Gibson GA-15 RVT Noisy Reverb Recovery

[Helmholtz]

OK, I'll look into that. My reading of 306V plate volts may have been incorrect or taken at a different time than I measured the bottom of the primary.

Please measure R16 and then cathode voltage. You could also have a grid side problem, like leaking capacitors or a bad R12 wiper contact, as the grid reference depends on the loudness pot.

BTW, it is normal to get increased hiss when V1B is not terminated by the reverb output transducer.

[bobloblaws]

Please measure R16 and then cathode voltage. You could also have a grid side problem, like leaking capacitors or a bad R12 wiper contact, as the grid reference depends on the loudness pot.

BTW, it is normal to get increased hiss when V1B is not terminated by the reverb output transducer.

Ahhh, I think it is the pot wiper, it is physically loose. I had ruled out the grid by disconnecting the lead coming from R14 to the eyelet board and hearing no change without cluing in that that was the evidence right there. I'm going to replace the pot, hopefully that does it.

[g1]

I don't think so.
The 2V drop would correspond to a plate current of around 3mA ,

The 12mA I calculated was based on his previous reading of 9V drop. Agree that with shorted turns there should be no output. Hopefully it is not the interstage transformer.

[bobloblaws]

The 12mA I calculated was based on his previous reading of 9V drop. Agree that with shorted turns there should be no output. Hopefully it is not the interstage transformer.

The good news is that it does not appear to be the interstage transformer, but wow, this is like playing whack-a-mole.

After putting in the new R12 pot I still had the loud hum. I bypassed the tone stack and jumpered straight from C1 to R12 pot and then straight from the wiper to V2A grid to rule out anything in the tone stack. Still a loud hum. That pot was grounded to one of the input jack tabs. I experimented with grounding it at different points and found grounding to low side of R16 worked best.

So then I redid what I had undid to the tone stack etc and turned my attention back to the reverb problem. Unfortunately, when I connected R15 back to the V2A grid there was now a really loud hum! So once again I experimented with grounding and found that grounding R16 over at the input jack tab where originally the volume pot (among other things, the ground tab on the pot is being used as a hub for grounds in and around the tone stack) was grounded reduced that hum greatly. However, the overall level of hum in the system is now greater than it was before that first really loud hum showed up yesterday that prompted me to change the volume pot etc. And now if I disconnect the reverb recovery cable there is also a very loud hum whereas before it would just add some hiss with a small amount of hum similar to what I was trying to fix in the first place.

So based on all that I gotta think the source of my problems is grounding, maybe it has been from the beginning? The way it was set up before I made the changes I just mentioned was the main ground point was a tab on the first input jack (not the same one as was used for the volume and tone stack). Then a wire running over to the eyelet board where R16, R27, R5, R4, and R19 are grounded via 2 jumpered side by side eyelets. Then a jumper over to an eyelet where R34, R35, and R23 are grounded along with V2B cathode. Then another jumper over to an eyelet grounding R16, C8, and C16. From there there is a long wire over to a terminal strip tab grounding the 4 filter caps. Earlier the power tube cathode resistor and bypass cap had also been grounded here but I moved that to another ground point at the end of the strip awhile back. What do you think, sound like a dog's breakfast? I still don't know what changed between two days ago and yesterday where I've had to start moving grounds around and seem to have worse problems than when I started.

[Chuck H]

Wow! I've been trying to move so I have been away. I had no idea my input would be confusing or end up as the main topic of the thread. That said... The guys here had my back pretty well I think. Basically you have grounds in the amp at the circuits and grounds that the tank uses that are to chassis from the shields. The reverb send and recovery circuits do not typically share a power node. You must examine the circuit for grounds that are redundant via the tank. It's actually very simple. As Helmholtz said, you DO need the shields grounded. They only need to be grounded at one end to work. Reverb tanks nearly always share the shield and signal grounds. This is the usual place to find a ground loop. A place where the signal ground is redundant because the tank uses the same ground point for both the shield and the signal and is referenced to the chassis but the same signal lead also has a ground reference from the board to the chassis.

[bobloblaws]

Wow! I've been trying to move so I have been away. I had no idea my input would be confusing or end up as the main topic of the thread. That said... The guys here had my back pretty well I think. Basically you have grounds in the amp at the circuits and grounds that the tank uses that are to chassis from the shields. he reverb send aTnd recovery circuits do not typically share a power node. You must examine the circuit for grounds that are redundant via the tank. It's actually very simple. As Helmholtz said, you DO need the shields grounded. They only need to be grounded at one end to work. Reverb tanks nearly always share the shield and signal grounds. This is the usual place to find a ground loop. A place where the signal ground is redundant because the tank uses the same ground point for both the shield and the signal and is referenced to the chassis but the same signal lead also has a ground reference from the board to the chassis.

Now my turn to say Wow. I wish I could say I followed all that. Sorry, I'm sure you are used to talking to people with IQ in double digits.

Basically you have grounds in the amp at the circuits and grounds that the tank uses that are to chassis from the shields.

When you say "grounds at the circuits" do you mean the cathode grounds?

Reverb tanks nearly always share the shield and signal grounds

When you say "signal ground" do you mean the same as "circuit ground"?

The reverb send and recovery circuits do not typically share a power node

But in this case they do. So?

You must examine the circuit for grounds that are redundant via the tank. ........ A place where the signal ground is redundant because the tank uses the same ground point for both the shield and the signal and is referenced to the chassis but the same signal lead also has a ground reference from the board to the chassis

Sorry, this is greek to me. What is "referenced to the chassis"? Connected to the chassis? Again, I don't understand the terminology "signal ground". In my little mind signal and ground are completely separated. I'm hoping I can learn something here but I need it dumbed down a bit.

[nosaj]

Try this link
nosaj

[Helmholtz]

I should add that at least one of the reverb cables needs to have its shield connected at both ends.That's necessary to close circuits, to allow for currents and to make sure the tank shell is grounded.

And using one cable with shield connected only at one end will only work, when send and return grounds are connected at the tank.

[bobloblaws]

Try this link
nosaj

Thanks nosaj. I've studied these concepts but maybe never quite comprehended them in practical terms. It's probably much simpler than it seems but the terminology confuses me. Chuck appears to make a distinction between chassis ground and signal ground (and maybe "grounds in the amp at the circuits"?). But if the circuit uses the chassis as ground (ground reference?) to me it stands to reason that ground is ground and I get lost when I hear terms talking about this ground and that ground and referenced to this ground etc. etc. so again, I feel like I need it explained in practical terms. If I research it I find resources like this https://www.youtube.com/watch?v=GMX2...hannel=PSAudio which is fine but it doesn't help me understand what Chuck is describing.

[g1]

to me it stands to reason that ground is ground and I get lost when I hear terms talking about this ground and that ground and referenced to this ground etc. etc. so again, I feel like I need it explained in practical terms.

I feel the same way, I'll take a stab at trying to give some practical explanation. All currents in the amp must return to the same 'one true ground', where all the electrons are sourced from. In this case, for the secondary side of PT, it is where the negative terminal of the main filter cap meets the HV winding center tap. Paths to this point will seem to be ground to us, but all have miniscule resistances depending on the thickness of the wire or trace. Where you have current and resistance you have voltage. These small voltages along the ground path can create ground loops. Maybe even more audible when you have duplicated grounds but at slightly different voltage. If we could measure nano-ohms and micro-volts it might seem more obvious to us.

[bobloblaws]

I feel the same way, I'll take a stab at trying to give some practical explanation. All currents in the amp must return to the same 'one true ground', where all the electrons are sourced from. In this case, for the secondary side of PT, it is where the negative terminal of the main filter cap meets the HV winding center tap. Paths to this point will seem to be ground to us, but all have miniscule resistances depending on the thickness of the wire or trace. Where you have current and resistance you have voltage. These small voltages along the ground path can create ground loops. Maybe even more audible when you have duplicated grounds but at slightly different voltage. If we could measure nano-ohms and micro-volts it might seem more obvious to us.

Thanks G. I'm OK with that, except maybe "duplicated grounds". I'm guessing that is the same as what Chuck called redundant grounds.

But I'm still scratching my head with
"You must examine the circuit for grounds that are redundant via the tank. ........ A place where the signal ground is redundant because the tank uses the same ground point for both the shield and the signal and is referenced to the chassis but the same signal lead also has a ground reference from the board to the chassis"

Signal ground. In the video I linked in post #39 the instructor explains how signal ground doesn't require an actual path to the earth, all well and good, battery is perfect example. He says ground/signal ground is "the lowest (huh?) point in a circuit that the designer has deemed ground". Even that is a little confusing as it seems completely arbitrary, but in practical terms I can accept that by convention it is the negative terminal of the battery, and in our case I'll go with your designation of the reservoir cap/PT CT.

So parsing his words in reverse fashion. Chuck is talking about a signal lead (e.g. a wire from reverb tank output transformer to recovery tube grid?) that has "a ground reference from the board to the chassis", which I would take to mean R26 in my V1B example which "references" the reverb return signal lead to the eyelet board ground rail that is ultimately connected to the chassis at the first input jack as I described in post #34. But he says there is a redundancy if there is also "A place where the ... tank uses the same ground point for both the shield and the signal and is referenced to the chassis. And that's where I'm lost. It sounds like he is saying the signal lead and shield are tied together at the same (ground) point and I have know idea why it is stipulated "referenced to the chassis". Is this to say the ground point is connected directly/physically to the chassis as opposed to via a ground rail, meaning the term "referenced" is used differently than when the chap in the signal ground video talks about referencing something to ground via a resistor?

Or am I misreading in the sense that when Chuck talks about examining the circuit he means not just the amp circuit but the tank as well? So how could it be that the tank (on it's own) use the same ground point for both shield and signal? And "referenced to the chassis" meaning? Tank chassis? Amp chassis via the shield? Either? If this is what he means is it just another way of saying get rid of R26? What about C13? For what it's worth I've experimented with removing those from the circuit with no success. I guess R23 would fall into that category as well.

Thanks to anyone everyone that is attempting to enlighten me, I realize it is an uphill battle!

[g1]

I think he may have meant the tank chassis, but I don't want to muddy anything up so I'll just give a side explanation. I'll try not to use 'chassis'.

The tank jacks for this particular tank both have their 'sleeve' (ground) contacts connected to the tank shell. Other tank schemes will have only one, or none, connected (by using insulating washers).
So, for this tank, you have a direct connection (the shell) from tank 'in' jack sleeve to tank 'out' jack sleeve. If both cables also have ground wires to the tank jacks, you have a redundant connection via the tank shell.

You can disconnect the ground wire at either the in or out jack, and the circuit will still get it's ground from the other end, through the tank shell.

[bobloblaws]

I think he may have meant the tank chassis, but I don't want to muddy anything up so I'll just give a side explanation. I'll try not to use 'chassis'.

The tank jacks for this particular tank both have their 'sleeve' (ground) contacts connected to the tank shell. Other tank schemes will have only one, or none, connected (by using insulating washers).
So, for this tank, you have a direct connection (the shell) from tank 'in' jack sleeve to tank 'out' jack sleeve. If both cables also have ground wires to the tank jacks, you have a redundant connection via the tank shell.

You can disconnect the ground wire at either the in or out jack, and the circuit will still get it's ground from the other end, through the tank shell.

OK, that is a much simpler explanation, but still, "If both cables also have ground wires to the tank jacks," I don't know what you mean. Any spring reverb system I've seen, including this one, uses simply a 2 conductor RCA cable with the sleeve as ground and the center pin as signal. How could an RCA cable possibly have an additional ground wire to the tank jack?

[g1]

Any spring reverb system I've seen, including this one, uses simply a 2 conductor RCA cable with the sleeve as ground and the center pin as signal. How could an RCA cable possibly have an additional ground wire to the tank jack?

Not an extra ground, I'm referring to the second conductor per cable. Where you have both tank jack sleeves connected, one of the cables can be single conductor (no ground).
Sort of like the classic Fender footswitch, one of the rca jacks is single conductor, no wire connected to sleeve.

[bobloblaws]

Not an extra ground, I'm referring to the second conductor per cable. Where you have both tank jack sleeves connected, one of the cables can be single conductor (no ground).
Sort of like the classic Fender footswitch, one of the rca jacks is single conductor, no wire connected to sleeve.

OK, I see what you mean. In any case, if all Chuck is saying is that if the hum I'm describing is due to both the input and output cables both having their shields attached to the reverb tank shell that the solution is to disconnect one of those shields from the tank, surely we've ruled that out as a possibility by unplugging the input cable and observing no change in the hum in question? I asked that same question in #24 and I know you have been skeptical yourself that it has anything to do with the tank.