Peavey Ultra 60 distorted overtones (clips attached) - filter caps?

[ACHIEVEIT]

Hello guys,

I recently got a Peavey Ultra 60 tube head. Nothing has been modded/tampered with as far as I can tell (opened and examined the boards). Even has 3 out of 4 original Peavey-branded preamp tubes. V1 (JJ ECC83) and the power tubes (Tung-Sol 6L6GC STR) must be newer.
The problem that I have is that there are distorted overtones (most easily noticed when note is abruptly muted) - there is this distorted trail that's always there. It is particularly prominent in the clean channel but it is also there in the Crunch and Ultra channels (it's just less apparent because of the distortion). I have made some short clips to demonstrate it.

I intend to change the filter capacitors as they are now 30 years old (they are original and the PCB dates back to 1991). I didn't spend too much time troubleshooting but here's the things I've tried already without any difference:
- Swapped all preamp tubes with known-good ones (didn't have a spare 6L6 set to test but it shouldn't be the culprit, see below)
- Visually examined the PCB/components and tested very crudely for any short on all electrolytic and most coupling capacitors
- Relays are energized correctly when I apply external voltage and in any case, setting crunch/ultra channels pre and post to 0 makes no difference (in other words, it is not a channel bleeding issue)
- Bypassed CR1 & CR2 (that act as a 'noise-gate' of sorts) - just for the sake of trying it
- Listening to just the preamp from either the FX Loop send or the Preamp Out exhibits the same behavior, while connecting the guitar to just the FX Loop Return (bypassing the preamp effectively) exhibits it to a much smaller degree. Assuming that it's present in both scenarios (although to a smaller degree in the power amp section), would it make sense to assume it's something that's shared by both -> power supply rail -> filter caps?

Could the filter caps be the culprit or should I be looking for something else?
Thanks for your input!

Guide to clips:
peavey_ultra_noise -> guitar to high input
peavey_ultra_noise_fx_return -> guitar straight to effects return (in order to test just the power section)
peavey_ultra_noise_fx_send -> guitar to high input, effects return to DI of sound card (in order to test just the preamp section)
peavey_ultra_noise_preamp_out -> guitar to high input, effects return to DI of sound card (in order to test just the preamp section)

[Jon Snell]

A common problem with a common probable cause;

[ACHIEVEIT]

Thanks a lot for the suggestion Jon Snell ! Went through my stash and don't have a similar (or close) capacitor at hand so I'll have to wait till the order (with all other electrolytics too) arrives.
For learning purposes, why are you suspecting that one specifically?

I am attaching the schematic/layout for everyone's convenience
Thanks!

[Jon Snell]

Thanks a lot for the suggestion Jon Snell ! Went through my stash and don't have a similar (or close) capacitor at hand so I'll have to wait till the order (with all other electrolytics too) arrives.
For learning purposes, why are you suspecting that one specifically?

I am attaching the schematic/layout for everyone's convenience
Thanks!

The last five that I had in for service had that particular capacitor just about to empty its contents onto the board and in fact one had squirted out and damaged a few components.
When they go high ESR, the voltages go unstable and due to the then lack of decoupling, the amplifiers then suffer from internal coupling issues.

[stokes]

While you're at it, get all those asian crap caps out of there.Cant tell you how many Fender Blues Deluxes,Devilles, Hot Rods etc I've had with various symptoms, including yours that were cured with a cap change. Even had a couple that had no symptoms at all that showed a marked improvement in tone and response by replacing with good quality caps.

[Helmholtz]

While you're at it, get all those asian crap caps out of there.

If you mean the IC caps, that's a US brand (meaning Illinois Capacitor).

These days most ecaps are manufactured in Asia, no matter what brand, quality reaching from poor to superior.

Early ecap failures in an amp typically mean that the amp designer didn't specify caps suitable for the application.

[tedmich]

The Illinois caps have easily reached their end of life in an amp this old, I'd replace them F&T if I wanted axials or if I wanted to boge in radials, something much better. The Ultra 60 internals look much like a Bravo (great amp) and very different from my Ultra 120, these are real sleepers IMO.

[g1]

The last five that I had in for service had that particular capacitor just about to empty its contents onto the board and in fact one had squirted out and damaged a few components.

The note on your photo also says the output tubes need replacing. Is that note for some other particular unit or is there a reason you suspect the power tubes in this amp?

[Jon Snell]

The note on your photo also says the output tubes need replacing. Is that note for some other particular unit or is there a reason you suspect the power tubes in this amp?

The power valves are possibly quite old and whilst the amp is being serviced, a check on their condition is warranted with a view to replace them for better quality types like Svetlana Winged C or anything not Chinesium. They are run very hot.

[stokes]

If you mean the IC caps, that's a US brand (meaning Illinois Capacitor).

These days most ecaps are manufactured in Asia, no matter what brand, quality reaching from poor to superior.

Early ecap failures in an amp typically mean that the amp designer didn't specify caps suitable for the application.

IC's are made in asia,thats why I called them asian crap caps.F&T are made in Germany and Sprague Atom and BMI are made in the USA.The Atoms and the BMI's are made in the same factory on the same machines and are essentially the same cap different labels.The BMI's are harder to come by and are about half the cost of the Atoms.I use the F&T's mostly,best cap for the money,I can get some BMI's but only in certain values.Atoms are extremely expensive but I would use them over any IC or nichicon etc, asian crap.

[Helmholtz]

I'm familiar with F&T caps and am happy they are available. I use them for vintage amp restoration where form factor matters.
That said, from spec data, own measurements and professional experience they seem to employ average 60s/70s technology.
(There were better ecaps even in the 70s, e.g. from Siemens, Philips, Rifa and others).
But that's probably what original customers like TAD or Ruby specified.

Many asian made (mostly Japanese brand) caps outperform F&T or Sprague caps, but typically are radial types.

Most cap manufacturers offer different quality levels.

Regarding quality it doesn't matter where a component or product it is manufactured, it's all about design, materials, process and QC.

[tubemonster]

Regarding quality it doesn't matter where a component or product it is manufactured, it's all about design, materials, process and QC.

[Chuck H]

I won't use Sprague Atoms any more. Too pricey and I've had two episodes where they under preformed in new builds. Though I expect the reason was that they were old and therefor "stale". This would be the fault of the vendor and not Sprague, but then there's the price too. I'd never use IC caps. At least not the ones you typically see in so many guitar amps. They seem to fail catastrophically, spitting their contents all over amp innards, typically die younger than others and are worse than average about not being charged on a regular basis by use. That's just my own limited experience with them. I've never had any problems with F&T caps.

And I'm a little bummed, stokes, that you lumped Nichicon into your "asian crap" category. Nichicon is a Japanese corporation which is not exactly the same as the "East Asian Manufacturing Machine" I crab about sometimes. They have set up manufacturing and distribution in China, Austria and elsewhere but the bulk of the raw materials are still made in Japan. The specs for their PW series is really good. I've had no trouble since I've been using them. Not that I use a whole bunch of capacitors, I don't. But as for the few amps I've built or rebuilt with these, so far so good. I've been retrofitting them into axial applications for older amps and I've modified my layouts to accommodate the radial design for new builds for about a decade. This said...

If I go back a long time to when I first started modding, fixing and otherwise messing around with amps I used a lot of "affordable" Xicon electrolytic capacitors from Mouser. I never had a problem with those and they were cheap to buy and I never even knew the detailed specs for them. Some of those amps surely need caps now as they would be over twenty years old. But for my experience and time in proximity to those amps they always worked just fine.

So I'm with Helmholtz on this one. It doesn't matter where a cap is made. You're better off researching brand and model reputations.

[ACHIEVEIT]

Hello again,

So I replaced all the electrolytic capacitors in the amp (even the small ones, like cathode bypass or the ones across the relays) but apparently that was not the issue at all (although I am glad I did change them).

So the problem persists and here's what my further tests revealed:
The distorted overtones are only present when the Reverb pot is above 0 (the higher, the worse -> in other words at 10 it's the worst) AND the footswitch is connected and the reverb is disabled from the footswitch.

If I disconnect the footswitch (the reverb is ON if the footswitch is disconnected), there is no distorted overtone regardless of the Reverb pot setting.
If I connect the footswitch, there is no distorted overtone IF the reverb is activated (regardless of the Reverb pot setting)
If I connect the footswitch, there IS distorted overtone IF the reverb is de-activated (the higher the Reverb pot setting, the worse the distorted overtone).

I have isolated the reverb part sub-circuit for your convenience:

I have written in RED the signal that is coming from the preamp, as well as the reverb sub-circuit output, which is then going to the FX loop sub-circuit.
My understanding is that the reverb sub-circuit is simply mixed with the signal from the preamp and then passed on to the FX loop sub-circuit.


The REV is going to the footswitch and upon inspecting the footswitch, the respective switch is simply grounding the REV (so I drew it in red for clarity).

I made some further measurements and Q1 GATE is as follows:
- Footswitch not connected OR footswitch connected and reverb ON: 0.01V
- Footswitch connected and reverb OFF: -4.78V

If my understanding of the circuit is correct, in the latter case, Q1 is not conducting and as such, the reverb pot should be rendered useless. However, that's not what is happening here - as mentioned, the higher the reverb pot setting, the worse the artifact (at 0, no artifact).

Misc thoughts:
Replaced all tubes with ones I know are working just for the sake of it and it makes no difference.
I have checked all resistor values in the reverb sub-circuit and are fine. No cap is shorted either.
The sub-circuit on the right (clamps, Q5, Q6, MAC97) should be responsible for momentarily muting the preamp signal when switching channels. Since I have no volume issues with the amp it should not be the culprit.
I had a spare 4558, so it was an easy test to do and it isn't the culprit.
Disconnecting the reverb pan RCA OUT removes the artifact, but I doubt the pan is to blame. Even if it was introducing noise, it should never reach past Q1 when Q1 is not conducting.
What's strange though is that I made a primitive test to Q1 and seems to be working fine. For example, R46 measures:
- Footswitch not connected OR footswitch connected and reverb ON: 4.61k => implies that Q1 is conducting (so what I measure is R46 in parallel to the reverb pot + Q1 DS resistance)
- Footswitch connected and reverb OFF: 10k => implies that Q1 is not conducting (so what I measure is R46 alone)

At this point I have run out of ideas and desperation is creeping in
Can it be that Q1 is to blame anyhow or that it is simply a design flaw (which I cannot explain though and have seen nowhere else reported)?
Can something else be the culprit?

Any help is greatly, greatly appreciated!

[tedmich]

Since the noise is when the reverb is bypassed I suspect one of the JFETs is bad, although this normally means you get muted.

These commonly go bad (Enzo has noticed this!) and are increasingly difficult to get replacements for some types: the J231 and J174 are only available from InterFET (via Mouser) for 6 fold more $ than the J111 which are still made by ON Semi/Fairchild for $0.50. Good luck!

PS apparently some will pay ~2000% more for "vintage NOS" J111's...