1974 Super Reverb - Channel Bleed

[410_Fanatic]

Hi, hope every one is well!

My 74 Super Reverb is passing signal when volume is at 0 and produces a distorted sound when playing at low volumes.  I suspect a grounding issue, but also noticed that if I pull V1 while playing on the vibrato channel, the issue goes away.  Similarly, if I pull V2 while playing the normal channel the problem is gone.  I suspect the sound that passes is actually bleeding through the power supply or ground to the other channel but so far no luck isolating it.

I have done some diagnostics to the limit of my knowledge and am looking for some pointers on where to look next.  Here is what I have tried:

• Tried with multiple tubes in V1 and V2
• Tested the pot, it only goes to ~1-2 ohms, not all the way to 0
• Tested all the preamp grounds to chassis - I get 0.2-0.3 ohms which is the same value I get when putting the lead together
• Replaced the shared cathode bypass cap on V1B & V2B
• Tack soldered known good 20uf filter caps across the preamp filter caps
• Tested the plate and cathode voltages for V1 and V2, all triodes are hovering around 270v for the plates and 2v for the cathode

The amp has had the power amp reverted to AB763 specs but is otherwise an AA270.  Here is a pic of the circuit board around the preamp in case there is anything obvious, but I would be happy to supply more pictures if needed.

It looks like my plate voltages on V1 and V2 are higher than the AA270 schematic.  Its also a bit of a mess in there, somebody before me burned the hell out of a lot of things.

Any ideas on where to look next?  Do amps bleed through the heater wires?

[TitaniumValhalla]

The board looks waxy. Have you checked if it has become conductive?

[tubeswell]

Check the pre-amp filter cap. From your description of the problem, it would seem to be open circuit/not grounding properly. You can tell by measuring the power rail VDC. What we're looking for is a step-change VDC at each separate filter cap node. But if the pre-amp filter cap voltage is the same as the next filter cap in the power rail, with all tubes etc installed, that would tell you that the power rail isn't decoupling properly, which could explain the 'mushiness' (i.e., signal being 'fed back' through the power rail between the different pre-amp stages, instead of being separately decoupled at each power supply node).

[410_Fanatic]

Good call - it's one of those wax covered boards from the mid-seventies, I turned the amp on and probed the board a bit, the most I saw was about 75 millivolts between plate resistors.  I am not sure if that is high or not.

Regarding testing the filter cap, is it okay to measure without power tubes?  If so, at node C I get 483v and at node D I get 445v.  The resistor between the two is 4.7K, and now that I look it at its a big carbon comp that might be original to the amp.

[tubeswell]

Regarding testing the filter cap, is it okay to measure without power tubes?  If so, at node C I get 483v and at node D I get 445v.  The resistor between the two is 4.7K, and now that I look it at its a big carbon comp that might be original to the amp.

This suggests the caps are decoupling (but how old do they look? Are they bulging? Is there any gunk coming out of the end(s)?)


Other thoughts:
1) channel bleed through open cathode bypass cap on the shared cathode of the channel TS recovery stages
2) DC leakage on the wax board

[410_Fanatic]

The filter caps are Sprague Atoms from 2010, no bulging that I can see.  Not super new though either.

I replaced the shared bypass cap on the TS recovery earlier today, didn't help at all.  Really starting to wonder if it is the board.

I have been on the fence about getting a board from Doug and maybe cleaning this one up a bit.  Might just be the excuse I need to pull the trigger.

[tdvt]

I understood that Fender used wax on the boards in the 70s but I never realized how much was used until I started work on a '79 Pro Reverb just last week.

It appears (at least on the one I am working on) that hot wax was misted on the populated board, coating everything, often building up quite thick when a component was close to the board, as the wax would fill the gap.

It is slow going, but after scraping off the really heavy stuff, I am having the best luck removing it with careful use of a heat gun, then blotting up the molten wax with rags for the more open & lots of Q-tip swabs for all the tight spots.

[Platefire]

Wow, your picture looks just like my 74 Pro Reverb regarding lead dress and  board. My board
Is wavy like the sea I cleaned mine with contact cleaner, alcohol and a lot of wiping. I checked my board for conductivity but couldn't detect any. Got rattling/distorted notes that sometimes mimics a busted speaker. Think I making progress, hope you get to the bottom of yours! Platefire

[410_Fanatic]

I understood that Fender used wax on the boards in the 70s but I never realized how much was used until I started work on a '79 Pro Reverb just last week.

It appears (at least on the one I am working on) that hot wax was misted on the populated board, coating everything, often building up quite thick when a component was close to the board, as the wax would fill the gap.

It is slow going, but after scraping off the really heavy stuff, I am having the best luck removing it with careful use of a heat gun, then blotting up the molten wax with rags for the more open & lots of Q-tip swabs for all the tight spots.

It really is amazing how much wax there is and I see wax on the leads of the components, so agree it must have been waxed after the board was assembled.  From what I've read it is pretty well impregnated in the board as well.

I will try scraping some of it and cleaning with alcohol between the various parts to see if that helps at all.  Fingers crossed.

[410_Fanatic]

Wow, your picture looks just like my 74 Pro Reverb regarding lead dress and  board. My board
Is wavy like the sea I cleaned mine with contact cleaner, alcohol and a lot of wiping. I checked my board for conductivity but couldn't detect any. Got rattling/distorted notes that sometimes mimics a busted speaker. Think I making progress, hope you get to the bottom of yours! Platefire

Absolutely, the contrast between the lead dress in this and a 64 I have is night and day.  That plastic wire is really squirrely and melts by breathing on it, frankly I am amazed Fender employees could work with it at all.

Your description of a busted speaker is pretty close to what I am experiencing, kind of a ratty distortion, but only at low volumes.  At first I thought it was coil rub but there was no difference when plugging into another cab.

[tubeswell]

at node C I get 483v and at node D I get 445v.  The resistor between the two is 4.7K, and now that I look it at its a big carbon comp that might be original to the amp.

Just out of interest, do you also get 483V at node B?

[410_Fanatic]

Just out of interest, do you also get 483V at node B?

I get 503v at node B when measuring without the output tubes (to align with my previous measurements).

I decided to remeasure all nodes with output tubes installed to understand what it looks like with a full load:

• Node A / Reservoir - 449v
• Node B - 446v
• Node C - 436v
• Node B - 397v

[tubeswell]

Just out of interest, do you also get 483V at node B?

I get 503v at node B when measuring without the output tubes (to align with my previous measurements).

I decided to remeasure all nodes with output tubes installed to understand what it looks like with a full load:

• Node A / Reservoir - 449v
• Node B - 446v
• Node C - 436v
• Node B - 397v

Not much of a drop between B and C. What’s the resistor value measure there?

[410_Fanatic]

Not much of a drop between B and C. What’s the resistor value measure there?

It's a 1K.

[tubeswell]

Try tacking another 20uF 500V in parallel across the C node filter cap

[410_Fanatic]

…and I see how little drop there is now.  The schematic is showing a 40v drop and my voltages are much higher than the schematic after that point.

[410_Fanatic]

Try tacking another 20uF 500V in parallel across the C node filter cap

Will do!  I’ll report back once I have a chance to test it out.

Thank you for bearing with me.

[410_Fanatic]

After tack soldering a new cap across the node C filter cap the crosstalk and low volume distortion remains, however all the voltages in the power supply rail dropped about 10v.

Here are my current voltages - the main reservoir and node B voltages are now closer to the schematic:

• Reservoir - 437v
• Node B - 435v
• Node C - 424v
• Node D - 388v

So I am still not seeing much of a drop between nodes B and C, but I was also wondering if the plate load on the PI being higher than the schematic (82K & 100K instead of the 47K in the aa270 schematic) would shift the expected voltage drop here.  But I am only seeing a quarter of the expected drop based on the schematic.