Page 1 of 1

What causes excessive power tube cathode voltage?

Posted: Sat 02/11/23 9:08 pm
by yello
I am working on a cathode-biased dual 6V6 amp for a friend (same guy as the el84 amp with the volume and tone pot interaction that need grid and screen resistors).

Most of the voltages throughout the amp are looking in a correct range to each other per the schematic, but one power tube is red plating and the cathode voltage measures 43 when it should be 19 (unless I am misunderstanding the schematic and it is supposed to be 19 per side so 38 total). Knowing if it is supposed to be 19v or 38v on schematic affects what I put into the bias calculator, which currently shows biased way too hot at 43v on cathode.

The other thing I noticed is that the plate voltages for each tube are different, one 6V6 is 316v, the other 6V6 is 302v. Screens are at 299v. Schematic shows 335v for the plates, so they are low and mismatched.

There is a large 500r 10w adjustable resistor (R40), currently set to 200r, that lowers voltage between 5y3 and B+1. I can adjust that easily to raise the voltage to the B+ rail, and think I need to do so to get the voltages before and through the B+ right.

After I adjust that resistor if necessary, and before I try a different set of tubes (and potentially damage them if they are red plating?), what should I explore to understand what might cause the red plating, even with the current low voltages, and the mismatch in plate voltages? Maybe it is the tubes, or the plates resistors that I need to test or measure?

FWIW, I went with a 330r shared bias resistor with a 50uf cap. Stock schematic shows 225r and 50uf, but I went with 330r to start, as a friend who likes these amps typically uses that value. I would think that would be helping my potential bias issue, not hurting it.

Schematic is here. I am currently testing with my variac set to 117v per the original schematic, although I will also need to test at full modern wall voltage of 122v to get it right for actual use.

http://www.acofs.org.au/part_5_files/Be ... iagram.PDF

Re: What causes excessive power tube cathode voltage?

Posted: Sun 02/12/23 10:12 am
by JMPGuitars
If the bias resistor is shared (it is), then what is the electrical difference between those two points (V5p8, V4p8)? Nothing. Unless there's a soldering issue. 19V is what the target is at both points, there's no halves, it's the same 19V.

Check your grid voltages. If those are different, then that might give you a clue.

Check your plate resistor values.

If your B+ is low, and you're still using a variac, don't adjust the B+ without switching to your actual wall voltage.

Thanks,
Josh

Re: What causes excessive power tube cathode voltage?

Posted: Sun 02/12/23 11:17 am
by yello
JMPGuitars wrote:
Sun 02/12/23 10:12 am
If the bias resistor is shared (it is), then what is the electrical difference between those two points (V5p8, V4p8)? Nothing. Unless there's a soldering issue. 19V is what the target is at both points, there's no halves, it's the same 19V.

Check your grid voltages. If those are different, then that might give you a clue.

Check your plate resistor values.

If your B+ is low, and you're still using a variac, don't adjust the B+ without switching to your actual wall voltage.

Thanks,
Josh
Ok, sounds like I am understanding the cathode correctly, should read 19v at either pin, and they both read 43 as the resistor is shared, so that numbers is too high. I had already double-checked all those soldered joints for problems, reflowed them, even removed the resistor and cap to remeasure values and they are spot on, not the problem.

I only have one 6V6 grid voltage written down, 7v. I think I was having trouble getting to the other 6V6 pin 5, or maybe the reading was off, I can't remember but can remeasure today.

Yep, not gonna mess with the B+ until I figure this out, and put it up to 122v from the wall, then figure out if adjustments are needed.

The 470k grid leak resistors and coupling caps are a single shared part, not sure what they are called. It is brown, and the parts are encased together inside, and has 5 wires coming from it to bridge the connections. Hopefully I can get a good measurement. I have a nagging thought something is off there, but don't know enough to know, thus will measure and report back!

Re: What causes excessive power tube cathode voltage?

Posted: Sun 02/12/23 11:54 am
by yello
JMPGuitars wrote:
Sun 02/12/23 10:12 am
Check your grid voltages. If those are different, then that might give you a clue.

Check your plate resistor values.
I just measured the grid voltages. One 6V6 measures -.001. The other measured at 7v, then slowly kept dropping down to 2v. Then a couple minutes later was up as high as 38v. This seems to be the problem area, and that is the tube that is red plating.

Unfortunately I can't measure the plate resistors as both .02 coupling caps and both 470k plate resistors are all together in a single part. I'm thinking at this point I need to just put in new coupling caps and plate resistors.

What does the above info I was able to measure tell me, is it likely a coupling cap or plate resistor issue? I gotta replace both either way given the stock part, but would love to learn to diagnose the issue better.

Re: What causes excessive power tube cathode voltage?

Posted: Sun 02/12/23 1:44 pm
by JMPGuitars
yello wrote:
Sun 02/12/23 11:54 am
I just measured the grid voltages. One 6V6 measures -.001. The other measured at 7v, then slowly kept dropping down to 2v. Then a couple minutes later was up as high as 38v. This seems to be the problem area, and that is the tube that is red plating.

Unfortunately I can't measure the plate resistors as both .02 coupling caps and both 470k plate resistors are all together in a single part. I'm thinking at this point I need to just put in new coupling caps and plate resistors.

What does the above info I was able to measure tell me, is it likely a coupling cap or plate resistor issue? I gotta replace both either way given the stock part, but would love to learn to diagnose the issue better.
In a cathode biased amp, that usually means the path to ground for the grid with higher voltage is problematic. In an 18W, I usually see around 10mV with decent soldering. If the voltage is higher (usually still less than 1V), that's a sign that the soldering on the path to ground is bad, assuming no component or socket issues.

Remove the tubes and measure the resistance to ground on both grids. It is a little odd that only one has higher grid voltage and both are red plating. I would inspect those sockets and all the soldering carefully.

You can measure the plate resistors if you measure at the tube socket pins, and an open point in the circuit (review schematic) before the resistor.

Thanks,
Josh

Re: What causes excessive power tube cathode voltage?

Posted: Sun 02/12/23 3:13 pm
by yello
JMPGuitars wrote:
Sun 02/12/23 1:44 pm
yello wrote:
Sun 02/12/23 11:54 am
I just measured the grid voltages. One 6V6 measures -.001. The other measured at 7v, then slowly kept dropping down to 2v. Then a couple minutes later was up as high as 38v. This seems to be the problem area, and that is the tube that is red plating.

Unfortunately I can't measure the plate resistors as both .02 coupling caps and both 470k plate resistors are all together in a single part. I'm thinking at this point I need to just put in new coupling caps and plate resistors.

What does the above info I was able to measure tell me, is it likely a coupling cap or plate resistor issue? I gotta replace both either way given the stock part, but would love to learn to diagnose the issue better.
In a cathode biased amp, that usually means the path to ground for the grid with higher voltage is problematic. In an 18W, I usually see around 10mV with decent soldering. If the voltage is higher (usually still less than 1V), that's a sign that the soldering on the path to ground is bad, assuming no component or socket issues.

Remove the tubes and measure the resistance to ground on both grids. It is a little odd that only one has higher grid voltage and both are red plating. I would inspect those sockets and all the soldering carefully.

You can measure the plate resistors if you measure at the tube socket pins, and an open point in the circuit (review schematic) before the resistor.

Thanks,
Josh
I might have had a terminology problem on my end. I was saying plate resistors but meant the 470k grid leak resistors going from power tube grids to ground. I measured them with tubes in (amp off) and one side reads open, the other side reads 892k. Same measurement with or without tubes. Both 6v6 pin 5s have no continuity to ground, and have proper grounding on the grounded leg of the 470k resistors.

Only one tube was red plating, and it is the socket with the high voltage and the open resistor.

Plate resistor terminology here I imagine is meant to be the phase inverter plate resistors, those measure within spec at 3.8k and 2.1k.

Re: What causes excessive power tube cathode voltage?

Posted: Sun 02/12/23 9:37 pm
by JMPGuitars
yello wrote:
Sun 02/12/23 3:13 pm
I might have had a terminology problem on my end. I was saying plate resistors but meant the 470k grid leak resistors going from power tube grids to ground. I measured them with tubes in (amp off) and one side reads open, the other side reads 892k.
Fix that.

Re: What causes excessive power tube cathode voltage?

Posted: Mon 02/13/23 9:37 am
by JMPGuitars
In case it's not clear, here's what's happening:

470 x 2 = 940. 940 x .95 = 893. So what you're seeing is two 470K resistors in series going to ground, with around 5% tolerance.
That means the wrong point is connected to ground. The junction of the two resistors should be connected to ground, which would make it so those two resistors are no longer in series to ground.

Re: What causes excessive power tube cathode voltage?

Posted: Mon 02/13/23 11:46 am
by yello
JMPGuitars wrote:
Mon 02/13/23 9:37 am
In case it's not clear, here's what's happening:

470 x 2 = 940. 940 x .95 = 893. So what you're seeing is two 470K resistors in series going to ground, with around 5% tolerance.
That means the wrong point is connected to ground. The junction of the two resistors should be connected to ground, which would make it so those two resistors are no longer in series to ground.
The part in question, which is a single part that contains both .02 coupling caps, and both 470k grid leak resistors, is in untouched stock format and wiring. One side of the part reads open (the side with the wierd voltage), the other side reads 893k. Seems to me the part has failed somewhere inside?

Here is a picture of the part, pencil pointing to it, not sure what these type of parts are called:

Image

Center wire of the part goes to ground, then the other 4 wires connect between the PI and Power tubes.

Here is a diagram of the part:

Image

1-2 and 4-5 would be the .02 coupling caps. 2-3 and 3-4 would be the 470k grid leak resistors. All encased in this one part.

Re: What causes excessive power tube cathode voltage?

Posted: Mon 02/13/23 5:08 pm
by JMPGuitars
You can try measuring each pin resistance to each other pin (1 to 2,3,4,5; 2 to 1,3,4,5;etc). If you find the values you're looking for, cool. Rewire it. But, if it's defective, replace it.

Re: What causes excessive power tube cathode voltage?

Posted: Sat 02/18/23 8:24 pm
by yello
JMPGuitars wrote:
Mon 02/13/23 5:08 pm
You can try measuring each pin resistance to each other pin (1 to 2,3,4,5; 2 to 1,3,4,5;etc). If you find the values you're looking for, cool. Rewire it. But, if it's defective, replace it.
I took out that multi-resistor-capacitor part. When I was disconnecting it I realized there was an additional part hidden inside, the 100k from the phase inverter plate pin 1 connected to B+.

I replaced the 100k, plus both .02 coupling caps, both 470k grid resistors, and also added 470r 3w screen stoppers and 1.5k grip stoppers for the 6V6. Additionally I changed R40 as the original was acting funny, thankfully I had a matching vintage original part on hand to match. It is set to 200r.

Good news, the biasing and red plating issue is resolved, and amp sounds great. At 117v in I am measuring 399 vdc from the rectifier into the R40 resistor, then 384vdc on B+ node 1, following by 375, 322, 283, and 215. The voltages are quite a bit higher now, not sure why. The same wall voltage and same rectifier tube is installed. Maybe changing R40 corrected something, but the voltage at the rectifier before that resistor is now 55v higher. I would expect to see about 360v from pin 8 of the rectifier but I am seeing 399v. Now it is too high, whereas before it was too low.

Maybe I need to try a new 5Y3 and remeasure?

With current tubes at 117v I am getting 381v on the 6V6 plates with 23v at cathode, so 12.6w at 90% dissipation. When I use regular wall voltage which is 122v I get 395v on the plates, 24.5v on cathode, so 13.6w at 97%.

Amp is sounding good, though I do wonder about what is causing the now much higher overall voltage (or previously was causing the much lower voltage)...

Also wondering, are 6V6's that have red plated for any length of time now bad/flawed, or will they function fine? I don't have a tube tester on hand.

Re: What causes excessive power tube cathode voltage?

Posted: Tue 02/21/23 9:48 am
by yello
JMPGuitars wrote:
Mon 02/13/23 5:08 pm
Rewire it. But, if it's defective, replace it.
I forgot to mention in my update above, that after installing the new parts, in addition to the mystery voltage jump, I now have a microphonic issue I need to track down (discovered right as I was wrapping up work on the amp for the weekend and didn’t have time to trace). The tubes weren’t noticed to be microphonic before - so wondering if higher voltages brought it out, or if my new components (all wired PTP on tube sockets) have created the issue.

Re: What causes excessive power tube cathode voltage?

Posted: Tue 02/21/23 11:44 am
by JMPGuitars
yello wrote:
Tue 02/21/23 9:48 am
I forgot to mention in my update above, that after installing the new parts, in addition to the mystery voltage jump, I now have a microphonic issue I need to track down (discovered right as I was wrapping up work on the amp for the weekend and didn’t have time to trace). The tubes weren’t noticed to be microphonic before - so wondering if higher voltages brought it out, or if my new components (all wired PTP on tube sockets) have created the issue.
Sounds fun. ;)

Try tapping tubes individually with a chopstick (gently) and see what you hear. If one is obviously bad, check the components and voltages. (EDIT: if only one is bad, try it in a different socket and see if the issue follows the tube or stays with the socket.)

Tubes can be robust or tubes can die. Test your voltages and see how it sounds. Without a tube tracer it's hard to know if the tubes are good or not without any obvious sound issues.

Your higher voltage could be the result of having fixed the issue so the voltage isn't being sucked down, or a number of other potential causes. Check your voltage before the rectifier and see if that seems normal or not before swapping the rectifier.

Thanks,
Josh

Re: What causes excessive power tube cathode voltage?

Posted: Sun 03/05/23 1:42 pm
by yello
JMPGuitars wrote:
Tue 02/21/23 11:44 am
yello wrote:
Tue 02/21/23 9:48 am
I forgot to mention in my update above, that after installing the new parts, in addition to the mystery voltage jump, I now have a microphonic issue I need to track down (discovered right as I was wrapping up work on the amp for the weekend and didn’t have time to trace). The tubes weren’t noticed to be microphonic before - so wondering if higher voltages brought it out, or if my new components (all wired PTP on tube sockets) have created the issue.
Check your voltage before the rectifier and see if that seems normal or not before swapping the rectifier.

Thanks,
Josh
Thats part of the problem, I'm not sure what the voltage is supposed to be before the rectifier. Schematic only lists voltage after the rectifier and after the first power dropping resistor.

I tried a different 5y3 and got the exact same voltages so thinking the tube itself isn't the problem.

What is a reasonable plate voltage for 6V6's?

Re: What causes excessive power tube cathode voltage?

Posted: Mon 03/06/23 9:51 am
by JMPGuitars
yello wrote:
Sun 03/05/23 1:42 pm
Thats part of the problem, I'm not sure what the voltage is supposed to be before the rectifier. Schematic only lists voltage after the rectifier and after the first power dropping resistor.

I tried a different 5y3 and got the exact same voltages so thinking the tube itself isn't the problem.

What is a reasonable plate voltage for 6V6's?
Get the voltages anyway. Don't you have more than 1 of these amps? You seem to have 30 of everything. 😉

Re: What causes excessive power tube cathode voltage?

Posted: Mon 03/06/23 10:41 am
by yello
JMPGuitars wrote:
Mon 03/06/23 9:51 am
yello wrote:
Sun 03/05/23 1:42 pm
Thats part of the problem, I'm not sure what the voltage is supposed to be before the rectifier. Schematic only lists voltage after the rectifier and after the first power dropping resistor.

I tried a different 5y3 and got the exact same voltages so thinking the tube itself isn't the problem.

What is a reasonable plate voltage for 6V6's?
Get the voltages anyway. Don't you have more than 1 of these amps? You seem to have 30 of everything. 😉
I did measure the voltages in the amp, with different rectifiers, and with different setting on R40 the variable 500r resistor, that is what I am exploring currently.

While my other similar amps aren't exact in the circuit, a few have the same PT and B+ scheme. I measured 3 of them yesterday, and came up with a range of voltages, and also reached out to a friend who measured his same amp and I got those voltages.

This amp in question does seem to have about 15-20 volts higher coming from the same type of PT as the other amps measured above. That said, tube dissipation is fine at about 90%, and the plate voltage on the RCA 6V6GT tubes is around 370. Amp seems to be happy and sounds good.