Phase Inverter Voltage seems off

[philbard]

Hello All,

First of all, hope everyone has a great holiday season this year. In the Pac NW we are getting pretty wet and on top of the pandemic it seems like we will never see the sun again, but there is a little hope for a white Christmas and that brightens things up a bit.

Just finished another 102 build and its up and running, but doesn't have the usual level of power. I have to set both the preamp and master volumes at noon to get it up to bedroom levels. Voltages are below, the one thing that seems off is the V3 (PI) A side plate voltage. Normally it is a close match to the B side plate but on this build is about 40v lower. Additionally it drops about 10-15v as I take the reading, the B side however does not and stays steady. The tube is a Sovtek, and I've swapped it for another that I know is reliable, no change. I've got the PI trimmer set to about 7v higher on the left leg (towards the power supply). When the tube is pulled the plate voltages on both sides rise to 415v and are steady. I've also tried routing the preamp output to another build's power amp via the effects loop jacks, and it ouputs the guitar signal at normal levels, so I know the preamp is fine.

Below are my voltages and a gut shot. Any feedback on what you might see going on appreciated...

B+1 431v
B+2 430v
B+3 412v
B+4 315v
B+5 308v

V1 12AX7A
1: 203v
3: 1.48v
6: 197v
8: 1.6v

V2 12AX7A
1: 194v
3: 1.71v
6: 186v
8: 1.84v

V3 Phase Inverter
1: 258v dropping to 248v in a few seconds
3: 62v
6: 288v
8: 62v

V4 6L6GC
3: 432v
4: 431v
5: -43.8v
9: 41.8mV

V5 6L6GC
3: 432v
4: 431v
5: -43.8v
9: 38.6mV

V6 6L6GC
3: 432v
4: 430v
5: -43.8v
9: 41.2mV

V7 6L6GC
3: 432v
4: 431v
5: -43.8v
9: 39.6mV

Final Top View.jpg

Cheers,

[Richard1001]

I can't see all color bands clearly but... What is the resistor value of the second anode resistor of V3 suppose to be? It seems to have the yellow band in the middle instead of at the end (multiplier band)

So what is the value of that resistor?

[philbard]

The plate load resistors for V3 are supposed to be 120K left (V3A) and 110K right (V3B). They measure 118.4K and 110.7K respectively so if I'm not mistaken should be supplying proper voltages. Resistance measurements from the trimmer wiper to the plate side of the load resistors are 120.7K left and 119.1K right, so pretty even overall.

[martin manning]

Have you' tried a different tube in the PI spot?

[philbard]

Hi Martin. Yes, swapped for another Sovtek that is working properly in another build and also a Tungsol. No change.

[philbard]

Just had a thought. Is there any risk to swapping the leads to the plates to see if the problem follows the swap?

[sluckey]

There's no risk, other than NFB becoming positive feedback. But you would need to swap the grids too. Maybe when you start soldering on that socket the problem will go away.

[philbard]

There's no risk. But you would need to swap the grids too. Maybe when you start soldering on that socket the problem will go away.

Yeah, I already reflowed the terminals, but perhaps there's a chance... Thx

[martin manning]

Yes, swapped for another Sovtek that is working properly in another build and also a Tungsol. No change.

Missed that in your first post :^/ Might be an oscillation going on. Can you scope it? Does your DMM have a frequency measuring function? Maybe you could pick it up with that, or maybe you could try tacking a small cap across the plates and see if ithat changes things.

[philbard]

Yes, swapped for another Sovtek that is working properly in another build and also a Tungsol. No change.

Missed that in your first post :^/ Might be an oscillation going on. Can you scope it? Does your DMM have a frequency measuring function? Maybe you could pick it up with that, or maybe you could try tacking a small cap across the plates and see if ithat changes things.

I have a scope, I'll put it on it later this afternoon, but I'm not experienced in spotting oscillations so will have questions...

[philbard]

Missed that in your first post :^/ Might be an oscillation going on. Can you scope it? Does your DMM have a frequency measuring function? Maybe you could pick it up with that, or maybe you could try tacking a small cap across the plates and see if ithat changes things.

FWIW, I put the scope across the OT secondary and turned on the Spectrum Analyzer. This is what I got at low master volume, above about 3 o'clock it flatlines. Not sure if this indicates anything...

Image1.png

[philbard]

Might be an oscillation going on. Can you scope it? Does your DMM have a frequency measuring function? Maybe you could pick it up with that, or maybe you could try tacking a small cap across the plates and see if ithat changes things.

Martin,
Jumpered a 470pf cap across the plates, V3A plate is steady at 267v and v3B plate lowered to 281v, also steady. Does this indicate there is some sort of oscillation? It did not help with the low volume character of the amp however, so maybe its something else that is causing that.

My OT transformer waveform now looks like this, but again, I'm unable to interpret the significance.

Image2.png

[martin manning]

This is with nothing plugged in to the amp input? There's a suspicious looking spike, but I can't tell what the frequency is. Put the probe on the PI outputs (AC coupled) after the PI coupling caps and see what's there. Turn the time base to 20us/div (50kHz) or so.

[philbard]

This is with nothing plugged in to the amp input? There's a suspicious looking spike, but I can't tell what the frequency is. Put the probe on the PI outputs (AC coupled) after the PI coupling caps and see what's there. Turn the time base to 20us/div (50kHz) or so.

Correct, no input. Here are the waveforms, probe after the coupling caps. Used autoset to get the time/divisions and voltages. The 470pf cap across the plates was removed.

V3A, 2us/div, 3v/div

V3A Coupling Cap 2us.png

V3B, 20us/div, 10mv/div

V3B Coupling Cap 20us.png

So I'm guessing we seeing a relatively large signal in V3A that is absent in V3B. Is this a correct interpretation?

[martin manning]

The A side will drive the B side through its cathode. Looks like ~250 kHz oscillation.