Parasitic suppressor data
Wes Stewart
As promised, I'm posting the data you all have been breathlessly awaiting.
It is on my home page (which is the first time I've done this) at:
http://www.azstarnet.com/~n7ws
I hope it works!
73, Wes -- N7WS
Chuck KD9JQ
FYI:
Wes...Your Home Web Page wants a userid and password!
Chuck KD9JQ
--
---------------------------------------------------------
Charles H. Reichert 847-358-3827 Home
KD9JQ 847-632-6669 Work
955 Concord Lane http://www.imaxx.net/~kd9jq
Hoffman Estates, IL. 60195 mailto:kd...@imaxx.net
---------------------------------------------------------
John Siegel
73 John
Wes Stewart
In article <32D909...@imaxx.net> Chuck KD9JQ <kd...@imaxx.net> writes:
>From: Chuck KD9JQ <kd...@imaxx.net>
>Subject: Re: Parasitic suppressor data
>Date: Sun, 12 Jan 1997 09:54:38 -0600
>Wes Stewart wrote:
>>
>> As promised, I'm posting the data you all have been breathlessly awaiting.
>>
>> It is on my home page (which is the first time I've done this) at:
>>
>> http://www.azstarnet.com/~n7ws
>>
>> I hope it works!
>>
>> 73, Wes -- N7WS
>FYI:
>Wes...Your Home Web Page wants a userid and password!
>Chuck KD9JQ
I dunno...
I called my ISP and they were baffled. The support guy did change the link
address to the complete path name so maybe that will help.
Others have said to input anything (or nothing ) for a password and it works.
Sorry about the problem.
73, Wes N7WS
w8j...@aol.com
>>Extracted from E-mail>>
The suppressor graphs (RL.GIF) seem to show two things:
1. The conventional (non-nichrome) suppressor gives a faster roll-off of
Q with frequency than the nichrome suppressors.
2. There's very little difference between the two grades of nichrome.
You obviously don't want to get publicly embroiled in this controversy,
and neither do I. However, it does seem that the distinguishing feature
of the nichrome suppressors is their low HF Q. The VHF Qs are all
pretty similar.
The VHF Q of the conventional suppressor could be made exactly the same
as the nichrome ones by changing the value of either L or R, but the HF
Q of the conventional suppressor would still be about a factor 2
higher.<<-<<
I don't blame anyone for avoiding Rich's wrath. But since I am already on
his top ten list, I have nothing much to lose, hi. So I'll post this and
accept the return attack it will almost certainly generate.
This conclusion above is the same conclusion I reached, nichrome
suppressors offer lower HF Q, and nearly identical VHF performance. Just
the slightest adjustment in resistance or inductance of the conventional
copper wire suppressor would result in less VHF Q without greatly changing
HF Q.
If the data is reviewed, the ratio of Q is almost unity at 200 MHz and
over 2:1 at 10 MHz. Just the slightest adjustment of R or L will make the
conventional suppressor equal the nichrome at VHF, while maintaining
better Q at HF where we want losses to be reduced.
For example, the AL-80B oscillates at about 170-180 MHz when the
suppressor is removed. At that frequency, Wes measured a Q of about one
(same as I posted weeks ago). The Q of a duplicate nichrome suppressor is
about .75. This very small Q change is diluted by circuit effects when the
suppressor is installed in an actual PA. The actual operating Q change
would be even less at VHF, and more at HF.
Rich's claim this small change in VHF Q is a great or worthwhile
improvement that cures all woes is completely false. Anyone believing that
would have to have a real problem understanding parallel and series L/R
circuits. Add one turn to the conventional suppressor or adjust the
resistor size, and you've gone even lower with Q than the nichrome.
Finally, I'd like to remind everyone how (over and over again) Measures
squaked about how he could SEE the difference in his dip meter, describing
all the sharp dips that went away when he made (what amounts to) a *very*
small Q reduction at the frequency of interest. Rich would have use
believe he measured a Q change of less than 0.5 with his GDO!
Remember and compare those claims of what he "saw first hand" to other
claims of things he "knew for a fact":
1.) The Miklos employment tale. He claimed he "called Eimac" and they told
him Buzz was never R+D manager. Yet Eimac even issued a letter (after
getting Buzz's permission) explaining Buzz was indeed R+D manager. Not
only that, VHF Handbooks list Buzz as Eimac's former R+D Manager. Even
after the employment confirming letter was out, Rich sent Buzz Miklos a
post card telling Buzz he was never R+D manager!
2.) The tale about Marvin Born. Rich said Marvin measured parasitics in an
AL1500. Turns out it wasn't even an AL1500, it was a dual 8877 PA built on
a old empty prototype chassis. When I talked to Marvin he said he was
positive Rich knew he was talking about a west coast manufactured 3CX3000
PA, and not a AL-1500.
3.) Rich's tale about 500 watt 8877 grids. Eimacs own catalog disagrees
and they even have an application note describing how improper tuning can
destroy the grid.
4.) Rich's story about a photon hitting the cathode of a tube ON STANDBY
and blowing up a PA. A physicist who works with that type of stuff every
day said it was bunk.
5.) The story about the variac and filament meter markings **on every
broadcast transmitter** and Rich's slam of every amateur manufacturer for
doing something the commercial manufacturers apparently don't often do.
Several people jumped in and proved he was telling another story.
6.) Measures' claim a certain TL-922 was damaged by a parasitic. The owner
showed up and said "no, it was a lightning arrestor causing the arc, I
removed it and the amp is fine".
7.) Measures' attempt to blame me for Wes' unintentional error. Rich was
completely aware of how this happened BEFORE he posted his pathological
accusation that I was trying to hide something, and that I misled Wes.
When all this is considered it becomes perfectly clear Measures presents
untrue statements as fact, even when he should have known in advance what
he says really isn't true.
We all make occasional mistakes, but this stuff goes far beyond once in a
while. It appears to be chronic or habitual.
73 Tom
w8j...@aol.com
Wes Stewart
In article <19970115191...@ladder01.news.aol.com> w8j...@aol.com writes:
>From: w8j...@aol.com
>Subject: Re: Parasitic suppressor data
>>>Extracted from E-mail>>
>The suppressor graphs (RL.GIF) seem to show two things:
The graphs referred to by Tom can now be found on my home page at:
Wes N7WS
R. L. Measures
In article <19970115191...@ladder01.news.aol.com>,
w8j...@aol.com wrote:
snip...
> The VHF Q of the conventional suppressor could be made exactly the same
> as the nichrome ones by changing the value of either L or R, but the HF
> Q of the conventional suppressor would still be about a factor 2
> higher.<<-<<
I know who Mr. Rauch is quoting above, but readers, other than Wes, N7WS,
probably do not know. So far, the person who made this statement has not
come up with a conventional suppressor that will reduce the VHF voltage
gain of an amplifier by the same amount as a nichrome suppressor. Sure, I
can come up with a suppressor design using a high-Q silver-strap Ls (with
a different amount of inductance) and a different amount of resistance for
Rs, which yields a much lower VHF Q than any of the suppressors Wes
measured for us. However, such a suppressor creates a 'gotcha'/land mine
that the naysayers don't seem to be aware of yet. .............Stay tuned
for the next exciting episode. (Note: this gotcha is discussed in an
article about VHF suppressors that appeared in the March, 1989 issue of
QST magazine.)
...snip...
> This conclusion above is the same conclusion I reached, ... snip
What are the values of Rs and Ls that will produce exactly the same VHF Q,
Mr. Rauch?
...snip...
> If the data is reviewed, the ratio of Q is almost unity at 200 MHz and
> over 2:1 at 10 MHz. Just the slightest adjustment of R or L will make the
> conventional suppressor equal the nichrome at VHF, while maintaining
> better Q at HF where we want losses to be reduced.
Is a 1% (0.087db) loss at 29MHz going to make a difference at the
receiving end? Is a 0.25% (0.022db) loss at 14MHz going to make a
difference? Every benefit has a trade-off. Is a 1% power increase at
29MHz worth a 2/3, or more, increase in VHF voltage gain?
> For example, the AL-80B oscillates at about 170-180 MHz when the
> suppressor is removed.
However, AL-80s seem to occasionally oscillate at roughly 160MHz with the
factory suppressor installed. Here's a clip from an e-mail I recently
received from an AL-80 owner who reportedly had two bandswitches fail.
"Hi Rich: Sorry for the delay but we were out of town. I am sending
you 2 sets of toasted AL-80A bandswitches. The first time it arc'ed over
, I
had no idea what the problem was. Since I only used it on 75 meters, I
thought it had wimpy bandswitch wafers in it and that was the problem. I
then soldered the bandswitch in the 75 meter position and of course then
when it arc'ed it mad more of a mess than the first time. I had a
chance to put your suppression kit in and what a difference. You of
course knew this would be the result, but I am most pleasantly surprised.
The amp tunes up nice and smooth now with no crackling and popping
sounds emanating from inside......."
When Rick's 'toasted' AL-80 bandswitches arrive, I will post photographs
of them on the Web for all to see. .............Once again, Mr. Rauch has
been proved to be correct---i.e., the tune capacitor in the AL-80 has no
resonances around 160MHz that could cause it to arc over from a VHF
parasitic oscillation. Congrats, Tom.
> At that frequency, Wes measured a Q of about one
> (same as I posted weeks ago). The Q of a duplicate nichrome suppressor is
> about .75. This very small Q change is diluted by circuit effects when the
> suppressor is installed in an actual PA. The actual operating Q change
> would be even less at VHF, and more at HF.
Now that we have Wes' measurements, we know that a fairly typical low Q
suppressor has an ESR of 4.32 ohms at 30MHz and an ESR of 0.88 ohms at
10MHz. Perhaps Mr. Rauch, or another 'recognized amplifier expert' can
show us how to calculate the effect this would have on the Q of the tank
circuit in an AL-80B at 10MHz, and at 30MHz? These calculations are beyond
my capabilities.
>
> Rich's claim this small change in VHF Q is a great or worthwhile
> improvement that cures all woes is completely false.
Granted, the claims about mirculously curing gout have not been verified
by a double-blind scientific study, but preliminary evidence suggests that
verification is absolutely, positively, 100% certain.
...snip...
> 1.) The Miklos employment tale. He claimed he "called Eimac" and they told
> him Buzz was never R+D manager. Yet Eimac even issued a letter (after
> getting Buzz's permission) explaining Buzz was indeed R+D manager. Not
> only that, VHF Handbooks list Buzz as Eimac's former R+D Manager. Even
> after the employment confirming letter was out, Rich sent Buzz Miklos a
> post card telling Buzz he was never R+D manager!
Curiously, Mr. Rauch does not quote what I wrote. In the postcard, I
told Mr. Miklos that someone on this newsgroup was providing information
about his prior employment at Varian-Eimac that, according to the
Varian-Eimac's personnel department, did not match his personnel records.
> 2.) The tale about Marvin Born. Rich said Marvin measured parasitics in an
> AL1500. Turns out it wasn't even an AL1500, it was a dual 8877 PA built on
> a old empty prototype chassis. When I talked to Marvin he said he was
> positive Rich knew he was talking about a west coast manufactured 3CX3000
> PA, and not a AL-1500.
True, it was retrofitted with a second 8877. As I recall, tank
modifications were made to the AL-1500's tank to double the power handling
ability and halve the input Z in order to match two tubes instead of one.
However, Mr. Rauch's statements are somewhat puzzling. First Mr. Rauch
tells us it was a 'dual 8877 PA', but then he tells us it was a Create
Electronics Co. 3CX3000A7 amplifier.
If Mr. Rauch is correct, Marvin Born ordered the WRONG suppressor
retrofit-kit. To explain: According to my records, Marvin ordered kit
s/n 4548, which was for an AL-1500 with an extra resistor set for a second
8877. As I recall, Marvin told me that the AL-1500 had no VHF parasitic
suppressors and that he could see intermittent VHF parasitic oscillations
at 87MHz on a Tektronix 454 oscilloscope as well as on a spectrum
analyzer. According to my records, Marvin Born sent a cheque to pay for
suppressor retrofit kit s/n 4548. However, a suppressor for an AL-1500
would prompty incinerate in a 3CX3000A7 amplifier. I just telephoned
Marvin to see if a mistake was made. Unfortunately, no one was home.
I'll try again later.
>
...snip...
> 7.) Measures' attempt to blame me for Wes' unintentional error. Rich was
> completely aware of how this happened BEFORE he posted his pathological
> accusation that I was trying to hide something, and that I misled Wes.
'Unintentional error'? To the best of my knowledge, Wes did not error
anywhere.
RE: hiding something: Mr. Rauch cancelled his post of 28 November,
sans-explanation. I believe the reason why Mr. Rauch cancelled this post
has something to do with the belated realization that a small change in
VHF Q affects a large change in VHF Rp.
> When all this is considered it becomes perfectly clear Measures presents
> untrue statements as fact, even when he should have known in advance what
> he says really isn't true.
Mr. Rauch names no such statements. However,, in his next reply, perhaps
he will take me to to task on the gout issue. .
> We all make occasional mistakes, but this stuff goes far beyond once in a
> while. It appears to be chronic or habitual.
>
Amen to that, Mr. Rauch.
A question: An Rs of 100 ohms is connected in parallel with an Ls of
85nH. Assuming that the intrinsic L in Rs is zero, and the intrinsic R in
Ls is zero, please calculate the currents in Ls and Rs if 1V at 160MHz
when applied. In your expert opinion, Mr. Rauch, is the resultant current
in Ls insignificant compared to the current in Rs?
POSTSCRIPT
Did anyone notice that Mr. Rauch avoided the subject of Rp? When
evaluating suppressor performance, Rp is is an important consideration
because Rp and tube mu determine VHF voltage gain. Since we can't change
the mu of the tube, Rp is the only game in town for VHF voltage-gain
reduction.
------------------
e-mail copy to Mr. Rauch
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <measures-160...@port17.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) wrote:
...snip...
> Here's a clip from an e-mail I recently
> received from an AL-80 owner who reportedly had two bandswitches fail.
> "Hi Rich: Sorry for the delay but we were out of town. I am sending
> you 2 sets of toasted AL-80A bandswitches. The first time it arc'ed over,
> I had no idea what the problem was. Since I only used it on 75 meters, I
> thought it had wimpy bandswitch wafers in it and that was the problem. I
> then soldered the bandswitch in the 75 meter position and of course then
> when it arc'ed it mad more of a mess than the first time. I had a
> chance to put your suppression kit in and what a difference. You of
> course knew this would be the result, but I am most pleasantly surprised.
> The amp tunes up nice and smooth now with no crackling and popping
> sounds emanating from inside......."
> When Rick's 'toasted' AL-80 bandswitches arrive, I will post photographs
> of them on the Web for all to see. .............Once again, Mr. Rauch has
> been proved to be correct---i.e., the tune capacitor in the AL-80 has no
> resonances around 160MHz that could cause it to arc over from a VHF
> parasitic oscillation. Congrats, Tom.
>
Rick's toasted AL-80 bandswitches arrived today. I photographed them,
saved the images as JPEG files, and made them available on my Web site at
http://www.vcnet.com/measures/
The images are available at Figures 15a and 15b.
BTW, replacement AL-80 bandswitches cost $58.22 (inv #02745).
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
> 2.) The tale about Marvin Born. Rich said Marvin measured parasitics in an
> AL1500. Turns out it wasn't even an AL1500, it was a dual 8877 PA built on
> a old empty prototype chassis. When I talked to Marvin he said he was
> positive Rich knew he was talking about a west coast manufactured 3CX3000
> PA, and not a AL-1500.
>
question was built on an AL1200 chassis, had two 8877s, now has one 8877,
and uses a low VHF Q parasitic suppressor intended for an AL-1500.
Mr. Rauch is correct in stating that Marvin's spectum analyzer detected a
parasitic-oscillation (at 78MHz) in a Create 3CX3000A7 g-g amplifier that
had vapourized various bandswith contacts and destroyed its VHF parasitic
suppressor. Marvin said he replaced the bandswitch, and replaced the
factory-stock parasitic suppressor with an all-nichrome, resistorless, low
VHF Q parasitic suppressor. No further arcing problems have been reported.
Congrats, Tom
--
--Rich-- ag6k, 805-386-3734
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>> 2.) The tale about Marvin Born. Rich said Marvin measured parasitics in
an
>> AL1500. Turns out it wasn't even an AL1500, it was a dual 8877 PA built
on
>> a old empty prototype chassis. When I talked to Marvin he said he was
>> positive Rich knew he was talking about a west coast manufactured
3CX3000
>> PA, and not a AL-1500.
>>
>I just had a conversation with Marvin Born. He said the amplifier in
>question was built on an AL1200 chassis, had two 8877s, now has one 8877,
>and uses a low VHF Q parasitic suppressor intended for an AL-1500.
>Congrats, Tom
It's not just this time Rich, it's nearly all the time. When your stories
are looked at, they almost always wind up being little tiny threads of
fact woven into big fabrications. It's just an attempt to support your
view of how the world works.
When any person repeatedly posts or distributes factoids that are almost
entirely figments of imagination, he looses all credibility. You've done
that over and over again.
Even when you admit being wrong, you still can't control yourself and
"slip in" another false claim!
Look at a small part of your history on this thread:
If Dave had not corrected you in public about the real cause of his
TL-922's arcing (a bad lightning arrestor), you likely would have posted a
picture of the capacitor on your web page and slammed Kenwood even more
than you have for parasitics. You declared the 922 cap damaged from
parasitics, sent Dave a kit, and almost got away with "another one"
except, before installing the kit, Dave found the problem. (By the way,
that was a problem you said could NEVER occur. You plainly said HF signals
or improper loading or antenna termination could NOT cause a PA to arc.)
Just hours after that, WB8BFS came on the thread and said his 922 arced
from a bad relay.
More recently you declared all amateur PA's poorly designed because they
had no rheostats like commercial transmitters, and declared no proper
engineer would purchase a transmitter without one. Several people rebuked
you for saying that, one of whom is a top notch engineer for a large
commercial BC transmitter manufacturer.
Now you've moved on to the AL-80 or something with more Richard L.
Measures picture "evidence". Once in a while you'll find a sucker who buys
into your fantasy, but your days of misleading large numbers of people are
over.
73 Tom
Joe Subich
In message <measures-ya0240800...@news.vcnet.com> -
meas...@mail.vcnet.com (R. L. Measures )Fri, 17 Jan 1997 16:15:27 -0800
writes:
:>
:>
:>Rick's toasted AL-80 bandswitches arrived today. I photographed them,
:>saved the images as JPEG files, and made them available on my Web site
:>at
:>
:>http://www.vcnet.com/measures/
:>
:>The images are available at Figures 15a and 15b.
The pictures are interesting and prove nothing.
The arcing visible in the pictures could just as easily been caused
by a mistuned amplifier running into a high VSWR. I'm sure someone
will correct me if I get it backward but in and underloaded amplifier
driving a high impedence load the voltage across the bandswitch (and
tuning capacitor) will soar nearly out of control. Underloading and
high feedline impedences are particularly common on 80 and 160 meters
as many amplifiers lack sufficient range in their output capacitors to
achieve proper loading ... particularly if the load impedence is very
capacitive.
Rule #1 ... if one hears hoofbeats don't expect zebras.
--
========================================================================
Joe Subich, W8IK ex-AD8I
<W8...@IBM.NET>
<W8...@AMSAT.ORG>
<sub...@ibm.net>
<71350...@compuserve.com>
========================================================================
w8j...@aol.com
In article <5brrbe$1tq6$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) writes:
>The arcing visible in the pictures could just as easily been caused
>by a mistuned amplifier running into a high VSWR. I'm sure someone
>will correct me if I get it backward but in and underloaded amplifier
>driving a high impedence load the voltage across the bandswitch (and
>tuning capacitor) will soar nearly out of control. Underloading and
>high feedline impedences are particularly common on 80 and 160 meters
>as many amplifiers lack sufficient range in their output capacitors to
>achieve proper loading ... particularly if the load impedence is very
>capacitive.
Hi Joe,
That's basically the idea.
Anything that underloads the PA for the amount of drive power applied
allows voltage to build up in the tank. The most common components ruined
are either the bandswitch or the tuning cap.
Dave had that happen with his 922 Kenwood, when his defective lightning
suppressors fired. Of course, Rich blamed it on a para$itic...and sold him
a kit.
I wonder why ?
73 Tom
comm...@aol.com
In article <19970119144...@ladder01.news.aol.com>,
w8j...@aol.com writes:
>>The arcing visible in the pictures could just as easily been caused
>>by a mistuned amplifier running into a high VSWR.
I have to disagree. Tom, I have a SX-42 receiver here with damaged
RF stage components -- arced bandswitches, burnt resistors, and
defective 6AG5 pentode tubes.
A few years back I was monitoring WWV during an intense solar
storm. Imagine my horror when a bandswitch section began arcing
over, carbonizing the wafer! IMHO this was caused by photon
bombardment. I will sell you a picture of the bandswitch damage
for your webpage, if you are interested.
Later that year, I had a long wire connected during an electrical storm.
Again, the wafer switches began arcing over. IMHO this was caused
by parasitic oscillations in the 6AG5 pentodes induced by the high
voltage transients present at the antenna terminals.
Hallicrafters used resistive loaded rf chokes in the screen supplies,
and 15 ohm resistors in the grid and plate leads. I suspect had they
used nichrome wire, these problems would have been avoided. The
last straw was when the parasitics took out a mica plate bypass
capacitor -- destroying the 1.2k plate resistor.
Tom, what is the saturated cathode current of a 6AG5? Folks, why
won't Tom answer this question?
One of the RF stages had damaged screen resistors as well and a
destroyed 6AG5. I will smash open the tube and sell you pictures
if you are interested.
Imagine my horror when I discovered sharp dips across the entire
spectrum with my Millen griddipper (The receiver covers 540-kHz to
108-MHz) while probing the RF compartments!!
I have one final question.... How does Rich measure the LOADED
Q of a PA tank circuit using a grid dipper? I see many comments
regarding the use of his nichrome parasitic suppressors. The most
common comment is "the vernier tuning effect" after the suppressors
are installed. This would seem to indicate the loaded Q of the HF
tank circuit is being compromised by the addition of the suppressors.
Isn't the loaded Q value an important parameter in designing a
tank circuit for optimum impedance transformation?
Peter Bertini K1ZJH
Senior Technical Editor, Communications Quarterly Magazine
"The Radio Connection," Popular Communications Magazine
-- proud members of the CQ family --
Gary Coffman
In article <19970119175...@ladder01.news.aol.com> comm...@aol.com writes:
>In article <19970119144...@ladder01.news.aol.com>,
>w8j...@aol.com writes:
>
>>>The arcing visible in the pictures could just as easily been caused
>>>by a mistuned amplifier running into a high VSWR.
>
>I have to disagree. Tom, I have a SX-42 receiver here with damaged
>RF stage components -- arced bandswitches, burnt resistors, and
>defective 6AG5 pentode tubes.
>
>A few years back I was monitoring WWV during an intense solar
>storm. Imagine my horror when a bandswitch section began arcing
>over, carbonizing the wafer! IMHO this was caused by photon
>bombardment. I will sell you a picture of the bandswitch damage
>for your webpage, if you are interested.
>
>Later that year, I had a long wire connected during an electrical storm.
>Again, the wafer switches began arcing over. IMHO this was caused
>by parasitic oscillations in the 6AG5 pentodes induced by the high
>voltage transients present at the antenna terminals.
I'd suggest *that* was the result of photon bombardment. Probably
about 2 MHz photons. :-)
Gary
--
Gary Coffman KE4ZV | You make it, | Due to provider problems
Destructive Testing Systems | we break it. | with previous uucp addresses
534 Shannon Way | Guaranteed! | Email to ke...@radio.org
Lawrenceville, GA 30244 | |
R. L. Measures
In article <19970118154...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
>
> If Dave had not corrected you in public about the real cause of his
> TL-922's arcing (a bad lightning arrestor), you likely would have posted a
> picture of the capacitor on your web page and slammed Kenwood ...snip...
Dave contacted me privately before the public announcement. About 4 years
ago, in my 8169 amplifier, I experienced a similar tank arc which was
caused by an arc in my 160m Hertz antenna's L-network tuner, 100 feet down
the coax feedline, out in the back yard.
According to Dave, he bought the TL-922 used. Dave said he noticed there
were arc marks on the tuning capacitor before he put the TL-922 into
service.
According to QST Staffer Dave Newkirk, the TL-922 at W1AW had a tune
capacitor that became so badly erroded from mysterious arcing, the tune
capacitor was replaced with a new unit. At Dave's request, I sent
suppressor retrofit kit s/n 293 (5/23/89) to to the folks in charge of
W1AW. The factory-stock suppressors were reportedly replaced with low
VHF-Q suppressors. According to my records, the ARRL paid for retrofit kit
s/n 293.
The burned bandswitch shown in "Parasitics Revisited" (9/90, 10/90 QST
magazine) was removed from a TL-922. The ham who owned this amplifier
lives in Phoenix, AZ. No lightning arrestor was connected to the TL-922.
The burned bandswitch was replaced with a new unit. The beautiful silver
plated factory-suppressors were replaced with ugly, low VHF Q suppressors.
I have spoken with a number of TL-922 owners who have had also had the 10m
and 15m contacts burn up on the bandswitch.
> More recently you declared all amateur PA's poorly designed because they
> had no rheostats like commercial transmitters, and declared no proper
> engineer would purchase a transmitter without one. Several people rebuked
> you for saying that, one of whom is a top notch engineer for a large
> commercial BC transmitter manufacturer.
W6TG said that all ten of the commercial amplifiers he takes care of have
continuously adjustable control of the filament voltage.
Unless my news server missed a post, the 'top notch engineer' failed to
answer the question about the peak inrush current through the rheostat
brushes, failed to explain why they were using carbon rheostat brushes
instead of graphite rheostat brushes, and failed to say whether or not the
amplifiers with the high rheostat-failure rates used a step-start circuit.
.
As I recall, the Collins KWS-1, 208-U... general coverage amplifiers, and
30S-1 amplifier use a rheostat to adjust filament voltage. I have never
heard of a rheostat failure in one of them. // Was the 'top notch
engineer' title awarded by a recognized amplifier expert? If so, who
recognized this expert? Based on my experiences, knowledgeable engineers
do not stonewall questions.
> Now you've moved on to the AL-80 or something with more Richard L.
> Measures picture "evidence". ...snip...
There is no 'or something', Mr. Rauch. According to the AL80A's owner,
Rick May, the original bandswitch burned up, and the first replacement
bandswitch burned up. He is currently using bandswitch #3, which is not
arcing. However, Rick said he is no longer using the AL-80's original VHF
parasitic suppressor. The photographs of the original and the replacement
AL-80A bandswitches with missing/vapourized contacts are available at:
http://www.vcnet.com/measures/
See Figure 15.
Rick's e-mail address is listed.
Seems like Mr. Rauch is suggesting that my two AL-80A bandswitch
photographs are somewhat less than authentic---i.e., fake..
Hmmm.
OJ swore that the picture evidence of him wearing size 12 Bruno Magli shoes
is a fake. Tom: In your expert opinion, are the news photographers, who
subsequently produced 30 other photographs of OJ wearing size 12 Bruno
Maglis, slamming OJ? IMO, prevaricators slam themselves. Is
impeachment slamming? A well-known teacher reportedly said that everything
that defiles a man comes out of his own mouth.
A while back, you groused about a lack of discussing technical issues on
this thread. How about discussing Wes Stewart's VHF Rp measurements? VHF
Rp is a determiner in VHF voltage gain, so discussing it seems appropriate.
BTW, Tom, what are the values of Rs and Ls in your new suppressor design
that will exhibit equal to or better VHF characteristics than any of the
suppressors Wes tested?
---------------------------
J. Fred Riley
meas...@mail.vcnet.com (R. L. Measures ) wrote:
> my 160m Hertz antenna
That would be a dipole?
I take it you are referring to me in your post when you say I
> failed to answer the question about the peak inrush current through the rheostat
>amplifiers with the high rheostat-failure rates used a step-start circuit
used that limited the inrush current to twice operating. No
step-start would be needed since the voltage comes up gradually as the
current drops. (As an aside, I experimented with the use of
thermistors as inrush current limiting devices. They were widely used
in many older TV sets. They worked quite well for me but their
lifetime was limited--which kept TV repairmen in business.)
>failed to explain why they were using carbon rheostat brushes
Well, I did fail to explain that but did, instead, point out that we
have noted a great deal of unexplained brush failure in filament
circuits when the same RMS current causes no failure in a commutating
supply transformer primary, e.g., a bias supply. (We never changed
the brush type or the filament control mechanism in the production
transmitter because it was out of production when we started seeing
the problem.) Forrest Cummings, the designer, may have been
influenced in his selection of filament adjust mechanisms by the
transmitter designs you mention below.
> As I recall, the Collins KWS-1, 208-U... general coverage amplifiers, and
>30S-1 amplifier use a rheostat to adjust filament voltage. I have never
>heard of a rheostat failure in one of them.
which had a strontium-oxide coated cathode while the KWS-1 used
(originally) 4X150A's. The 4CX1000A tube was still a new item when
the 30S-1 was designed and the tube manufacturer had not been
able--has never been able--to get more than 6000 hours out of that
type cathode when operated at full current (big proviso). The tube
was also oriented at an angle rather than mounted vertically as in the
KWS-1. Good heater/cathode temperature control was considered
essential to maximize the life expectancy when these amplifiers were
used in commercial service (higher duty cycle). High cathode
temperature could lead to oxide flaking and internal shorts. Because
the filament transformer was located in a difficult-to-access area, no
taps were provided and yet the amplifier had to operate over a wide
range of voltages. That's why the heater control was incorporated.
The choice of a rheostat was made, I assume, because of cost and size.
The second issue is whether there have been rheostat failures in these
Collins transmitters with heater/cathode tubes. And I can't comment
on that except from personal experience. I've owned two 30S-1's and
one KWS-1 and never had any problem. The low-capacity filament
transformer of the 30S-1 limited the surge current of the heater--just
as it did in the filaments of the more-recently designed MW
transmitter I used as an example. You realize, of course, that the
fact I never had any failures doesn't mean there weren't any. I can
tell you, however, that rheostat failures on other broadcast
transmitters (the Gates FM-20H, for example) were quite common and
that we haven't used them in a new commercial design for twenty years.
I hope this post clarifies any questions about my previous post.
Fred, W8OY
w8j...@aol.com
>Dave contacted me privately before the public announcement. About 4
years
>ago, in my 8169 amplifier, I experienced a similar tank arc which was
>caused by an arc in my 160m Hertz antenna's L-network tuner, 100 feet
down
>the coax feedline, out in the back yard.
If Dave contacted you privately before the public announcement, why did
you go ahead announce the arcing was caused by a parasitic? If you
experienced an arc in your tuner that caused your PA's tank capacitor to
arc, why didn't you ever indicate or admit an arcing problem could be load
related?
In article <measures-301...@port20.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>The voltages that can result from an intermittent vhf oscillation are
>impressive. For instance the TL-922 uses a tuning capacitor that tests
>out at well over 5000vdc.
>The max. RF peak voltage across this capacitor
>is normally around 2700v with the electric-mains supplying 240v to the
>922. Other manufacturers of similar amplifiers typically use a tune
>capacitor with only 60% of the air gap in the 922's tune capacitor, so
the
>safety factor is awesome. However, when the 922 oscillates at 140 MHz,
>the tune cap. can arc. If you are lucky enough to see this happen, the
>arc is purple---a strong indicator that vhf voltage is present.
In article <measures-231...@port19.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
> I have seen large air gaps arc for no apparent
>reason. An arced TL-922 tune capacitor is currently being shipped to me
>from NM for testing. I will test it and post the results here. When
this
>tune capacitor was replaced with a new unit, it also arced.
>IMO, the
>mysterious intermittent arcing is VHF parasitic related. .
In article <measures-231...@port10.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>This afternoon I tested David Holtcamp's TL-922 tune capacitor. The
>breakdown voltage was about 6100V. David observed that this capacitor
>arced during operation on the CW tap.
>The HF voltage present on the CW
>tap is around 2000V IMO, the intermittent arcing of the 6000v+ tune cap.
>was due to a largely-unloaded VHF parasitic oscillation.
Three times you indicted the 922 arcing had to be a parasitic. It was your
opinion. It was your conclusion.
You posted that conclusion knowing full well that it could be a load
problem, you admit you did so **after** Dave indicated in E-mail it was
an arcing lightning arrestor.
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>About 4 years
>ago, in my 8169 amplifier, I experienced a similar tank arc which was
>caused by an arc in my 160m Hertz antenna's L-network tuner, 100 feet
down
>the coax feedline, out in the back yard.
You intentionally omitted evidence you discovered four years earlier that
contradicts your public posture that parasitics are the only cause of tank
circuit arcing.
You misled us about the AL-1500 that you NOW admit was not even an
AL-1500. (Marvin said there is no way you didn't know that from the very
beginning, he said he was very clear about what the PA was.) You did the
same with your claims about conversations with Eimac.
Rich, how can anyone trust you?
73 Tom
Roy Lewallen
I'm really reluctant to touch this tar baby, but my curiosity has been
aroused.
Why do VHF parasitics cause band switches to arc?
Is a VHF parasitic oscillation the only thing which can? (I assume so,
since the presence of arcing bandswitches seems to be presented as adequate
proof of VHF parasitics.)
Has anyone ever actually directly detected the presence of VHF oscillation
(with a simple wavemeter, for example) in one of these amplifiers, then
confirmed that the condition no longer exists after applying a fix?
Is this science or art?
Roy Lewallen, W7EL
R. L. Measures
In article <5c23r9$9t9$2...@nadine.teleport.com>, w7...@teleport.com (Roy
Lewallen) wrote:
> I'm really reluctant to touch this tar baby, but my curiosity has been
> aroused.
>
> Why do VHF parasitics cause band switches to arc?
Might there be fireworks if a 1000W, 100MHz oscillator were operated with
virtually no load?
In a typical HF amplifier, the tank circuit is a low-pass Pi filter that
blocks VHF energy. Thus, if VHF regeneration appears at the anode, it has
no where to go. IMO, something would be likely to give.
Bandswitches do not necessarily arc during a VHF parasitic. When Heath
was using a 3500v-rated tune-capacitor in the SB-220, bandswitches rarely
arced, but reports of tune capacitor arcs were common. I own such an
amplifier, and the tune-cap. has numerous pits from arcing. However, when
Heath started using a tune-capacitor with a higher v-rating, reports of
bandswitch arcing in such amplifiers became fairly common.
Arcs typically prefer to jump the gap with the lowest breakdown voltage.
However, like Murphy said, Nothing is as simple as it looks. Thanks to
tests performed by Mr. Rauch, another factor appears to be the internal VHF
resonances in the tune-capacitor. Based on his measurements, it seems that
a tune-capacitor which does not exhibit a VHF resonance near the frequency
of parasitic oscillation is less likely to arc than are contacts on the
bandswitch. (see Figure 15 on
<http://www.vcnet.com/measures/>) In these photographs, the 10m and 15
meter contacts are partially vapourized. The amplifier from which the two
burned bandswitches were removed has no VHF resonance near the frequency of
oscillation (c.160MHz). IMO, burned 10m & 15m bandswitch contacts is the
signature of a VHF parasitic oscillation. [See kaput bandswitch in
"Parasitics Revisited", 9/90 and 10/90 QST. Also see my first post on 1/17
(<measures-160...@port17.vcnet.com>,)]
> Is a VHF parasitic oscillation the only thing which can? (I assume so,
> since the presence of arcing bandswitches seems to be presented as adequate
> proof of VHF parasitics.)
There are other sources of tank arcs. An antenna-tuner capacitor arc
caused an arc in the Johnson 226 roller-inductor in my general-coverage
homebrew 8169 amplifier. David Holtcamp, K5KH, recently discovered that a
defective lightning arrestor installed in the feedline caused a tank arc in
his TL-922. Moths, spiders, and heavy lint accumulation can cause arcs. I
had an occasional spider-problem in an 8171 amplifier that uses a 9100v
anode supply.
The best protection against peripheral amplifier damage, as the result of
arcs, is a fault-current limiting (a.k.a. 'glitch') resistor, in series
with the anode supply. Such a resistor must be able to survive discharging
the high voltage filter capacitor. Svetlana recommends a glitch R of c.25
ohms for a 1500w anode-dissipation tube. Glass-coated, wire-wound
resistors can be used for this purpose, provided that they have a high
length/diameter ratio. Eimac began recommending such resistors
around1986.
> Has anyone ever actually directly detected the presence of VHF oscillation
> (with a simple wavemeter, for example) in one of these amplifiers, then
> confirmed that the condition no longer exists after applying a fix?
Marvin Born reportedly saw a 78MHz parasitic oscillation on a spectrum
analyzer and on an oscilloscope in a Create Electronics 3CX3000A7 g-g
amplifier. After replacing the burned bandswitch and installing a 100%
nichrome VHF parasitic suppressor, no arcing was observed in the
replacement bandswitch. However, I am certain that low-level damped wave
oscillations at 78MHz still existed in this amplifier whenever the anode
current decreased or increased suddenly. The best way to observe this
phenomenon is to send 50wpm dits with a CW keyer.
I do not have a spectrum analyzer, so I have to depend on those who do.
A customer who bought a suppressor retrofit-kit for an SB-220 reported that
he could see damped-wave VHF ringing in the anode circuit when the
amplifier appeared to be stable. No mystery there, Roy,---ringing a
tuned-circuit with a current pulse is basicly how spark transmitters
generated RF in days of yore.
> Is this science or art?
Some say art. Some say black art. Some say $cam. Some say unsliced
bologna. Some say science. However, it's nothing new. F.E. Handy wrote
about the method on page 72 in the 1926 *Radio Amateur's Handbook*.
I find it somewhat curious that one of the naysayers who claim that low
VHF Q suppressors are the worst kind of 'wurst', rejects areas of science
that do support his conclusion. This person is seemingly unfamiliar with
algebra and scientific notation, and rejects accepted AC Circuit Analysis
methods whenever they fail to support his conclusions.
When Wes Stewart's scientific measurements indicated that a conventional
VHF suppressor would create 63% more VHF voltage gain than a similar low
VHF-Q suppressor, this person switched into ignore-mode and
smokescreen-mode.
------------------------------
e-mail copy to Mr. Lewallen
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <19970120114...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
...snip...
.............
...snip..
Bzzzzzzzzzzzzzzzzzzt. (cough, cough).........and the sensitivity control
is set on zero. Amazing........... (cough)
Mr. Rauch:
Wes Stewart, N7WS, appears to have gone to considerable trouble making
scientific measurements of conventional copper vs. Ni-Cr-Fe and Ni-Cr
suppressors with a Hewlett-Packard Model 4191A RF Impedance Analyzer. IMO,
discussing Wes' suppressor measurements could resolve the core issue in
this debate to the satisfaction of most. .
-----------------------------
w8j...@aol.com
R. L. Measures
In article <19970123051...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
>
> >Dave contacted me privately before the public announcement. ...snip....
>
> If Dave contacted you privately before the public announcement, why did
> you go ahead announce the arcing was caused by a parasitic?
...snip...
You would do well to check the dates and times of these events in Will's
archive, Mr. Rauch.
>
...snip...
> >IMO, the
> >mysterious intermittent arcing is VHF parasitic related. .
>
...snip...
>
Mr. Rauch---FYI, on the Internet, 'IMO' is an abbreviation for 'in my
opinion'. Is an opinion the same as a conclusion?
There are obviously lots of things that can cause tank arcs. I provide
room and board for a male cat named B.W., who has all of his original
equipment, who, when the mood suits him, emits a rather impressive cloud of
conductive mist out of his stern end, that smells quite unlike Chanel
Number Five®.
Why don't you ask David Holtcamp what I told him before and after he
discovered that the lightning arrestor was the instigator of the tank arc.
How do you explain the arc pits that David observed on the tune capacitor
prior to the use of said lightning arrestor? Why did the TL-922 at W1AW,
and numerous other TL-922s, apparently stop arcing after the vhf parasitic
suppressors were replaced by low vhf-Q/low vhf-Rp suppressors?
> You posted that conclusion knowing full well that it could be a load
> problem, you admit you did so **after** Dave indicated in E-mail it was
> an arcing lightning arrestor.
Those who would like to see if Mr. Rauch has his facts straight can look in
Will's archive.
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
>
> >About 4 years
> >ago, in my 8169 amplifier, I experienced a similar tank arc which was
> >caused by an arc in my 160m Hertz antenna's L-network tuner, 100 feet
> down
> >the coax feedline, out in the back yard.
>
> You intentionally omitted evidence you discovered four years ......snip......
Nah, this was discussed at length on the air at the time it happened. I
have written about it in e-mail.
>
> You misled us about the AL-1500 ...snip...
ROFL, L ,L, L........how many times did I ask you whether or not the
AL-1500, which you say you designed, has a vhf parasitic oscillation
suppressor? I'll bet a large pizza with 3 toppings I asked you at least
six times, Mr. Rauch. Sure the amp. Marvin was working on was built on a
AL-1200 chassis, housed in an AL-1200 cabinet, and it currently has one
8877 instead of one 3CX1200A7. Curiously, when Marvin made this
clarification to me on the telephone, he had no idea whatsoever who Tom
Rauch, W8JI, is. Within an hour after talking to Marvin on the telephone,
I posted the correction. Most fortunately, since I am NOT a
self-recognized amplifier expert, I don't have a title to defend, so
admitting blunders is no-prob. at all
>
> Rich, how can anyone trust you?
Some folks seem to. Some folks do not.
Did you design the Heath SB-1000 amplifier?
Is the bandswitch in the SB-1000 interchangeable with the bandswitch in
the AL-80?
IMO, at this point, we need to discuss Wes, N7WS's parasitic suppressor
measurements..
--
--Rich-- ag6k, 805-386-3734
comm...@aol.com
In article <Uyv4lCAr...@ifwtech.demon.co.uk>, "Ian White, G3SEK"
<G3...@ifwtech.demon.co.uk> writes:
>
>Roy Lewallen wrote:
>>I'm really reluctant to touch this tar baby, but my curiosity has been
>>aroused.
>>
>>Why do VHF parasitics cause band switches to arc?
>>
>>Is a VHF parasitic oscillation the only thing which can? (I assume so,
>>since the presence of arcing bandswitches seems to be presented as
adequate
>>proof of VHF parasitics.)
>
>Two very fair questions, to which I'd like to add:
>
>When the arcs occurred, were the switches actually using the contacts
>that arced, or were they set to another band?
>
>If the arcs really were due to VHF currents flowing in the
>inductor/bandswitch combination, there surely has to be a strong
>spurious VHF resonance somwhere in there. Without a VHF resonance in the
>inductor, how could it develop VHF voltages/currents high enough to
>create an arc? Has anybody detected such a thing with a GDO on/around
>the pi-tank coil?
>
>If there is no detectable VHF resonance, that argues very strongly
>against the arcs having been caused by VHF energy.
>
>>
>>Has anyone ever actually directly detected the presence of VHF
oscillation
>>(with a simple wavemeter, for example) in one of these amplifiers, then
>>confirmed that the condition no longer exists after applying a fix?
>>
>>Is this science or art?
>
>Darn good question.
>
>73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
>
My two questions: Why should the presence of spurious
VHF resonances be taken as proof that the amplifier is
being damaged by VHF parasitics? Where is the feedback
path, and are the phase angles such that oscillation are
suppressed or enchanced?
I can understand the plate tuning capacitor breaking down, but
the bandswitch -- even on ten meters -- is downline of the
tank coil and in parallel with the loading capacitor. To my simple
mind, this sure looks like a lowpass filter.... What are
the impedances of these components at the parasitic frequency,
and what will the VHF voltages be at the bandswitch when the
parasitic occurs? Why should the bandswitch arc over before the
breakdown voltage of the tuning capacitor is exceeded? Maybe the
cause might be instead related to high SWR or mistuning?
What is the main component of the "VHF" resonance? Is it
the plate strap combined with the internal plate assembly
within the PA tube?
Why doesn't the parasitic oscillation manifest itself whenever the
amplifier is in use? If it is a very sporadic occurence, how do we
know for sure if any bandaid fixes are in reality curing the problem?
Ie: How do we know the same variables that caused the initial
incident to happen are being duplicated?
How can someone armed only with a grid-dip meter, and a neon lamp --
sans spectrum analyzer, network analyzer, Q-meter, etc. determine
what is the real cause of any amplifier flashovers, and then market
a UNVERSAL fixall?.
Without laboratory measurements and scientific analysis, it
is bad science.
Ian White, G3SEK
Roy Lewallen wrote:
>I'm really reluctant to touch this tar baby, but my curiosity has been
>aroused.
>
>Why do VHF parasitics cause band switches to arc?
>
>Is a VHF parasitic oscillation the only thing which can? (I assume so,
>since the presence of arcing bandswitches seems to be presented as adequate
>proof of VHF parasitics.)
Two very fair questions, to which I'd like to add:
When the arcs occurred, were the switches actually using the contacts
that arced, or were they set to another band?
If the arcs really were due to VHF currents flowing in the
inductor/bandswitch combination, there surely has to be a strong
spurious VHF resonance somwhere in there. Without a VHF resonance in the
inductor, how could it develop VHF voltages/currents high enough to
create an arc? Has anybody detected such a thing with a GDO on/around
the pi-tank coil?
If there is no detectable VHF resonance, that argues very strongly
against the arcs having been caused by VHF energy.
>
>Has anyone ever actually directly detected the presence of VHF oscillation
>(with a simple wavemeter, for example) in one of these amplifiers, then
>confirmed that the condition no longer exists after applying a fix?
>
>Is this science or art?
Darn good question.
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
Professionally:
IFW Technical Services Clear technical English - world-wide.
David Holtkamp
w8j...@aol.com wrote:
> If Dave contacted you privately before the public announcement, why did
> you go ahead announce the arcing was caused by a parasitic? If you
> experienced an arc in your tuner that caused your PA's tank capacitor to
> arc, why didn't you ever indicate or admit an arcing problem could be load
> related?
And,
> You posted that conclusion knowing full well that it could be a load
> problem, you admit you did so **after** Dave indicated in E-mail it was
> an arcing lightning arrestor.
Tom, I don't want to get in the middle of this in the least, but as I
recall, I posted Rich privately slightly before I posted it publicly
(attempting to be polite to him, since I knew his diagnosis and what I
found were not in agreement). I remember reading Rich's last message
posted *before* he got my email or the public posting - they had passed
by each other in the Internet Ether by some minutes or hours - I don't
remember exactly.
Just info to try and set the precise record straight.
Best Wishes es 73s de David (K5KH)
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>
>Mr. Rauch---FYI, on the Internet, 'IMO' is an abbreviation for 'in my
>opinion'. Is an opinion the same as a conclusion?
Please show me where your published documents state IMO. I can't find the
disclaimer that it is just an opinion.
>equipment, who, when the mood suits him, emits a rather impressive cloud
of
>conductive mist out of his stern end, that smells quite unlike Chanel
>Number Five®.
Thank you for sharing that important technical information. I don't
believe most normal people allow cats to whiz on their equipment, and then
publicly compare arcing cat urine to prefume.
> Why don't you ask David Holtcamp what I told him before and after he
<snip>
> Curiously, when Marvin made this
>clarification to me on the telephone,
<snip>
Richard, let's suppose your new story is factual or truthful. That being
the case, you posted things before having facts.
You go so far as to say things like "Marvin Born measured the parasitic in
an AL-1500 with a spectrum analyzer" and use that as an accusation against
my abilities or one of the typical Measure's slams on a manufacturer's
product. Yet now you admit you made a mistake, and misrepresented fact.
Rather than admitting you fibbed or intentionally changed data, you want
us to believe you just made a simple mistake and didn't have all the
facts. Same with the TL-922.
If you post conclusions, suggestions, and/or form opinions without having
heard the facts, how can we trust anything you say? Either we can trust
you to be accurate and thorough, or we can't. IMO, if we can't trust
everything a sole source presents, we can't trust anything presented by
that source.
Surely even you can see it makes no difference if your stories are
premature, an incorrect conclusion, or outright pathological lies. The
only thing that matters is the information presented is often wrong, and
you (for what ever reason) can't be trusted to supply accurate data. This
isn't just one or two things Rich, it's many items.
1.) The dissipation of an 8877 grid. It is not 500 watts in the Eimac data
tables Rich.
2.) The failures of 8877's. Eimac did have a manufacturing problem in the
80's that you blamed on parasitics. You were caught on that one.
3.) Photons causing arcs. Even you admit QST couldn't be duped into
printing that one.
4.) Your claim the peak tank voltage could never exceed the supply rail.
You argued that point with Mario and lost.
5.) Your claim about Buzz Miklos' employment, and that Eimac released data
to you. You were caught on that one.
6.) Your claim about the variacs and filament controls in BC transmitters.
You were caught be several people telling that one.
7.) Your "opinion" about Dave's 922. You were caught on that one.
8.) Your early statements about Marv Born's measurements. You were caught
on that one.
It's easy to find eight things that are untrue, claims you argued about
and defended. The reason *why* you present false claims over and over
doesn't matter. The bottom line is your claims can't be trusted.
73 Tom
R. L. Measures
In article <19970123222...@ladder01.news.aol.com>,
comm...@aol.com wrote:
...snip...
> My two questions: Why should the presence of spurious
> VHF resonances be taken as proof that the amplifier is
> being damaged by VHF parasitics?
Because there are clues that such an event probably took place, Peter.
What else would be likely to destroy a grid-I meter. it's shunt-R, short a
cathode-bias zener diode, cause the parasitic-suppressor R to suddenly
increase in resistance by several times without showing any external sign
of damage, and/or summarily weld a 3-500Z's filament to its grid?
The type of VHF parasitic-oscillation that causes such damage probably
lasts under a millisecondand typically creates a big bang From my
experiences, such an event causes an immediate failure of one or more
amplifier components, so having a long look at one is quite unlikely.
.
> Where is the feedback
> path, ...
The feedback path is between the anode and the input element of the
amplifier tube. In a 3CX1200A7, the feedback C is around 0.8 pF. At the
typical anode-circuit VHF resonance, the Xc is around 2000 ohms between the
anode/output and the cathode/input.. In your opinion, is 2000 ohms of
reactance significant at, say, 4000v?
> ...and are the phase angles such that oscillation are
> suppressed or enchanced?
I don't know.
> I can understand the plate tuning capacitor breaking down, but
> the bandswitch -- even on ten meters -- is downline of the
> tank coil and in parallel with the loading capacitor.
I have never heard a report of a bandswitch L contacts arcing while an
amplifier was being used on 10m. Afterall, on 10m, all of L contacts are
closed. // The typical HF tank-L is hardly a VHF RF-choke. By measuring
the VHF self-resonances in the tank-L with a dipmeter when the bandswitch
is in the 80m and160m positions---and by. measuring the anode-circuit's VHF
resonance at the stud on the DC blocking capacitor, a clue might be
provided. If any of these resonances are sympathetic, an interaction is
seemingly possible.
According to measurements made by Mr. Rauch, VHF self-resonances in the
tune C appear to influence whether or not a parasitic arc is likely to
occur therein.
> To my simple
> mind, this sure looks like a lowpass filter....
>What are
> the impedances of these components at the parasitic frequency,'''
I don't know.
> and what will the VHF voltages be at the bandswitch when the
> parasitic occurs?
The measured breakdown voltage of a TL-922's or a SB-220's or an AL-80's
ceramic bandswitch exceeds 5500V. The RF peak voltage in these amplifiers
is under 2800v. In the TL-922, the measured breakdown voltage of the tune
C is 6100v. I have seen evidense of arcing in all of these components that
can not be satisfactorily explained by other means If you want to see some
photos of burned bandswitches, see Figure 15 at
http://www.vcnet.com/measures/
>Why should the bandswitch arc over before the
> breakdown voltage of the tuning capacitor is exceeded? Maybe the
> cause might be instead related to high SWR or mistuning?
Possibly some are. However, some are probably not. Have you read the
report from Rick May, KC0HH?
>
> What is the main component of the "VHF" resonance?
The unchangeable constant is the anode-C.
> Is it the plate strap combined with the internal plate assembly
> within the PA tube?
That is part of the total L. However, every L, between the anode C and the
tune C is part of the VHF resonance. Naturally, the total L also includes
the chassis-L between the tune C ground and the place where the grid is
grounded to the chassis. For instance, in a SB-220, the total L is about
160nH.
> Why doesn't the parasitic oscillation manifest itself whenever the
> amplifier is in use?
Why didn't the Space Shuttle Challenger self-destruct before Mission 25?
> If it is a very sporadic occurence, how do we
> know for sure if any bandaid fixes are in reality curing the problem? ...
I don't believe that there is a 100% sure-cure for VHF
parasitic-oscillation. However, I believe I know how to go about making
them less likely. IMO, a parasitic suppressor design that exhibits a
relatively low VHF-Rp will produce less VHF voltage-amplification, which
should be less likely to support regeneration. In practice, this approach
seems to work.
...snip...
Since voltage amplification is basicly Mu*Rp, in your opinion, Peter,
would a higher or a lower VHF-Rp be desirable if one were trying to improve
VHF stability?
[see Wes' suppressor measurements at
<http://www.azstarnet.com/~n7ws/suppress.html>]
> How can someone armed only with a grid-dip meter, and a neon lamp --
> sans spectrum analyzer, network analyzer, Q-meter, etc. determine
> what is the real cause of any amplifier flashovers, and then market
> a UNVERSAL fixall?.
Neon tends to glow purple at VHF and glow orange at HF. I would not take a
dipper near an amplifier that was plugged into the electric-mains.
>
> Without laboratory measurements and scientific analysis, it
> is bad science.
>
Is naysaying good science? What happened when Pasteur suggested that it
might be a good idea for doctors to wash their hands before performing
surgery?
One of the naysayers in this debate claimed that in a VHF
parasitic-suppressor whose Rs is 100 ohms, and whose Ls is 85nH, at a
frequency of 160MHz, insignificant current flows in Ls compared to the
current in Rs. In your opinion, Peter, is this good science or bad
science?
---------------------------
e-mail copy to Mr. Bertini
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <19970119144...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <5brrbe$1tq6$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
> Subich) writes:
>
> >The arcing visible in the pictures could just as easily been caused
> >by a mistuned amplifier running into a high VSWR. I'm sure someone
> >will correct me if I get it backward but in and underloaded amplifier
> >driving a high impedence load the voltage across the bandswitch (and
> >tuning capacitor) will soar nearly out of control. Underloading and
> >high feedline impedences are particularly common on 80 and 160 meters
> >as many amplifiers lack sufficient range in their output capacitors to
> >achieve proper loading ... particularly if the load impedence is very
> >capacitive.
>
...snip...
Mr. Subich's reply did not show up on my news server. Thanks for the
relay, Mr. Rauch.
Joe: I received the following e-mail message from Rick after he saw your
reply regarding his two toasted AL-80 bandswitches:
quoting---
Rich
..."... I can tell you are wasting your time trying to convince these
people vhf parasitics exist in form to destroy bandswitches. If they are
blaming it on operator error or non-resonant antennas then they are also
barking up the wrong tree. Between myself and the phased delta loops I
use on 80 meters, my Drake L4-B hasn't had any band switch or output
capacitor problems what so ever. All I know is that I put about 20 hours
on the AL-80A the past couple of weeks and it hasn't barked one time.
This is with the same rig and antenna I have used for the past 7 years. I
guarantee that prior to putting the suppressors in, the arcing would have
happened. ..."
73' Rick KC0HH
end of quote.
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <19970124174...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
> >
...snip...
Mr. Rauch:
We need to discuss Wes Stewart's measurements of the parallel-equivalent
resistance (Rp) for different types of VHF parasitic-oscillation
suppressors. .
--
--Rich-- ag6k, 805-386-3734
w8j...@aol.com
>.."... I can tell you are wasting your time trying to convince these
>people vhf parasitics exist in form to destroy bandswitches. If they are
>blaming it on operator error or non-resonant antennas then they are also
>barking up the wrong tree. Between myself and the phased delta loops I
>use on 80 meters, my Drake L4-B hasn't had any band switch or output
>capacitor problems what so ever. All I know is that I put about 20 hours
>on the AL-80A the past couple of weeks and it hasn't barked one time.
>This is with the same rig and antenna I have used for the past 7 years. I
>guarantee that prior to putting the suppressors in, the arcing would have
>happened. ..."
>
>
>73' Rick KC0HH
How can we believe anything you say?
R. L. Measures
w8j...@aol.com
>Mr. Rauch:
>We need to discuss Wes Stewart's measurements of the parallel-equivalent
>resistance (Rp) for different types of VHF parasitic-oscillation
>suppressors. .
>
>
Discuss or argue? Will you agree to remain factual or will you throw in
all sorts of unrelated and unconfirmable data?
If you agree to stay with the facts at hand, I'd be happy to. If this is
just another platform for slamming a manufacturer or person, or if it is
filled with hearsay we can't confirm, then I decline.
73 Tom
Harry Brown
I just ran across this discussion and have no expertice on the
subject. However it occured to me that the reason for the arcing
occuring was because the switch was thrown. The impedances could
really be momentarily messed up during this short time and cause
conditions to occur that bring about oscillations. Very high rf
current flow when the contacts break contact could be the source.
If I'm completely off base let me know kindly please.
73, Harry, W3IIT
R. L. Measures
The Kenwood Service Dept in Compton, California used to tell TL-922
owners' whose bandswitches arced and burned up, that the arcing was caused
by rapidly switching the bandswitch back and forth while transmitting.
This would be like Chrysler telling minivan owners' whose transmissions
failed that the failure was caused by rapidly shifting back and forth
between gears while travelling at highway speed.
To the best of my knowledge, bandswitch contact-arcing always occurs across
open contacts. For example, in Rick (KC0HH) May's two, toasted, AL-80
bandswitches shown in Figure 15 at
http://www.vcnet.com/measures/
the amplifier was solely being used on 75m when bandswitches #1 and #2
arced and burned up their 10m and 15m contacts. When bandswitch #3
started arcing, Rick realized that something peculiar was going on. Rick
reported that the arcing ceased after he modified the VHF-suppression
circuitry in the amplfiier.
IMO, 10m and 15m contact arcing in a bandswitch is indicative of
intermittent VHF parasitic oscillation---especially when another indicator
is found, such a VHF parasitic-suppressor resistor that measures more than
double its marked value of R.
On page 33 of "Parasitics Revisited" in the October 1990 issue of QST,
the 10m and 15m contacts have literally evaporated from the TL-922
bandswitch. The replacement bandswitch in this TL-922 continued to arc
until the beautiful silver-plated Trio-Kenwood VHF suppressors were
replaced with ugly, but lossy, Ni-Cr-Fe VHF suppressors.
-----------------------
e-mail copy to Mr. Brown
--
--Rich-- ag6k, 805-386-3734
K1BQT
Hi Rich--
This being a newsgroup, I'd rather see the promised in-depth technical
discussion of what your after-market product actually does or doesn't
do--rather than have to sort through this never-ending stream of anecdotal
lip-flap. If you simply wish to advertise your product, wouldn't it be
better to use a webb page??
Rick K1BQT
Bill Turner
Hold on there, Rick. Rich didn't get on this thread to advertise his product.
I was the original poster clear back in October and it's been going ever
since. There has been a lot of good technical fact and opinion along with the
lip-flap. Sorting out which is which is a whole 'nother story, however.
73, Bill W7TI
w...@eskimo.com
R. L. Measures
In article <32f0b14b...@eskinews.eskimo.com>, w...@eskimo.com (Bill
Turner) wrote:
------------------
It's hardly my product, Rick, since I am not the only person who sells low
vhf-Q suppressor retrofit kits.
Shortly after the original article about lossier VHF suppressors was
published in the October, 1988 issue of QST magazine, I was contacted by an
attorney who happens to be W5NTQ. He told me that I could apply for a U.
S. patent of low VHF-Q suppressors within one year of copyrighted
publication, and he recommended that I do so. I declined, so the idea fell
into public domain. Thus, no individual or company can ever patent it.
For some of us, it's just a hobby, Rick.
Bill: I agree that there has been a substantive technical discussion. If
Rick has been seeing a "never-ending stream of anecdotal lip-flap", he must
be skipping over the technical parts. Sure, the blather about fake photos
(of arced bandswitches), , fabricated (KC0HH) testimony , and claims such
as "it's the entire staff at Eimac" sounds a bit like a take-off on the O.
J. Simpson civil trial, but that sort of thing SHOULD be easy to skip
over---except for those who are mezmerized by such. OTOH, a purely
technical discussion might be as dry as last Saturday's toast.
Rick: who made: " the promised in-depth technical discussion"?
RE: lack of technical topics: I have repeatedly tried to engage one of
the naysayers in this debate in a technical discussion of Wes Stewart's
scientific measurements of Rp in various types of VHF parasitic-oscillation
suppressors, but so far I have been stonewalled, Rick.
Rp should be the core-issue here since VHF voltage amplification is
basicly Mu*VHF-Rp. More VHF Rp equals more VHF gain in an HF amplifier.
Less VHF-Rp equals less VHF gain in an HF amplifier..
------------------------
e-mail copy to Bill W7TI
--
--Rich-- ag6k, 805-386-3734
w8j...@aol.com
In article <5cjud9$p...@news1.voicenet.com>, hbr...@voicenet.com (Harry
Brown) writes:
>I just ran across this discussion and have no expertice on the
>subject. However it occured to me that the reason for the arcing
>occuring was because the switch was thrown. The impedances could
>really be momentarily messed up during this short time and cause
>conditions to occur that bring about oscillations. Very high rf
>current flow when the contacts break contact could be the source.
>
>
Hi Harry,
Actually a bandswitch can fail from many causes.
Improper tuning, a fault in the load, improper lead dress around the
switch, dirt or deposits (one common cause is if the rotor rubs the
ceramic, it deposits metalic trails along the ceramic) and so on.
One person even has a fettish about parasitics, he even $ells kits to
"cure" parasitics even though there is no evidence they ever occur. He
admits his own PA arced from a load fault, and ignores other people who
have said the same thing.
Good science tells us if the switch arcs, it is from RF near the operating
frequency. The tank capacitor has almost zero reactance over the upper VHF
range, it looks just about like a short circuit.
source ---------------{{{{-------{{{{{{{{{{{------->
I ^ ^
tune cap === ^---switch--^
I
ground
It is impossible to develop enough voltage across the few ohms of tune cap
reactance shunting the source to ground to allow the switch, downstream
through even more low pass components, to arc.
The cause of switch arcing HAS to be excessive voltage near the
fundamental operating frequency, or one of the other causes listed above.
I think that you'll see this makes sense, unless you believe in
pathological science.
73 Tom
K1BQT
Hi Bill--
Wooops...sorry about that. I've been following this thread for several
months, but didn't get in on the beginning of it. I guess, in the hope of
learning something meaningful and central to the issue, I was attempting
to goad Rich into focusing on substance. For me, the "infomercial" stuff
gets real old after a while--and detracts from the debate!
Rick K1BQT
R. L. Measures
In article <19970129145...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <5cjud9$p...@news1.voicenet.com>, hbr...@voicenet.com (Harry
> Brown) writes:
>
> >I just ran across this discussion and have no expertice on the
> >subject. However it occured to me that the reason for the arcing
> >occuring was because the switch was thrown. The impedances could
> >really be momentarily messed up during this short time and cause
> >conditions to occur that bring about oscillations. Very high rf
> >current flow when the contacts break contact could be the source.
> >
Apparently, it happens, Harry. In the letter from an ARRL Director (which
is published on my Web site) he describes such an event.
Quoting
"We both lost perfectly good 8877s due to internal flashover within the
tube itself most probably caused by an improper load (wrong antenna) being
switched to the amplifier. In my case I had a guest operator who
inadvertently switched in a 15M Yagi while operating on 20M. There was a
loud "bang", and the 8877 along with the bias circuitry, RF plate choke,
and both meters were destroyed. I didn't need a network analyzer as is
suggested by Telewski to understand the problem! The wrong antenna
presented the tube with an unrealistic plate impedance prompting the
parasitic VHF or UHF oscillation. KD9Q's amplifier, which was commercially
built, experienced a similar situation."
(Note: the 'internal flashover' he mentioned seems to be commonly caused
by loose gold inside the 8877) In the recent shipment of three, kaput,
8877s I received from Lon, K5JV, two of the 8877s had loose particles of
gold, which created abnormally high internal leakage. When gold lands on
the ceramic insulator between the anode and the grounded grid, IMO,
internal flashover is not impossible. In 1986, Eimac's now-retired Chief
Specs Engineer, W. B. Foote, told me that the 8877 design engineers
determined that gold evaporates from the 8877's grid during an oscillation
condition. I asked Foote what the frequency of oscillation was? He said
possibly VHF, possibly UHF.
>
> Hi Harry,
>
> Actually a bandswitch can fail from many causes.
>
> Improper tuning, ...
When someone on this thread measured the maximum peak tune-capacitor
voltage during mistuning in a SB-220, the result was 3700v-p. I tried
mistuning my SB-220 and nothing I could do would arc the tune capacitor or
arc the bandswitch. No surprise there since the measured breakdown voltage
of the switch exceeds 5500v. I mistuned my TL-922, and nothing appeared to
arc.
...a fault in the load,
True. However, there was apprently no load-fault in Rick (KC0HH) May's
case. (see Figure 15)
> improper lead dress around the
> switch,
Bzzzzzzzzzzt.......cough, cough There are no leads around the bandswitch
in a SB-220, TL-922, or any other amplifier I have seen the insides of.
>...dirt or deposits (one common cause is if the rotor rubs the
> ceramic, it deposits metalic trails along the ceramic) and so on.
Tom: I have never seen an amplifier that uses a bandswitch whose rotor
rubs on the ceramic. Which amplifier uses a bandswitch with a rotor that
rubs on the ceramic?
> One person even has a fettish about parasitics, he even $ells kits to
> "cure" parasitics even though there is no evidence they ever occur.
...snip...
Sounds a bit kinky.
Evidence:
1. "Parasitics Revisited", 9/90 & 10/90 QST magazine.
2. Figure 15 and Figure 16(new, courtesy of W6IEG) at
<http://www.vcnet.com/measures/
> It is impossible to develop enough voltage across the few ohms of tune cap
> reactance shunting the source to ground to allow the switch, downstream
> through even more low pass components, to arc.
Tom are you saying that tune capacitors do not have internal VHF
resonances? As I recall, someone measured the internal VHF resonances in a
tune capacitor and posted the results on this thread. Does anyone remember
who made these measurements?
> The cause of switch arcing HAS to be excessive voltage near the
> fundamental operating frequency, ...snip...
Does it make sense that fundamental (HF) energy could wipe out the VHF
suppressor resistor, Rs, at the instant when HF-energy wipes out the
bandswitch?
Factory-stock SB-220s have a penchant for tank arcing and wiping out Rs
when operated on 40m. The SB-220's suppressor-inductor, Ls, has about 3.5
ohms of reactance at 7MHz. In my opinion, it would take around 20v to wipe
out Heath's Rs. Since Rs and Ls are in parallel, in order to do this,
around 6a rms at 7MHz would have to pass through Ls. However, Ls is in
series with the 5pf of internal anode-C in the 3-500Z. At 7Mz, 5pF has
4500 ohms or reactance. From E=I*R, 6a*4500 ohms =s 27,000v-rms. Does
27,000v make sense? ........... Another problem with the HF-energy theory
is that the 10m contact is usually the most damaged, while the 15m contact
is usually damaged less, and the 20m contact even less so. This pattern of
damage is not consistent with HF energy. However, it is consistent with
what I would expect from VHF energy. In the photos of arced bandswitches
on my Web site, such is apparent.
> I think that you'll see this makes sense, unless you believe in
> pathological science.
>
> 73 Tom
BTW, last week on KTLA-TV, which is Channel 5 in LA, newsman Hal Fishman
reported that a TV-satellite which relays CNN and two other TV networks was
permanently knocked out by a front-ejection solar disturbance, and that a
number of military satellites were damaged at the same time.. Fishman
mentioned a previous front-ejection solar disturbance which forced a Space
Shuttle mission to return home prematurely a few years ago. ...Cosmic
photons? Any comments, Mr. Rauch?
Joe Subich
In message <measures-ya0240800...@news.vcnet.com> -:>
:>
:>True. However, there was apprently no load-fault in Rick (KC0HH) May's
:>case. (see Figure 15)
:>
You can't confirm or disprove that statement, it has no probative value.
The statement is based entirely on your customer's claim with NO
independent corroboration.
:>
:>Bzzzzzzzzzzt.......cough, cough There are no leads around the :>bandswitch in a SB-220, TL-922, or any other amplifier I have seen the
:>insides of.
:>
What are the wires running from the coil to the switch? They certainly
look like leads to me.
:>........... Another problem with the HF-energy theory is that the 10m
:>contact is usually the most damaged, while the 15m contact is usually
:>damaged less, and the 20m contact even less so. This pattern of
:>damage is not consistent with HF energy. However, it is consistent
:>with what I would expect from VHF energy. In the photos of arced
:>bandswitches on my Web site, such is apparent.
The pattern is entirely consistent with HF energy. The Plate tuning
coil represents an impedence to RF of ANY frequency. Thus the (RF)
voltage across the coil (impedence) will be proportional to the
magnitude of the impedence from a given (high impedence) point to a
reference (low impedence) point. Since, in the case of an output
network, the low impedence point is the loading capacitor (or antenna
connector) the HF voltage across the coil, and thus the bandswitch, is
highest at the 10 meter contact, somewhat less on 15, still less on 20,
etc. The higher the voltage the bigger and more destructive the arc ...
:>BTW, last week on KTLA-TV, which is Channel 5 in LA, newsman Hal :>Fishman reported that a TV-satellite which relays CNN and two other TV
:>networks was permanently knocked out by a front-ejection solar
:>disturbance, and that a number of military satellites were damaged at
:>the same time.
Fishman reported rumor and speculation. AT&T, the owner and operator of
Telstar 401 (the "TV satellite" in question) have stated that they
believe the problem to be due to a meteor(ite) impact which physically
damaged the spacecraft's main power source. Otherwise, there would be
some residual RF emission (one of several emergency beacons) and
absolutely no RF has been detected from the craft.
If this had been due to circuit damage from high levels of charged solar
particles, other TV satellites of the same design would also have been
damaged, particularly those like Telstar 302 which were in orbital
positions further to the east and which would have received even higher
levels of charged particle bombardment.
Pathological science makes good stories for tabloid TV just like it
sells snake oil for amplifiers and RCA still doesn't mark the "normal
operating voltage range" on their filament meters.
--
========================================================================
Joe Subich, W8IK ex-AD8I
<W8...@IBM.NET>
<W8...@AMSAT.ORG>
<sub...@ibm.net>
<71350...@compuserve.com>
========================================================================
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>Any comments, Mr. Rauch?
>-----------------------------
Only that we have been round and round this tree many times. Going around
the tree once more will make no difference. Even if we cut the tree down,
you'll keep right on running around the same old circle howling out the
same old tune.
You'll continue to present false claims as fact, just as you always have.
Some people will swallow your tall tales of pathological science, and
others won't.
Not much either of us can do about that.
73 Tom
J. Fred Riley
meas...@mail.vcnet.com (R. L. Measures ) wrote:
>HF arcs are typically blue, not purple.
This ought to be carved in stone somewhere.
Fred, W8OY
comm...@aol.com
In article <5cui5j$14g8$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) writes:
>:>damaged less, and the 20m contact even less so. This pattern of
>:>damage is not consistent with HF energy. However, it is consistent
>:>with what I would expect from VHF energy. In the photos of arced
>:>bandswitches on my Web site, such is apparent.
>
>The pattern is entirely consistent with HF energy. The Plate tuning
>coil represents an impedence to RF of ANY frequency. Thus the (RF)
>voltage across the coil (impedence) will be proportional to the
>magnitude of the impedence from a given (high impedence) point to a
>reference (low impedence) point. Since, in the case of an output
>network, the low impedence point is the loading capacitor (or antenna
>connector) the HF voltage across the coil, and thus the bandswitch, is
>highest at the 10 meter contact, somewhat less on 15, still less on 20,
>etc. The higher the voltage the bigger and more destructive the arc ...
>
>
Dear Rich....
This is the point I ws trying to make, and is the reason I mentioned
ten-meters,
which would be a worst case example.
I believe Tom Rauch has shown that the plate tuning capacitor has very
low reactance at VHF, and is actually slightly inductive at those
frequencies.
If a VHF resonant circuit was formed by the internal capacitance of the
tube,
combined with the series inductance of the Plate capacitor and the
connecting
straps -- the VHF voltage at the plate tuning capacitor would be
substantially
reduced based on the ratio of the inductances of the tuning capacitor and
its connecting strap to the DC blocking capacitor.
73
Peter
R. L. Measures
In article <19970201175...@ladder01.news.aol.com>,
comm...@aol.com wrote:
> In article <5cui5j$14g8$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
> Subich) writes:
>
> >:>damaged less, and the 20m contact even less so. This pattern of
> >:>damage is not consistent with HF energy. However, it is consistent
> >:>with what I would expect from VHF energy. In the photos of arced
> >:>bandswitches on my Web site, such is apparent.
> >
> >The pattern is entirely consistent with HF energy. The Plate tuning
> >coil represents an impedence to RF of ANY frequency. Thus the (RF)
> >voltage across the coil (impedence) will be proportional to the
> >magnitude of the impedence from a given (high impedence) point to a
> >reference (low impedence) point. Since, in the case of an output
> >network, the low impedence point is the loading capacitor (or antenna
> >connector) the HF voltage across the coil, and thus the bandswitch, is
> >highest at the 10 meter contact, somewhat less on 15, still less on 20,
> >etc. The higher the voltage the bigger and more destructive the arc ...
> >
Peter...
>
> This is the point I ws trying to make, and is the reason I mentioned
> ten-meters,
> which would be a worst case example.
Two bandswitches in Rick May's amplifier burned up their 10m contact while
the amplifier was being operated on 80m into an antenna that exhibited no
anomalies when driven by a Drake L-4B amplifier, at approximately twice the
power level of the amplifier that burned up the two bandswitches. Rick
installed bandswitch #3. It too intermittently arced like bandswitch #1
and #2. Rick decreased the VHF-Rp of the parasitic suppressor. He said
that bandswitch #3 has not burned up from mysterious arcing. If you have
any questions for him, Rick's e-mail address is listed on my Web site.
Since December of 1989, more than 4500 suppressor retrofit kits have been
sold by me. The guarantee: If the suppressor retrofit kit does not remedy
the parasitic-symptom, the customer does not pay.
You did not address the issue of the damaged/kaput VHF suppressor resistor
that often coincides with mysterious bandswitch arcing.
> I believe Tom Rauch has shown that the plate tuning capacitor has very
> low reactance at VHF, and is actually slightly inductive at those
> frequencies.
As I recall, Mr. Rauch measured the internal VHF resonances in a tuning
capacitor. All tuning capacitors have them, Peter. Do VHF resonant
circuits always exhibit low reactance at VHF?
A question: Peter--If you were designing an HF amplfier, would you want a
VHF parasitic oscillation suppressor that exhibited more (or less) VHF-Rp
than another type of VHF suppressor?
----------------
e-mail copy to Peter
--
--Rich-- ag6k, 805-386-3734
K1BQT
Rich Writes:
"Since December of 1989, more than 4500 suppressor retrofit kits have been
sold by me."
After looking in on this debate for several months, it appears that some
of the better technical minds on here have the correct answers about why
amplifiers fail--but Rich has the money!
Rick -- K1BQT
R. L. Measures
In article <19970202052...@ladder01.news.aol.com>, k1...@aol.com
(K1BQT) wrote:
------------------
Rick:
If you believe that Gamma-rays, X-rays and Cosmic-rays are 'charged
particles', count yourself among the 'better technical minds'.
If you believe that Wes' measurements of the characteristics of various
vhf parasitic oscillation suppressors are virtually meaningless, you have
yet another reason to count yourself among the 'better technical minds'.
If you don't have a clue as to how the algebraic expression (E1/E2)^23.4
was derived from Eimac's statement that a 3% rise in filament voltage
causes a 50% decrease in the emissive life of the tungsten-dicarbide layer
in a thoriated-tungsten cathode, count yourself among the 'better technical
minds'.
If you believe that the formula for calculating the maximum RF current in a
DC blocking capacitor, which appears on page 13.13 in the new *Radio
Amateurs Handbook*, is correct, count yourself among the 'better technical
minds'.
If you scored 4 out of 4, Rick, Congrats---you are thereby qualified to be
a bona fide 'Recognized Amplifier Expert'--Apprentice. (contact Mr. Rauch
or Dick Erhorn for further information)
---
It is true that, at this point, I seem to be burdened with more money than
I will ever need in this life---and if Mr. Rauch makes good on his threat
to sue me for libel, there is a pretty fair chance that this burden will
increase even further.
It can be rightfully argued that this Apple Macintosh 2ci computer and my
HP Laserjet-3 printer was paid for with monies from the business of selling
mail-order VHF suppressor kits. I dutifully-deducted a fraction of the
cost of this equipment from my taxable income (Section 179 of the U. S. Tax
Code) because I use said equipment in the business . However, I have not
made a profit from selling suppressor retrofit-kits during the last 8
years. It's just a hobby, Rick.
----------------------------
e-mail copy to Rick, K1BQT
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <19970201154...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
>
> >Any comments, Mr. Rauch?
> >-----------------------------
>
> Only that we have been round and round ......
...snip...
Mr. Rauch has seemingly come down with a bad case of photonphobia.
In November, Mr. Rauch said:
__QUOTE
>The former *R+D engineering manager*, Buzz Miklos, of the Salt Lake City
>Varian plant has agreed to speak to one or two people on the telephone
>about your theories. He said he will also come on line, but since he just
>took a position as engineering manager with another tube manufacturer in
>PA, it will be sometime in January before he is able to connect directly
>to this newsgroup.
__END OF QUOTE
Mr. Miklos did not show up on this Newsgroup 'sometime in January', as
proffered. No letter from Eimac has since been presented by Mr. Rauch
that refutes the information provided by the Eimac/Varian Salt Lake
division's personnel department regarding the job title of Mr. Joseph P.
Miklos.
----
Mr. Rauch is somewhat understandably tiptoeing around the issue of Rp.
IMO, Rp is the central issue in this discussion. Thanks to Wes (N7WS)
Stewart's scientific measurements, we know what Rp is for the two principal
types of VHF parasitic-oscillation suppressors at a number of frequencies
in the VHF range. (see Figure 12 at
< http://www.vcnet.com/measures/>)
In Wes' measurements, Rp stands for parallel-equivalent resistance. For a
VHF parasitic suppressor, Rp is representative of the VHF shunt load
resistance presented to the anode of the amplifier tube.
Since voltage gain is basically Mu*Rp, more VHF-Rp means more VHF voltage
gain. Conversely, less VHF-Rp means less VHF voltage gain.
The choice is yours, folks. If you want more VHF-Rp in your HF amplfier,
suppressors made from copper are for you. If you want even more VHF-Rp, go
for silver-strap suppressors, such as those found in the Kenwood TL-922
amplifier, and in Henry Radio amplifiers.
OTOH, if you want less VHF-Rp in your amplifier, according to Wes'
measurements, a nickle-chromium alloy VHF suppressor will produce the
desired result---which is really nothing new. In the 1926 Edition of the
*Radio Amateur's Handbook*, Mr. F. E. Handy wrote about it.
Mr. Handy was not the only person to arrive at this conclusion before I
wrote the October1988 *QST* article "Improved Anode Parasitic-Suppression
for Modern Amplifier Tubes". Around the time when I was in middle school,
a ham club on the East Coast decided on a club-project of building copies
of the first (4) 811A g-g linear amplifier that appeared in a ham magazine.
All of the builders, except for one, followed the plans to the letter. The
exception was Dr. William Riley., K4GXA. Riley concluded that, since the
job of the VHF suppressors is to suppress, maybe a resistive conductor
might be a good way to go. So, instead of using copper wire for Ls, Riley
used hay-bailing wire for Ls. All of the club-project amplifiers
eventually worked. However, over the years, mysterious things reportedly
happened to the 811As in the other amplifiers. From time to time, sudden
tube failures occured wherein 811A filaments were found in pieces in the
bottom of the glass envelope. These tube failures were typically
accompanied by a big-bang and an apparent arc. When Bill Riley eventually
sold his amplifier 15 years later, his was reportedly the only amplifier
that was still using the original (4) 811As. Bill told me that when the
other amplifier builders first saw his rusty bailing-wire suppressors, they
were amused. I suppose you could say that Bill Riley had the last laugh.
comm...@aol.com
> Subich) writes:
>
> >:>damaged less, and the 20m contact even less so. This pattern of
> >:>damage is not consistent with HF energy. However, it is consistent
> >:>with what I would expect from VHF energy. In the photos of arced
> >:>bandswitches on my Web site, such is apparent.
> >
> >The pattern is entirely consistent with HF energy. The Plate tuning
> >coil represents an impedence to RF of ANY frequency. Thus the (RF)
> >voltage across the coil (impedence) will be proportional to the
> >magnitude of the impedence from a given (high impedence) point to a
> >reference (low impedence) point. Since, in the case of an output
> >network, the low impedence point is the loading capacitor (or antenna
> >connector) the HF voltage across the coil, and thus the bandswitch, is
> >highest at the 10 meter contact, somewhat less on 15, still less on 20,
> >etc. The higher the voltage the bigger and more destructive the arc
...
> >
Measures writes:
> Peter...
> You did not address the issue of the damaged/kaput VHF suppressor
resistor
> that often coincides with mysterious bandswitch arcing.
I am not questioning the occurence of VHF parasitics -- and the fact
that they may occur at the same time other amplifier failures occur.
I question the "fact" that parasitics cause bandswitch
failure or tuning capacitor arcing. Consider Mr. Subich's and
Mr. Rauch's statements concerning the lowpass filter effect of the
ten-meter portion of the tank coil (a very high impedance at VHF).
How can damaging VHF voltages be present at the bandswitch? For
the lower HF bands, more (plate tank) inductance is involved,
and thus more impedance to the flow of VHF energy.
>>> I believe Tom Rauch has shown that the plate tuning capacitor has very
>>> low reactance at VHF, and is actually slightly inductive at those
>>> frequencies. (earlier post by Bertini)
> As I recall, Mr. Rauch measured the internal VHF resonances in a tuning
> capacitor. All tuning capacitors have them, Peter. Do VHF resonant
> circuits always exhibit low reactance at VHF?
No. I don't believe this is correct. I believe Tom showed the
capacitor to exhibit slight Xl at VHF frequencies. Not resonance.
Are you confusing reactance and impedance? Impedance is determined
by the resitive and reactive components of a device at a specific
frequency. For the tuning capacitor to be resonant at a given
frequency, it would need both capacitive and inductive reactances.
Given only it has a "slight" inductive reactance, its impedance
at VHF is most likely very minimal. I believe Mr. Rauch has
REPEATEDLY made this point. It is unlikely any further comments
on my part will sway you from your beliefs.
I do not understand your question regarding resonance and
low reactance. A resonant circuit may imply a high impedance
in the case of parallel resonance, or low impedance in the
case of series-resonance. It depends on
whether you are building a trap or bandpass circuit..
Rich, are you measuring series or parallel resonances with you
dip meter?
>A question: Peter--If you were designing an HF amplfier, would you want
a
>VHF parasitic oscillation suppressor that exhibited more (or less) VHF-Rp
>than another type of VHF suppressor?
Rich, I would honestly want a suppressor that did not compromise
the operating Q of the amplifier at HF frequencies.
I would select values of L, R and C (for the suppressor)
that were effective at VHF, and had minimal effect at HF.
That's where I would put my Q and RX meters to work.
I would also heed some of the other advice gathered in this thread,
and provide arc-over electrodes and current limiting resistors
in the plate supply.
73, Peter
w8j...@aol.com
>
>As I recall, Mr. Rauch measured the internal VHF resonances in a tuning
>capacitor. All tuning capacitors have them, Peter. Do VHF resonant
>circuits always exhibit low reactance at VHF?
>
>
Mr. Measures has a very selective memory, and a great ability to misapply
simple and clear facts. His track record clearly proves this.
In this case, Mr. Measures is trying to distort the fact the capacitor
goes through a minimum reactance at 150-200 MHz as evidence it must go
through a maximum impedance at the same point.
He misquotes me just as freely as he has misquotes Varian, Dave Holtcamp,
Marvin Born, and others.
Rich, you "forgot" I measured the voltage distributed through the tank.
Nearly anyone can make this measurement. Connect a VHF signal source at
the anode and excite the tank. Measure (that word is becoming distasteful)
the voltage along the tank with a high impedance RF probe, and you will
see it behaves exactly as common sense would tell us. It is nothing at all
like Mr. Measures would have us believe.
No one can be that stubid in the face of so much contrary evidence or even
if conventional circuit theory is applied. This parasitic arcing theory of
Measures must be an intentional hoax. In that case, any effort to
communicate or argue with him is a waste of time.
If you clearly prove him wrong a dozen times in a dozen ways.....he'll
keep going and going and going ....just like the mechanical Energizer
Bunny on TV.
73 Tom
K1BQT
Hi Rich--
Speaking of formulas, how about this one: Obscure researcher makes an
amazing discovery that works as a cure-all for a myriad of ills. Then, to
market it, he attacks the establishment that somehow failed to understand
and recognize the "real" truth. When confronted, he counters by crying
conspiracy--and identifies a powerful special interest group who has much
to lose by admitting he's right.
Just like E/I=R, this is a proven formula. Only, this formula is well
know for marketing healing crystal, odd potions, and electric blood-vessel
stimulators. Instead of burned switches, I think I'd rather see those
elusive spectral displays that nobody seems able to produce (or reproduce)
under controlled conditions. If the science is good, why would anyone
need to conjure up anecdotal evidence, create enemies, and cry conspiricy?
Rick K1BQT
Wes Stewart
In article <measures-ya0240800...@news.vcnet.com> meas...@mail.vcnet.com (R. L. Measures ) writes:
>From: meas...@mail.vcnet.com (R. L. Measures )
>Subject: Re: good source for amplifier tips
>Date: Sun, 02 Feb 1997 06:11:13 -0800
[snip]
Rich Measures writes
-
>Mr. Rauch is somewhat understandably tiptoeing around the issue of Rp.
> IMO, Rp is the central issue in this discussion. Thanks to Wes (N7WS)
>Stewart's scientific measurements, we know what Rp is for the two principal
>types of VHF parasitic-oscillation suppressors at a number of frequencies
>in the VHF range. (see Figure 12 at
>< http://www.vcnet.com/measures/>)
> In Wes' measurements, Rp stands for parallel-equivalent resistance. For a
>VHF parasitic suppressor, Rp is representative of the VHF shunt load
>resistance presented to the anode of the amplifier tube.
> Since voltage gain is basically Mu*Rp, more VHF-Rp means more VHF voltage
>gain. Conversely, less VHF-Rp means less VHF voltage gain.
I tried starting a new thread on this issue, but didn't get any takers, so as
Roy would say, I'll take a swing at this tar baby here.
I tried to connect to Rich's web site without success, so I can't address what
is posted there. However, I have updated my page
(http//www.azstarnet.com/~n7ws) with some charts which directly compare two
suppressor networks.
For these measurements, the network is necessarily measured by exciting one
end with respect to ground and measuring the characteristics. In this
situation, the network is measured in a controlled (or well known)
environment. The instrumentation is "smart" enough to remove any internal
errors and present "corrected" measurement data. THIS IS NOT how the
suppressor network is used in an amplifier.
The network is not an isolated device who's characteristics are independent of
its environment. It is connected to the anode, which has an unknown source
impedance, on one side and the output tank circuit, which has an unknown load
impedance, on the other side.
Until these source and load impedances are known (at every frequency of
interest) the effects of the suppressor network cannot be surmised.
I know this will sound heretical, but the more I look at this, the more
I question the need for these things. If someone can present me a circuit
model in which the addition of a few ohms loss resistance can lower the loop
gain enough to prevent an oscillation, I'd love to see it.
Mmmm... love them tar babies.
R. L. Measures
In article <19970202143...@ladder01.news.aol.com>,
comm...@aol.com wrote:
...snip...
>I am not questioning the occurence of VHF parasitics -- and the fact
>that they may occur at the same time other amplifier failures occur.
>I question the "fact" that parasitics cause bandswitch
>failure or tuning capacitor arcing. Consider Mr. Subich's and
>Mr. Rauch's statements concerning the lowpass filter effect of the
>ten-meter portion of the tank coil (a very high impedance at VHF).
>How can damaging VHF voltages be present at the bandswitch? For
>the lower HF bands, more (plate tank) inductance is involved,
>and thus more impedance to the flow of VHF energy.
Peter: Have you ever checked a tank L with a dipmeter for the presence of
VHF resonances?
>
>>>> I believe Tom Rauch has shown that the plate tuning capacitor has very
>>>> low reactance at VHF, and is actually slightly inductive at those
>>>> frequencies. (earlier post by Bertini)
As I recall, Tom Rauch found some internal VHF resonances. However, this
is nothing new since variable-capacitor resonances are discussed in the
32nd (1955) Edition of *The Radio Amateur Handbook*.
>...snip... For the tuning capacitor to be resonant at a given
>frequency, it would need both capacitive and inductive reactances.
Do the tuning capacitor plates have inductance?
>Given only it has a "slight" inductive reactance, its impedance
>at VHF is most likely very minimal.
Quite true, Peter---unless the tune capacitor is resonant or near
resonance. Before this discussion began, I hadn't given much thought to
how VHF voltage might be able to arc a tuning capacitor. Thanks to Mr.
Rauch's measurements of a tune capacitor from an AL-80, I think I have an
inking.
Do you remember the problem I proposed of generating over 10,000vrms from
a 120v, 60Hz source with only one capacitor and one inductor? In my
solution to this problem, I picked an L/C combination that was
series-resonant about 0.6Hz away from 60Hz. Assuming that the L and C had
zero ESR, the circuit produced over 10,000vrms. Sure, it took around 100A
to do it. However, the typical filter capacitor in a typical 1500w amateur
radio amplifier has about 10 ohms of ESR. How much current is 2800V/10
ohms?
So why does the 10m bandswitch contact sometimes arc.? An open 10m contact
probably has around 1 pF. The typical 10m section of a tank L has around
0.9uH. The series-resonant frequency is 170MHz. In an AL-80, this is in
the neighborhood of the anode circuit's resonance at 160MHz. Thanks to Mr.
Rauch's measurements, we know that the AL-80's tune capacitor has no
resonances in that neighborhood. I have not heard of any case where an
AL-80 tune capacitor allegedly arced. However, I have heard of a number of
cases where AL-80 10m bandswitch contacts mysteriously arced and
evaporated. If you're interested, photos of same are available at Figure
15 on my Web site.
>
...snip...
>Rich, are you measuring series or parallel resonances with you
>dip meter?
I do not know of a way of differentiating between a series-resonance dip
and a parallel-resonance dip with a dipmeter.
>>A question: Peter--If you were designing an HF amplfier, would you want a
>>VHF parasitic oscillation suppressor that exhibited more (or less) VHF-Rp
>>than another type of VHF suppressor?
>
>Rich, I would honestly want a suppressor that did not compromise
>the operating Q of the amplifier at HF frequencies.
Did you accept someone's word that this was the case, or did you calculate
a typical tank Q reduction, based on Wes' ESR measurement for a Ni-Cr Ls
suppressor at 30MHz (4.32 ohms) and/or at 10MHz (0.88 ohms)? If you did
the calculations, Peter, I would appreciate seeing how to do it, and how to
calculate what the probable increase in harmonic output would be. I don't
have a clue as to how to proceed. OTOH, I have heard of no cases where a
low VHF-Q suppressor caused increased TVI to channel 2 during 10m
operation---which would, according to Wes' measurements, would undoubtedly
be THE worst-case. To date, I know of no instance where an FCC citation
resulted.
Mr. Rauch made the claim that a serious HF Q degradation would result
from using a Ni-Cr VHF suppressor.
Mr. Rauch presented me with the following problem. F=160MHz: An L of
85nH is paralleled with 100 ohms. How does the VHF current divide?
According to AC Circuit Analysis, the calculated current through the 85nH L
is about 10% higher than the calculated current through the 100 ohm
resistor. However, Mr. Rauch concluded that these calculations are
invalid. Do you accept Mr. Rauch's conclusion?
>I would select values of L, R and C (for the suppressor)
>that were effective at VHF, and had minimal effect at HF.
>That's where I would put my Q and RX meters to work.
>
I haven't seen a VHF suppressor design that uses a C. What does this
circuit look like?
>I would also heed some of the other advice gathered in this thread,
>and provide arc-over electrodes and current limiting resistors
>in the plate supply.
>
In the Sept. 1994 QST critique of the Jan. 1994 QST article "The Nearly
Perfect Amplifier", my suggestion about using a peak fault-current limiting
resistor in the anode supply was pooh-poohed by the (self) 'recognized
amplifier expert'.
An arc-over electrode sounds like a good idea, Peter. However, since tune
capacitors, as well as open bandswitch contacts seem to arc during some
intermittent VHF parasitic oscillations, IMO, at least two
arc-over/spark-gap electrodes are indicated. Adding a low-L low-ohm MOF
resistor in series with each spark-gap might help to limit the peak fault
current. The Matsushita 3w-rated MOF resistors we use in our suppressor
retrofit-kits might do the job. According to Wes, N7WS, the30 ohm units
have around 9nH/0.009uH of internal L, and from my own tests I know that
they have remarkable overload capacity.
----------------------------------
e-mail copy to Mr. Peter Bertini
--
--Rich-- ag6k, 805-386-3734
J. Fred Riley
meas...@mail.vcnet.com (R. L. Measures ) wrote:
>In the Sept. 1994 QST critique of the Jan. 1994 QST article "The Nearly
>Perfect Amplifier", my suggestion about using a peak fault-current limiting
>resistor in the anode supply was pooh-poohed by the (self) 'recognized
>amplifier expert'.
I don't think the use of a current-limiting resistor was the subject
of criticism. It was, rather, the type of resistor used. A
current-limiting resistor is almost always a good idea. Using the
wrong kind of resistor is almost always a bad idea. Using a resistor
that's too "small" is sorta like using the low-voltage fuse in a HV
line.
Fred, W8OY
J. Fred Riley
meas...@mail.vcnet.com (R. L. Measures ) wrote:
>Dr. William Riley., K4GXA...had the last laugh.
On 75 meters we refer to him as Doc Bull, if that tells you anything.
Fred, W8OY
w8j...@aol.com
Hi Wes,
Good to see someone taking this thread back to science.
In article <n7ws.253...@azstarnet.com>, n7...@azstarnet.com (Wes
Stewart) writes:
>For these measurements, the network is necessarily measured by exciting
one
>end with respect to ground and measuring the characteristics. In this
>situation, the network is measured in a controlled (or well known)
>environment. The instrumentation is "smart" enough to remove any internal
>errors and present "corrected" measurement data. THIS IS NOT how the
>suppressor network is used in an amplifier.
Absolutely.
>The network is not an isolated device who's characteristics are
independent
>of
>its environment. It is connected to the anode, which has an unknown
source
>impedance, on one side and the output tank circuit, which has an unknown
load
>impedance, on the other side.
Ahh, but I do know those impedances. And you can too if you have a PA and
a Network Analyzer. They are no big secret.
The effect of the other circuit componets is to bring Q and impedance of
both parasitic suppressors to higher but converging values.
>Until these source and load impedances are known (at every frequency of
>interest) the effects of the suppressor network cannot be surmised.
I measured Q in a pair of 3CX800's,and with a suppressor Q of unity
outside the PA the Q climbed to over three with the suppressor in the PA.
Adding the Measure's voo doo nichrome suppressor exactly as described in
his paper resulted in almost the same Q, except just ever so slightly
higher than the stock suppressor.
I will give Rich credit, his nichrome did lower the UNloaded Q of the
tank nearly 50% on one frequency of interest, ten meters. So it isn't like
it's complete voo doo. If the PA oscillates on ten meters, it might settle
it down.
>I know this will sound heretical, but the more I look at this, the more
>I question the need for these things. If someone can present me a
circuit
>model in which the addition of a few ohms loss resistance can lower the
loop
>gain enough to prevent an oscillation, I'd love to see it.
It can indeed Wes. Even a Q of three or four is enough to damped the
system at VHF.
Some PA's are totally stable without ANY suppressor, and some require a
large suppressor. There are so many variables involved a universal
suppressor system is impossible to define.
This is why I had to spend almost as much $$$ as Rich makes off his snake
oyl cure to get the right equipment. While I could do it by cut and try,
it's much nicer to know what you are really doing.
Now if I had uncanny ESP that allowed me to, just by holding a part in my
hand, see the cause of the parasitic, if I had a rare Q measuring GDO, if
Varian would have whispered the hiiden truth in MY ear in one of the
hundreds of conversations I've had with them that they ONLY told Mr.
Measures (NEVER have they told this parasitic stuff to anyone else in the
world...just their best friend Mr. Measures. Strange, isn't it?), if I
thought oscillators were self-destructive gongs waiting to be rung by a
photon from the sun, if I knew so little about field theory that I
believed a parasitic could melt the glass of a Chinese tube, than I would
probably just throw something in and call it a done cure.
I'm not blessed with those features, and do not (through experience with
pathological liars in my family) understand how to predict when someone
else is lying. So I'm stuck, I just have to make actual measurements.
If you'd like to have a PA to measure, let me know. You hit the tar baby.
I need to go away and do something more constuctive. You do not get to
keep it though, hi.
73 Tom
comm...@aol.com
In a message R. L. Measures writes:
<< Peter: Have you ever checked a tank L with a dipmeter for the presence
of
VHF resonances?
No. What would it prove? My dipmeter is the most unused item in the
shop.
>
>>>> I believe Tom Rauch has shown that the plate tuning capacitor has
very
>>>> low reactance at VHF, and is actually slightly inductive at those
>>>> frequencies. (earlier post by Bertini)
< As I recall, Tom Rauch found some internal VHF resonances. However,
this
< is nothing new since variable-capacitor resonances are discussed in the
< 32nd (1955) Edition of *The Radio Amateur Handbook*.
>...snip... For the tuning capacitor to be resonant at a given
>frequency, it would need both capacitive and inductive reactances.
< Do the tuning capacitor plates have inductance?
The question should be "do tuning capacitors have a parallel
resonance at VHF?"
>Given only it has a "slight" inductive reactance, its impedance
>at VHF is most likely very minimal.
< Quite true, Peter---unless the tune capacitor is resonant or near
< resonance. Before this discussion began, I hadn't given much thought
to
< how VHF voltage might be able to arc a tuning capacitor. Thanks to Mr.
< Rauch's measurements of a tune capacitor from an AL-80, I think I have
an
inking.
< Do you remember the problem I proposed of generating over 10,000vrms
from
< a 120v, 60Hz source with only one capacitor and one inductor?
Nothing new there, Rich... there is a whole new disipline based on that
principle. It's called switching power supplies.
< So why does the 10m bandswitch contact sometimes arc.? An open 10m
contact
< probably has around 1 pF. The typical 10m section of a tank L has
around
< 0.9uH. The series-resonant frequency is 170MHz. In an AL-80, this is
in
< the neighborhood of the anode circuit's resonance at 160MHz. Thanks to
Mr.
< Rauch's measurements, we know that the AL-80's tune capacitor has no
< resonances in that neighborhood. I have not heard of any case where an
< AL-80 tune capacitor allegedly arced.
Bingo!!!!! Of course not!! It is virtually a dead short at VHF! Sort of
kills
the possibility of VHF energy arcing the bandswitch.......
< However, I have heard of a number of
< cases where AL-80 10m bandswitch contacts mysteriously arced and
< evaporated. If you're interested, photos of same are available at
Figure
< 15 on my Web site.
The winds are a shifting. Last week, when I mentioned ten meters, you
went on to say this happened on 40 meters, not ten meters.
>>A question: Peter--If you were designing an HF amplfier, would you
want a
>>VHF parasitic oscillation suppressor that exhibited more (or less)
VHF-Rp
>>than another type of VHF suppressor?
>
>Rich, I would honestly want a suppressor that did not compromise
>the operating Q of the amplifier at HF frequencies.
< Did you accept someone's word that this was the case, or did you
calculate
< a typical tank Q reduction, based on Wes' ESR measurement for a Ni-Cr Ls
< suppressor at 30MHz (4.32 ohms) and/or at 10MHz (0.88 ohms)?
Rich, I am confused. You are marketing this product, and have been
for years. You mean that thanks to Wes' calculations you only now
have some idea of what they doing?
< the probable increase in harmonic output would be. I don't
< have a clue as to how to proceed. OTOH, I have heard of no cases where
a
< low VHF-Q suppressor caused increased TVI to channel 2 during 10m
< operation---which would, according to Wes' measurements, would
undoubtedly
< be THE worst-case. To date, I know of no instance where an FCC citation
< resulted.
Spectrum analyzer? They are fairly cheap.
< Mr. Rauch made the claim that a serious HF Q degradation would result
< from using a Ni-Cr VHF suppressor.
Yes, it should. While the impedance of your suppressor would change
to a lower value at HF due to a decreasing XL, the resistive component
of the wire remains constant -- the resistive element will absorb power
at HF. Why waste time silver plating tanks coils when we are going to
place resistors (nichrome or otherwise) in series? There is a difference
between loaded and unloaded Q...
< Mr. Rauch presented me with the following problem. F=160MHz: An L of
< 85nH is paralleled with 100 ohms. How does the VHF current divide?
< According to AC Circuit Analysis, the calculated current through the
85nH L
< is about 10% higher than the calculated current through the 100 ohm
< resistor. However, Mr. Rauch concluded that these calculations are
< invalid. Do you accept Mr. Rauch's conclusion?
Rich, the math and caculations have been gone over, and over, and
over....
Do you disagree with Rauch's math? Argue with him.
>I would select values of L, R and C (for the suppressor)
>that were effective at VHF, and had minimal effect at HF.
>That's where I would put my Q and RX meters to work.
>
< I haven't seen a VHF suppressor design that uses a C. What does this
< circuit look like?
Mr. Rauch described a suppressor using RCL components in an earlier
post. I did not retain a copy. I am sure Mr. Rauch would reply if asked
--
nicely:)
>I would also heed some of the other advice gathered in this thread,
>and provide arc-over electrodes and current limiting resistors
>in the plate supply.
>
< In the Sept. 1994 QST critique of the Jan. 1994 QST article "The Nearly
< Perfect Amplifier", my suggestion about using a peak fault-current
limiting
< resistor in the anode supply was pooh-poohed by the (self) 'recognized
< amplifier expert'.
I am surprised. I would use them regardless of Mr. Rauch's alleged
comments. I think you had a good idea there.
< An arc-over electrode sounds like a good idea, Peter. However, since
tune
< capacitors, as well as open bandswitch contacts seem to arc during some
< intermittent VHF parasitic oscillations, IMO, at least two
< arc-over/spark-gap electrodes are indicated. Adding a low-L low-ohm MOF
< resistor in series with each spark-gap might help to limit the peak
fault
< current. The Matsushita 3w-rated MOF resistors we use in our suppressor
< retrofit-kits might do the job. According to Wes, N7WS, the30 ohm units
< have around 9nH/0.009uH of internal L, and from my own tests I know that
< they have remarkable overload capacity.
The peak limiting resistance belongs in the plate supply feed. You refuse
to
accept the fact that a plate tuning capacitor can be effectively
a dead short at VHF, but readily assume that a Matsushita MOF is a pure
resistance, free of resonances at VHF?
Let's put one on a RX meter and see what it looks like at VHF first.
73, Peter
----------------------------------
Phil Clements
In article <measures-ya0240800...@news.vcnet.com>, meas...@mail.vcnet.com (R. L. Measures ) says:
>
>.....snip
>>I would select values of L, R and C (for the suppressor)
>>that were effective at VHF, and had minimal effect at HF.
>>That's where I would put my Q and RX meters to work.
>>
>
>I haven't seen a VHF suppressor design that uses a C. What does this
>circuit look like?
>
In an article in the January, 1974 issue of Ham Radio, W4QQ has a 4-1000A
5-band Kilowatt Linear that uses a suppressor consisting of:
A 20 pf ceramic capacitor (Centralab 852S) which is a mini-doorknob with
axial leads in parallel with 4-470 ohm 2 watt resistors and 4 turns #12
wire, 1/2 inch diameter, resonated at 110 mhz. I built that amp project
in 1981, and it has performed flawlessly ever since, with the original parts
and tube.
(((73)))
Phil, K5PC
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>Quite true, Peter---unless the tune capacitor is resonant or near
>resonance. Before this discussion began, I hadn't given much thought to
>how VHF voltage might be able to arc a tuning capacitor. Thanks to Mr.
>Rauch's measurements of a tune capacitor from an AL-80, I think I have an
>inking.
.
Measures now comes out and admits he hadn't given much thought to how a
VHF parasitic could cause an arc in a tuning capacitor. He just went right
ahead and said it could for years and years, without even thinking about
it.
Now, after YEARS of making his claims, he admits he is just beginning to
think. He now has an "inking."
What is an inking? Peter, do they ink people at Belvue? Answer the
question Peter.
Do you think Rich checked out other parts of his theories as well as the
plate capacitor he admits never even thinking about?
I wonder if this has anything to do with the stuff he incorrectly posted
from Born? And about Miklos job? And from Holtcamp on the 922? Or about
the filament meters and variacs? Or those 500 watt 8877 grids? Or his cat
and a popular perfume?
Do you suppose Rich never bothers to check anything out? Is it possible he
just makes stuff up or thinks he hears something, and decides in his mind
it's true enough to repeat as a fact before he even understands it?
73 Tom
Ian White, G3SEK
R. L. Measures wrote:
>>I would also heed some of the other advice gathered in this thread,
>>and provide arc-over electrodes and current limiting resistors
>>in the plate supply.
>>
>In the Sept. 1994 QST critique of the Jan. 1994 QST article "The Nearly
>Perfect Amplifier", my suggestion about using a peak fault-current limiting
>resistor in the anode supply was pooh-poohed by the (self) 'recognized
>amplifier expert'.
>
Let's look at what was actually printed.
Rich's suggestion to use a 10R resistor for up to 3kV was criticized
because it isn't enough to make a significant difference to the internal
resistance of the power supply. The specific point was that the ESR of
the electrolytic capacitors is likely to be at least 10R, and then
there's the RF choke and other series components.
I wasn't mentioned in QST, but Eimac Application Bulletin #17 recommends
50R for "small" transmitting tubes that run at about 2kV.
In other words, the criticism was not about the recommendation to use a
resistor; the point was that 10R isn't high enough.
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
R. L. Measures
In article <19970203015...@ladder01.news.aol.com>,
comm...@aol.com wrote:
>
...snip...
> < Do you remember the problem I proposed of generating over 10,000vrms
> from
> < a 120v, 60Hz source with only one capacitor and one inductor?
>
> Nothing new there, Rich... there is a whole new disipline based on that
> principle. It's called switching power supplies.
The switching power supplies that I am familiar with switch DC at a 50KHz
or so rate into a toroidial transformer, whose secondary windings are
rectified, filtered, and sampled by a comparator that proportionally
controls the switcher to maintain the desired DC output voltages.
It seems to me that a closer analogy is the L-network. An L-network is
basically a series L/C circuit that is operated slightly off-resonance. In
the problem that used one L in series with one C to generate 10kV at 60Hz,
the two reactances were one ohm apart at 60Hz. Thus, with 120v @60Hz
applied, over 100A of current passed through L and C. Since the reactances
were approx. 100 ohms, the voltage across either reactance exceeded the
goal of 10kV.
> < So why does the 10m bandswitch contact sometimes arc.? An open 10m
> contact
> < probably has around 1 pF. The typical 10m section of a tank L has around
> < 0.9uH. The series-resonant frequency is 170MHz. In an AL-80, this is
>
> < in the neighborhood of the anode circuit's resonance at 160MHz. Thanks to
> < Mr. Rauch's measurements, we know that the AL-80's tune capacitor has no
> < resonances in that neighborhood. I have not heard of any case where an
> < AL-80 tune capacitor allegedly arced.
>
> Bingo!!!!! Of course not!! It is virtually a dead short at VHF! Sort of
> kills
> the possibility of VHF energy arcing the bandswitch.......
Sure, a circuit that is series-resonant is a virtual dead short, but ONLY
at the frequency of resonance. Is a circuit that is series-resonant at
170MHz a virtual dead-short at 160MHz? Not in my opinion. It seems to me
that this situation is similar to the example of generating 10kV from a
120v, 60Hz source with an almost series-resonant circuit. ........Please
double-check your bingo card before coming to the pay-window,
bingo-players.
> < However, I have heard of a number of
> < cases where AL-80 10m bandswitch contacts mysteriously arced and
> < evaporated. If you're interested, photos of same are available at
> Figure
> < 15 on my Web site.
>
> The winds are a shifting. Last week, when I mentioned ten meters, you
> went on to say this happened on 40 meters, not ten meters.
Different amplifers, Peter.
>
> < Did you accept someone's word that this was the case, or did you calculate
> < a typical tank Q reduction, based on Wes' ESR measurement for a Ni-Cr Ls
> < suppressor at 30MHz (4.32 ohms) and/or at 10MHz (0.88 ohms)?
>
> Rich, I am confused. You are marketing this product, and have been
> for years. You mean that thanks to Wes' calculations you only now
> have some idea of what they doing?
I talked about how VHF parallel-equivalent resistance (Rp) determined VHF
voltage gain in "Parasitics Revisited" (9/90-10/90 QST). I knew that Rp
decreased with Ni-Cr alloy conductors as frequency increased, but I did not
have the test equipment to make measurements at the frequencies where VHF
parasitic-oscillations typically take place. Seven years later, Wes
Stewart made the measurements that I needed when I wrote the manuscript.
>
> < Mr. Rauch made the claim that a serious HF Q degradation would result
> < from using a Ni-Cr VHF suppressor.
>
> Yes, it should. ...snip...
We have 4.32 ohms in series with 5pF (the anode-C of a 3-500Z) of the
total tune C of 5pF +15pF=20pF at 30MHz. At 30MHz, 5pF has about 1100 ohms
of reactance. How can 4.32 ohms of R in series with 1100 ohms of Xc cause
a 'serious HF Q degradation' when 75% of the HF tank current at 30MHz does
NOT pass through the 4.32 ohm resistance? Please show us the calculations,
Mr. Bertini and/or Mr. Rauch.
> < Mr. Rauch presented me with the following problem. F=160MHz: An L of
> < 85nH is paralleled with 100 ohms. How does the VHF current divide?
> < According to AC Circuit Analysis, the calculated current through the
> < 85nH L is about 10% higher than the calculated current through the 100 ohm
> < resistor. However, Mr. Rauch concluded that these calculations are
> < invalid. Do you accept Mr. Rauch's conclusion?
>
> Rich, the math and caculations have been gone over, and over, and
> over....
> Do you disagree with Rauch's math? Argue with him.
>
Mr. Rauch most likely did NO mathematical calculations, Peter. If he had,
he assuredly would not have given me a problem that disproved his
intransigent assertions that insignificant VHF current flows in Ls.
In the 9/94 QST (in regards to VHF suppressors) Mr. Rauch said:
---"The coil's reactance increases with frequency, and at VHF most of the
signal path is through the resistor. It is plainly evident that the
dominant component at VHF is the resistor, not the coil."---
Apparently, self-recognized 'experts' are reluctant to acknowledge mistakes
they make.
So, Peter: Do you accept Mr. Rauch's statement regarding the VHF current
in Ls as fact? If you do not accept it, then why do you accept Mr. Rauch's
assertion that "...changing the resistance of the suppressor's coil
radically affects the HF circuit Q." (9/94 QST)
IMO, any electronic engineer who rejects AC Circuit-Analysis, and instead
solves such problems by intuition, quite probably does not have both oars
in the water.
..snip...
BTW, I am still waiting for one of the naysayers to provide us with the
values of a copper-wire suppressor that equals the performance of a
suppressor made from resistance-wire. ........ The time has come to stand
and deliver. What will it be--deliver or quiver?
---------------------------
e-mail copy to Mr. Peter Bertini, K1ZJH
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <5d3ue5$a...@dallas1.connect.net>, phil...@connect.net (Phil
Clements) wrote:
> In an article in the January, 1974 issue of Ham Radio, W4QQ has a 4-1000A
> 5-band Kilowatt Linear that uses a suppressor consisting of:
>
> A 20 pf ceramic capacitor (Centralab 852S) which is a mini-doorknob with
> axial leads in parallel with 4-470 ohm 2 watt resistors and 4 turns #12
> wire, 1/2 inch diameter, resonated at 110 mhz. I built that amp project
> in 1981, and it has performed flawlessly ever since, with the original parts
> and tube.
>
Interesting, Phil. Was the anode-circuit resonance in this amplifier also
110MHz? Did the author explain the purpose of the 20pF capacitor? How
much DC anode voltage do you use? Does this amplifier/suppressor perform
satisfactorily on 10m?
--
--Rich-- ag6k, 805-386-3734
Ian White, G3SEK
R. L. Measures wrote:
>BTW, I am still waiting for one of the naysayers to provide us with the
>values of a copper-wire suppressor that equals the performance of a
>suppressor made from resistance-wire. ........ The time has come to stand
>and deliver.
Rich is referring to me; but I'm not a naysayer - I'm a skeptic.
This all arose from an off-newsgroup discussion after I'd plotted the
N7WS measurements of the three different suppressors on graphs (that's
two of Rich's NiCr models and a conventional one from W8JI).
The Q vs frequency graphs are plotted on a log-log scale to emphasize
the ratios between the three different units.
>What will it be--deliver or quiver?
Neither.
As Wes, N7WS points out in his revised web pages (http://www.azstarnet.c
om/~n7ws):
"at VHF, there are only slight differences between the effective
parameter values for the two types of networks."
All that I was saying to Rich was that if he wanted to make the
conventional suppressor have *exactly* the same Q as one of his, at any
desired frequency, he only needed to shift the whole curve up or down by
a small amount (this being a useful property of the log scale).
I never did offer to "provide us with the values of a copper-wire
suppressor that equals the performance of a suppressor made from
resistance-wire" so Rich would have waited a long time for that!
Meanwhile Wes has already done what he asked, on his revised web page:
"Some mathematical manipulation can demonstrate that shunting the
conventional suppressor with additional resistance will bring the curves
into better alignment. An additional 250 ohm resistor in parallel with
the network will cause the Rp curves to nearly overlap the nichrome
curves above about 30 MHz. Below 30 MHz, the Rp of the adjusted
conventional type will be higher than the Rp of the nichrome device."
So now Rich can cross this one off his list.
Joe Subich
In message <measures-ya0240800...@news.vcnet.com> -:>
:>
:>Do these wires run around the bandswitch? In every amplifier I have
:>seen, the bandswitch leads follow a direct route to the corresponding
:>tank inductor taps I have never seen a case of bandswitch contacts
:>arcing to anything but to open contacts, and to the rotor.
:>
Definitions of "around" include "running in all directions" and
"somewhat
close to" both of which would apply to nearly every bandswitch I have
ever seen. No legitimate amplifier builder would wrap leads from the
coil to the switch completely along the circumference of the switch as
you wold imply.
Yes, there are wires around the bandswitch in that there are wires
running in all directions from the switch.
:>
:>HF arcs are typically blue, not purple. HF arcs are unlikely to cause
:>interference to a particular TV channel. HF arcs are probably not going
:>to peg a VHF field strength meter in the same room. IMO, HF is unlikely
:>to masquerade as VHF on a spectrum analyzer.
:>
I don't know of many commercial amplifiers, particularly amplifiers by
manufacturers you have so regularly libeled, that would allow observing
the color of of any arc while operating.
An arc contains all manner of harmonics. Since you claim to have
measured
high-Q VHF resonances in the plate circuits of many commercial amlifiers
it is entirely possible that a high-Q resonance would allow the
radiation
of sufficient VHF energy to be detected on a particular TV channel or
peg
a sensitive VHF field strenght meter.
The only reported and documented observation by spectrum analyzer is
that
by Marvin Born on a nototiously unstable Create 3CX3000 amplifier ... an
amplifier that *YOU* have insisted on claiming (many times) was an
Ameritron AL-1500.
:>
:>You did not explain why a load fault was not subsequently discovered in
:>the antennas or feedlines. Why did the mysterious arcing cease after
:>the VHF-Rp of the VHF parasitic-oscillation suppressors was decreased?
:>
The problem need not be confined to the antennas and/or feedline. It
could also have been an antenna switch, the T/R relay, a coax cable, or
a lightning arrestor (or are you going to hide that story again?).
:>
:> Are X-rays and Gamma-rays charged particles, Joe?
:>
Not a subject of this discussion ... AT&T claim the solar event was not
the cause of the failure.
:>
:>Are you saying that geosynchronous satellites do not travel in and out
:>of Earth's shadow, Joe?
:>
Of course they do ... however, if this had been caused by a solar event
it would nost sretainly have also damaged other TV satellites of the
same
or similar design within a few degrees of T401. However, since all of
this flies in the face of your pseudo-science you will ignore it or
distort it as usual.
========================================================================
Joe Subich, W8IK ex-AD8I
<W8...@IBM.NET>
<W8...@AMSAT.ORG>
<sub...@ibm.net>
<sub...@compuserve.com>
========================================================================
w8j...@aol.com
>
> A 20 pf ceramic capacitor (Centralab 852S) which is a mini-doorknob with
> axial leads in parallel with 4-470 ohm 2 watt resistors and 4 turns #12
> wire, 1/2 inch diameter, resonated at 110 mhz. I built that amp project
> in 1981, and it has performed flawlessly ever since, with the original
parts
> and tube.
>
> (((73)))
> Phil, K5PC
W4QQ is a smart man. He understands parasitics and parasitic suppressors.
The 4-1000 is a pretty unstable tube, so his method is a good one..
73 Tom
R. L. Measures
In article <5d622c$9m8$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) wrote:
> In message <measures-ya0240800...@news.vcnet.com> -
> meas...@mail.vcnet.com (R. L. Measures ) writes:
...snip...
>
> I don't know of many commercial amplifiers, particularly amplifiers by
> manufacturers you have so regularly libeled, that would allow observing
> the color of of any arc while operating.
Apparently you are unfamiliar with the definition of the word 'libel'---and
you have never looked into the top of a Heath SB-220 amplifier.
...snip...
> :>You did not explain why a load fault was not subsequently discovered in
> :>the antennas or feedlines.
Sorry for the inadvertent omission, Mr. Subich. No load fault was
subsequently discovered in the antennas or feedlines because there was no
load fault in the antennas or feedlines.
In other words, the arcing that destroyed the original bandswitch and the
replacement bandswitch in Rick May's AL-80 can not be explained by a load
fault. The only thing that was altered after bandswitch #3 was the VHF-Rp
of the VHF suppressor in the AL-80. The original suppressor has an Rp of
about 166 ohms at the anode resonant frequency of160MHz. The replacement
suppressor has an Rp that is around 101 ohms at 160MHz. Since voltage gain
is essentially Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain. //
The e-mail address of Rick May is listed on my Web page if you need more
answers.
>Why did the mysterious arcing cease after
> :>the VHF-Rp of the VHF parasitic-oscillation suppressors was decreased?
> :>
>
> The problem need not be confined to the antennas and/or feedline. It
> could also have been an antenna switch, the T/R relay, a coax cable, or
> a lightning arrestor (or are you going to hide that story again?).
>
If this had been the case, then why didn't a problem show up when the Drake
L4B was connected to the same load? If there was a problem with the things
you listed, then why was there no apparent problem when the AL-80 was
placed back in service with its new VHF suppressor?
> :> Are X-rays and Gamma-rays charged particles, Joe?
> :>
> Not a subject of this discussion ... snip...
You were the one who used the term 'charged particles', Mr. Subich.
........Mr. Miklos, where are you?
----------------------------------
e-mail copy to Mr. Subich
--
--Rich-- ag6k, 805-386-3734
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>
> We have 4.32 ohms in series with 5pF (the anode-C of a 3-500Z) of the
>total tune C of 5pF +15pF=20pF at 30MHz. At 30MHz, 5pF has about 1100
ohms
>of reactance. How can 4.32 ohms of R in series with 1100 ohms of Xc
cause
>a 'serious HF Q degradation' when 75% of the HF tank current at 30MHz
does
>NOT pass through the 4.32 ohm resistance? Please show us the
calculations,
>Mr. Bertini and/or Mr. Rauch.
You have to watch out at every turn Peter, just like when he said claimed
I said using a resistor was a bad idea. Keep in mind he just says things.
The stuff above is a good example, Mr. Measures ignores several things.
Let me give you a hint to ask him about just one of them, the effects of
the series inductance in the path from the tank to the anode.
Arguing a complex circuit with five or ten major reactances should only
take a few months of time. It will take at least five exchanges of
spitting and hissing to get path inductance correct. Bet on it.
73 Tom
w8j...@aol.com
In article <K5BBkKA6...@ifwtech.demon.co.uk>, "Ian White, G3SEK"
<G3...@ifwtech.demon.co.uk> writes:
>
>Meanwhile Wes has already done what he asked, on his revised web page:
>
>"Some mathematical manipulation can demonstrate that shunting the
>conventional suppressor with additional resistance will bring the curves
>into better alignment. An additional 250 ohm resistor in parallel with
>the network will cause the Rp curves to nearly overlap the nichrome
>curves above about 30 MHz. Below 30 MHz, the Rp of the adjusted
>conventional type will be higher than the Rp of the nichrome device."
>
>So now Rich can cross this one off his list.
Mr. Measures magic nichrome suppressor is equal to a the Q a stock AL-80A
suppressor with a 71 ohm resistor would have, instead of the 100 ohm
resistor used.
Say, isn't that some improvement ?? Years of promoting how much lower the
VHF Q is, and it's almost nothing.
The PA goes from a oscillating arcing nightmare, to a stable smooth
perfect PA by adding an "invention" that lowers anode SYSTEM (look, I
said system) VHF Q from about 3 to a new total of 2.9! Not only that,
Measures SWORE he could measure that change clearly on his super dooper Q
indicating GDO that is better than any network analyzer.
Rich, why not sell your customers a 68 ohm resistor for $50. It's a bigger
improvement yet, so you should be able to charge even more $$$ for it!
Maybe QST will run another article, describing how a change to a 68 ohm
resistor will:
1.) Stop melting glass in Chinese tubes.
2.) Stop bandswitch arcing.
3.) Stop other tube failures like damaged, melted. bent or twisted grids
or filaments.
4.) Stop meter failures.
5.) Stop plate choke failures.
ALL while making the PA notably easier to tune.
Perhaps someone in California is just having fun and at any moment is
going to shout, "April fool"?
73 Tom
w8j...@aol.com
In article <32f53900....@news.airmail.net>, jfr...@airmail.net (J.
Fred Riley) writes:
>
>I don't think the use of a current-limiting resistor was the subject
>of criticism. It was, rather, the type of resistor used. A
>current-limiting resistor is almost always a good idea. Using the
>wrong kind of resistor is almost always a bad idea. Using a resistor
>that's too "small" is sorta like using the low-voltage fuse in a HV
>line.
>
>
That's correct Fred. The non-surge rated resistor Measures suggested was a
ten ohm 700 volt rated component.
FOr about two dollars more, the user could purchase a 35 ohm surge rated
componet that would withstand fault impact without damage, and wouldn't
arc internally from end to end.
The other problem was with Mr. Measures calculations. He assumed the ESR
of the supplya and the entire poath was zero ohms in his example, when in
fact it is much higher. In a typical PA like the AL-1200 the fault path
ESR is about 12 ohms as is WITHOUT any resistor.
Adding a ten ohm resistor would lower fault current from 291 amperes to
159 amperes IF the resistor did not arc internally. Big deal.
Adding a correct style 35 ohm resistor would lower fault current to 75
amperes. That's a good improvement.
Adding the wrong type of 10 ohm resistor is a waste of time, unless the PA
has a big oil filled cap, low resistance plate choke, and ypou are very
lucky in a fault condition and the resistor does not arc internally.
Like in other areas, Measures clearly understood the point I made but
chose to distort the data. It's a habit I guess.
73 Tom
w8j...@aol.com
> In other words, the arcing that destroyed the original bandswitch and
the
>replacement bandswitch in Rick May's AL-80 can not be explained by a load
>fault. The only thing that was altered after bandswitch #3 was the
VHF-Rp
>of the VHF suppressor in the AL-80. The original suppressor has an Rp of
>about 166 ohms at the anode resonant frequency of160MHz. The replacement
>suppressor has an Rp that is around 101 ohms at 160MHz. Since voltage
gain
>is essentially Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain. //
>The e-mail address of Rick May is listed on my Web page if you need more
>answers.
What a large change! From 166 ohms to 101 ohms! No wonder the suppressor
cures everything from arcing tubes to exploded meters.
Seriously, the problem is Rich doesn't understand the system well enough
to realize the "large" change is diluted and the expected effects reversed
by the rest of the system.
A HIGHER impedance suppressor actually reduces the tendency to of a system
to oscillate much MORE than a lower impedance suppressor. Here's
why....and pay attention to this carefully because it will show how little
Measures understands the very thing he has become a guru of....
Measures claims above:
>of the VHF suppressor in the AL-80. The original suppressor has an Rp of
>about 166 ohms at the anode resonant frequency of160MHz. The replacement
>suppressor has an Rp that is around 101 ohms at 160MHz. Since voltage
gain
>is essentially Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain. //
Here's the path:
tube-----L1-----sup-------L2--------XL------bandswitch
^
Plate tune
The tube is a time varying resistance in shunt with several pF of C (not
exactly the 5 pF of pure C measured in a four foot box with only a
vertical transmission line used in the Eimac data). From the tube to the
suppressor we have series L and shunt distributed C represented as L1
(meaning line 1) . From the suppressor to the tank we have series L and
shunt distributed C trepresented as L2 (or line 2).
From line two we have a fairly low reactance to ground, and then the path
branches off to XL (the tank inductor) and the bandswitch.
At different frequencies the total impedance of this path varies widely.
At lower frequencies the tank and tube impedances dominate, at higher
frequencies the tube (including grid lead impedance), L1, L2, sup, and
plate tune dominate.
Here's what we can all clearly see. If the suppressor is a LOW impedance
compared to the entire path of the system it will have VERY little effect
on Q or circuit loading. It only has a major effect when it's impedance is
HIGH compared to the path impedance.
Using logic nearly anyone is capable of understanding, we can solve this
problem by going to the extremes of impedance and looking at the system.
As a reminder, Measures claims above:
>of the VHF suppressor in the AL-80. The original suppressor has an Rp of
>about 166 ohms at the anode resonant frequency of160MHz. The replacement
>suppressor has an Rp that is around 101 ohms at 160MHz.
He "thinks" this lowers gain, because he considers it the ONLY element in
the circuit. He considers it a direct connection in series with the anode
to voltage source, and everthing else in the system is vanished from view!
Take Mr. Measures claim that a lower impedance helps to the extreme, and
build a suppressor with the lowest possible parallel equivalent impedance.
That suppressor would be a short circuit. If we replace a conventional
suppressor with a direct short the PA will be more stable, according to
Mr. Measures logic, "Since voltage gain is essentially Mu*Rp, decreasing
VHF-Rp decreases VHF voltage gain."
HE said that, not me. He said what amounts to just shorting the suppressor
out will produce the lowest VHF gain and most stability!!!!!!!
Isn't that amazing? The guy who knows so much more than anyone else about
suppressors and knows EVERYTHING everyone else does wrong argues clearly
(in his own words) the best possible suppression comes from the lowest
possible impedance suppressor.
That clearly means a direct short (NO suppressor at all) is the best
suppressor, and a pure 1000 ohm resistance in series with the anode would
INCREASE VHF gain!!!
Unreal!
Anyone with even the most elementary experience with audio amplifiers
knows quite well that a LOW value anode to source resistor INCREASES high
frequency response in the amplifier, while a high series resistance LOWERS
the high frequency rsponse.
Mr. Measures, on the other hand, argues that his magic nichrome has a
lower impedance, and thus LOWERS VHF gain! He clearly argues with less
impedance, the PA is more stable. Exactly the opposite of what the most
elementary common sense tells us happens!
Which is it Measures? Do we want a lower impedance, like your suppressors
offer? Do we want a higher value of impedance like the 80A uses? Will an
unstable amplifier be more stable with a zero impedance suppressor, or
will adding turns (increasing the inductance and impedance) improve
stabilty?
Does anyone out there use the Measures method of removing turns until the
PA QUITS oscillating? All my life, I've done the opposite. If you look at
the most unstable tubes, they ALL use the higher values of suppressor
impedance. Those idiot designers.
Even W4QQ made a big mistake (according to Measures), because he parallel
resonanted his suppressor..... guaranteeing maximum impedance at the
parasitic frequency. I wonder why that worked, since it disagrees with the
Measures theory of stability.
If you look carefully at that big bunny in the battery
commercial....you'll see the springs on the drum are nichrome. At least I
have an inking they are.
Rich, you demanded I discuss this "Since voltage gain is essentially
Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain." thing with you. You
were right, we did need to talk about it. I had no idea you didn't
understand circuit analysis related to gain. But I really think you need
to figure out how the tuning capacitor works first, before you argue gain.
Later on, in a year or two, we may be up to the point of discussing gain.
In the mean time, take your hi-fi amplifier and increase the load
resistances through all the stages, that will really improve HF response.
In the mean time, if you run across a PA that oscillates, lower the
suppressors resistance and inductance!!! That will work too. (Seriously to
anyone left reading this, don't do that. Such things only work if you live
in a pathological science world.)
Rich, it's very clear you have no idea how suppressors work. You just say
things.
73 Tom
comm...@aol.com
> However, the typical filter capacitor in a typical 1500w amateur
> radio amplifier has about 10 ohms of ESR. How much current is 2800V/10
>
> In the Sept. 1994 QST critique of the Jan. 1994 QST article "The Nearly
> Perfect Amplifier", my suggestion about using a peak fault-current
limiting
> resistor in the anode supply was pooh-poohed by the (self) 'recognized
> amplifier expert'.
?
Rich, I would rather discuss why you tried to BS me and this newsgroup
regarding the merits of using a peak fault-current limiting resistor
in the anode supply path.
You claimed that Mr. X pooh-poohed the idea. But, isn't it a
fact that Mr. X's concerns on this matter were instead directed
at your claims that a 10-ohm resistor with a 700-volt
rating would be adequate for the purpose?????
Weren't your original claims based on your assumption
that the plate supply had an ESR of zero ohms? Isn't a figure
of 16 ohms closer to reality, Rich? Or, are you
choosing to ignore the added resistance of the plate choke
and meter shunts? The winds are shifting again. Go check your
smoke detector batteries.
Why are you out-and-out misleading us, Rich?
> So why does the 10m bandswitch contact sometimes arc.?
Spiders? Dust balls/cat hairs?? Photons??? Go for it!!
> An open 10m contact
> probably has around 1 pF. The typical 10m section of a tank L has
around
> 0.9uH. The series-resonant frequency is 170MHz. In an AL-80, this is
in
> the neighborhood of the anode circuit's resonance at 160MHz. Thanks to
Mr.
> Rauch's measurements, we know that the AL-80's tune capacitor has no
> resonances in that neighborhood.
Yo ho, Rich!!! Helloooooooo?
Are you adjusting reality to fit your opinions again?
Didn't Mr. Rauch ALSO show that the plate tune
capacitor had such a low value of XL that it would be
impossible to develop the levels of VHF you claim?
Isn't the tune capacitor shunting
the RF path to ground before the bandswitch? Is the gap
across between the rotor/stator plates that much further
apart than the the bandswitch contacts? Take the XL and the
DC resistance of the capacitor, calculate the impedance at
at 160 MHz, and tell me what the voltage drop will be. Don't
go on another tangent about non-existent resonances.
>
>>A question: Peter--If you were designing an HF amplfier, would you want
a
>>VHF parasitic oscillation suppressor that exhibited more (or less)
VHF-Rp
>>than another type of VHF suppressor?
>
Rich, I would honestly want a suppressor that did
not compromise the operating Q of the amplifier at
HF frequencies.
> Did you accept someone's word that this was the case, or did you
calculate
> a typical tank Q reduction, based on Wes' ESR measurement for a Ni-Cr Ls
> suppressor at 30MHz (4.32 ohms) and/or at 10MHz (0.88 ohms)? If you did
> the calculations, Peter, I would appreciate seeing how to do it, and how
to
> calculate what the probable increase in harmonic output would be. I
don't
> have a clue as to how to proceed.
I am sure the harmonic output relating to Q is the most
easily solved portion of the puzzle. Various handbooks show
computer modeled values for Pi and Pi-L networks for a variety
of tube load impedances and the harmonic suppression based
on the selected value of loaded Q.
I am not experienced in the design of RF power amplifiers.
But, I have given some thought to solving the reduction of loaded
Q in a the plate tank circuit based on the HF Q of the suppressor.
There is no simple solution to your question based on what you
presented. But then again, you chose to neglect numerous variables.
The effect of the tube plate lead inductance, the fact that the
tube is a varying resistor in operation, the reactance of the
plate blocking capacitor, interconnecting straps, grid
bypassing scheme -- all these and more would come into
the final equation in one form or another.
> blah, blah, snip >>>Do you accept Mr. Rauch's conclusion?
Rich, this is what I accept. I accept Wes' calculations of your
suppressor. I accept both Mr. Rauch's and Ian Smith's observation
that a conventional suppressor -- with slight adjustments -- can be
made to be as effective at VHF as a suppressor wound with either rusty
bailing wire or nichrome.
Rich, do you accept Wes's measurements as being accurate?
Mr. Rauch has stated from day-one that a simple adjustment would
permit a conventional suppressor to perform as well as your nichrome
wound suppressor. Now that Wes's data is available, others are
reaching the same conclusion. Are they all part of a great conspiracy?
> I haven't seen a VHF suppressor design that uses a C. What does this
> circuit look like?
I believe Phil, K5PC, answered this a month or so back when I posed
a query regarding the use of RCL components in suppressors. He pointed
out that this was done by W4QQ in a 1974 Ham Radio project. I also
understand that the "self-proclaimed" amplifier expert
has used RCL suppressors in some of his Ameriton designs.
W4QQ was able to reduce what would have been a rather large value
of XL by parallel resonating the suppressor to 110 MHz, the resonant
frequency of the 4-1000. Tom has used fixed capacitors in series with
the suppressor resistor to cancel the XL of that device. (series
resonance) This eliminates RF voltage drop across the XL component
of the resistor, allowing full dissipation of VHF energy in the
resistor. Clear enough? Care to muddy the waters?
R. L. Measures
In article <19970204154...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <32f53900....@news.airmail.net>, jfr...@airmail.net (J.
> Fred Riley) writes:
>
> >
> >I don't think the use of a current-limiting resistor was the subject
> >of criticism. It was, rather, the type of resistor used. A
> >current-limiting resistor is almost always a good idea. Using the
> >wrong kind of resistor is almost always a bad idea. Using a resistor
> >that's too "small" is sorta like using the low-voltage fuse in a HV
> >line.
> >
> >
>
> That's correct Fred. The non-surge rated resistor Measures suggested was a
> ten ohm 700 volt rated component.
Actually it's rated at 10 volts maximum--i.e., E=(R*P)^0.5 = 100^0.5=10v.
However, job at hand is typically discharging a 25microfarad capacitor that
is charged up to around 2800v. In round numbers, that's about 100
watt-seconds (a.k.a. joules). In the beginning, we used a cheaper 10 ohm
10w resistor, but they broke down in this application. We switched over to
a 2-inch long, glass-coated design that seems to be able to handle the job
without breaking down between the turns of resistance-wire. For 3kV or
more, we recommend using two of them in series to keep the voltage gradient
per turn within the capability the resistors.
> FOr about two dollars more, the user could purchase a 35 ohm surge rated
> componet that would withstand fault impact without damage, and wouldn't
> arc internally from end to end.
Cesivid quoted me a price of just under $10,000 for 1000 surge-rated units.
Cesivid Co.'s telephone number is 716 286 7610.
>
> The other problem was with Mr. Measures calculations. He assumed the ESR
> of the supplya and the entire poath was zero ohms in his example, when in
> fact it is much higher. In a typical PA like the AL-1200 the fault path
> ESR is about 12 ohms as is WITHOUT any resistor.
The article says: "During a major glitch, the anode (plate) current meter
is subjected to a current surge as the HV filter capacitors discharge.
Such a current is several hundred peak amperes. Several thousand volts
divided by 'zero ohms' is not several hundred amperes. Try dividing
several thousand volts divided by 10 ohms and see what you get, Mr. Rauch.
The added10 ohm glitch resistor came later.
12 ohms seems pretty high. There must be lots of ESR in the AL-1200's
filter capacitors. With 0.7a of ripple current, you could be dissipating
plenty of heat in the electrolytic filter capacitors. Which capacitor
manufacturer made them?
> Adding a ten ohm resistor would lower fault current from 291 amperes to
> 159 amperes IF the resistor did not arc internally. Big deal.
It is a big deal to us because we use 200a-maximum glitch protection diodes.
> Adding a correct style 35 ohm resistor would lower fault current to 75
> amperes. That's a good improvement.
True enough. But how low do we need to go? Svetlana recommends 10 ohms to
25 ohms for a 1500w tube. Would anybody like to guess whether or not
Ameritron/MFJ amplifiers use a 'correct style' resistor?
> Adding the wrong type of 10 ohm resistor is a waste of time, unless the PA
> has a big oil filled cap, low resistance plate choke, and ypou are very
> lucky in a fault condition and the resistor does not arc internally.
The previous paragraph does not appear to make sense.
Where is Mr. Miklos?
BTW, Tom it looks like your spell-checker has developed some ESR.
-------------------------------
e-mail copy to Mr. Tom Rauch
--
--Rich-- ag6k, 805-386-3734
Wes Stewart
In article <5d8s7o$sm6$1...@news-s01.ny.us.ibm.net> sub...@ibm.net (Joe Subich) writes:
>From: sub...@ibm.net (Joe Subich)
>Subject: Re: good source for amplifier tips
[snip]
>*SO WHAT!* Even Wes Stewart, who made the measurements you are so
>fond of quoting now states that the supressors do not operate in the
>parallel mode (Rp is not significant) but rather operate as a SERIES
>component in a COMPLEX network which includes the tube output
>capacitance, the stray capacitance from the plate blocker to ground,
>the parasitic inductance in the plate blocker, stray reactances in
>the plate choke, etc.
Hold on there counselor. You are putting words in my mouth. I did not state
that suppressors do not operate in a parallel mode and that Rp is not
significant. Clearly, the way a suppressor is configured, it is a parallel
network. Equally clear is that it operates in series with the remainder of
the circuitry.
Every parallel network has an equivalent series one (an vice versa), which
when measured independently is indistinguishable from the other. This does
not mean that the equivalent networks behave the same when imbedded in another
network.
I offered to make the measurements as an unbiased third party. This is what I
do for a living. This does not mean that I don't have an opinion, just that I
try to keep my opinions from coloring my judgement; again part of my job.
I published the data for everyone to use in making his own judgement. Frankly,
I'm not personally convinced that extrapolating some isolated laboratory
measurements into a definitive statement about the relative "goodness" of a
particular circuit is good science. Sounds more like lawyerin' to me:)
[snip]
Respectfully,
N7WS
comm...@aol.com
> IMO, any electronic engineer who rejects AC Circuit-Analysis, and
instead
> solves such problems by intuition, quite probably does not have both
oars
> in the water.
How's the water, Rich?
BTW, I am still waiting for one of the naysayers to provide us with the
values of a copper-wire suppressor that equals the performance of a
suppressor made from resistance-wire. ........ The time has come to stand
and deliver. What will it be--deliver or quiver?
Tom showed you how and why months ago. Ian White reached the
same conclusion based on measurements you endorse...
---------------------------
comm...@aol.com
>
>> A 20 pf ceramic capacitor (Centralab 852S) which is a mini-doorknob
with
>> axial leads in parallel with 4-470 ohm 2 watt resistors and 4 turns
#12
>> wire, 1/2 inch diameter, resonated at 110 mhz. I built that amp
project
>> in 1981, and it has performed flawlessly ever since, with the original
>parts
>> and tube.
>>
> Interesting, Phil. Was the anode-circuit resonance in this amplifier
also
>110MHz? Did the author explain the purpose of the 20pF capacitor? How
>much DC anode voltage do you use? Does this amplifier/suppressor perform
>satisfactorily on 10m?
>
>--
>--Rich-- ag6k, 805-386-3734
>
Rich
The suppressor is at best a compromise between VHF and HF performance.
The designer must keep the inductance at a minimum, otherwise the
designed loaded Q of the HF tank will be compromised. The higher
the plate resonance, the smaller the inductor maybe made and still be
effective at VHF. Most modern tubes have resonances in the area of
160 or 180 MHz I believe. But, the 4-1000 has
a plate resonance at 110-MHz. Rather than using a large value of L in
the suppressor (and comprising 10 meters) the designer opted to use
a rather small value of L and parallel resonanted the suppressor
LC components at 110 MHz.
At 110-MHz, where Xc and Xl cancel, only the suppressor resistor
is in the path for VHF energy. The coil and capacitor are resonant, and
present a very high impedance path for VHF. An ideal situation. Look at
the
series circuit equivalent, with L, C, and R in series. L and C
reactances cancel, the impedance is determined by the value of
R.
Would you rather use a clever scheme like this in a 4-1000 design,
or would you opt for a suppressor made with nichrome windings,
and compromise the ten-meter tank Q?
Rich, why do you ask us to accept your claims at face value,
despite the fact that Wes's measurements show that nichrome
suppressors offer negliable improvement over conventional designs --
while
on the other hand where those measurements allow others to
show that they compromise tank Q on ten-meters, we should assume
those facts to be of no consequence?
You have admitted in this forum that until Wes undertook some
basic and fundamental measurements of your suppressors you had
no idea how they actually performed -- despite having marketed them
for the past 13 years..
You have admitted that you do not own a Q-meter, RX meter, Network
analyzer or spectrum analyzer. But, through good engineering and science
(and a crystal ball or two??) you have all the answers using a dipmeter
and by watching pretty arcs. Observations based on random occurences,
in uncontrolled environments combined with an infinate number of variables
was not acceptable science when I went to school. But, perhaps the
sciences are more liberal these days..
You have also admitted that you have no idea how to calculate
the effect of suppressors on the loaded Q of a HF tank coil. Where
is your credibility? What happened to good engineering and science?
When an industry recognized, practicing engineer questions
your findings, or posts his views, why is he
misquoted, subjected to obscurantism,
and vilified? I have been following this thread in the hopes of
learning and honing my own limited skills. I tried to keep an open
mind, being careful not to involve personalities or questioning
financial motives. I don't have all of the answers, but as the smoke
slowly clears I think I can pretty well determined who does.
73
Joe Subich
In message <measures-ya0240800...@news.vcnet.com> -:>
:>Apparently you are unfamiliar with the definition of the word
:>'libel'
As a member of the Fifth Estate, I am most certainly familiar with
the concept of libel and can assure you that publication (including
posting on the internet) of charges such as you make may very well
be considered libel. I'm suprised that Messers Erhorn, Henry,
Collins, et al. haven't considered legal action. Since none of them
are "public figures" in the legal sense, they do not have the burden
of proving malice ...
Much of what you have claimed has been proven false by third parties.
You post as fact information you know to be untrue (Marvin Born's
alleged AL-1500). You asssign impure motives to anyone who dares to
disagree with you. And you ridicule anyone who shows the errors in
your pseudo-science.
:>---and you have never looked into the top of a Heath SB-220
:>amplifier.
Yes. Since it consists of a double case for shielding, has a fairly
bright set of tube filaments (though certainly not bright enough to
read the newspaper), and has a fairly pronounced redish-orange glow
from the plates at normal power output ... no person of normal vision
and color perception is going to be able to tell with any certainty
the difference between a blue and pruple arc -- particularly an
unexpected and transient arc.
:> The only thing that was altered after bandswitch #3 was the VHF-Rp
:>of the VHF suppressor in the AL-80.
Are you absolutely certain that every other portion of the system was
unchanged? Are you certain that a different jumper cable or antenna
switch was not used? Are you certain that rewiring the anode circuit
didn't result in some minor (but significant) change to a TR relay?
:>The original suppressor has an Rp of about 166 ohms at the anode
:>resonant frequency of160MHz. The replacement suppressor has an
:>Rp that is around 101 ohms at 160MHz. Since voltage gain is
:>essentially Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain.
*SO WHAT!* Even Wes Stewart, who made the measurements you are so
fond of quoting now states that the supressors do not operate in the
parallel mode (Rp is not significant) but rather operate as a SERIES
component in a COMPLEX network which includes the tube output
capacitance, the stray capacitance from the plate blocker to ground,
the parasitic inductance in the plate blocker, stray reactances in
the plate choke, etc.
N7WS further states on his web page:
"Some mathematical manipulation can demonstrate that shunting
the conventional suppressor with additional resistance will
bring the curves into better alignment. An additional 250 ohm
resistor in parallel with the network will cause the Rp curves
to nearly overlap the nichrome curves above about 30 MHz. Below
30 MHz, the Rp of the adjusted conventional type will be higher
than the Rp of the nichrome device."
Perhaps you can explain why adding a 250 Ohm resistor in parallel
with a conventional supressor would eliminate VHF parasitics?
Phil Clements
In article <32f36a75...@news.airmail.net>, jfr...@airmail.net (J. Fred Riley) says:>This ought to be carved in stone somewhere.
>
>
At what frequency does the arc change from blue to purple? How about
the good old white arcs we see in the springtime?
(((73)))
Phil, K5PC
>Fred, W8OY
>
>JFR...@AIRMAIL.NET FREDW...@AOL.COM
Phil Clements
In article <19970205012...@ladder01.news.aol.com>, comm...@aol.com says:
>
>In article <measures-ya0240800...@news.vcnet.com>,
>meas...@mail.vcnet.com (R. L. Measures ) writes:
>
>>
Sorry Rich, the article below was apparantly to me but I don't have
access to it on the newsgroup! (Doesn't appear) Will answer your
questions now.
>>>
>> Interesting, Phil. Was the anode-circuit resonance in this amplifier
>also
>>110MHz?
Did the author explain the purpose of the 20pF capacitor?
No, he didn't, but it is to resonate the coil and resistors @ 110 mhz
How much DC anode voltage do you use? Does this amplifier/suppressor perform
>>satisfactorily on 10m?
>>
>> I use 6000 volts on the anode. It was the first suppressor that I found to
be successful on 10m in the late 70's @ 1500 watts without using a big
Glowbar resistor as R in a suppressor.
(((73)))
Phil, K5PC
Ian White, G3SEK
R. L. Measures wrote:
>12 ohms seems pretty high. There must be lots of ESR in the AL-1200's
>filter capacitors. With 0.7a of ripple current, you could be dissipating
>plenty of heat in the electrolytic filter capacitors. Which capacitor
>manufacturer made them?
>
In a stack of eight capacitors, that's an ESR of 1.5R each - for
300-400V units that looks pretty normal.
>> Adding a ten ohm resistor would lower fault current from 291 amperes to
>> 159 amperes IF the resistor did not arc internally. Big deal.
>
>It is a big deal to us because we use 200a-maximum glitch protection diodes.
>
Are you seriously recommending diodes with a surge rating of only 200A
to handle a surge of up to 159A? Those diodes are safety-critical
components - not just equipment safety but human safety too. For the
little extra cost of fitting a higher-rated component, I would never
trust a surge rating where safety is involved.
As a basic safety precaution I would limit the maximum possible surge
current to a much lower value (such as 40A - see below) and choose the
protection diodes on the basis of their CONTINUOUS current rating. Then
I can be sure that those safety diodes will hang in there until a fuse
or breaker blows.
>> Adding a correct style 35 ohm resistor would lower fault current to 75
>> amperes. That's a good improvement.
>
>True enough. But how low do we need to go? Svetlana recommends 10 ohms to
>25 ohms for a 1500w tube.
Eimac recommends 50R for a 2kV supply, so that even the surge current
cannot possibly exceed 40A. I'll go with Eimac.
When it comes to safety with HV supplies, there's no such thing as
overkill.
Roy Lewallen
In article <n7ws.254...@azstarnet.com>,
n7...@azstarnet.com (Wes Stewart) wrote:
>Hold on there counselor. You are putting words in my mouth. I did not
state
>that suppressors do not operate in a parallel mode and that Rp is not
>significant. Clearly, the way a suppressor is configured, it is a
parallel
>network. Equally clear is that it operates in series with the remainder
of
>the circuitry.
>
>Every parallel network has an equivalent series one (an vice versa), which
>when measured independently is indistinguishable from the other. This
does
>not mean that the equivalent networks behave the same when imbedded in
another
>network.
As written, this could cause people to reach wrong conclusions.
Every parallel network has an equivalent one (and vice versa) which is
indistinguishable ONLY AT A SINGLE FREQUENCY. This qualification is very
important. If you have, for example, a parallel network, you get a
different series equivalent circuit at each frequency. So you can't just
replace one with the other, then do your analysis at a different frequency.
For analysis at other than a single frequency, you have to use the correct
network, not an "equivalent", or else find and use a different "equivalent"
at each frequency.
The equivalent networks DO behave the same when embedded in another network
-- but again, only at a single frequency.
Roy Lewallen, W7EL
R. L. Measures
In article <5d8s7o$sm6$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) wrote:
> In message <measures-ya0240800...@news.vcnet.com> -
> meas...@mail.vcnet.com (R. L. Measures ) writes:
> :>
> :>Apparently you are unfamiliar with the definition of the word
> :>'libel'
>
> As a member of the Fifth Estate, I am most certainly familiar with
> the concept of libel and can assure you that publication (including
> posting on the internet) of charges such as you make may very well
> be considered libel. I'm suprised that Messers Erhorn, Henry,
> Collins, et al. haven't considered legal action. Since none of them
> are "public figures" in the legal sense, they do not have the burden
> of proving malice ...
>
Apparently, being a member of Fifth Estate will do little to improve one's
legal acumen.
------
According to American Heritage Dictionary: "name-drop"
To mention casually the names of illustrious or famous people in order to
imply that one is on familiar terms with them, intended as a means of
self-promotion.
------
I published a photograph of a burned bandswitch that was removed from an
ETO amplifier. Are you suggesting that---like the "fake" photo of the
Bruno Magli shoes---this photo was also fake? Would you like to
speculate on whether or not this bandswitch would be introduced as evidence
in the event that Mr. Erhorn sued me for libel?
Regarding the formula on page 13.13 in the *Radio Amateur's Handbook*, do
you believe that the maximum RF current in an 8877 amplifier's DC blocking
capacitor is 46 amperes rms when the tube is operating at 4kV and 1a?
>
> :> The only thing that was altered after bandswitch #3 was the VHF-Rp
> :>of the VHF suppressor in the AL-80.
>
> Are you absolutely certain that every other portion of the system was
> unchanged? Are you certain that a different jumper cable or antenna
> switch was not used? Are you certain that rewiring the anode circuit
> didn't result in some minor (but significant) change to a TR relay?
>
I must confess that I did not fly to Colorado and check these things for
myself, Mr. Subich. I'm not 'absolutely certain' that Rick May even exists
because I have never met him face to face. Hmmmmm. Are you absolutely
certain that Rich Measures is writing this, Joe? If Rick May said that
nothing else was changed, this was probably the case---even though he is
married to my sister.......................not. Rick's experience with an
AL-80 is not unique. Sure AL-80s don't eat as many bandswitches as
TL-922s, but they do eat them on occasion.
During my 1990 phone conversation with Mr. Rauch, he insisted that AL-80
bandswitch failures were attributed to bad operators, and/or cheap coax,
and/or bad antennas. I don't think so.
Mr. Rauch told us that Mr. Miklos was an "R+D engineering manager".
Eimac's personnel department says otherwise. Mr. Rauch said that Mr.
Miklos would appear on this Newsgroup in January.
>
> :>The original suppressor has an Rp of about 166 ohms at the anode
> :>resonant frequency of160MHz. The replacement suppressor has an
> :>Rp that is around 101 ohms at 160MHz. Since voltage gain is
> :>essentially Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain.
>
> *SO WHAT!* Even Wes Stewart, who made the measurements you are so
> fond of quoting now states that the supressors do not operate in the
> parallel mode (Rp is not significant) ...snip...
I believe that everything in the VHF-resonant anode circuit's signal path
makes some degree of difference, but some things more than others.. Except
within the city limits of Conyers, GA, U.S.A., roughly half of the VHF
current flows through Rs and half of the VHF current flows through Ls.
IMO, that is not something that should be dismissed out of hand, even by a
self-recognized amplifier expert. Those who say that the suppressor does
not count probably should not be using a suppressor.in the first place.
The next time you talk to an AL-1500 owner, ask him/her if an 8877 failed
during the 12 month warranty period. To the best of my knowledge, the
AL-1500 is the only 8877 amplifier that does not use a VHF
parasitic-oscillation suppressor.
> N7WS further states on his web page:
>
> "Some mathematical manipulation can demonstrate that shunting
> the conventional suppressor with additional resistance will
> bring the curves into better alignment. An additional 250 ohm
> resistor in parallel with the network will cause the Rp curves
> to nearly overlap the nichrome curves above about 30 MHz. Below
> 30 MHz, the Rp of the adjusted conventional type will be higher
> than the Rp of the nichrome device."
>
> Perhaps you can explain why adding a 250 Ohm resistor in parallel
> with a conventional supressor would eliminate VHF parasitics?
>
100 ohms in parallel with 250 ohms = 71 ohms. Decreasing Rs from 100 ohms
to 71 ohms reduces the parallel-equivalent resistance (Rp) at VHF. Why not
try using a 1 ohm Rs and see how the parallel-equivalent calculation for Rp
comes out at 100MHz, Joe? Post the results, and we'll all have a look.
For a refresher course in parallel-equivalency, you might try the
VHF-suppressor article in the March, 1989 issue of QST magazine.
In our low VHF Q parasitic-oscillation suppressor for 3-500Zs, we use two
Matsushita 100 ohm MOF resistors in parallel, so Rs is 50 ohms, not 100
ohms. The low VHF Q suppressor that Wes, N7WS, measured had one Matsushita
100 ohm MOF resistor. Presumably Wes, used 100 ohms instead of 50 ohms to
preclude a stenorian cries of FOUL from the Rauchian encampment.
Using the same value of Rp, 100 ohms, at 100MHz, Mr. Rauch's suppressor
produced over 60% more Rp than did the Ni-Cr-Fe alloy suppressor. Since
VHF voltage amplification is basically Ae=Rp*Mu, the question is: does one
want more VHF Ae or less VHF Ae in one's amplifier?
Does it make sense to use silver or copper conductors in the design of an
electrical circuit that supposed to dampen regeneration?
Wes Stewart
In article <5dambu$lon$3...@nadine.teleport.com> w7...@teleport.com (Roy Lewallen) writes:
>From: w7...@teleport.com (Roy Lewallen)
>Subject: Re: good source for amplifier tips
[SNIP]
>As written, this could cause people to reach wrong conclusions.
Yes it could.
>Every parallel network has an equivalent one (and vice versa) which is
>indistinguishable ONLY AT A SINGLE FREQUENCY. This qualification is very
>important. If you have, for example, a parallel network, you get a
>different series equivalent circuit at each frequency. So you can't just
>replace one with the other, then do your analysis at a different frequency.
>For analysis at other than a single frequency, you have to use the correct
>network, not an "equivalent", or else find and use a different "equivalent"
>at each frequency.
True. I should have so stated. It gets dangerous when one mixes
constant reactance values with actual component values. If fixed reactance
values are used for all calculations, then the frequency doesn't matter. When
we use real component values, the frequency certainly makes a difference.
>The equivalent networks DO behave the same when embedded in another network
>-- but again, only at a single frequency.
Correct again. This is where I went back to real components.
I've got to stop writing this stuff after midnight. As you first noted, this
is a real tar baby.
>Roy Lewallen, W7EL
N7WS
R. L. Measures
In article <19970205000...@ladder01.news.aol.com>,
comm...@aol.com wrote:
> > IMO, any electronic engineer who rejects AC Circuit-Analysis, and
> instead
> > solves such problems by intuition, quite probably does not have both
> oars
> > in the water.
>
> How's the water, Rich?
I am not the person who dismissed the solution to the 100 ohm Rs, 85nH Ls,
F=160MHz problem. I calculated the answers according to standard AC
Circuit-Analysis. These answers were summarily rejected by Mr. Rauch, who
apparently solves such problems by intuition.
> BTW, I am still waiting for one of the naysayers to provide us with the
> values of a copper-wire suppressor that equals the performance of a
> suppressor made from resistance-wire. ........ The time has come to stand
> and deliver. What will it be--deliver or quiver?
>
> Tom showed you how and why months ago. Ian White reached the
> same conclusion based on measurements you endorse...
Apparently I missed it, Peter. What are the values of Ls and Rs?
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <19970205012...@ladder01.news.aol.com>,
comm...@aol.com wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
...snip...
--
> while
> on the other hand where those measurements allow others to
> show that they compromise tank Q on ten-meters, we should assume
> those facts to be of no consequence?
>
Who are the 'others' who have shown what 4.32 ohms of R in series with 1100
ohms of anode Xc does to harmonic suppression at 30MHz?
--
--Rich-- ag6k, 805-386-3734
w8j...@aol.com
In article <5d9a28$q...@dallas1.connect.net>, phil...@connect.net (Phil
Clements) writes:
>>
> At what frequency does the arc change from blue to purple? How about
> the good old white arcs we see in the springtime?
>
>
I've looked all through my books. I can't find that either. My spark plug
wires arc blue, as does my 160 meter mobile antenna coil. Maybe my IC-706
or MSD car ignition are putting out UHF parasitics?
Or maybe this is just something someone just said, without having facts.
Was the person that started this color thing a reliable source noted for
accuracy?
73 Tom
comm...@aol.com
>Who are the 'others' who have shown what 4.32 ohms of R in series with
1100
>ohms of anode Xc does to harmonic suppression at 30MHz?
>
>
Rich, your figures are meaningless. There is no way to calculate
the reduction of the tank Q with the information you are supplying.
Haven't you been following this thread?? Didn't Tom and others just
recently post what is really involved in making those
measurements?
Perhaps your ISP provider is having problems with the newsgroups and
you are missing some important posts. Might I suggest you try using
WWW.DejaNews.Com to research earlier material
posted in this thread.
Best 73
R. L. Measures
In article <5d9b4o$q...@dallas1.connect.net>, phil...@connect.net (Phil
Clements) wrote:
> In article <19970205012...@ladder01.news.aol.com>,
comm...@aol.com says:
> >
> >In article <measures-ya0240800...@news.vcnet.com>,
> >meas...@mail.vcnet.com (R. L. Measures ) writes:
> >
> >>
> Sorry Rich, the article below was apparantly to me but I don't have
> access to it on the newsgroup! (Doesn't appear) Will answer your
> questions now.
> >>>
> >> Interesting, Phil. Was the anode-circuit resonance in this amplifier
> >also
> >>110MHz?
>
> Yes, it has been a well established fact for the 4-1000A...
>
> Did the author explain the purpose of the 20pF capacitor?
The anode-resonance is somewhat dependant on the amount of external L that
is connected to the anode. Since the anode-C is around 9pF for a
4-1000A/8166, 220nH would create a resonance at 110MHz.
> No, he didn't, but it is to resonate the coil and resistors @ 110 mhz
>
True enough if Ls is 110nH. However, a 110nH Ls is going to produce a
substantial voltage drop at 29MHz due the anode-C of 9pF---which, at 29MHz,
has about 600 ohms of reactance. At 6kVDC, the AC anode voltage would be a
bit over 3600Vrms. Thus, the current through Ls/Rs would be about 6Arms at
29MHz. IMO, this would dictate that Rs be rather robust, lest it combust.
>
> How much DC anode voltage do you use? Does this amplifier/suppressor
perform
> >>satisfactorily on 10m?
> >>
> >> I use 6000 volts on the anode. It was the first suppressor that I found to
> be successful on 10m in the late 70's @ 1500 watts without using a big
> Glowbar resistor as R in a suppressor.
>
Thanks.
It's an interesting idea, Phil. I wonder why none of the amplifier
manufacturers are using the technique---especially Ameritron since Mr.
Rauch gave W4QQ high marks. On the other hand, I see that W4QQ is not on
Mr. Rauch's recognized (and self-recognized) Amplifier Experts List in the
9/94 issue of QST magazine.
-----------------------------
e-mail copy to Mr. Clements
--
--Rich-- ag6k, 805-386-3734
Bill Turner
Does anyone know WHY grids are gold-plated? If evaporation of gold is a
potential problem, why not just make them from a high-temperature metal such
as tungsten and be done with it? Obviously, there has to be some advantage,
but what?
73, Bill W7TI
w...@eskimo.com
R. L. Measures
In article <19970206041...@ladder01.news.aol.com>,
comm...@aol.com wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
>
> >Who are the 'others' who have shown what 4.32 ohms of R in series with
> 1100
> >ohms of anode Xc does to harmonic suppression at 30MHz?
> >
> >
>
> Rich, your figures are meaningless. There is no way to calculate
> the reduction of the tank Q with the information you are supplying.
> Haven't you been following this thread?? Didn't Tom and others just
> recently post what is really involved in making those
> measurements?
The figures are not mine. The figures are from Wes' measurements, from
Eimac's specs for the 3-500Z, and from worldwide-accepted (except in
Conyers, Georgia, U.S.A.) AC Circuit-Analysis calculations.
Since no one seems to be able to calculate how Wes' measured value of ESR
(4.32 ohms) for Ls/Rs would affect the harmonic output of a 30MHz signal,
and no one has apparently measured the harmonic output of an amplifier that
uses a Ni-Cr-Fe alloy VHF suppressor, how are we to know if the sky is
really falling, Peter?
> Perhaps your ISP provider is having problems with the newsgroups and
> you are missing some important posts. Might I suggest you try using
> WWW.DejaNews.Com to research earlier material
> posted in this thread.
>
Thanks for the tip, Peter.
On the vcnet news server, I am currently seeing 11 posts by Mr. Rauch
between 1/29 and the present. I currently see 7 posts by yourself in
February. How does this compare with what you see on your news server?.
---------------------------
e-mail copy to Mr. Bertini.
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <32fbc058...@eskinews.eskimo.com>, w...@eskimo.com (Bill
Turner) wrote:
Gold has low emissivity. Thus, when the grid heats up from the flow of
normal grid current, primary electron-emission from the grid is minimal.
Also, gold conveniently has a high boiling point---and, as an added bonus,
gold helps to reduce the ESR of the grid. ESR becomes progressively more
important as the frequency increases. The 8877's anode-to-grid C is 10pF.
At 250MHz, 10pF has a reactance of 65 ohms. When operated from a 4000v
anode supply, an 8877 has about 2700Vrms at the anode. Thus, I-grid
(rms-ac)=2700V/65 ohms = 41.5A-rms, or about 0.43A-rms for each of the 96
grid wires. Due to skin-effect, 0.43A-rms @ 250MHz is a considerable
amount of current for a wire whose circumferance is only about 1.1mm.
Thus, if the 8877's grid wires were not gold-plated, serious grid heating
would probably result at such frequencies.
RE: Gold-evaporation: In the last batch of kaput 8877s I received from
Lon Cottingham, K5JV, two of the three 8877s contained condensed gold
particles/'melt-balls'.
However, I do not believe that gold-evaporation is likely in HF amplifiers
if the VHF-Rp (parallel-equivalent R) of the VHF-resonant anode circuit is
made as low as is practicable. To explain: since voltage amplification
(Ae) is basically Mu*Rp, if the VHF-Rp is relatively low, VHF-Ae is also
relatively low---which should reduce the likelyhood of VHF-regeneration.
However, none of the foregoing applies in one area of the State of Georgia.
Placing the tune capacitor close to the anode in order to increase the
anode resonant frequency also contributes to VHF stability. Fortunately,
this applies throughout Georgia.
For more information, try Dick Erhorn's Chapter 13 in the 1995--1997
Editions of the *Radio Amateur's Handbook*. If you don't have a copy yet,
the soft-cover version is available from the ARRL for $44, inc. UPS.
ground-delivery
------------------------------
e-mail copy to Mr. Turner
--
--Rich-- ag6k, 805-386-3734
Joe Subich
In message <measures-ya0240800...@news.vcnet.com> -
meas...@mail.vcnet.com (R. L. Measures ) writes:
:>
:>According to American Heritage Dictionary: "name-drop"
:> To mention casually the names of illustrious or famous people in order :>to imply that one is on familiar terms with them, intended as a means of
:>self-promotion.
:>------
"Name-drop" is not the issue here ...
Webster's New World Dictionary of the American Language: libel (n)
any written or printed statement ... tending to expose a person to
public ridicule or contempt or to injure his reputation in any way.
You have regularly implied that various commercial amplifier builders
are ripping off amateurs with inferior product.
Your claims do not hold up when examined by a network analyzer ...
--
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>I am not the person who dismissed the solution to the 100 ohm Rs, 85nH
Ls,
>F=160MHz problem. I calculated the answers according to standard AC
>Circuit-Analysis. These answers were summarily rejected by Mr. Rauch,
who
>apparently solves such problems by intuition.
It would be nice if you showed fault in my analysis of your claim a low Rp
results in more stability, instead of launching into another lie.
Rich, you plainly claimed this:
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
> In other words, the arcing that destroyed the original bandswitch and
the
>replacement bandswitch in Rick May's AL-80 can not be explained by a load
>fault. The only thing that was altered after bandswitch #3 was the
VHF-Rp
>of the VHF suppressor in the AL-80. The original suppressor has an Rp of
>about 166 ohms at the anode resonant frequency of160MHz. The replacement
>suppressor has an Rp that is around 101 ohms at 160MHz. Since voltage
gain
>is essentially Mu*Rp, decreasing VHF-Rp decreases VHF voltage gain. //
You are plainly saying a lower Rp suppressor reduces gain, and citing that
as an example why your nichrome works miracles.
A month ago it was Q, you even advertise it very plainly as a low Q cure.
Now you change your fire-chant from muttering low-Q low-Q to muttering Rp
Rp.
If a lower Rp suppressor results in more stability, the ultimate low Rp in
a suppressor would occur with a short. The highest Rp and most unstable
suppressor would be an open.
According to your theory, if we connect a 5000 ohm resistor in series with
the anode it would make the PA less stable than with a zero ohms series
resistor. According to your theory, removing all suppressors from all
equipment, and replacing them with shorts, will result in maximum
stability.
Rich, do you wish to correct your statement? Or does it stand? I'll bet
some follower's faces are redder than a glowing 3-500Z anode. Can anyone
with real world PA experience support Rich's theory? If we replace a
series resistance with a dead short at the tube's anode connection point,
does stabilty improve as Richard claims?
In article <measures-ya0240800...@news.vcnet.com>,
meas...@mail.vcnet.com (R. L. Measures ) writes:
>Who are the 'others' who have shown what 4.32 ohms of R in series with
1100
>ohms of anode Xc does to harmonic suppression at 30MHz?
Measures law of simplification of circuits:
A complex circuit can be understood by picking two random parameters that
give the desired answer and ignoring all others
73 Tom
w8j...@aol.com
In article <measures-ya0240800...@news.vcnet.com>,
>In article <32fbc058...@eskinews.eskimo.com>, w...@eskimo.com (Bill
>Turner) wrote:
>
> Does anyone know WHY grids are gold-plated? If evaporation of gold is a
> potential problem, why not just make them from a high-temperature metal
>such
> as tungsten and be done with it? Obviously, there has to be some
>advantage,
> but what?
The grids are made from tungsten or other high temperature materials Bill.
The gold is necessary to prevent poisoning of the grid by cathode
materials as they migrate from the cathode.
The gold is generally used only in a metal oxide cathode tube. In a
conventional directly heated tube, the grid is almost always not plated.
When the grid is not plated, grid dissipation is MUCH higher. That's
because the grid can be run much hotter without the very soft low melting
point gold getiing hot and contaminating the tube. If the gold gets hot,
or is hit hard by electrons, it evaporates (it can do that without even
"melting") and deposits on the cathode and other places where it ruins the
tube.
Conductivity has nothing at all to do with it, the gold is put there
simply because without gold the grid of a MOX cathode tube would quickly
be ruined by contamination.
Gold is a necessary but troublesome evil.
73 Tom
R. L. Measures
In article <19970207010...@ladder01.news.aol.com>, w8j...@aol.com
wrote:
> In article <measures-ya0240800...@news.vcnet.com>,
> meas...@mail.vcnet.com (R. L. Measures ) writes:
...snip...
> It would be nice if you showed fault in my analysis of your claim a low Rp
> results in more stability, instead of launching into another lie.
>
If your 'amplifier expert' analysis determined that the voltage
amplification of a triode is basically NOT equal to Mu*R, this discussion
has reached impasse.
----------------------------
e-mail copy to Mr. Rauch
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <5ddtkm$23lk$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) wrote:
> In message <measures-ya0240800...@news.vcnet.com> -
> meas...@mail.vcnet.com (R. L. Measures ) writes:
> :>
> :>------
> :>According to American Heritage Dictionary: "name-drop"
> :> To mention casually the names of illustrious or famous people in order
:>to imply that one is on familiar terms with them, intended as a means of
> :>self-promotion.
> :>------
>
> "Name-drop" is not the issue here ...
True enough, Joe, but some of the paticipants in this discussion used this
ploy.
> Webster's New World Dictionary of the American Language: libel (n)
> any written or printed statement ... tending to expose a person to
> public ridicule or contempt or to injure his reputation in any way.
>
Bzzzzzzzzzzt. .........
My copy of the Webster's New World Dictionary says: LIBEL n. 1. Any
false and malicious written or printed statement....... The key
ingredient in the definition of the word LIBEL is the word FALSE. It
appears that you adjusted the defination, Mr. Subich. ...........BTW, is
that a bullet hole in the toe of your left shoe?
For example: The 8877 has a maximum filament voltage rating of 5.25v. If
I state that a certain 8877 amplifier manufacturer applies 5.90v to the
filament of the 8877 when the amplifier is operated from 240.0v mains---and
the actual filament voltage is 5.0 volts, then the statement is libelous.
However, if the filament voltage measures 5.90v, the statement that the
filament voltage is 5.90v is not libelous---even though the manufacturer's
reputation was damaged by the disclosure of such information.
Let's say that YOU bought such an amplifier, Joe. When you measured the
filament voltage, it was 5.90v. However, when you telephoned the
manufacturer and voiced concern, you were told that 5.90v was perfectly
harmless to the 8877, and that it might be best to leave the engineering to
recognized amplifier experts.
...snip...
Gary Coffman
In article w...@eskimo.com (Bill Turner) writes:
>
>Does anyone know WHY grids are gold-plated? If evaporation of gold is a
>potential problem, why not just make them from a high-temperature metal such
>as tungsten and be done with it? Obviously, there has to be some advantage,
>but what?
That's a good question, Bill. The underlayment for the grid is a tungsten
alloy, and that's a pretty poor conductor for a metal, so skin resistance
would be rather high if it were bare. One would think that silver plate
would be used, or even a copper plate, but they may be too reactive with
the getter (and other things). Gold is fairly corrosion resistant as well
as a fairly good conductor. It also plates easily. Besides, you know you
can always charge more for gold plated hardware. :-)
But the main reason appears to be to reduce primary grid emission. In
a tube with an oxide cathode, barium evaporates from the cathode over
time, and can bond to the grid. This increases grid emission due to the
low work function of barium (that's why it is used on the cathode, to
increase emission). Barium doesn't bond readily with gold, which is
primarily why gold is used to plate the grid in tubes with an oxide
cathode.
Gary
--
Gary Coffman KE4ZV | You make it, | Due to provider problems
Destructive Testing Systems | we break it. | with previous uucp addresses
534 Shannon Way | Guaranteed! | Email to ke...@radio.org
Lawrenceville, GA 30244 | |
R. L. Measures
In article <5ddtkm$23lk$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) wrote:
> In message <measures-ya0240800...@news.vcnet.com> -
> meas...@mail.vcnet.com (R. L. Measures ) writes:
> :>
> :>------
> :>According to American Heritage Dictionary: "name-drop"
> :> To mention casually the names of illustrious or famous people in order
:>to imply that one is on familiar terms with them, intended as a means of
> :>self-promotion.
> :>------
>
> "Name-drop" is not the issue here ...
True enough, Joe, but some of the paticipants in this discussion used this
ploy.
> Webster's New World Dictionary of the American Language: libel (n)
> any written or printed statement ... tending to expose a person to
> public ridicule or contempt or to injure his reputation in any way.
>
Bzzzzzzzzzzt. .........
My copy of the Webster's New World Dictionary says: LIBEL n. 1. Any
false and malicious written or printed statement....... The key
ingredient in the definition of LIBEL is the word FALSE. It appears that
you adjusted the definition, Mr. Subich. ...........BTW, is that a bullet
hole in the toe of your left shoe?
For example: The 8877 has a maximum filament voltage rating of 5.25v. If
I state that a certain 8877 amplifier manufacturer applies 5.90v to the
filament of the 8877 when the amplifier is operated from 240.0v mains---and
the actual filament voltage is 5.0 volts, then the statement is libelous.
However, if the filament voltage measures 5.90v, the statement that the
filament voltage is 5.90v is not libelous---even though the manufacturer's
reputation was damaged by the disclosure of such information.
Let's say that YOU bought such an amplifier, Joe. When you measured the
filament voltage, it was 5.90v. However, when you telephoned the
manufacturer and voiced concern, you were told that 5.90v was perfectly
harmless to the 8877, and that it might be best to leave the engineering to
recognized amplifier experts.
...snip...
R. L. Measures
In article <19970208011...@ladder01.news.aol.com>,
comm...@aol.com wrote:
snip...
> > N7WS further states on his web page:
> >
> > "Some mathematical manipulation can demonstrate that shunting
> > the conventional suppressor with additional resistance will
> > bring the curves into better alignment. An additional 250 ohm
> > resistor in parallel with the network will cause the Rp curves
> > to nearly overlap the nichrome curves above about 30 MHz. Below
> > 30 MHz, the Rp of the adjusted conventional type will be higher
> > than the Rp of the nichrome device."
>
>
>
I don't know, Peter.
----------------------------
e-mail copy to Mr. Bertini
--
--Rich-- ag6k, 805-386-3734
R. L. Measures
In article <5dgp49$2k2i$1...@news-s01.ny.us.ibm.net>, sub...@ibm.net (Joe
Subich) wrote:
...snip...
So according to Joe Subich, OJ was libeled by the 30 photographs of him
wearing Bruno Magli shoes, and he was libeled by the photo of him wearing
Bruno Magli shoes that appeared in a football newsletter some months prior
to the N. Brown and R. Goldman murders.
What's wrong with this picture?
------------------------------------