5f2a bias voltage too high?

[big-daddy-59]

okay I built a 5fd2a with the following mods:
used the 330 volt taps on power transformer
changed the rectifier tube to 5AR4
changed filter caps to 20/20/10/10 from 16/16/8/8
changed the resistor (r11 on the weber schematic)to 27k from 22k
and I'm using a 6L6GC power tube

on first power up the cathode bypass cap(c5 per weber schem.)blew up,it's value was 25u25v.

cut it out, waiting on a higher voltage replacement.

fired it up again tonight for short periods to measure voltages

with all tubes out got 360VAC from PT.
with only rectifier in got the following

v3 pin 2 497VDC
pin 4 357VAC
pin 6 357VAC
pin 8 497VDC

v2 pin 3 497VDC
pin 4 476VDC
pin 6 -3.6
pin 8 0

v1 pin 1 430VDC
pin 2 0
pin 3 0
pin 6 430VDC
pin 7 0
pin 8 0

with all tubes in

v3 pin 2 376Vdc
pin 4 323 vac
pin 6 323Vac
pin 8 377Vdc

v2 pin 3 315Vdc
pin 4 288Vdc
pin 6 .1vdc
pin 8 71.5Vdc

v1 pin 1 254Vdc
pin 2 00
pin 3 00
pin 6 254Vdc
pin 7 00
pin 8 00


shortly after shutting it off after the last round of measurements I noticed a burning smell and upon inspection found that the circuit board under the 470 ohm cathode bias resistor (R8 per weber schem) was charred.

[Robert M. Martinelli]

okay I built a 5fd2a with the following mods:
used the 330 volt taps on power transformer
changed the rectifier tube to 5AR4
changed filter caps to 20/20/10/10 from 16/16/8/8
changed the resistor (r11 on the weber schematic)to 27k from 22k
and I'm using a 6L6GC power tube

on first power up the cathode bypass cap(c5 per weber schem.)blew up,it's value was 25u25v.

cut it out, waiting on a higher voltage replacement.

fired it up again tonight for short periods to measure voltages

with all tubes out got 360VAC from PT.
with only rectifier in got the following

v3 pin 2 497VDC
pin 4 357VAC
pin 6 357VAC
pin 8 497VDC

v2 pin 3 497VDC
pin 4 476VDC
pin 6 -3.6
pin 8 0

v1 pin 1 430VDC
pin 2 0
pin 3 0
pin 6 430VDC
pin 7 0
pin 8 0

with all tubes in

v3 pin 2 376Vdc
pin 4 323 vac
pin 6 323Vac
pin 8 377Vdc

v2 pin 3 315Vdc
pin 4 288Vdc
pin 6 .1vdc
pin 8 71.5Vdc

v1 pin 1 254Vdc
pin 2 00
pin 3 00
pin 6 254Vdc
pin 7 00
pin 8 00


shortly after shutting it off after the last round of measurements I noticed a burning smell and upon inspection found that the circuit board under the 470 ohm cathode bias resistor (R8 per weber schem) was charred.

Your V2 cathode voltage is way out of whack, some 71 VDC across a 470 Ohm resistor mean some 151 mAmps are flowing through it in quiescent conditions. No wonder the area close to the resistor is charred, the cathode resistor has overheated by dissipating some ( 0.151 A * 71 V ) 10,7 Watts! I'm surprised it didn't burn to ashes! Have you measured if the resistor is still within the specs? I would expect it to have drifted on the high side...

A 6V6 in a single ended configuration like yours should have, say, 13 to 15 VDC at its cathode, while a 6L6 should be in the 20 to 22 VDC range IMHO.

I strongly suspect the problems you're facing have something to do with the mods, as you can't simply alter the + B, throw in a different ( bigger ) tube even if it has the same pinout and hope for the operating conditions to be in the safe area. Mods should not be performed without doing some reckoning first IMHO.

As to the preamp tube ( V1 ), both sections are cathode biased, so the voltage at the cathodes ( pins 3 and 8 ) cannot be zero, it should be some 1.35 VDC for both according to my calculations ( if the cathodes really were at GND potential the grids ( pins 2 and 7 ) couldn't be "negative" with respect to their cathode so the grids could not control the electrons' flow and each triode would behave like a diode conducting as hard as it can ).

Another thing I'd like to point out is the impedance mismatch you introduced, as your 6L6 is probably in the 4 K output impedance range, while the " original" 6V6 was probably more in the 6 K area ( though this is not our main concern at the moment, and not a severe one, I thought it was worth mentioning ).

Hope this helps

Best regards

Bob

[tubeswell]

Yeah - by those results I'd guess that the cathode resistor was a 1 or 2Wer and it went out of whack. I agree with Bob - and I think you should be trying a something like a 5W 220R-270R cathode resistor with a 6L6CG and be getting around 20ish V, and have the OT set up for a 3k2 - 4k load resistance. Make sure you've got something like a 63V bypass cap in there (once you change the cathode resistor).

Also maybe increase the pre-amp power rail supply resistor - 250ish V for the V1 plates, whilst within spec for a 12AX7, is a bit on the high side for this type of amp. Try getting it to around 180V on V1A and see if the output tube plate voltage comes up any. JM2CW

[Robert M. Martinelli]

@ tw,
I agree about the preamp tube plate voltages,( I didn't deal with it because I focused on the output stage problems ) you're absolutely right, they're high for that kind of amp.

@ b-d-59
Cathode bias works on the principle of lifting the cathode potential from GND by means of a resistor which causes a voltage drop due to the current flowing through it. This way the control grid becomes "negative" if seen from the cathode, ( always remember the cathode is THE reference point for ALL tubes ) and this enables it to control the tube's operation.

The cathode resistor is part of a voltage divider in which the first leg is the tube, and the second leg is the resistor itself. The higher the cathode resistor, the more voltage will be dropping across it, so the cathode voltage will be higher ( and LESS voltage will be dropping across the tube - e.g. if your plate voltage is 370 V and your cathode voltage is,say, 10 VDC then 360 V will be dropping across the tube, if you increase the cathode voltage to 20VDC by increasing the cathode resistor's value, then only 350 VDC will be dropping across the tube ). This means that, all other voltages being the same, increasing the cathode resistor's value shifts the bias on the "cold" side, while decreasing it shifts the bias on the "hot side".

The cathode resistor also acts as a current limiter, because the higher the plate-to-GND resistance ( which, as we've just seen is the sum of the tube's resistance plus the cathode resistor ) the lower the current flowing from the plate to GND.

This also means the "cathode bias" method is a self-regulating one to some degree, because as the current through the tube tends to increase, so does the cathode voltage ( Ohm's law ), and, as we've seen, less voltage will be dropping across the tube, tending to reduce the current, thus shifting the operating point on the "cool" side some. This phenomenon is called "cathode sag" and it is typical of a "cathode bias" arrangement ( it does not occur with an "external" bias arrangement, either fixed or adjustable ).

As to the cathode resistor's dissipated power rating, you have to multiply the voltage drop across it by the cathode current flowing through it. To determine the cathode current ( which, it's worth remembering, is the sum of plate and screen current in power tetrodes and pentodes ) simply divide the cathode voltage by the cathode resistor value.

( e.g. with a cathode voltage of, say, 22V and a, say, 220 Ohm cathode resistor, your cathode bias current will be 22/220=0.1 Amp, so your cathode resistor will have to dissipate a minimum of 20VDC*0.1Amp=2W ).

This, BTW, is, together with the values you posted, what I used to state that your 470 Ohm 1W cathode resistor's probably gone south. ( 71 VDC/470 Ohm = 0.151 A ------> 0.151 A *71V=10,7W ------> Ouch!!!

Your amp is a class "A" one, so the current swing between quiescent and operating conditions is limited ( ideally zero ) - a safety margin of 2.5 times seems reasonable, so I'd use a 5 W cathode resistor, even though a 7W or a 10 W one are not out of question as they add some "peace of mind" to the design.

Sorry for the wordy explanation, but that's the best I can do with my rather poor English, I hope you'll forgive me.

Hope this helps

Best regards

Bob

[Wakculloch]

Have you tried a different 6l6 in it. I've had a few duff ones that go way over current.

[Robert M. Martinelli]

I've had the same problem with some 6L6s and 5881s ( Sovtek ), so I second that, but I'd also make myself sure everything is OK ( voltages, biasing network etc. ) before throwing in another tube. If something is wrong ( e.g. biasing ) the new tube is at risk of living a brief though extremely glorious life, much "a la" James Dean.

HTH

Best regards

Bob

[big-daddy-59]

is it normal for there to be a large voltage drop thru the output transformer?
I've got 376 volts for B+ but only 315V for plate voltage (measured plate to ground)? Could there be something wonky with the OT?

I tried a 240 ohm 5watt resistor on the cathode last night and while the cathode voltage dropped it was not enough. I 'm still getting plate currents of 100+ milliamps and the plate voltage to ground dropped some more.

[Robert M. Martinelli]

is it normal for there to be a large voltage drop thru the output transformer?
I've got 376 volts for B+ but only 315V for plate voltage (measured plate to ground)? Could there be something wonky with the OT?

I tried a 240 ohm 5watt resistor on the cathode last night and while the cathode voltage dropped it was not enough. I 'm still getting plate currents of 100+ milliamps and the plate voltage to ground dropped some more.

Hi again, b-d-59,
you're overloading the OT IMHO. a 60 VDC drop across it is way too high.
Remember, you are using a 6L6GC, while the transformer has been sized to deal with 6V6s, and, since you're drawing more current, the voltage drop across the OT is obviously higher. a 60+ VDC voltage drop at 100+ mAmps mean the OT primary winding alone is dissipating some 7 W ( quiescent ). Be careful as the OT could overheat and fail under these conditions.

You don't state what the cathode voltage is with a 240 Ohms resistor, but, since you state it is still low, and you are well over 100 mAmps of plate current, I'd definitely try a 330 Ohms cathode resistor, as your target is to get some 20 VDC at the cathode, with a plate current around 60-65 mAmps in quiescent conditions ( if you are taking the current reading at the cathode, you must remember some 5-8 mAmps will be coming from the screen grid, so your target cathode current will be somewhat higher, say some 70 mAmps ).

Hope this helps

Best regards

Bob

[tubeswell]

Ahhh yes I forgot about making sure that the heater winding has enough juice in it for the tubes. A 6L6 draws 0.9A from the heater, whereas a 6V6 draws only 0.45A. If your PT heater winding is only rated for 1A, then it won't do any good trying to run a 6L6 on this amp. You could run a JJ6V6s - that is somewhat like a 6L6 in tonal characteristics, and can take higher plate voltages (and dissipation) than a more typical 6V6. That is if teh heater is the problem. You might want to check your filter caps are okay by tacking a 2nd cap in parallel with each one in turn, to see if that brings the voltage up (just in case one or more of the filter caps is shot) - but you would be getting a hummier noise if they were failing

[Robert M. Martinelli]

Ahhh yes I forgot about making sure that the heater winding has enough juice in it for the tubes.

Me too!!!
Must be the age.....

Good point!

Cheers

Bob

[MWJB]

Big daddy 59 - Something is wired wrong/wrong cathode resistor value/poor ground connection at the cathode resistor? You shouldn't have 71vdc at the cathode of the power tube, especially not with a 330ohm cathode resistor. I'd expect 30-ish with a 470ohm cathode resistor. As you are using a smaller cathode resistor than 470ohms, your cathode voltage should also be less than what it would be with a 470ohm.

Robert Martinwelli Wrote: "A 6V6 in a single ended configuration like yours should have, say, 13 to 15 VDC at its cathode, while a 6L6 should be in the 20 to 22 VDC range IMHO." These estimates are WAY low with 330-0-330 VAC PT and a GZ34.

Tubeswell wrote: "If your PT heater winding is only rated for 1A, then it won't do any good trying to run a 6L6 on this amp." the stock spec for a 125P1B PT is 2A of heater current, fine for 1x6L6 & 1x12AX7. Big Daddy 59 is using the Weber brand which according to Ted's site has MORE current capacity than the Fender spec. Heater current is not an issue.

Big Daddy 59, your cathode resistor should be rated at 10W (& should be 470ohms, plate current around 60mA, no more than 70mA).

[Robert M. Martinelli]

Hi MWJB,
he said he was having some 71 VDC at the cathode with a 470 Ohm 1W resistor in the first place, and the area around the resistor was charred ( a proof that current was indeed flowing through the cathode resistor ), this is why we suspected that resistor to have gone out of whack, because 71 VDC/470 Ohms = 0,151 Amps, which multiplied by 71 VDC is 10.7 W ( a little too much to bear for a 1 W resistor IMHO ).

As to my estimates, I beg your pardon, but I think they're not far from reality, you just said that you expect 30 VDC at the cathode with a 470 Ohm resistor, so my 22 volts with a 330 Ohms resistor could be in the ballpark ( and the cathode current should be some 70 mAmps ). According to my estimate, having some 370 VDC at the plate, some 350 VDC are dropping through the tube, so, if the plate current alone is around 60-65 mAmps the plate has to dissipate more or less some 22 W which is well within the specs ( 6L6GC are rated at 30 W AFAIK ). I don't want to say I'm right, mine are just estimates, but, if you think I'm wrong, or you think I made some bad assumption, I kindly ask you to explain the rationale for that, so that I can correct myself and learn something from my mistakes .

As to the cathode resistor dissipated power, even with 30 VDC at the cathode and 70 mAmps the resistor would have to dissipate "only" 2.1 W, anyway, I too advised him to use up to a "10 Watter" for peace of mind.

Cheers

Bob

[MWJB]

"he said he was having some 71 VDC at the cathode with a 470 Ohm 1W resistor in the first place," This alone is evidence of a problem....a 470ohm should not yeild anything like 71vdc at approx 400vdc B+. There is a wiring error/duff resistor/poor ground connection...this is not normal operation for these values.

A 330-0-330VAC B+ winding with adequate B+ & heater current supply should yeild around 400vdc at the plate with a 470ohm. If you try and run many current production 6L6 at 30W/70mA plate dissipatation, you will be changing tubes between bars. 22-23W/60mA would be safer.

Yes, the cathode resistor will dissipate around 2w, double that for a minimum safe level and 5W would probably do it, but 10W will fit & last forever.

[big-daddy-59]

OK MWJB my power tranny puts out 330-0-330 100ma 6.3V 4A 5V 3A
so I should be good there. I do get 400+ volts at pin 3 on the power tube socket with the 6L6 out And I've got a few values for the cathode resistor...470R, 240R , 180R All are 5 watt. The filter caps are grounded to the power transformer mounting bolts, I think the cathode ground is tied in with the preamp grounds, I'll have to check later, don't have it in front of me right now. Could the tube/ output transformer be the cause of my troubles? The power amp is almost the same as a Kendrick Ruffneck which I believe is a proven design.

[Robert M. Martinelli]

"he said he was having some 71 VDC at the cathode with a 470 Ohm 1W resistor in the first place," This alone is evidence of a problem....a 470ohm should not yeild anything like 71vdc at approx 400vdc B+. There is a wiring error/duff resistor/poor ground connection...this is not normal operation for these values.

A 330-0-330VAC B+ winding with adequate B+ & heater current supply should yeild around 400vdc at the plate with a 470ohm. If you try and run many current production 6L6 at 30W/70mA plate dissipatation, you will be changing tubes between bars. 22-23W/60mA would be safer.

Yes, the cathode resistor will dissipate around 2w, double that for a minimum safe level and 5W would probably do it, but 10W will fit & last forever.

Hi Again MWJB,
Maybe I didn't manage to express my point of view very clearly, due to my poor English, if so I beg your pardon again , but I never said I wanted to run the tube at 70 mAmps and 30 W, I said that with some 70 mAmps of CATHODE current I expected the PLATE current to be in the 60-65 mAmps range ( the remainder being screeen current ). I based my estimate on the plate voltage b-d-59 gave us in the first place ( in the 370 VDC range ), and with that value and some 20-22 VDC at the cathode I expected some 350 VDC to drop across the tube, and the tube to dissipate 22-23 W ( which more or less matches the value you indicated as safe for "current production" tubes ).

If now the plate voltage is higher, as b-d-59 is now saying, then I agree that the cathode resistor will have to be higher to elevate the cathode voltage some, limit the current, lower the voltage dropping across the tube, bias the tube cooler and limit the tube's PD, so I' d move towards a 390 Ohms resistor, or a 470 Ohms one as per your advice, even though I find it would set the tube bias a tad low ( but safer, as you wisely noted ).

Being the amp a class A one, the current swing ( ideally zero ) will be low, but present ( I'd say in the 8-10 mAmps range at full volume ) so there will also be a little "cathode sag effect" while playing with the amp fully cranked IMHO.

As to the initial 70 VDC at the cathode, we all already agreed that was not normal, so I have nothing to add about it.... ( I still think that "charred" resistor was out of whack/gone south though, b-d-59, can you confirm that? ).

Peace

Bob

P.S. again, sorry for my wordy and prolix posts - I have to force my lazy brain to overwork, since I have to think "technical" and moreover in English ( which is not my native language, as you might have already guessed ) at the same time, and that's way too much work for my worn out neurons.