Strange Vox AC30 problems

[Jimmy74]

Ok the center hole of V12 is just a junction for the various shields and the ground wire which goes to the reverb board.... which is grounded to the preamp board via 2 shielded wires. Therefore the center hole of V12 is not grounded at all to the chassis if there are no wires attached to it or if I lift the wire going to the safety earth point. I will separate the reverb section from the junction on the preamp board and attach the shielded wires right to point B, that way reverb ground does not have any junctions with the preamp board ground and goes straight to point B. I have tied the HV CT right to the negative lug on the main filter cap and passes a wire from there to point B. I will do the same with the preamp filter caps and the reverb HV cap. I'll leave the heater CT and power transformer shield grounded at the same place for now and see what happens. Do you still advise using the input jacks ground source as true ground and not the main safe ground point?

thanks

[Jimmy74]

Ok I made point B tying together at this point:

- the whole preamp board
- the whole reverb section
- negative lug of the main filter cap
- negative sides of the two 32+32uf preamp caps

I made the true ground point to which the input jacks and Point B are attached to.... on the mounting screw of V4.
I put the power tubes, V5 and V12 in and turned the amp on and I am still getting hum. I will proceed to taking out the junction on the center hole of V12 and attaching the reverb footswitch & tank shields right on the reverb board ground junction as is everything else in the reverb circuit. And I will also proceed to attaching the bias section ground to point B and also the 47uf reverb HV cap.

So this is what it will look like:
http://www.flickr.com/photos/7383552...28676/sizes/l/

keeping my fingers crossed.

The OT secondary common is attached at a junction to chassis + speaker

The wire from the rectifier connects at a junction point, on the output circuit board, where the choke is connected and a wire then follows through to the main filter cap.

thanks

[R.G.]

Ok I made point B tying together at this point:
- the whole preamp board
- the whole reverb section
- negative lug of the main filter cap
- negative sides of the two 32+32uf preamp caps
I made the true ground point to which the input jacks and Point B are attached to.... on the mounting screw of V4.
I put the power tubes, V5 and V12 in and turned the amp on and I am still getting hum.

ARGGGHHH!!!!!!

I will proceed to taking out the junction on the center hole of V12 and attaching the reverb footswitch & tank shields right on the reverb board ground junction as is everything else in the reverb circuit. And I will also proceed to attaching the bias section ground to point B and also the 47uf reverb HV cap.

That's getting there.

keeping my fingers crossed.

Me too. I'm going to feel really silly if this doesn't get it. We've eliminated so much as the possible causes, it has to be in here somewhere.

The OT secondary common is attached at a junction to chassis + speaker

That ought to be fine since this amp does not use feedback from the outputs.

The wire from the rectifier connects at a junction point, on the output circuit board, where the choke is connected and a wire then follows through to the main filter cap.

That's a bit of a problem. The rectifier wire needs to go directly to the + pin of the first filter cap. But the inserted noise should be well under the ripple.

[Jimmy74]

Well I did take out the junction on the center hole of V12, soldering the shields to ground right on point B, passed the biar grounding to point B too. BUT I'm still getting hum plus now when I turn the reverb pot up, out comes another hum much worse than it was before, maybe I should this junction of shields to the ground junction on the reverb board?

I don't want to alarm myself or you, but I have heard that the PCB boards that Rose Morris used in Vox AC30 were of very poor material and not compliant to voltages or heat found in tube amps, do you know if this is true?

thanks

[Jimmy74]

I went forward and attached the reverb shield junction to the shield and ground junction on the reverb board(which all follow straight through to point B. I also rewired the HV off the rectifier right to the first filter cap together with output transformer CT and one wire from the choke..... but nothing, I'm still getting that same old lovely hum.... where else do I look?

thanks

[R.G.]

Well I did take out the junction on the center hole of V12, soldering the shields to ground right on point B, passed the biar grounding to point B too. BUT I'm still getting hum plus now when I turn the reverb pot up, out comes another hum much worse than it was before, maybe I should this junction of shields to the ground junction on the reverb board?

Wow.
Yes, put it back to the reverb board. One fundamental precept of engineering is that if reality conflicts with theory, reality wins.

I don't want to alarm myself or you, but I have heard that the PCB boards that Rose Morris used in Vox AC30 were of very poor material and not compliant to voltages or heat found in tube amps, do you know if this is true?

I don't know. I've actually had an ongoing campaign to get myself an AC30, but never ran into one I could afford, and I've never repaired one of that era, so maybe.

However, there has been so much baseless condemnation of PCBs that I have this sensitive spot that says "prove it!" when someone says they're wholesale bad. It's a personal quirk I try to watch, as it impedes my judgment.

If the PCBs in the amps are actually of insufficient stuff, they would show it physically. The PCBs would be physically bent and warped, there would be cracks, burnt areas, and other visible signs of deterioration. Perhaps traces lifted from the substrate. In general, paper/phenolic would be substandard. PP is usually light brown, and opaque. Glass epoxy is generally light greenish and may be quite translucent. Glass epoxy is resistant to higher temperatures, and is in general sturdier stuff.

Both are marginal in the stock 1/16th inch (0.062") thickness. Glass epoxy gets really useful in 0.093" to 0.125" thickness. At 0.125, it gets into the range of structural materials.

I went forward and attached the reverb shield junction to the shield and ground junction on the reverb board(which all follow straight through to point B. I also rewired the HV off the rectifier right to the first filter cap together with output transformer CT and one wire from the choke..... but nothing, I'm still getting that same old lovely hum.... where else do I look?

I'm going to feel really stupid if I've led you down the garden path on this. I know I'm not supposed to say that, but it's true.

Lemme think back:
- we isolated the hum source to V12B by cutting off all signal at tubes before and after V12B, one by one. The only thing which killed the hum dead was grounding V12B's grid.
- grounding a grid forces the input signal to a tube to be zero. It does not clamp the plate, so power supply issues would still come through. So it's not the plate circuit of V12B, it's in the grid circuit.
- the grid of V12B is fed from three places: two input channels and the reverb recovery circuit on V12A. Grounding those outside the mixing resistors did nothing. Again, it's inside the mixing resistors and before the plate of V12B.
- I took that last plus your note that you'd done tube swapping to indicate that there was something odd about the way tube V12 is wired/hooked up to ground. So we went off chasing grounds. It's beginning to look like that may have not been it either. You've done a ton of grounding work, and no help has come from it. Rats.

Reverb circuits in general are prone to hum, because you're adding another high gain amplifier stage by running a signal out of the chassis, doing something to it to drop the signal level, then running it back into the chassis and amplifying it. That's a fruitful situation for generating hum. But shorting the reverb recovery pot should have eliminated the hum if it was coming back in on the reverb return cable.

The mixer is another place that's a good source for hum, because of that 3.3M resistor running the impedance of the grid circuit up. It makes it prone to stray hum pickup. Could be wire placement of that wire. Good wiring practice says to put the mixing resistors as close to the tube socket grid pin as you can get them. I can't tell from this distance where they actually put them. So that's a maybe.

Routing of the heater power wires is another remaining possibility, as is stray emission from the heater. These are down the probability list, which is why I didn't suggest them earlier.

Try this. As you know, tube amps run down a little bit after the AC power is turned off. Bring the amp up, get it set up so the hum is clearly audible, and listening carefully, flip off the AC power switch. Does the hum quit the instant the AC power goes off or does it continue a little bit as the power runs down in the DC filter caps?

If I had the amp, I'd follow where the AC heater leads run to see if they come near that high impedance grid, and possibly patch in either DC heater power to the V12 position (or all of them), but I have external power supplies that makes this easy. I might also experiment with elevating the heater CT to some positive DC voltage if the DC heater setup caused the hum to quit. These last are pretty difficult for someone without a bench full of instruments to do easily. Not because they can't, but because it's much more involved.

From here on, the work is getting more difficult to do.

[Jimmy74]

Well I took out the heater hum trim pot and decided to ground the heater CT, but that didn't seem to change anything and as I have already gone through the entire heater chain, which are all twisted together and are far away from the grid on V12 (which is the last tube in the heater chain in the power/output part of the chassis). When I turn the amp on the hum starts to come out once bias starts to reach the power tubes, say after 10 seconds ( I don't have a variac) but the hum does disappear as soon as I switch the amp off (sometimes with a pop). All I would need to convert AC to DC is 4 normal 1N4007 diodes for the rectifier with 2 smoothing caps(would 2200uf 25v be ok?) right?

There don't seem to be any obvious signs on the boards anywhere, the weakspot I heard about was actually where the bias resistor is, and the board would completely burn up from the heat created by that resistor. I might want to try getting the bias cap and resistors off the board and placing them elsewhere just for safety even thought that may not affect the hum. The 3.3M resistor is actually as close as possible to pin 7 on V12.

Maybe a silly thought, but would it be possible to completely exclude the reverb section (V12 and V11)? Just to make sure the hum isn't coming from the power supply

thanks

[R.G.]

Well I took out the heater hum trim pot and decided to ground the heater CT, but that didn't seem to change anything

OK. Keep that in mind, as moving the heater CT is oen thing we may try later.

I have already gone through the entire heater chain, which are all twisted together and are far away from the grid on V12 (which is the last tube in the heater chain in the power/output part of the chassis).

That's good.

When I turn the amp on the hum starts to come out once bias starts to reach the power tubes, say after 10 seconds...
but the hum does disappear as soon as I switch the amp off (sometimes with a pop).

That simple test tells you that the hum is coming from the AC power line into the amp, not picked up from outside the amp or in external ground loops, etc. It could still be heater hum, AC power line leakage, or power supply ripple, but it's from inside the amp.

All I would need to convert AC to DC is 4 normal 1N4007 diodes for the rectifier with 2 smoothing caps(would 2200uf 25v be ok?) right?

DC heaters are more complicated than that. Yes, you could make DC that way, but you lose two diodes of voltage, about 1.4V, with a bridge rectifier, from the peak voltage. 6.3Vac for heaters (if that's what your PT actually produces) is 8.91V peak minus 1.4V in two diodes or 7.5Vdc. That's too much for your heaters. You'd need a dropping resistor, unless your heater winding sagged enough to drop off the other 1.5V; maybe you could stick in more diodes to lose some voltage. Or it could sag too low. Making DC heater voltages out of 6.3V heater windings is tricky.

It's probably better to find a way to make a +12V to +20V DC volt source, and tie the heater CT there.

Let's explore other stuff first.

There don't seem to be any obvious signs on the boards anywhere, the weakspot I heard about was actually where the bias resistor is, and the board would completely burn up from the heat created by that resistor.

Yeah, there's a lot of heat coming out of that cathode resistor. But if the board looks OK there, it probably is OK. There would be burn marks.

I might want to try getting the bias cap and resistors off the board and placing them elsewhere just for safety even thought that may not affect the hum.

Good idea for later.

The 3.3M resistor is actually as close as possible to pin 7 on V12.

OK. That's not it.

Maybe a silly thought, but would it be possible to completely exclude the reverb section (V12 and V11)? Just to make sure the hum isn't coming from the power supply

That's not a silly idea, and it's worth trying. I would first just pull out V11. Hum still there?
Then I'd ground the grid of V12A by shorting the reverb footswitch. That prevents V12A from getting any signal, but leaves any power supply hum on its plate.

[Jimmy74]

just to answer your last question, all the tests we've done so far were done with V5 (PI) - V12 (reverb recovery) and the power tubes fitted in so V11 was already pulled out.

[Jimmy74]

Instead of using 3 x 5watt 47 ohm resistors for bias would a 3 watt 150 ohm resistor take the load or does it have to be 5 watts? Also the footswitch shielded wire is attached to the same place as the reverb return shielded wire, how exactly should I ground the grid for V12A?

thanks

[R.G.]

I called up a couple of my amp-tech friends. The consensus is that "yep, those reverb AC30s are humming *#$^*&'s."

Here are some tidbits of advice.
- the *exact* place where the CT of the heaters is grounded makes a difference, even if properly star grounded. It's worth messing around with moving the heater CT to different places with the one wire that you now have to ground it, moving that to different places that ought to be "ground".
- the *exact* balance of the heater CTs makes a difference in some amps. Resistor CTs are better than winding CTs for this. Maybe that hum balance pot was a good thing to leave in.
- The *exact* wiring and location of the reverb tank wires makes a difference. Some tanks have the send side wire either not shielded and a single wire to the tank, on the theory that the tank shell and receive side is grounded by the return wire. Some have the send wire shielded and shield connected one end, some with shield connected both ends. Ditto on the return wires. This is usually determined by experiment in the particular amp model. If the return wire runs near the PT, it will pick up hum. Period.

Three times 5W resistors for the bias resistor is a good idea. It spreads the heat out, and keeps the temperature at any one point down. Bigger wattage resistors is OK.

Also the footswitch shielded wire is attached to the same place as the reverb return shielded wire, how exactly should I ground the grid for V12A?

In a perfect world, those would go to the "ground" side of the V12A cathode resistor.

[Jimmy74]

I called up a couple of my amp-tech friends. The consensus is that "yep, those reverb AC30s are humming *#$^*&'s."

Here are some tidbits of advice.
- the *exact* place where the CT of the heaters is grounded makes a difference, even if properly star grounded. It's worth messing around with moving the heater CT to different places with the one wire that you now have to ground it, moving that to different places that ought to be "ground".
- the *exact* balance of the heater CTs makes a difference in some amps. Resistor CTs are better than winding CTs for this. Maybe that hum balance pot was a good thing to leave in.
- The *exact* wiring and location of the reverb tank wires makes a difference. Some tanks have the send side wire either not shielded and a single wire to the tank, on the theory that the tank shell and receive side is grounded by the return wire. Some have the send wire shielded and shield connected one end, some with shield connected both ends. Ditto on the return wires. This is usually determined by experiment in the particular amp model. If the return wire runs near the PT, it will pick up hum. Period.

Three times 5W resistors for the bias resistor is a good idea. It spreads the heat out, and keeps the temperature at any one point down. Bigger wattage resistors is OK.


In a perfect world, those would go to the "ground" side of the V12A cathode resistor.

Ok the heater CT is grounded to chassis together with the power transformer shield wire, they are bolted to the chassis on one of the power transformer mounting screws.

The heater hum pot did work but when I tried turning it full either way it would introduce another hum which my guess was unbalanced heater hum, it did not in anyway change the hum I'm getting now.... so I pulled the pot out and decided to ground the heater CT together with the power transformer shield.... which didn't seem to change anything anyway.

The send/return reverb cables are both shielded but only the return cable has the shield grounded to the reverb board which then follows to point B. They are both the original cables with the old type RCA plug. These cables are situated on the output transformer side of the chassis therefore as fas as possible from the power transformer.

I did decide to make a small separate board for the bias parts ( 3 x 47ohm resistors 5W + 1 x 220uf 25v cap) and will fit it in the only empty space that is left between the main filter cap and the reverb board. One question for this, should the bias cap be in parallel only to the first 47ohm resistor or to all 3?

The shield of the reverb footswitch and the reverb return cable are tied together and go to the reverb board ground to point B ( as shown in the last schematic I posted up) so all I hae to do is short pin 2 on V12 to ground?

thanks

[R.G.]

Ok the heater CT is grounded to chassis together with the power transformer shield wire, they are bolted to the chassis on one of the power transformer mounting screws.

We've just about exhausted the systemic stuff by now. So from here on, it's going to be much more experimental. There are zillions of ways for hum to get in.

Take that heater CT wire loose from the chassis where it's connected. Connect it to
(a) the local star point on the V12 tube socket
(b) the major star point on V4.
Does the hum even change?

The send/return reverb cables are both shielded but only the return cable has the shield grounded to the reverb board which then follows to point B. They are both the original cables with the old type RCA plug. These cables are situated on the output transformer side of the chassis therefore as fas as possible from the power transformer.

Get out the ohmmeter. Find out where the shield/ground signal goes in the reverb tank.
(1) Probe the tank shell with one ohmmeter lead, and the reverb board ground point where the shield attaches. Is this continuous?
(2) Is one side of each of the transducer coils inside the tank connected to ground?
(3) Is the shield on the input wire connected to ground somehow?
(4) Are the RCA jack connectors clean and tight?

I did decide to make a small separate board for the bias parts ( 3 x 47ohm resistors 5W + 1 x 220uf 25v cap) and will fit it in the only empty space that is left between the main filter cap and the reverb board. One question for this, should the bias cap be in parallel only to the first 47ohm resistor or to all 3?

Bias cap needs to parallel the whole string of bias resistors. The cap goes from cathodes of the output tubes to ground. It's job is to make the resistance of the bias resistors invisible to the tubes. Cooking any cap with heat from the bias resistors is a Bad Thing. This will eventually kill the caps. The resistors need to go somewhere where they can't heat capacitors.

The shield of the reverb footswitch and the reverb return cable are tied together and go to the reverb board ground to point B ( as shown in the last schematic I posted up) so all I hae to do is short pin 2 on V12 to ground?

Yes.

[Jimmy74]

I put in the new bias board paralleling the bias cap along the whole bias resistor chain (that is one thing that was wrong up until now, the bias cap was in parallel only to the first resistor).

I grounded pin 2 on V12 and I am still getting hum.

I will forward to doing the other tests now

thanks

[Jimmy74]

We've just about exhausted the systemic stuff by now. So from here on, it's going to be much more experimental. There are zillions of ways for hum to get in.

Take that heater CT wire loose from the chassis where it's connected. Connect it to
(a) the local star point on the V12 tube socket NO CHANGE
(b) the major star point on V4. NO CHANGE



Get out the ohmmeter. Find out where the shield/ground signal goes in the reverb tank.
(1) Probe the tank shell with one ohmmeter lead, and the reverb board ground point where the shield attaches. Is this continuous?YES
(2) Is one side of each of the transducer coils inside the tank connected to ground?
(3) Is the shield on the input wire connected to ground somehow?
(4) Are the RCA jack connectors clean and tight?Yes the reverb works so that means that the connections are fine

I'll check points 2 and 3 tomorrow.

Maybe a silly thought, but would it be possible to completely exclude the reverb section (V12 and V11)? Just to make sure the hum isn't coming from the power supply

That's not a silly idea, and it's worth trying. I would first just pull out V11. Hum still there? Then I'd ground the grid of V12A by shorting the reverb footswitch. That prevents V12A from getting any signal, but leaves any power supply hum on its plate.

What have I found out if I'm getting hum after this has been done? (which is actually what the reverb footswitch does)