Bias circuit loading?

[turbofeedus]

Hey all, working on a new build.
Building a JCM800 2204 into this sundown Formula 50 chassis.
There isn't a public schematic for the Formula 50, but it's the same PT as this sundown.
I've built and rebuilt this bias supply a couple times (hence the messiness), the current iteration is the marshall capacitor-coupled supply from JCM900:

bias supply.png

Photo Oct 01, 5 15 03 PM.jpg

Photo Oct 01, 5 15 09 PM.jpg

Photo Oct 01, 5 15 19 PM.jpg

B+ is ~450V unloaded, bias supply delivering -52V to -32V.
I can power up to full line voltage, set the bias on the output tubes to 60% dissipation, no problem.
The moment any input signal is applied (touching tip on input jack, sine wave, guitar, doesn't matter), the current through the tube skyrockets.
Bit stumped here, I thought maybe the bias supply was being loaded down by the input signal, causing the negative voltage to go more positive and cause the excessive draw.
(Side note; I know the cap for the bias supply is not ideal, it's what I had handy. I have a more appropriately rated cap on order)

[pdf64]

…
The moment any input signal is applied (touching tip on input jack, sine wave, guitar, doesn't matter), the current through the tube skyrockets.
…

Are you sure that’s different to any other similar class AB amp?
eg what happens to the bias supply voltage?

[FUCHSAUDIO]

I'm betting you have another issue. Maybe oscillation or instability ? That's a pretty time-honored bias supply (I use it too), and it should be fine and dandy.
What happens to your bias voltage and B+ when the tubes take-off ? If it's still stable, you have another issue somewhere... do you own a scope as the instability could be outside of your hearing range.

[turbofeedus]

…
The moment any input signal is applied (touching tip on input jack, sine wave, guitar, doesn't matter), the current through the tube skyrockets.
…

Are you sure that’s different to any other similar class AB amp?
eg what happens to the bias supply voltage?

I suppose it's the magnitude of current draw that's alarming. I've seen current draw increase across the 1Ω change during playing, but never this drastically, and never with the master and preamp volumes both dialed almost completely down. Something's definitely not right.

I'm betting you have another issue. Maybe oscillation or instability ? That's a pretty time-honored bias supply (I use it too), and it should be fine and dandy.
What happens to your bias voltage and B+ when the tubes take-off ? If it's still stable, you have another issue somewhere... do you own a scope as the instability could be outside of your hearing range.

I made a video showing the events here:


The bias voltage gets more positive with some input signal applied, although the change in bias voltage seems like it wouldn't be enough to account for the huge increase in current flow.
Also this video is shown with a bulb limiter in series, that’s why the plate voltage is lower.
I do have a scope, what should I probe?

[bepone]

The moment any input signal is applied (touching tip on input jack, sine wave, guitar, doesn't matter), the current through the tube skyrockets.

can you explain what means skyrockets=? i see on the video only 8 mA and current is going to 20ma which are peanuts=?
all normal except for the little drop in bias, but i would increase 10uF to 47uF caps in bias circuit and see again.

bias would be more stable if you have normal low impedance source for the bias, like normal transformer secondary....

during operation is acceptable from 150-to 200mA for EL34, and note that this fluke is measuring DC plus AC components together so it is not perfect measuring method

[Stevem]

Hook up your voltmeter set for AC volts across your speaker or better yet load resistor ( even if it’s only a 10 water) and check how many watts it’s driving out.

20 ma isn’t even 3 watts depending on the load resistor your using!

[turbofeedus]

The moment any input signal is applied (touching tip on input jack, sine wave, guitar, doesn't matter), the current through the tube skyrockets.

can you explain what means skyrockets=? i see on the video only 8 mA and current is going to 20ma which are peanuts=?
all normal except for the little drop in bias, but i would increase 10uF to 47uF caps in bias circuit and see again.

bias would be more stable if you have normal low impedance source for the bias, like normal transformer secondary....

during operation is acceptable from 150-to 200mA for EL34, and note that this fluke is measuring DC plus AC components together so it is not perfect measuring method

Here is second video at full wall voltage, no current limiter.
I should also mention the volume and master are both at 2.

[turbofeedus]

Hook up your voltmeter set for AC volts across your speaker or better yet load resistor ( even if it’s only a 10 water) and check how many watts it’s driving out.

20 ma isn’t even 3 watts depending on the load resistor your using!

I'm measuring around 500mVAC, that's at the same bias conditions as the second video I made.
8Ω resistor

[Stevem]

If your where to for example read 12 vac into a 8 ohm load that would be all of 18 watts.

[bepone]

Here is second video at full wall voltage, no current limiter.
I should also mention the volume and master are both at 2.

friend but this is apsolutelly normal! what is not normal is that you are doing it without the load! or you have some resistor for the load?
what is happening here that your finger is antenna and you are inputting all the garbage from the wall (60Hz mixed with all the harmonics) to the input in the amp and your amp is giving a lot of Watts output in this moment!
if you are not having load connected you will burn the output transformer!

try to connect speaker and try to touch the input jack, tell me about the sound, how loud is? fully=?

and p.s. this current is below 150mA what i have mention before...so it is nothing to worry about?

p.s.2 if your bias is dropping too much, buy one cheap transformer 40VAC 10W and mount it external, so you can have proper bias circuit , this x cap will explode one day, sooner or later......

[turbofeedus]

If your where to for example read 12 vac into a 8 ohm load that would be all of 18 watts.

it's much less than that, less than 1VAC into 8Ω resistor

friend but this is apsolutelly normal! what is not normal is that you are doing it without the load! or you have some resistor for the load?
what is happening here that your finger is antenna and you are inputting all the garbage from the wall (60Hz mixed with all the harmonics) to the input in the amp and your amp is giving a lot of Watts output in this moment!
if you are not having load connected you will burn the output transformer!

try to connect speaker and try to touch the input jack, tell me about the sound, how loud is? fully=?

and p.s. this current is below 150mA what i have mention before...so it is nothing to worry about?

p.s.2 if your bias is dropping too much, buy one cheap transformer 40VAC 10W and mount it external, so you can have proper bias circuit , this x cap will explode one day, sooner or later......

sorry if I didn't show it, there is an 8Ω resistor on the OT secondary. You can see if briefly at the beginning of the second video under the amp.
Under normal conditions I would bias for about 34mA across the bias sense resistor, at 430V plate that's 14.6W, or about 60% max dissipation.
I connected a speaker, the output is very low.

Also thank you everyone for helping!

[bepone]

do you have also signal generator? maybe you can see if there is something strange in output signal at ful power!
- maybe your bias circuit is weak
- maybe your layout (180 deg from ususal) creating the problem, oscillations
- maybe your resistors are too big in bias circuit
- maybe you rewire your output transf. 16 ohm output to 8 ohm load now(increase of current)
- maybe your coupling caps to the output tubes are leaking DC

need to measure bias and currents on full power and compare this to datasheet max values

[turbofeedus]

do you have also signal generator? maybe you can see if there is something strange in output signal at ful power!
- maybe your bias circuit is weak
- maybe your layout (180 deg from ususal) creating the problem, oscillations
- maybe your resistors are too big in bias circuit
- maybe you rewire your output transf. 16 ohm output to 8 ohm load now(increase of current)
- maybe your coupling caps to the output tubes are leaking DC

need to measure bias and currents on full power and compare this to datasheet max values

The bias supply not delivering the necessary current was on my list, but this is the stock bias supply for many amps, as to which Mr.Fuchs alluded.
I did measurements on the OT prior to install, the primary is 4kΩ impedance with the printed 4Ω and 8Ω taps.
If the coupling caps were leaking, I would expect to see the bias voltage drifting positive on the grids of the output tubes. It's actually pretty steady as the previous video showed.

I was about to post some screenshots of showing a 1kHz sine moving through the amp, but my 2213 decided to stop triggering on any signals this morning
Forecast calls for downpours I suppose.
Guess it's finally time to bite the bullet and grab one of those new budget Hantek scopes.
Anyway here's some more shots of the build.

Photo Oct 02, 10 35 17 AM.jpg

Photo Oct 02, 10 35 22 AM.jpg

Photo Oct 02, 10 35 26 AM.jpg

Photo Oct 02, 10 36 44 AM.jpg

Photo Oct 02, 10 37 09 AM.jpg

Photo Oct 02, 10 37 25 AM.jpg

[sluckey]

There's nothing wrong with your bias supply. The current requirement is almost nothing.

Nice looking bracket for the bias pot.

I notice your grounding relies on the mounting screws for several of the terminal strips. But I don't see that you removed the paint down to bare metal on the chassis. If this is not causing you any problems today, it most likely will tomorrow.

[turbofeedus]

There's nothing wrong with your bias supply. The current requirement is almost nothing.

Nice looking bracket for the bias pot.

I notice your grounding relies on the mounting screws for several of the terminal strips. But I don't see that you removed the paint down to bare metal on the chassis. If this is not causing you any problems today, it most likely will tomorrow.

Good eye sluckey, I did actually remove the paint on the opposite side, hard to see in the pic.