift. The shunt inductor could even be eliminated, if a
> >
> > cooperative transformer winder could be found, to wind the second
> > filament
> > transformer with the exact required value of primary inductance.
>
> Wouldn't a bridge rectifier for a buck and a couple of 4700
> microfarad caps at 35 volts for $3.50 apiece be a bit cheaper and
> easier?
What could be simpler and easier than a transformer and a capacitor?
> > With this scheme, the 120 Hz component of DHT hum could be eliminated
> > in a
> > PSE amplifier with only the addition of a second filament heating
> > transformer (of custom design) and a series capacitor. This would
> > eliminate the need to resort to DC heating and the resultant sonic
> > compromises.
>
> > Any thoughts on this idea?
>
> Yeah, this transformer could cost about $150 (custom made? Maybe
> more), if it is feasible. But, the $8.00 alternative that everyone else
> uses... K.I.S.S.
What could be simpler than a transformer and a capacitor? Plus there is
no "sand" to possibly corrupt the sound. I haven't gone through the math,
but I would guess a standard transformer has more inductance than needed.
If that is true, you could just use a standard transformer with an extra
inductor in parallel with the primary. There may be other clever ways to
generate the required 90 degree phase shift.
Regards,
John Byrns
TubeGarden
I am now listening to 50CD6G in enhanced trode mode. For the 50 volt filaments,
I am using an old Lambda 100 Vdc regulated power supply. $15. plus AC cord
:-)
I have seen posts of people saying other people like AC heater "sound".
No one has described what they like about it.
DHTs for RF transmitters do not have overlap of useful bandwidth between RF and
AC. Audio does.
All the current at one end, pretty funny :-)
John Byrns
tubeg...@aol.com (TubeGarden) wrote:
> Hi,
>
> I am now listening to 50CD6G in enhanced trode mode. For the 50 volt
filaments,
> I am using an old Lambda 100 Vdc regulated power supply. $15. plus AC cord
> :-)
>
> I have seen posts of people saying other people like AC heater "sound".
>
> No one has described what they like about it.
>
> DHTs for RF transmitters do not have overlap of useful bandwidth between
RF and
> AC. Audio does.
But they do operate as nonlinear devices which causes a multiplicative
effect between the RF and the 120 Hz hum, generating sidebands which will
be demodulated as hum by the radio receiver.
Regards,
John Byrns
Gabe Velez
> I don't believe that all Direct Heated Triodes are made for DC heating.
> Some large directly heated transmitting tubes are designed for AC
> heating. I think there are even a few designed for a polyphase supply to
> the filaments, to reduce the hum. A lot of people claim DC heating
> reduces longevity because of the higher current density at the positive
> end of the filament, I don't know if this is true or not.
Sounds reasonable. Sort of a "current crowding". But I think it may be at
the negative side of the filament, since they are being attracted to the
positive plate. I guess.
I know very little about RF type tubes. Just that they are huge, run on
thousands of volts and run cherry red plates *normally*. I wonder, though if
the consequence you mentioned on another post of AC hum being demodulated at
the radio is not moot. The tubes run class C, dont they? How much of the 60
hertz is going to get to the "air" (R.F.)?
> Sorry, that was a typo, I meant to say "full wave rectification effect".
OK. But most full wave rectifiers I have seen have two plates to one
filament. A triode has one plate to the filament making for a half wave
rectifier. I guess.
> >didn't need the phasing circuit I suggested. There is no 60 Hz hum in an
> AC heated DHT, it is 120 Hz, and various other harmonics.
I don't know. One plate one filament still sounds like a half wave
rectifier to me. How would it become a full wave rectifier?
> I don't know if there are sonic benefits to AC heating or not, but there
> are a lot of people that think there is, so I am interested in what the
> reason might be. That's one of the reasons I was casting about for ways
> to suppress the 120 Hz hum of an AC heated DTH, to remove that as a
> variable in the equation.
I wonder if it is related to the experiments Steve Bench did regarding
lowering the voltage supply of a DHt to reducce distortion? If we did use an
unrectified 6 VAC, then the average voltage would be lower than a 6vDC
current, would it not? Especuialy since you claim the rectification properties
of the tube causing the voltage to be rectified. With no filter the average
voltage would be lower, no? So,the sonic properties should improve, and the
longevity, because now the filament is working with lower actual voltage. So,
the conclusion would be to use a lower DC voltage to acheive the same results.
> What could be simpler and easier than a transformer and a capacitor?
Just putting a cap in line... how could that make two waves out of phase?
Unless I missed that you meant an *additional* transformer with a cap to lag
the phase 90 degrees, then parallel the secondaries. I could try this myself.
I have a power trannie that has two filament taps (aside from the filament
one). Wouldn't it merely delay the one that is there (assuming single phase AC
as the source) I would think that you could use a transformer with three
primaries. Then use one common to the three, one winding direct, one with a
capacitor and one with an inductor. Then you would have the necessary three
phase AC which I agree would simulate DC without the "sand" rectifier. and it
would increase the effective voltage. But how much would a three winding
primary transformer cost (or a three winding secondary for that matter)? Or
one could use three inductors to make two phases. But then there is cost. (I
have an AC analysis book at home, and I seem to recall that it can be done
that way, but it is sorta complicated.)
But then one introduces the distortin products and harmonics now of the
cap and coil, no? how about resonance frequencies. They will occur.
A spike in the AC line could result in a temporary oscillation of the
power supply and possibly, no it will blow out the precious and delicate
filament. I know, I have made crossovers and power supply filters using caps
and coils that without proper damping caused oscillations that brought the
speaker or amp to punch real loud. Not good.
Although, if one were adventurous one could tap the 220 volt AC line
coming into the house generally used for hot water heaters, heating furnaces,
clothes dryers and electric ovens and use the two phases coming from them,
they are 110 volts at 180 degree differences with respect to neutral. But be
careful. Its phase difference can kill and hold you there do do so. (A friend
of mine almost died that way. His wife smartly pushed him out of the way using
a bat.)
> What could be simpler than a transformer and a capacitor? Plus there is
> no "sand" to possibly corrupt the sound. I haven't gone through the math,
> but I would guess a standard transformer has more inductance than needed.
> If that is true, you could just use a standard transformer with an extra
> inductor in parallel with the primary. There may be other clever ways to
> generate the required 90 degree phase shift.
Phasor math is sorta complicated, isnt it? ;-) But with good filtering the
sand is worth its salt (har har har!) Besides, the ac-dc DHT radios (using the
1 volt filaments) I have use selenium rectifiers for the filaments.
This is a very interesting thread! I think I will use the thoughts on my
web site and refer to the contributers. Thanks!
Experiment. That is where the most fun is with this stuff.
As always,
--
Have Fun!
Gabe
Remove the *nospam* from my email address.
John Byrns
gvelez@*nospam*richmond.infi.net wrote:
> Hi John,
>
> > I don't believe that all Direct Heated Triodes are made for DC heating.
> > Some large directly heated transmitting tubes are designed for AC
> > heating. I think there are even a few designed for a polyphase supply to
> > the filaments, to reduce the hum. A lot of people claim DC heating
> > reduces longevity because of the higher current density at the positive
> > end of the filament, I don't know if this is true or not.
>
> Sounds reasonable. Sort of a "current crowding". But I think it may be at
> the negative side of the filament, since they are being attracted to the
> positive plate. I guess.
Yes, I think you are right, I got it backwards. It is so hard to keep
this stuff straight while typing. At any rate, whichever end of the grid
would have the least negative bias, would be the end of the filament with
the greater current.
That's also sort of the reason that the hum is 120 Hz (and other
harmonics) with a center tapped filament transformer. If you think about
it the filament is sort of symetrical, and the same conditions apply when
either end is positive, or negative, there by doubling the hum frequency,
plus those other troublesom harmonics of course.
Regards,
John Byrns
André Jute
> I have seen posts of people saying other people like AC heater "sound".
>
> No one has described what they like about it.
I don't think you have that right, Al. What people usually say is that
they don't like sound of DC on filaments, and that AC is the least bad
alternative.
As for regulated supplies, after going to enormous trouble to design and
match parts for a fil regulator board for a manufacturer, I decided I
didn't like the sound of regulated supplies because they have
bleached-out sound. (I think it is Herb Reichert who once wrote that he
'despised' the sound of regulated filament supplies. I don't want to go
that far, but I know where his head is screwed on.)
But if you insist, I will say that AC fils sound rougher, more real,
less like that acid studio sound I hate because it sacrifices musical
integration for clarity and "sound-staging", an audiophile concept with
no concert-room reality.
Interestingly, I was reading the MIT Radlabs 18, Valley & Wallman in the
lavatory a couple of days ago in the section on filament balancing,
which for some crazy reason is included in the chapter on direct
coupling, and their story on the influence of filament voltage variation
is that it is of greater importance than was tentatively concluded in
the recent thread on the subject on RAT. If I built amps by the book,
after reading V&W I'd build only regulated supplies.
Andre
--
Andre Jute an...@indigo.ie COMMUNICATION JUTE
--see our pages for music lovers, writers and audiophiles at
http://indigo.ie/~andre/ComJuteF1.html
TubeGarden
try the 50CD6G tubes. I had a Lambda surplus power supply for 100 Vdc and
hooked up the fils in series. At first it was just OK, but by late last night I
was wondering if I had some magic tubes. Then I realized this was my first
brush with regulated filaments, so today I hooked up the VA (EF86) and CF
(SV83) to another Lambda and set it to 6.30 Vdc and now I am listening to
Schickele Mix waltzes on the radio.
These Lambda's are surplus units that were $15. per and I have been stubbing my
toes on for about a year or so. See, it pays to buy stuff and keep it around
until you get a clue. I have a 36 Vdc left unused...hmmmm Bias???
Anyway, it is very smooth in my amp, so for the moment it is the worst filament
supply I know of, except for all the rest :-)
The 50CD6G Raytheons sound very solid in etm and I am glad I spent the $.04 ea
for them :-)
And these chilly winter evenings in Phoenix, when the temps outdoors drop below
60 degrees F. All these power supplies are like a little electric fireplace.
Um-pa-pa. Um-pa=pa. Peter is a great human being. He just said he hopes to
write music that well when he is dead...
Essence d'valse,
Bob C
TubeGarden wrote in message
<19981128202640...@ng-cc1.aol.com>...
>Yup. I stumbled into regulated filament supplies yesterday when I wanted
to
>try the 50CD6G tubes. I had a Lambda surplus power supply for 100 Vdc and
>hooked up the fils in series. At first it was just OK, but by late last
night I
>was wondering if I had some magic tubes. Then I realized this was my first
>brush with regulated filaments, so today I hooked up the VA (EF86) and CF
>(SV83) to another Lambda and set it to 6.30 Vdc and now I am listening to
>Schickele Mix waltzes on the radio.
>
>These Lambda's are surplus units that were $15. per and I have been
stubbing my
>toes on for about a year or so. See, it pays to buy stuff and keep it
around
>until you get a clue. I have a 36 Vdc left unused...hmmmm Bias???
>
>Anyway, it is very smooth in my amp, so for the moment it is the worst
filament
>supply I know of, except for all the rest :-)
>
>The 50CD6G Raytheons sound very solid in etm and I am glad I spent the $.04
ea
>for them :-)
>
Hi Al,
Very interesting that you experience with regulated filaments is very
different from Jute's. I have a friend, a real engineer who designed a real
regulated filament supply and it sounds very good. Truth does sounds better
than fiction. Could it be that you really did it, and he just makes it up?
Of course...That's it...
It is very sad that he keeps posting his fantasy stories, which can
discourage the less intrepid from experimenting and trying things
themselves.
BC
len...@my-dejanews.com
I have since rectified my 6.3ac with a CLC using 22,000 on either side
of a car stereo alternator whine choke(7amp). I get about 5.8-9 volts now
however.
Hum is now pretty much nil and inaudible.
Personally it sounds a whole lot cleaner now.
I really don't know how people can build those 300b amps with AC filaments.
Maybe its the large current draw(4+amps) of the SV811 which really
causes the hum to be coupled into the output.
Grover Gardner
improves the sound of the DC supply. The answer to your question is that at 5
volts and 1.2 amps, the 300B AC hum can be balanced out sufficiently for most
users, to around 3-4mV of noise. At 6.3 volts/2A, I find the Vaic tubes too
noisy for comfort. The Svetlanas are definitely unlistenable with AC
filaments. Your solution is a very good one. For the proper filament
voltage, however, you might track down a 15 or 16vct tranny, from Mouser for
instance, use half the secondary for 7.5 or 8 volts, and you should get the
proper voltage for the 811.
len...@my-dejanews.com wrote:
--
Grover Gardner
gro...@postoffice.att.net
O'Connor
>
> Hi John,
>
> > I don't believe that all Direct Heated Triodes are made for DC heating.
> > Some large directly heated transmitting tubes are designed for AC
> > heating.
Transmitting tubes like the 811A and 572B have been used in push-pull
output stages for audio using AC filaments to good result. The key, of
course, is the inherent balance in the circuit, but some fine tuning is
required to null the hum out completely (or next to it).
I have used the Svetlana SV572-160 and 572Bs in such circuits and have
tried both AC and DC filaments. I like the AC connection for its
simplicity and true-ness to classic designs. DC does make hum reduction
a tiny bit simpler and I think you can achieve better performance if you
need it, and it is also easier to incorporate turn-on surge limiting.
For single-ended circuits, you really must use DC on directly heated
cathodes.
These are my observations for what they are worth.
Kevin O'Connor
TubeGarden
My experience with regulated filaments is one circuit in one amp with one set
of tubes. It was so positive I am happy to report it. I will continue to build
and report as I go along this path, as there is literally very little else I
can do these days.
I was disappointed that my carefully filtered DC was so easily trounced.
But then, I was really surprised when adding 0.01 uF 600V PP caps between the
incoming AC legs and ground was so noticeable on my double Pi CLCLC B+ filters.
The guys who designed the old Zenith table radios were on the ball :-)
I have come to picture filters not as fine mesh screens, but more like the
posts in front of stores which block trucks from crashing the entrance. Coarse.
Both "Ultimate Ella" and the RR Janacek made fresh performances last night.
I love this hobby.
Happy Ears!
John Byrns
gro...@postoffice.att.net wrote:
> Way to go, Lennard. The choke filtering is a great solution and really
> improves the sound of the DC supply. The answer to your question is that at 5
> volts and 1.2 amps, the 300B AC hum can be balanced out sufficiently for most
> users, to around 3-4mV of noise.
Hi Grover,
Would 3-4mV of noise really be tolerated by "most users"? You may be
right, but I find the hum and noise of my "Power Amplifier Without Power
Transformer" to be marginal at 0.3-0.4mV, a noise level fully 20 dB
lower. What do other readers think? How much hum do you tolerate in the
output of your amplifiers?
Regards,
John Byrns
Grover Gardner
tolerable on speakers on moderate efficiency. On Lowthers, no :-). I've tried my
amps and other AC-heated homebrew and commercial amps on my own and friends'
systems and noise was not a seriosu problem. But I'm sure it depends on your
speakers, your room and the ambient noise-level you have in your listening
environment.
John Byrns wrote:
--
Grover Gardner
gro...@postoffice.att.net
TubeGarden
I manage to endure the horrible racket my 18 uV noise floor makes. But, I am
making progress :-)
A Dougla6
John Byrns wondered:
>Would 3-4mV of noise really be tolerated by "most users"? You may be
>right, but I find the hum and noise of my "Power Amplifier Without Power
>Transformer" to be marginal at 0.3-0.4mV, a noise level fully 20 dB lower.
>What do other readers think? How much hum do you tolerate in the
>output of your amplifiers?
20 mV is full volume from my p-p 6BX7 amp into WE 555 drivers on
8-cell IPC horns. With no CD playing, I don't hear anything from the horns,
even close up. Of course with a 500Hz cutoff they wouldn't reproduce
much hum anyway. The bass channels are split off ahead of the amp
by L-C crossovers, following the reprint article in SP #3, except for
choosing a load impedance to suit the toroids I had on hand.
Cheers, Alan
Gabe Velez
> Yes, I think you are right, I got it backwards. It is so hard to keep
>
> this stuff straight while typing. At any rate, whichever end of the
> grid
> would have the least negative bias, would be the end of the filament
> with
> the greater current.
Totally understandable. I get happy fingers, too. The ADD doesn't help
either
> That's also sort of the reason that the hum is 120 Hz (and other
> harmonics) with a center tapped filament transformer. If you think
> about
> it the filament is sort of symetrical, and the same conditions apply
> when
> either end is positive, or negative, there by doubling the hum
> frequency,
> plus those other troublesom harmonics of course.
Ah, there is where my thinking was amiss. I was thinking a straight
source with no center tap. Here I would agree with you. With a canter
tap the filament would see, relative to ground, two AC waves 180 degrees
out of phase that would give a full wave rectifier effect. Makes sense.
Thanks for clearing that up for me!
Have Fun!
Gabe
Bob Miller
matter what the reference level and with a small power se amp much worse.
Bob
John Byrns wrote:
> In article <3660F780...@postoffice.att.net>,
> gro...@postoffice.att.net wrote:
>
> > Way to go, Lennard. The choke filtering is a great solution and really
> > improves the sound of the DC supply. The answer to your question is that at 5
> > volts and 1.2 amps, the 300B AC hum can be balanced out sufficiently for most
> > users, to around 3-4mV of noise.
>
> Hi Grover,
>
> Would 3-4mV of noise really be tolerated by "most users"? You may be
> right, but I find the hum and noise of my "Power Amplifier Without Power
> Transformer" to be marginal at 0.3-0.4mV, a noise level fully 20 dB
> lower. What do other readers think? How much hum do you tolerate in the
> output of your amplifiers?
>
> Regards,
>
> John Byrns
Ned Carlson
>I have seen posts of people saying other people like AC heater "sound".
>
>No one has described what they like about it.
I still don't understand this either, but I ain't tried it.
Grover says he can hear a difference, I believe him, but I suspect
some other folk are just trying too hard to hear a difference.
>
>DHTs for RF transmitters do not have overlap of useful bandwidth between RF and
>AC. Audio does.
Huh? I don't get that one, Al. Many of the DHT's I've heard of come
with specs for AF and RF operation (811A, 211, 833, 3-500Z, 3CX1000),
and all were/are expected to run with AC filaments. IIRC, some
transmitters used 833's for both PA *and* modulator, and I'm pretty
sure both sides ran AC filaments, too.
Ned Carlson Triode Electronics "where da tubes are!"
2225 W Roscoe Chicago, IL, 60618 USA
ph 773-871-7459 fax 773-871-7938
12:30 to 8 PM CT, (1830-0200 UTC) 12:30-5 Sat, Closed Wed & Sun
http://www.triodeel.com
Your Start Page for Tube and Tube Amp info on the net...
http://www.triodeel.com/tlinks.htm
Ned Carlson
wrote:
>I don't believe that all Direct Heated Triodes are made for DC heating.
I think Gaberoo must only collect *really* old radios...I had a
McMillan and an RCA console ca. 1929 that both used '45's
with AC filaments..not to mention the Atwater Kent model 40,
which had AC filament on a single-ended (parallel-feed, IIRC)
'71A..of course the typical speakers used with the AK40 probably
had such limited low frequency response that no one noticed
the hum.
>Some large directly heated transmitting tubes are designed for AC
>heating.
That's an understatement.
I think there are even a few designed for a polyphase supply to
>the filaments, to reduce the hum.
I'm looking at my Eimac book, the only ones that use polyphase
heating that I can see are rated 400KW or more. However, the
notes for one (8974, a monster 1.5 megawatt tetrode) does say the
dual filament can be used to reduce hum by feeding the AC in
quadrature.
A lot of people claim DC heating
>reduces longevity because of the higher current density at the positive
>end of the filament, I don't know if this is true or not.
I dunno either. Maybe we could test a bunch of 3V4 pulls from
Zenith Transoceanics and see if any exhibit abnormal wear from
their filaments being used on DC. Of course, when those radios were
made, 3V4's cost about a buck apiece and Zenith probably didn't care.
>It does work, but it wasn't really my idea. This scheme was used in the
>high powered RF amps of some old AM broadcast transmitters, which used PSE
>RF finals, it wasn't necessary in the PP modulator, because of the
>cancellation effect provided by the PP operation.
But what you are talking about isn't a cancellation effect...90
degrees can't exactly *cancel*, right? It just keeps the hum from
the tube filaments from adding together.
It occurred to me that
>it might be a clever idea to apply this scheme to DHT PSE amps. Of course
>those transmitters had a 3-phase supply available, so it was a simple
>matter to generate the 2-phase filament currents with a Scott-T,
Wouldn't those be 60 degrees out of phase, not 90?
TubeGarden
and AC. Audio does.
>
>Huh? I don't get that one, Al. Many of the DHT's I've heard of come with specs
for AF and RF operation (811A, 211, 833, 3-500Z, 3CX1000), and all were/are
expected to run with AC filaments.
Oh. OK. Nevermind.
I have been using surplus Lambda DC power supplies for filaments. Cheap and
easy. Local reclamation shop sells them for about the price of a 62K uF 16V
filter cap. The 100V doea a great job with a series pair of 50CD6A, sounds very
nice.
SBench
>
>I dunno either. Maybe we could test a bunch of 3V4 pulls from
>Zenith Transoceanics and see if any exhibit abnormal wear from
>their filaments being used on DC. Of course, when those radios were
>made, 3V4's cost about a buck apiece and Zenith probably didn't care.
>
with open filaments were always "broken" at the same place. That
could *either* be due to the DC effect or to the manufacturing method.
I guess the answer depends on which side of the argument you wanna
take.
>
>>It does work, but it wasn't really my idea. This scheme was used in the
>>high powered RF amps of some old AM broadcast transmitters, which used PSE
>>RF finals, it wasn't necessary in the PP modulator, because of the
>>cancellation effect provided by the PP operation.
>
>But what you are talking about isn't a cancellation effect...90
>degrees can't exactly *cancel*, right? It just keeps the hum from
>the tube filaments from adding together.
-----> Sure it is. The hum, being due to the instantaneous power (heating)
in the filament is modulated at twice the operating frequency (max heating
occurring somewhere after the positive or negative peak of the sine wave).
So the angle gets multiplied to 180 degrees, effectively. 90 degrees
(quadrature) is an interesting phenomenum. Note that 2 transformers
oriented 90 degrees "cancel" their coupling effects. Actually in that
case, ignore is probably a better word than cancel.
>
>
> It occurred to me that
>>it might be a clever idea to apply this scheme to DHT PSE amps. Of course
>>those transmitters had a 3-phase supply available, so it was a simple
>>matter to generate the 2-phase filament currents with a Scott-T,
>
>Wouldn't those be 60 degrees out of phase, not 90?
-----> or 120 degrees. (3*120=360). Actually 3 phase, for those who
haven't studied the advantages) has a very interesting property, which
makes it useful in a number of conditions............ the "instantaneous"
power (at any point in the conduction cycle) is constant. So, in principle
3 phase fed filaments produce *no* hum, and rectified 3 phase has
*no* ripple. (Of course that's in principle. In the real world, the
phases are never exact, either in amplitude or phase relationship, so
there is always a residual ripple component).
>Ned Carlson Triode Electronics "where da tubes are!"
>2225 W Roscoe Chicago, IL, 60618 USA
>ph 773-871-7459 fax 773-871-7938
>12:30 to 8 PM CT, (1830-0200 UTC) 12:30-5 Sat, Closed Wed & Sun
>http://www.triodeel.com
Best Regards,
Steve
Check my web page .. <A
HREF="http://members.aol.com/sbench101/">http://members.aol.com/sbench101</A>
Remove the .gov to EMail me
Acrosound
>not to mention the Atwater Kent model 40,
>which had AC filament on a single-ended (parallel-feed, IIRC)
>'71A..of course the typical speakers used with the AK40 probably
>had such limited low frequency response that no one noticed
>the hum.
Hey Ned: I have two questions if you would be so kind.
First... what in the world does IIRC mean. Is this some special
type of power supply arrangement :=) ?
second... do you know what other tubes the AK model 40 used?
Do you per chance have a schematic for this amp on your website?
third... ooops... do RAT's know that the Atwater Kent Museum
is in Philadelphia? On 3rd street in the Old City section? Come
visit :=) !!!
Mike LaFevre
A Dougla6
Mike asked:
>Hey Ned: I have two questions if you would be so kind.
>First... what in the world does IIRC mean. Is this some special
>type of power supply arrangement :=) ?
Sorry, I can't resist. Inverted Interstage Resistance Coupled of course,
IIRC.
>second... do you know what other tubes the AK model 40 used?
>Do you per chance have a schematic for this amp on your website?
26s mostly, one 27 detector (27s were more expensive than 26s
at the time, being quite new) and an 80 rectifier.
>third... ooops... do RAT's know that the Atwater Kent Museum
>is in Philadelphia? On 3rd street in the Old City section? Come
>visit :=) !!!
Yup, but Arthur didn't put any radios there, other than perhaps the
AK travelling display from the late 20s that was stored in the museum
basement for decades.
Cheers, Alan
John Byrns
sbe...@aol.com.gov (SBench) wrote:
[snip]
> >>It does work, but it wasn't really my idea. This scheme was used in the
> >>high powered RF amps of some old AM broadcast transmitters, which used PSE
> >>RF finals, it wasn't necessary in the PP modulator, because of the
> >>cancellation effect provided by the PP operation.
> >
> >But what you are talking about isn't a cancellation effect...90
> >degrees can't exactly *cancel*, right? It just keeps the hum from
> >the tube filaments from adding together.
> -----> Sure it is. The hum, being due to the instantaneous power (heating)
> in the filament is modulated at twice the operating frequency (max heating
> occurring somewhere after the positive or negative peak of the sine wave).
Interesting, I had assumed the hum was due to the varying potential of the
different parts of the filament over the AC cycle changing the grid bias,
it never occurred to me that it might be a heating effect. I guess I
could do a small experiment, if it is the heating effect, as you say, then
the hum should remain at 120 Hz, even when one end of the filament supply
is grounded, rather than the center tap, if the hum changes to 60 Hz, then
it is electric effects.
> So the angle gets multiplied to 180 degrees, effectively. 90 degrees
> (quadrature) is an interesting phenomenum. Note that 2 transformers
> oriented 90 degrees "cancel" their coupling effects. Actually in that
> case, ignore is probably a better word than cancel.
> >
> >
> > It occurred to me that
> >>it might be a clever idea to apply this scheme to DHT PSE amps. Of course
> >>those transmitters had a 3-phase supply available, so it was a simple
> >>matter to generate the 2-phase filament currents with a Scott-T,
> >
> >Wouldn't those be 60 degrees out of phase, not 90?
> -----> or 120 degrees. (3*120=360). Actually 3 phase, for those who
> haven't studied the advantages) has a very interesting property, which
> makes it useful in a number of conditions............ the "instantaneous"
> power (at any point in the conduction cycle) is constant. So, in principle
> 3 phase fed filaments produce *no* hum, and rectified 3 phase has
> *no* ripple. (Of course that's in principle. In the real world, the
> phases are never exact, either in amplitude or phase relationship, so
> there is always a residual ripple component).
I think you have this wrong, I never thought about the power being
constant over the cycle in a three phase system, but it sounds plausible
to me. I know it is true that a three phase rectifier does produce ripple
on the output though. You also have to specify whether you are talking
about a 3 pulse rectifier, a 6 pulse rectifier, a 12 pulse rectifier, or
what? Each one produces successively less ripple on the output.
I think someone was also wondering how 90 degree, or quadrature currents
are obtained from a three phase supply. In a two phase power system ther
are two currents separated by 90 degrees , and two transformers connected
in an arrangement called a Scott-T (SP?) can be used to convert three
phase currents to two phase, or two phase currents to three phase. Two
phase systems can be very useful, but I think three phase transmission is
more economical of copper.
Regards,
John Byrns
John Byrns
sbe...@aol.com.gov (SBench) wrote:
> Hi John.
> 3 phase is indeed constant power, work the integral.
Too much work, as I said, constant power sounds plausable, so I have no
reason to doubt what you said about the power.
> On the ripple, you're right. The unfiltered ripple on a
> 3 phase setup, full wave rectified, is 4% of the rectified DC
> voltage (Source: Electronic Designers HB).
By full wave, I assume you mean a 6 pulse rectifier. A 12 pulse rectifier
feed from a 3 phase supply will have even less ripple.
> BTW, in addition to the economics of the copper, the power issue
> allows a smoother start and run performance on three phase motors
> WRT to their single phase counterparts. (And you get the delight
> of having the motor running backwards if you get the phases mixed up).
It's the same thing with a 2 phase motor, reverse one of the phases, and
the motor will run backwards.
Regards,
John Byrns
SBench
<snip>
>nteresting, I had assumed the hum was due to the varying potential of the
>different parts of the filament over the AC cycle changing the grid bias,
>it never occurred to me that it might be a heating effect. I guess I
>could do a small experiment, if it is the heating effect, as you say, then
>the hum should remain at 120 Hz, even when one end of the filament supply
>is grounded, rather than the center tap, if the hum changes to 60 Hz, then
>it is electric effects.
-----> Good test!
<snip>
>> >Wouldn't those be 60 degrees out of phase, not 90?
>> -----> or 120 degrees. (3*120=360). Actually 3 phase, for those who
>> haven't studied the advantages) has a very interesting property, which
>> makes it useful in a number of conditions............ the "instantaneous"
>> power (at any point in the conduction cycle) is constant. So, in principle
>> 3 phase fed filaments produce *no* hum, and rectified 3 phase has
>> *no* ripple. (Of course that's in principle. In the real world, the
>> phases are never exact, either in amplitude or phase relationship, so
>> there is always a residual ripple component).
>
>I think you have this wrong, I never thought about the power being
>constant over the cycle in a three phase system, but it sounds plausible
>to me. I know it is true that a three phase rectifier does produce ripple
>on the output though. You also have to specify whether you are talking
>about a 3 pulse rectifier, a 6 pulse rectifier, a 12 pulse rectifier, or
>what? Each one produces successively less ripple on the output.
>
>I think someone was also wondering how 90 degree, or quadrature currents
>are obtained from a three phase supply. In a two phase power system ther
>are two currents separated by 90 degrees , and two transformers connected
>in an arrangement called a Scott-T (SP?) can be used to convert three
>phase currents to two phase, or two phase currents to three phase. Two
>phase systems can be very useful, but I think three phase transmission is
>more economical of copper.
>
>
>Regards,
>
>John Byrns
Hi John.
3 phase is indeed constant power, work the integral.
On the ripple, you're right. The unfiltered ripple on a
3 phase setup, full wave rectified, is 4% of the rectified DC
voltage (Source: Electronic Designers HB).
BTW, in addition to the economics of the copper, the power issue
allows a smoother start and run performance on three phase motors
WRT to their single phase counterparts. (And you get the delight
of having the motor running backwards if you get the phases mixed up).
Ned Carlson
>Ned wrote:
>
>>not to mention the Atwater Kent model 40,
>>which had AC filament on a single-ended (parallel-feed, IIRC)
>>'71A..of course the typical speakers used with the AK40 probably
>>had such limited low frequency response that no one noticed
>>the hum.
>
>Hey Ned: I have two questions if you would be so kind.
>
>First... what in the world does IIRC mean. Is this some special
>type of power supply arrangement :=) ?
If I Recall Correctly
>
>second... do you know what other tubes the AK model 40 used?
4 26's, 1 each '27, 71A, 80.
>Do you per chance have a schematic for this amp on your website?
No, but this place does, plus lots of other AK's...
http://www.nostagiaair.org
Ned Carlson Triode Electronics "where da tubes are!"
2225 W Roscoe Chicago, IL, 60618 USA
ph 773-871-7459 fax 773-871-7938
12:30 to 8 PM CT, (1830-0200 UTC) 12:30-5 Sat, Closed Wed & Sun
http://www.triodeel.com