My first build - AB763 Deluxe Reverb clone - opinions please

[Greg_L]

You could try going to a 220pF cap.

Other than that, my hunch is it’s likely to be OT or lead dress related.

If it is HF oscillation caused by inter winding capacitance in that OT, you could try one more thing which is slugging the dominant FB loop pole with 220pF grid to cathode on each 6V6.

Well check this out and tell me what you think....


That's V2b grid. Vibrato channel, vol knob around 5-6 is where this happens. I'm injecting a 1k sine about 300 mv pk to pk into input 1, and that's the weirdness I'm getting on the 2nd stage.

Here's a little video sweeping the volume knob on V2.

What do yall think of that kink in my sine wave?

That does not happen on the normal channel.

This is probably not related to my brittle noise, but it caught my attention anyway.

[The Dude]

I see what appears to be high frequency noise/oscillation riding on the sine wave. Some generators are not that "pure". You might scope the generator output and make sure it's clean. That said, I've got a '68 Custom Deluxe Reverb Reissue in the shop now that's doing the same thing. I'll be following along and when I get to it, I'll post anything helpful I come up with.

[Greg_L]

I see what appears to be high frequency noise/oscillation riding on the sine wave. Some generators are not that "pure". You might scope the generator output and make sure it's clean. That said, I've got a '68 Custom Deluxe Reverb Reissue in the shop now that's doing the same thing. I'll be following along and when I get to it, I'll post anything helpful I come up with.

Thanks, that would be awesome.

My "generator" is actually an app on my phone. I adjust the volume and measure the output before I put it into the amp. I don't get that kink on the normal channel though. The signal through the normal channel stays pretty pure throughout the gain stages. I only get that kink on the vibrato's first gain stages and it's only halfway through the volume sweep. At low volume, the signal stays intact. And at max volume the signal is again a smooth sine wave. It's just in the middle it gets goofy.

[galaxiex]

Thanks, that would be awesome.

My "generator" is actually an app on my phone. I adjust the volume and measure the output before I put it into the amp. I don't get that kink on the normal channel though. The signal through the normal channel stays pretty pure throughout the gain stages. I only get that kink on the vibrato's first gain stages and it's only halfway through the volume sweep. At low volume, the signal stays intact. And at max volume the signal is again a smooth sine wave. It's just in the middle it gets goofy.

I'm just spitballin' here...

I have a 1970 Traynor YGM III that had similar looking sine wave in the middle of the vol pot sweep.
Turned out to be the reverb circuit.
If I disconnected the tank leads it cleaned up.
I don't remember from earlier in the thread if you tried that?

Edit; ahhh NVM, I see you pulled the reverb tube.

In my case it was reverb circuit oscillating.

[Greg_L]

I'm just spitballin' here...

I have a 1970 Traynor YGM III that had similar looking sine wave in the middle of the vol pot sweep.
Turned out to be the reverb circuit.
If I disconnected the tank leads it cleaned up.
I don't remember from earlier in the thread if you tried that?

No I haven't, but I sure will. Thanks!

[tubeswell]

Well check this out and tell me what you think....


That's V2b grid. Vibrato channel, vol knob around 5-6 is where this happens. I'm injecting a 1k sine about 300 mv pk to pk into input 1, and that's the weirdness I'm getting on the 2nd stage.[/video]


What do yall think of that kink in my sine wave?

That does not happen on the normal channel.

So if you swap V1 and V2 around, does that grid weirdness stay on the same pin/socket?

[dstrat]

It seems to me your seeing distortion from over driving the first triode with that 300mv.

I would try a more sane input voltage and see if its still there..

[pdf64]

...What do yall think of that kink in my sine wave?
That does not happen on the normal channel....

Is the bright off?
Try it again with V3 reverb driver removed.

I guess that parasitic capacitances are acting to add more harmonics (from later in the circuit, eg reverb send) under certain conditions of impedance at the grid.
Being low signal level and high impedance, the preamp grid wires are very sensitive 'receivers', so lead dress is will affect this sort of thing.

300mV p-p doesn’t seem an excessive input signal, equates to ~100mV rms.

[Greg_L]

So if you swap V1 and V2 around, does that grid weirdness stay on the same pin/socket?

I'll try it. I think I got the same thing with the old tubes, but I can't remember now.

It seems to me your seeing distortion from over driving the first triode with that 300mv.

I would try a more sane input voltage and see if its still there..

Okay thanks. But I think my generator input is less than what a guitar pickup kicks out, no? I had a guitar on the scope just to see how hot my pickups peaked and it put out well over 500mv pk to pk with hard picking. But, as always, it's entirely possible I did that wrong. Lol.

Is the bright off?
Try it again with V3 reverb driver removed.

I guess that parasitic capacitances are acting to add more harmonics (from later in the circuit, eg reverb send) under certain conditions of impedance at the grid.
Being low signal level and high impedance, the preamp grid wires are very sensitive 'receivers', so lead dress is will affect this sort of thing.

300mV p-p doesn’t seem an excessive input signal, equates to ~100mV rms.

Yes, bright switch off. Everything I'm doing is with the bright switch off. I'll give it a try, thanks. I'll try just about anything. I did play the amp with V3 removed. My initial problem was still there - the brittleness. Since that brittleness happens from both channels, I don't think my V2 sine wave kink is related. I think they're two separate issues.

So today's first round of testing will be more tube rolling, play with the input voltage, try to address lead dress better, and scope test with V3 removed.

[Chuck H]

I suppose it's possible to get a glimpse of what an amp may be doing at note attack by cranking up the signal generator. But the amp won't ever have to actually respond to a guitar that way. The difference is that an amp actually can manage peaks to some degree with more grace than a constant signal at that peak level. So the scope shot result can be a little misleading. That is, testing with a constant signal at pick attack levels won't show you how the amp responds to transients. It will show you how the amp responds to attempting to amplify a constant signal at the higher transient level. Not the same thing. I'm sure there are ways to test with a pulse and trigger capture the shot. I haven't worked out how to do that and it may require bench gear most average builders don't have.

[dstrat]

+1 ^^^ Chuck said what I was thinking but better then I could at the time.
My point was turn down the generator to the point where the sine is clean and note the voltage.
Driving it harder is no problem but does not reveal anything other then tube does distort.
imo,
the attack noise your trying to fix could be PI, power tubes, OT trfm.

[Greg_L]

I suppose it's possible to get a glimpse of what an amp may be doing at note attack by cranking up the signal generator. But the amp won't ever have to actually respond to a guitar that way. The difference is that an amp actually can manage peaks to some degree with more grace than a constant signal at that peak level. So the scope shot result can be a little misleading. That is, testing with a constant signal at pick attack levels won't show you how the amp responds to transients. It will show you how the amp responds to attempting to amplify a constant signal at the higher transient level. Not the same thing. I'm sure there are ways to test with a pulse and trigger capture the shot. I haven't worked out how to do that and it may require bench gear most average builders don't have.

Mr Rob Robinette suggested I scope the amp while playing it. That is going way way way above my pay grade, but I suppose it's worth a shot.

Another simple variable that I just considered is my recording preamp. It's unlikely to be the problem, but I've recorded all of my test sound clips using the same channels on this preamp interface. I need to try a different preamp just to make sure that the noise isn't on the recording end of things. God, wouldn't that be a bitch if I'm chasing an amp bug that isn't actually an amp bug at all? That would be infuriating and relieving all at the same time.

+1 ^^^ Chuck said what I was thinking but better then I could at the time.
My point was turn down the generator to the point where the sine is clean and note the voltage.
Driving it harder is no problem but does not reveal anything other then tube does distort.
imo,
the attack noise your trying to fix could be PI, power tubes, OT trfm.

That's what I've been thinking all along. It's gotta be output stage related since the noise is on either channel and the tone stack doesn't affect it at all...

[Greg_L]

Okay mi amigos....more scope testing has yielded a finding.

First, I dropped my input signal to about 100 v pk to pk. That cleaned up the kink in V2.
Pulled V3 - reverb driver - and re-tested at V2b grid and speaker output - smooth clean sine wave at both locations. No more kink.
Max boosted input signal with V3 removed - still no kink. A little clipping at max volume, but the kink was not there.
Put V3 back in, disconnect reverb tank - no kink.
Reconnect reverb tank - kink is back on grid and at output.

So I'm thinking it's gotta be in that reverb driver/recovery circuit, and this.....


That is V3. See that little loop connecting the plates? Chopsticking that thing drastically alters the shape of the output sine wave and kink shape. Moving the grid and cathode wires near the loop did nothing. But jiggling that loop around made huge differences, even going into full-on rampant oscillation. So I'm thinking I'll need to re-route that thing somehow and re-test.

Am I on the right track here?

[Greg_L]

Well, I went ahead and flipped the plate-to-plate jumper wire back around behind the tube socket and my signal got MUCH better. There is only the tiniest bit of kink in the wave, I mean barely there, and it's not until I wind the volume way up to like 9-10.

Additional chopsticking and moving wires around yielded no noticeable effects at all.

Just for grins I jumper-wired a 460pf cap across the plate-to-cathode on V3b. I think that may be a Silverface thing? But what the hell, I'll give it a shot. Holy crap that totally smoothed it out and I didn't lose any amplitude. My signal seems pure....or at least kink-free and smooth like a swedish girl's hair.

So maybe I'm making progress.

Yall please keep the ideas and opinions coming. I'm taking it all in and I appreciate it very much.

[pdf64]

Yes, that loop will be a significant transmitter.
Shorten it right back and go over the top or around the other side.
As it is, it's awful close and parallel to V2 6-8 wires.