Vox AC30 clone (D. Manley) unstable output

[Sansteeth]

Hello everyone,
I have this Vox AC30 clone on my bench, extremely well-built by Denis Manley (France). It came in the shop for a revision, everything works fine, I'm getting 292Vdc on the plate of the power tubes, 10V across the 50R cathode resistor, which I find quite high but apparently that's how the folks at Vox wanted it to be played...
I'm getting a clean and stable sine wave on the scope up until the grids of the EL84 that clips on the positive side if I push it hard but still a very stable trace.
However, at the output, the same signal is giving me a pretty unstable trace and I'm not sure what could be causing this in the output section.
It happens regardless of the Top Cut potentiometer's position (the 250k pot/.0047uF on the schematic)
Here's a quick video of what the scope trace looks like.
Thanks a lot for your help!

all the best,
greg
https://drive.google.com/file/d/1vRUKj3UZV1Ua-sz9c0inbKBJQVcMYC5d/view?usp=drive_link

[sluckey]

However, at the output, the same signal is giving me a pretty unstable trace and I'm not sure what could be causing this in the output section.

https://drive.google.com/file/d/1vRUKj3UZV1Ua-sz9c0inbKBJQVcMYC5d/view?usp=drive_link

May just be a scope issue. How are you triggering the scope?

Access denied on your video link. Probably need to share, or make public.

[Sansteeth]

Hey Sluckey,
thanks for your response
definitely not ruling out an issue with my test equipment but I'm getting the same behavior with my Iiwatsu and my Tektronix.
Here's a link that people should be able to play:
https://www.youtube.com/shorts/H-Z7Cjgl4N8

Thanks a lot

[shooter]

the signal does look like it's one the verge of losing trigger/sync. the sinewave itself looks good, no asymmetry, odd clipping, weird runners along the wave.
guessing your AC coupled, almost looks like pulsing DC is trying to breakthrough?

[Sansteeth]

Hey Shooter, thanks for your response,
yes I'm AC coupled, I get the same behavior on both scopes, with different frequencies (and different signal shapes) on the signal generator. I tried a different dummy load I have lying around, same results...
I just tried monitoring the output on a 50W tube amp I had with the exact same setup and I'm getting a stable and clean trace on the output. So I'd tend to rule out my test equipment...

[shooter]

I don't believe it's the test equipment, I think there is something there, you got a spectrum analyzer laying around, or does one of your scopes have FFT feature?


have you scoped the input feeding the PI? 
have you "changed" the environment around the amp, one amp might not "pick up" floaties in the air, while another will.

[Sansteeth]

Unfortunately, I don't have a spectrum analyzer nor a scope with FFT, only good old analog scopes but that kind of issue makes me feel like I'm missing a lot of info by not having those things...

Signal at the input of the PI is super clean and stable, all the way up to the grid of the power tubes.
I mean, the second amp was in its headsleeve but the build inside is an absolute mess, whereas the Vox clone has the chassis on my bench but the build is one of the cleanest I have seen.
Lights off, pushed away from my variac, no difference to the trace on the output...

[sluckey]

So I'd tend to rule out my test equipment...

Did not mean to imply it could be a problem with the scope. I asked how you were triggering the scope. When chasing a signal through an amp it's a good idea to reliably trigger the scope so you don't have to fiddle with trigger levels as you look at different points in the circuit. I always use a Tee on the sig gen output and connect one cable to the scope ext. trig input. Set the scope to EXT TRIG and set for reliable triggering directly from the sig gen. Then the display will be rock solid no matter what point you are viewing the signal in the amp.

[sluckey]

I just saw the yt video. The scope is in sync. It's the amplutude that's changing. Probably changing at a 60hz rate. Slow the time base way down and you're display will look like an AM modulated waveform. Now sync the scope on "line" and the modulation will become steady if it's a 60Hz modulation.
.

[Sansteeth]

No I know you were not implying that but I was, I've had my x10 attenuation on my homemade dummy load that started acting up a couple of weeks ago and my scopes are not exactly brand new.
Manual or ext triggering give me the same results.
It's not getting steady on the "line"set up, however when I slow it down and use the cursors on my Tektronix (gotta love these) I'm getting 50Hz modulation (picture attached).

[Sansteeth]

Ughh okay my brain isn't working right here  .  I 'm based in France, so yeah 50Hz is your 60Hz.
Would that confirm the theory that it's just picking up the funny stuff in the air? But it's only happening in the power section, which should be the least prone to picking up weak frequencies in the air, right?

[sluckey]

That's more likely a grounding issue between the scope and amp. Doesn't necessarily indicate any problem with the amp. Maybe the OT secondary does not have a good earth ground to the chassis and house wiring or maybe the scope doesn't have a good earth ground. Use a ground clip on your probe and connect to the amp chassis. Then the ground connection does not depend on the earth connections to the scope or the amp.

[Sansteeth]

Actually connecting the ground clip to the amp chassis gave me the same results, but connecting it to the ground of the output jack (which is isolated from the chassis), I'm getting a stable trace on the (with a very small 100Hz modulation).
Looks like we got a winner! Thanks a lot Sluckey!
So the symptoms I am seeing have only something to do with the interfacing of my scope and the DUT? Or should I look into the grounding scheme of the amp?

[sluckey]

Actually connecting the ground clip to the amp chassis gave me the same results, but connecting it to the ground of the output jack (which is isolated from the chassis), I'm getting a stable trace on the (with a very small 100Hz modulation).

The schematic shows the OT secondary connected to chassis ground. I'd inspect that connection. It may be an actual wire or it may just be grounding through the mechanical connection of the output jack. Either way, it ain't properly grounded to chassis.

[glass54]

Hi Sans
Just for interest sakes, can you measure the AC component at C39 and C40 (positive terminals). I don't know what multimeter you have, but make sure it is at least 400V rated input. (you have already indicated that +HT is 292V DC.
I don't suspect your Oscilloscope but I'm suspicious of Hum/Buzz induced problems. You might consider temporary tacking a 32u 450V, across C39, then C40, then C42 (one at a time) with the oscilloscope probe at the output.
Regards
Mirek

[Sansteeth]

The schematic shows the OT secondary connected to chassis ground. I'd inspect that connection. It may be an actual wire or it may just be grounding through the mechanical connection of the output jack. Either way, it ain't properly grounded to chassis.

Yes, mistake on the part of the  builder I guess, the output jack was isolated from the chassis and that's where the secondary wiring was going to. Removing the Switchcraft isolating washer made a big difference! Thanks a lot for the pointer.

Hi Sans
Just for interest sakes, can you measure the AC component at C39 and C40 (positive terminals). I don't know what multimeter you have, but make sure it is at least 400V rated input. (you have already indicated that +HT is 292V DC.
I don't suspect your Oscilloscope but I'm suspicious of Hum/Buzz induced problems. You might consider temporary tacking a 32u 450V, across C39, then C40, then C42 (one at a time) with the oscilloscope probe at the output.
Regards
Mirek

Mirek, thanks for chiming in! You are into something, tacking a 32uF cap across c39 reduces ripples by half on the scope. But honestly when I put the cap on a switch, the sonic difference is negligible. I 'll talk with the customers to see what he wants to do.
Thank you all for the pointers, super helpful!

[glass54]

Removing the Switchcraft isolating washer made a big difference!

Thanks for the feedback Sans. I imagine we all assumed the that the "common" on Output Tx was wired correctly as per schematic.
I have been caught before with incorrect O/P Tx wiring.
Regards
Mirek