[TMC] This darn amp

Wed Oct 20 17:31:20 EDT 2021

Hi Gary!

Still haven't made it to the shop yet today! :-( Answers in Italics

 Ok, what band are you doing the test on?
*I'm testing inside the 40M band.*

An open screen bypass cap would probably cause the amp to go into
oscillation.
*That is a big concern, but I am not seeing any oscillation. I was/am
hoping that greatly reduced output would be the symptom :-) Waveforms look
clean, but I only have a 100Mhz scope so something could be happening in
the stratosphere that I cannot see I guess. I have only verified one tube
socket bypass cap is open. Still haven't checked the other one yet. Going
to do that as soon as I can get out there and do it.*

If the screen current goes negative then the screen voltage will rise
provided there is a high resistance in the screen supply.
The screen meter being at 1/3 scale is because they have a resistor in the
meter circuit that makes the meter draw some current, to where they
indicate zero current, so that the meter can show positive or negative
screen current.
Y*es, I understand that, and from what I can tell the circuit is working
correctly even though I never see a change in screen current at least
nothing that I can be sure is a change. That is very puzzling.*

A quick indication of good tubes is to drive them rather hard and turn the
plate tuning off resonance to see how high the plate current goes.
When the amp is properly loaded you will not get much plate current dip but
you should see a peak in screen current. If you are getting a large plate
current dip then it is not loaded properly
*Hmmmm.....I have to process this. Can you relate the first sentence to the
second and third? I am at the limit of my understanding and I am getting
confused. Plate current dips nicely with tune, output peaks nicely with
load. The amp is driving a 50 ohm resistive load. I can repeat this process
while increasing the drive until I draw so much plate current that the HV
breaker trips. Nowhere in this process do I see screen current ever change.*

*This has me stumped which is why I am reaching out. And why I very much
appreciate the help. It's a tetrode amp with negative feedback and multiple
drive stages. It's all new to me. Although it's "just an amp" there is a
lot more to it than triode amps I have tinkered with thus far.*

*I will report back after checking the bypass cap in the second final tube
socket.*

*Thanks again!*

*Chris*

On Wed, Oct 20, 2021 at 2:02 PM Gary Schafer <garyschafer at largeriver.net>
wrote:

> Ok, what band are you doing the test on?
>
> An open screen bypass cap would probably cause the amp to go into
> oscillation.
>
> If the screen current goes negative then the screen voltage will rise
> provided there is a high resistance in the screen supply.
> The screen meter being at 1/3 scale is because they have a resistor in the
> meter circuit that makes the meter draw some current, to where they
> indicate zero current, so that the meter can show positive or negative
> screen current.
>
> A quick indication of good tubes is to drive them rather hard and turn the
> plate tuning off resonance to see how high the plate current goes.
> When the amp is properly loaded you will not get much plate current dip
> but you should see a peak in screen current. If you are getting a large
> plate current dip then it is not loaded properly
>
> 73
> Gary  K4FMX
>
> ------------------------------
> *From:* Chris Bolkan [mailto:bolkyboats at gmail.com]
> *Sent:* Wednesday, October 20, 2021 3:04 PM
> *To:* Gary Schafer
> *Cc:* tmc collector's group
> *Subject:* Re: [TMC] This darn amp
>
> Hi Gary!
>
> Thanks for the quick reply. Answers in Italics. I still haven't been out
> to the garage today so don't know any more than the last reply but.......
>
> if you had a shorted screen bypass capacitor then you wouldn't have proper
> screen voltage when measuring at the tube socket.
> *Agree. The cap is NOT shorted. It is open or appears so. It only measures
> 4pf, not 1000pf. I believe the built-in cap in one tube socket is defective
> (open). Will check the other socket shortly.*
>
> If there is a high resistance in the screen voltage circuit the screen
> voltage will fall as the screens draw current.
> *Good to know! I did not know if the screen voltage would rise or fall. I
> assumed it could do either since screen current could go either positive or
> negative on the meter. I am going to check to see if there is high DC
> resistance in the screen supply circuit as you suggested earlier. At this
> point I know the AC impedance of the screens to ground is high because of
> the failed bypass cap built into the tube socket. Do you know the effect of
> a high AC impedance on the screen grid, i.e. failed bypass cap? I do not.*
>
> It is sounding like you don't have enough drive to the amp. There are two
> input jacks, one is 100 mill watts and the other is 1 watt. If you put too
> much drive power in you will burn out the load resistor(s) at the input.
> *I seem to get adequate drive, if by that you mean adequate drive voltage
> displayed on the "drive" meter setting. Drive can EASILY be set to pull so
> much plate current that the HV breaker trips off. Problem is the amp is
> only putting out about 70 watts when it is drawing 300ma of plate current.*
>
> A few, 5 to 10 ma of NEGATIVE screen current is normal when the amp is
> fully tuned up.
> *No matter what I do adjusting the amplifier I do not see any change in
> screen current in the "screen current" meter position. This is very
> frustrating as I have to the best of my ability determined that the meter
> is actually measuring screen current. It's kind of a funky circuit that
> measures screen current where a resistor divider has zero screen current
> displayed at about 1/3 scale on the meter. it all seems to be working
> correctly, but I never see a change in screen current on the meter. Hope
> that description was not too confusing.*
>
> Check your driver meter readings as the manual shows.
> *Drive is monitored per manual. I have noticed this....when the amp is
> outputting power, the drive reading will be say 20-30 on the drive meter.
> When I unkey the amp (bias the finals off or kill the HV) the drive as
> reading as indicated on the meter rises significantly*
>
> Try the SSB and CW positions of the bias switch.
> *There is no SSB/CW switch. The amp I have is a PAL350A.*
>
> Thanks again for the questions and suggestions. This is frustrating but I
> am learning a lot, and in the end that is what it is all about. I
> appreciate everyone helping me through this.
>
> best regards,
>
> Chris
> KI7NXY
>
> On Wed, Oct 20, 2021 at 12:29 PM Gary Schafer <garyschafer at largeriver.net>
> wrote:
>
>> If you had a shorted screen bypass capacitor then you wouldn't have
>> proper screen voltage when measuring at the tube socket.
>>
>> If there is a high resistance in the screen voltage circuit the screen
>> voltage will fall as the screens draw current.
>>
>> It is sounding like you don't have enough drive to the amp. There are two
>> input jacks, one is 100 mill watts and the other is 1 watt. If you put too
>> much drive power in you will burn out the load resistor(s) at the input.
>>
>> A few, 5 to 10 ma of NEGATIVE screen current is normal when the amp is
>> fully tuned up.
>>
>> Check your driver meter readings as the manual shows.
>> Try the SSB and CW positions of the bias switch.
>>
>> 73
>> Gary  K4FMX
>>
>> ------------------------------
>> *From:* Chris Bolkan [mailto:bolkyboats at gmail.com]
>> *Sent:* Wednesday, October 20, 2021 10:13 AM
>> *To:* garyschafer at largeriver.net
>> *Cc:* tmc collector's group
>> *Subject:* Re: [TMC] This darn amp
>>
>> Hi Gary!
>>
>> Great questions. I'll do my best to answer.
>>
>> 1) Check that your screen voltage at the socket is holding steady as you
>> drive it. Could be a high resistance in the supply circuit.
>> *I have not checked this. I did find a power supply wiring error which I
>> corrected and that brought the screen voltage up to spec. I then measured
>> proper screen voltage at the socket. I have not however verified that the
>> screen supply is stiff at the socket. I will apply some resistive loading
>> and verify that the voltage remains the same. *
>> *I have begun checking the capacitor on the screen pin in the sockets. I
>> have only started this effort and found right before bed last night that
>> one of the sockets (haven't checked the other yet) has only 4pf of
>> capacitance. The pin is not leaky to ground. It's like the cap in the
>> socket just opened up. I do not know what the effect of this bypass cap
>> being open can be. If it is degenerative feedback this might be my problem
>> but I have to check the other socket first and learn what the effect of a
>> high impedance screen to ground is.*
>>
>> 2) What is your plate idle and screen idle current?
>> *Plate idle current is set to 90ma on V203 and then to 160ma with V204.
>> Screen current reads 0*
>>
>> 3) Also check the band switch, may be the TANK is way off frequency.
>> *The tank tunes nicely dipping plate current with dune and peaking output
>> with load*
>>
>> 4) Do you get any dip in plate current when you turn the plate tune cap?
>> *Yessiree a nice dip*
>>
>> 5) Shorted output padder cap?
>> *What is a padder cap?*
>>
>> 6) Do you get a peak in output power when tuning the plate tune cap?
>> *Yes, I think so but I can't positively remember. I have been looking at
>> the plate current when dipping and the scope in parallel with the dummy
>> load when loading for peak. I **dip the plate current with the tune and **load
>> for peak output and dip the plate and load for peak etc while bringing up
>> the drive. I can't actually say if the output was peaking at the dip. I
>> should know but I don't. Sorry.*
>>
>> Obviously I need to sort out the tube socket screen bypass cap thing, and
>> I can easily check for screen supply stiffness at the screen pin on the
>> socket. Can you tell me the effect of high impedance on the screens?  There
>> are inductors in series with the screens so without the bypass capacitance
>> the screen(s) would be high impedance.
>>
>> Thank you for the great questions Gary. I appreciate everyone's help.
>>
>> Chris
>> KI7NXY
>>
>> On Tue, Oct 19, 2021 at 11:04 AM Gary Schafer <garyschafer at largeriver.net>
>> wrote:
>>
>>> Check that your screen voltage at the socket is holding steady as you
>>> drive it. Could be a high resistance in the supply circuit.
>>>
>>> What is your plate idle and screen idle current?
>>>
>>> Also check the band switch, may be the TANK is way off frequency.
>>> Do you get any dip in plate current when you turn the plate tune cap?
>>>
>>> Shorted output padder cap?
>>> Do you get a peak in output power when tuning the plate tune cap?
>>>
>>> 73
>>> Gary. K4FMX
>>>
>>>
>>> Sent from my iPhone
>>>
>>> > On Oct 19, 2021, at 10:31 AM, Chris Bolkan <bolkyboats at gmail.com>
>>> wrote:
>>> >
>>> > Hi TMC lovers!
>>> >
>>> >
>>> > Got the power supply corrected. Now puts out a nice solid 360V to the
>>> final
>>> > screens and all other voltages are correct.
>>> >
>>> > The good: All voltages correct, finals now bias nicely just as I think
>>> they
>>> > should and it tunes up nicely.
>>> >
>>> > The bad: At 240ma plate current, single tone with a drive of 20, it
>>> puts
>>> > out 120V p-p into a 50 ohm load. That calculates out to 35 watts RMS
>>> and is
>>> > almost exactly what the inline wattmeter shows. Even when I pump it up
>>> to
>>> > over 300ma it only puts out just 50-70 watts. I don't like doing that
>>> for
>>> > long as occasionally the breaker trips because I have it set
>>> conservatively
>>> > until I figure out what is wrong.
>>> >
>>> > So correcting the screen voltage and making it all "right" still did
>>> not
>>> > change the output power. That's what it was before even with what I
>>> > consider a profound change.
>>> >
>>> > This is also puzzling: There is no change in screen current (on the
>>> meter)
>>> > no matter what I am doing tuning this thing up. The meter reads very
>>> close
>>> > to 0 (about a third of the way up the scale as it should, but nothing
>>> I do
>>> > changes screen current. I verified the proper screen voltage on the
>>> final
>>> > tube socket pins so I know the sockets are not shorted. Since the
>>> screens
>>> > get their voltage off the measurement divider, if the current is
>>> changing
>>> > it should register.
>>> >
>>> > I still have not checked the capacitance on the screen pins of the
>>> sockets
>>> > yet. If those were somehow low would that have a degenerative feedback
>>> > effect? Even so it seems highly unlikely that both sockets have failed
>>> in
>>> > exactly the same way. This is so confusing.
>>> >
>>> > I've been posting a bit lately about my PAL 350. I've not seen a single
>>> > comment so I guess I don't know if the posts are even going out. Well,
>>> > here's the latest. Any ideas would be greatly appreciated!
>>> >
>>> > Chris
>>> > ______________________________________________________________
>>> > TMC mailing list
>>> > Home: http://mailman.qth.net/mailman/listinfo/tmc
>>> > Help: http://mailman.qth.net/mmfaq.htm
>>> > Post: mailto:TMC at mailman.qth.net
>>> >
>>> > This list hosted by: http://www.qsl.net
>>> > Please help support this email list: http://www.qsl.net/donate.html
>>>
>>>