My first build - AB763 Deluxe Reverb clone - opinions please

[Greg_L]

What reverb tank? I mean the actual numbers.
Are both tank jacks grounded?
perhaps there is a ground loop thru the tank.

Type 4
8 ohms input impedance
2250 output impedance
Long decay
Input insulated/output grounded
No lock
Horizontal open side down mounting

My RCA jacks are grounded via their mounting contact with the chassis. Did I create a ground loop by having the RCA jacks chassis grounded?

[Chuck H]

Looking at the photo's it seems like your reverb transformer may be grounded at the buss near the input? It's typically grounded at the chassis with the RCA jacks. Grounding is the next thing to suspect when lead dress changes don't seem to improve stability.

[Greg_L]

Looking at the photo's it seems like your reverb transformer may be grounded at the buss near the input? It's typically grounded at the chassis with the RCA jacks. Grounding is the next thing to suspect when lead dress changes don't seem to improve stability.

Ah, yes it is grounded with the preamp grounds near the inputs. I've got the preamps/buss bar grounded at a bolt near the inputs and the reverb transformer is grounded there as well. I can change that and see what happens.

[tubeswell]

Most pans have one RCA jack sleeve isolated from the pan chassis, which stops ground loop hum through the pan circuit.

If you’re still getting the HF ‘disturbance’, then I would still look at lead dress. Particularly keeping plate and grid leads to tube sockets physically separated by an inch, or where it’s necessary to have them closer, having them criss-cross at right angles - to minimise EM induction that could otherwise generate oscillation. Also keep sensitive signal wires in shielded cable or laid close on the chassis.

Also, consider the OT. If you have used a ‘non-standard’ OT (e.g. with more interleaving than usual), you might be developing some Pr:Sec HF phase shift (as a result of increased interwinding capacitance) which could produce HF oscillation. That’s why I suggested 220pF grid-to-cathode at each 6V6 socket. YMMV

[Greg_L]

Most pans have one RCA jack sleeve isolated from the pan chassis, which stops ground loop hum through the pan circuit.

If you’re still getting the HF ‘disturbance’, then I would still look at lead dress. Particularly keeping plate and grid leads to tube sockets physically separated by an inch, or where it’s necessary to have them closer, having them criss-cross at right angles - to minimise EM induction that could otherwise generate oscillation. Also keep sensitive signal wires in shielded cable or laid close on the chassis.

Also, consider the OT. If you have used a ‘non-standard’ OT (e.g. with more interleaving than usual), you might be developing some Pr:Sec HF phase shift (as a result of increased interwinding capacitance) which could produce HF oscillation. That’s why I suggested 220pF grid-to-cathode at each 6V6 socket. YMMV

Oh yeah, I forgot about that. My head's been swimming.

I've moved wires back and forth, up and down, none of that has any effect on the tiny little kink in my waveform. I'm kind of starting to think maybe that's just how it is? The amp doesn't squeal, hiss, hum, or pop or anything. It's very very quiet at idle, and it sounds great cranked. The only issue was that little brittle sound that has sent me down this bottomless rabbit hole! And the little kink totally goes away with the reverb pan unplugged. The kink was much worse before I moved that plate-to-plate jumper on V3. And I'm talking a tiny little bump on the waveform. It's not the squiggly psycho oscillation.

I certainly can try little caps across the power tubes. So that's it? Just a 220pf from grid to cathode on each tube socket?

[galaxiex]

I don't remember if you have a shielded wire from the reverb pot to the board, (the center lug on the pot).
If not... might be something to try.
Ground the shield at one end only.

[Greg_L]

I don't remember if you have a shielded wire from the reverb pot to the board, (the center lug on the pot).
If not... might be something to try.
Ground the shield at one end only.

I do not have that wire shielded, but I can try it. I have lots of shielded wire. Shielding ALL of the grid wires has been in the back of my mind for a few days.

[tubeswell]

I certainly can try little caps across the power tubes. So that's it? Just a 220pf from grid to cathode on each tube socket?

Probably worth a try, but it’s just a hunch - me being on the other side of the world’n’all, I can’t really get up close myself to see the OT for myself with a scope’n’all etc.

[Greg_L]

Probably worth a try, but it’s just a hunch - me being on the other side of the world’n’all, I can’t really get up close myself to see the OT for myself with a scope’n’all etc.

I think if you just start digging you might get here eventually.

[tubeswell]

I’ve dug myself into enough holes already. I’m staying put.

Even if it’s an OT that has been carelessly scatterwound, albeit with the correct number of turns and interleaving, it could result in sound dissatisfaction. Can’t really tell without at least trying something that removes the possible ‘interference’ pathway.

Snubbing the dominant HF NFB pole with 220pF might reveal something (or it might not)

[Greg_L]

I’ve dug myself into enough holes already. I’m staying put.

Even if it’s an OT that has been carelessly scatterwound, albeit with the correct number of turns and interleaving, it could result in sound dissatisfaction. Can’t really tell without at least trying something that removes the possible ‘interference’ pathway.

Snubbing the dominant HF NFB pole with 220pF might reveal something (or it might not)

I'm going to try your suggestions when I get the amp back in the loud room.

[pdf64]

So grid stoppers on V3 didn’t help with the kink?

[Greg_L]

So grid stoppers on V3 didn’t help with the kink?

Whoa, I didn't try that. I must have missed that suggestion or it got lost in the clusterf*** that is my head right now.

The grids are linked on V3 across the socket and the actual grid wire comes off pin 7. Would I use just one grid stopper at pin 7?

Never mind, I found your suggestion. Sorry. 22k/grid.

What's the best way to implement this idea if the grids are linked?

[Greg_L]

Grid stoppers on V3 are very helpful, eg 22k / grid.

So this might have helped a tiny bit. I put a 22k on pin 7 of V3 and cleaned up some solders. Here's my kinky kinks....



Zoomed in a little...this is taken off V2b grid - reverb tank/tube installed. Volume around 8.


Same signal, reverb tank disconnected.



Backed out view - reverb in circuit


Reverb disconnected




And just for reference....here's where it was a few days ago.

[g1]

I bet if you sweep through some freqs with a dial type sig gen, you will see the kink get more or less severe. I just consider it an artifact of the reverb circuit?