AA864 Musings Help Needed

[TexasTone]

Customer has an amp, the AA864 Single Channel, as bydwinstoodwood and SLucky.  The only changes are a switchable light/heavy NFB and a 1.2k Rk on V2-A.

I checked all the voltages, and everything was spot-on, and we had a clean 40+ Watts, and the amp sounded loud, clean, excellent.  In fact, it was so strong, I used a 220k in front of the 500k Gain control to tame it a bit.  So, I tacked one in line.

He was playing a few SRV licks, and then all of a sudden - "BLLLRRTTT" - very loud!

So, here's where I am. 

On powering it on for troubleshooting, there was a noticeable 120/240Hz hum.  I found the ground lug for the power ground - HV and Heater center taps, can cap, bias, and output grounds, everything but preamp ground - was loose, so I tightened it.   I started removing tube and found that removing V1 and V2 made no difference in the hum.  Removing V3 silenced everything.  I moved the ground for the V3 tail, AC shunt, and NFB from the preamp ground to the power ground.  This had no effect on the hum.

I put all the preamp tubes back in, along with new 12AT7 in, replacing the EHX with a JAN Philips 12AT7WC.  The amp played, but there was still a hum, and now the entire chassis was microphonic!

• V1 - JJ ECC83
• V2 - EHX 12AY7 (put in to tame the gain a bit.
• V3 - JAN 6189/12AU7 (this was a troubleshooting step due to a noticeable 120/240 Hz hum)
• RESULT - Non hum, but the amp sounded very compressed, and not clean.

• I subbed a tested food JJ ECC83 for V2.  Seemed more compressed.
• I subbed a tested food JJ ECC83 for V3.   Very microphonic
• I subbed an EHX 12AY7 for V3, and V2 and V3 were very microphonic.
• I subbed an EHX 12AT7 for V3.  Pronounce hum, and V2 and V3 were very microphonic.
• Put the 12AU7 back in V3 and no hum.  Amp not playing as loud, and very compressed sounding.
• I bypassed the 220k resistor.  Low hum, but compressed and raspy sound.
• The only way to get rid of the 120/240 Hz hum was to put a 12AU7 in V3.  Not a real solution.
• Putting known good ECC83 in V1 and V2, and a 12AY7 in V3, resulted in the whole chassis being microphonic
• Putting three known good ECC83 tubes in sounded good, with little hum, but both the volume and gain controls would pop loudly when turned down below the 9 O'Clock position (basically turning them off).

This had me a bit flustered at the moment, especially the 120 and 240 Hz hum, the entire chassis being microphonic, and the loud popping on the volume and gain.  I'm using the "Spectroid" Android app to measure the hum.  Noticeable in -30dB, quiet is -60dB.

I'm suspecting perhaps a bad filter cap.  This amp has a JJ 40-20-20-20.

Any help is appreciated. 

Thanks.

[Willabe]

I found the ground lug for the power ground - HV and Heater center taps, can cap, bias, and output grounds, everything but preamp ground - was loose, so I tightened it.

That's not good. That might have blown/stressed something.

Is that ground lug on 1 of the PT bolts?

It's common for a PT bolt to loosen if it has a ground lug under it. That's why you shouldn't use the PT bolts for the ground lug. It should have it's own dedicated bolt with a star washer on both sides of the chassis, inside/outside. And you use 2 nuts and lock them together with 2 wrenches. Tighten the 1st nut down then tighten the 2nd nut down onto the 1st, lock them together with the 2 wrenches. And use a drop or 2 of lock tite. That's a solid ground now that wont loosen.

there was a noticeable 120/240Hz hum.

240Hz is just an octave of 120Hz, kill the 120Hz most likely the 240Hz will be gone too. We look at/listen for 60Hz/120Hz.   
 

I put all the preamp tubes back in, along with new 12AT7 in, replacing the EHX with a JAN Philips 12AT7WC.  The amp played, but there was still a hum, and now the entire chassis was microphonic!

Naw, it's not the chassis, it's a microphonic tube. Tubes in a socket mounted in the chassis, bang the chassis it goes through the socket to the tube.

   I subbed a tested food JJ ECC83 for V2.  Seemed more compressed.
   I subbed a tested food JJ ECC83 for V3.   Very microphonic

What does "tested food" mean? Spell check or something?

That JJ ECC83/12AX7 seems to be bad.
   

Putting three known good ECC83 tubes in sounded good, with little hum, but both the volume and gain controls would pop loudly when turned down below the 9 O'Clock position (basically turning them off).

This had me a bit flustered at the moment, especially the 120 and 240 Hz hum, the entire chassis being microphonic, and the loud popping on the volume and gain.

I'm not sure what this means. You turn the pot and at a certain point the amp pops and no sound? Modern pots often when you get to around 9:00 you get no sound, but they don't pop. 

I'm suspecting perhaps a bad filter cap.  This amp has a JJ 40-20-20-20.

Measure at each cap for dcv and post those readings. Should be posted as filter cap; node A, node B, node C, node D. 

Then discharge the cap cans caps and measure each cap 1 at a time for resistance to ground. And post those 4 numbers. 

[Willabe]

You seem to have at least 1 microphonic tube that you think is good.

Swapping in different tube types, 12AY7, 12AU7, etc is not going to fix the hum, your wasting your time.

Get a good set of 12AX7's and a good 12AT7 and use those and those only until you fix the problem.

[Willabe]

Customer has an amp, the AA864 Single Channel, as by dwinstoodwood and Slucky.

Which channel? Bass or normal?

[SEL49]

Schematic may be useful...

https://sluckeyamps.com/misc/AA864_Musings.pdf

[TexasTone]

Is that ground lug on 1 of the PT bolts?

No.  There is one stand alone ground lug for power grounds (this screw was loose). 

• The 4/20/20/20 cap can
• Output jack ground (output jacks are isolated)
• HV center tap
• Heater center tap
• Bias circuit ground
• PT shield
• Standby switch shunt cap
• LTPI (moved from preamp ground for troubleshooting

All the preamp, input jack, and control grounds tie to one lug near V1.  The input jacks are isolated.

240Hz is just an octave of 120Hz, kill the 120Hz most likely the 240Hz will be gone too. We look at/listen for 60Hz/120Hz.

I measured the hum with the Android Spectroid app.  120Hz measured about -30dB, 240Hz was slightly less.  "No hum" meant that the 120Hz was at -65dB and not noticeable.
 

Naw, it's not the chassis, it's a microphonic tube. Tubes in a socket mounted in the chassis, bang the chassis it goes through the socket to the tube.

What I mean is that the entire chassis sounds 'live', almost resonant.  Touch any component, any part of the chassis, and it goes live.

What does "tested food" mean? Spell check or something?

Yes. Tested good.

I have about a dozen NOS JAN Sylvania 12AT7 tubes and maybe a half-dozen new Eurotubes tested 12AX7 tubes.

I'm not sure what this means. You turn the pot and at a certain point the amp pops and no sound? Modern pots often when you get to around 9:00 you get no sound, but they don't pop. 

This means that when you turn down the Volume or Gain controls from Noon to Off, there was a loud POP! at about the 9 O'clock position.

Measure at each cap for dcv and post those readings. Should be posted as filter cap; node A, node B, node C, node D. 

When I measured them in circuit (Fluke 177], I got 42.9µF on the 40, 59µF on the first 20, 62.8 on the second 20 (they're separated by a 1k resistor), and when I checked the third 20, the meter kept cycling from 46 to 52.  I'll be taking it out and checking it out of circuit later.

[TexasTone]

Customer has an amp, the AA864 Single Channel, as by dwinstoodwood and Slucky.

Which channel? Bass or normal?

Normal

[Willabe]

       Output jack ground (output jacks are isolated)
   HV center tap .............

   
You all ready told us all that. I just wanted to know if the ground was on a PT bolt.

I measured the hum with the Android Spectroid app.  120Hz measured about -30dB, 240Hz was slightly less.  "No hum" meant that the 120Hz was at -65dB and not noticeable.

Again, you already told us how you measured. And I told you to forget about the 240Hz. Kill the 120Hz and most likely the 240 will be gone too.

And you don't need to measure with that app anymore, you know what the frequency is. And you'll know when the humm is gone by using your ears now.
 

Naw, it's not the chassis, it's a microphonic tube. Tubes in a socket mounted in the chassis, bang the chassis it goes through the socket to the tube.

What I mean is that the entire chassis sounds 'live', almost resonant.  Touch any component, any part of the chassis, and it goes live.

Yes, we get it. It's probably a bad tube.

This means that when you turn down the Volume or Gain controls from Noon to Off, there was a loud POP! at about the 9 O'clock position.

Not sure about this. Sounds like somethings discharging very quickly, like a cap with a charge on it. 

Measure at each cap for dcv and post those readings. Should be posted as filter cap; node A, node B, node C, node D. 

When I measured them in circuit (Fluke 177], I got 42.9µF on the 40, 59µF on the first 20, 62.8 on the second 20 (they're separated by a 1k resistor), and when I checked the third 20, the meter kept cycling from 46 to 52.  I'll be taking it out and checking it out of circuit later.

No, I asked for the DCV at each cap.

[Willabe]

I'm suspecting perhaps a bad filter cap.  This amp has a JJ 40-20-20-20.

Measure at each cap for dcv and post those readings. Should be posted as filter cap; node A, node B, node C, node D. 

Then discharge the cap cans caps and measure each cap 1 at a time for resistance to ground. And post those 4 numbers.

This  ^^^^  above  ^^^^  is what I asked you to measure and post. I did not ask for cap uF measurements.

[TexasTone]

Measure at each cap for dcv and post those readings. Should be posted as filter cap; node A, node B, node C, node D. 

Then discharge the cap cans caps and measure each cap 1 at a time for resistance to ground. And post those 4 numbers.

Initial Testing 12/21/2024:
C1 40µF  "A"   466 VDC
C2 20µF         466VDC
C3 20µF  "B"   453 VDC
C4 20µF  "C"   416 VDC
Node "D"         363 VDC

After discharge, resistance to ground:
All     0.009 VDC
   
C1 40µF "A"   OL
C2 20µF      OL
C3 20µF  "B"   OL
C4 20µF  "C"   OL

[dwinstonwood]

Hello TexasTone,
This amp was my idea. Steve (sluckey) guided me through the build and gave me some great advice. I drew the original schematic and Steve volunteered by drawing the layout and a revised schematic.

It is really nothing more than the Bass channel of the Fender AA864 Bassman with an additional 500K-A "Gain" pot in front of the third triode.

In fact, I was going to build a Trainwreck Express when I realized that they're the really same thing (with some parts value changes).

What you are describing sounds like a layout and grounding issue, and not a circuit problem, assuming the electrolytics and wiring are all good.

Correct me if I'm wrong, but I think you said the input jacks are isolated. If you're using Switchcraft 12A's then ground them to the chassis and ground the preamp to the input jack. One ground point.

I'll have to look at your power amp, but it should all have one ground point, too. If using NFB, ground the speaker jack to the filter cap of the phase inverter which also grounds to the input jack along with the preamp.

Here is a grounding scheme that I now use and is dead quiet with zero hum. Notice that everything from the point of the NFB back is grounded at the input jack. Reservoir cap, Choke, and first power tubes filter cap have their own ground very close to the PT (small ground loop away from preamp).

I've stopped using Heater center tap grounds. I use two 100 Ohm resistors and ground them at the power tube cathode. This makes the heater AC ride on top of the DC power tube cathode voltage (and is a smart thing to do when using a Cathode Follower, too).

This is a great amp. Like you said, brilliant cleans with a lot of body and substance. Get the grounding and lead dress worked out and it's an AA864 BF Bassman with a very usable premap gain control.

David

[Willabe]

If using NFB, ground the speaker jack to the filter cap of the phase inverter which also grounds to the input jack along with the preamp.

Yes, after you run the OT ground fly lead directly to the speaker jacks ground lug, then you run a wire from there over to the PI's B+ filter caps ground lead. But you do not ground that PI filter cap node with the preamp ground.

Merlin, Kevin O'Connor, Randall Aiken, and others, all say, that if the amp has NFB applied to the PI, then you ground the PI B+ filter cap with the power amp ground. Because the PI is now referenced to the OT's ground which is grounded with the power amp ground.

If the amp has no NFB loop, then the PI gets grounded with the preamp ground because it is not referenced to the OT's ground. 

Yes, I know your amp is dead quite and you did the opposite.
 
https://www.valvewizard.co.uk/Grounding.html

https://www.aikenamps.com/index.php/grounding

[TexasTone]

Thank you all for your input.

This amp has a four section can capacitor, 40/20/20/20µF.  I pulled the can and checked all four nodes and resoldered everything.

• The HV CT and the bias supply ground go to the ground terminal of the can.
• The ground terminal of the can goes to a dedicated ground lug via a 2" 18ga wire.
• The OT ground (isolated jacks), heater CT and PT shield go to the ground lug.

C1 sits between the diode HV DC out and the standby switch.
The Standy switch feeds the OT CT and the choke.
C2 sits between the choke out and a 1K 3W resistor to C3.
C3 feeds the power tubes g2 grids via 470Ω 3W resistors, and a 4.7k 3W to C4.
C4 feeds the V3 (PI) and V2.
A 22K 1W after C4 feeds V1.

The Gain ground connects to the Volume ground, which connects to the two (isolated) input jacks, and then to a dedicated ground lug via a 3" wire.

The Mid grounds to the pot.

The PI (with NFB) was grounded to the input ground.  I moved it to the power ground, but no change in hum.  I disconnected the NFB and no change.

I've used 6 different tubes in the PI with no change to the hum.

I chop sticked pretty much everything.  The two power tube heaters are in phase and are the three preamp tubes. 

Right now, there are no tubes in V1 and V2 and the 120Hz hum is pronounced.

At this point I'm wondering if there's an under-board wiring issue, e.g. bias wire crossing the power tube grids.

[Willabe]

So, this amp was working fine with no 120Hzhumm?

Your friend was playing amp made loud noise. When trouble shooting the amp had a 120Hz humm. You checked the amp for voltages, they were good, but found the power ground was loose. You tightened the ground, but now the humm remains.

It seems like that loose ground power lug being loose and losing that ground connection while playing through the amp, something got stressed/shocked pretty badly? If so, you have to find what part it is. 

Have you tried a different set of power tubes? Tube could have gone bad.

This amp has a four section can capacitor, 40/20/20/20µF.

As far as you saying nothing was changed except for a switchable light/heavy NFB and a 1.2k Rk on V2-A.

That's not true, who ever built the amp used a 4 section cap can and changed the grounding some. Sluckey's layout should have been followed exactly, as it was a proven build. That cap can messed up the grounding a little. But, if it was working fine before, ok. They got away with it.

The PI (with NFB) was grounded to the input ground.  I moved it to the power ground, but no change in hum.  I disconnected the NFB and no change.

I only brought this up about the PI/NFB ground because of what dwinstonwood posted about it.

[TexasTone]

Latest update.  Issue is isolated to V2.  There are only four components attached to V2.

Pin 6 is the plate, with a 100k load resistor and the LTPI coupling cap.
Pin 7 is the grid, which comes from the Gain control.
Pin the cathode, connected to a 1.2k resistor to the preamp ground buss.

BYPASS V2
1.  Desolder and cable from Gain pot wiper (that goes to V2-7).  Ground the desoldered cable end.
2.  Desolder C8 (LTPI coupling cap) from the BLU plate lead from V2-6.
3.  Connect GAIN control wiper to the front end of C8.
4.  V2 is now bypassed, and the grid of V2 (pin 7) is grounded.

Power on, both Gain and Volume at "0".
RESULT:
1. No hum, just an initial quick buzz when the amp was switched from Standby to Run.  No sound from speaker.
2.  Turn Volume to 10 O'clock.  No pop.  No sound from speaker.
3.  Turn Gain to 10 O'clock.  No pop.  No sound from speaker.
4.  Turn up guitar volume.  Amp plays great.  Except for the guitar plating, the amp is dead silent; you wouldn't even know it was turned on.  SOUNDS PERFECT!

Therefore, I can only see the following as causes.

1.  Heater phase wiring.  All three preamp tubes are wired in parallel, along with the push-pull power tubes.  Since V2 has only one preamp section, the signal coming from V2 is inverted only and not reverted to normal.
2.  Ck should be grounded to power amp buss instead of preamp buss.
3.  Bad tube socket.  The amp had Belton 9-pin top mounts.  I find this the least likely.
4.  Bad resistor.  The plate is a 100k Vishay PRO2 2W metal film (100.9 kΩ), and the cathode is a 1W carbon film 1187 Ω.  So, not likely.

I plan to revert the wiring to put V2 back in circuit, and the test items 1 & 2.

I could be wrong, so I appreciate all the input!

[TexasTone]

I made the following changes, with the listed results.
1. Moved V2 Rk from preamp ground to power amp ground.  The 120 Hz hum increased 6 dB, so I reverted it back to the preamp ground.
2. Reversed the heater wires coming from V3 (LTPI) to V2, leaving the wiring to V1 alone, i.e. V3 and V1=GRN on 4 & 5, BLK on 9.  V2 pin 9 = GRN from V3 4&5, BLK to V1 9, and pins 4 & 5 = BLK from V3-9 and GRN from V3-4&5.  This made the hum much louder by 12dB.

All changes have been reverted.

With V2 completely bypassed, the amp has zero hum and/or hiss and sounds great.  When bypassed. the only parts not used are the tube itself, Rk and Ra, and the grid is grounded.
With V2 in circuit, the hum is pronounced and the Gain and Volume both pop loudly when turned up.

I've used at least a dozen tubes and while they contribute varying amounts of hum and microphonics, they all have the same issues, so I'm left with what?

Both V2 Rk (100.9 kΩ) and Ra test 1187 Ω good, so, the only thing I haven't tried is replacing the tube socket itself, but maybe there's something else going on.  The hum and pop are obvious grounding issues.  Perhaps moving the Gain pot ground or mid resistor ground (current to the mid pot) will help.

Thanks for all the advice!