Silverface Fender Twin has nasty bass distortion

[g1]

I have read here, several times, of people replacing filters in their BF/SF amps and experiencing more fidelity. More LF, HF and better dynamics. And I've also read that this isn't always appreciated by the user. The tone changes such that the amps can become harsh or muddy when clipping. That you desctibe the problem as an out of tune artifact, and that the problem persisted with the old caps back in makes think there is an actual problem. I have never heard of new filter caps changing bias significantly. You may need to take this amp to a shop where a tech with proper bench gear can scope it to isolate the odd distortion. This sounds terribly frustrating for you. And diagnosing it without test results from a remote location is proving impossible.

Further confounding the issue, there were previously bad/worn out parts and incorrect value resistors. The problem could have been there before but masked by other issues. The preamp voltages would have been very different with the other resistors in there.
Also, putting the old caps back in won't be the same unless you marked the exact location each cap was in originally. An "open" cap that was doing nothing before in node "x" may do different things in node "y".
So putting the old caps back in is not the same as putting them back in their original locations.
Also seeing as you mentioned bias, did you replace the bias filter cap?

[Gtr_tech]

I'd check for any DC leakage from coupling caps. You shouldn't have any DC on the grid of a gain stage. What are the coupling cap values off the PI going to the output tube grids? Some of the SF amps had larger values here than they actually needed for a guitar amp. If they are .22 or .1u, try .047 and see how that does. You can go down to a .022 for even tighter bass response.

[Whymeworry]

I replaced all other Electrolytic caps on the amp, including Pre-amp caps, and the bias caps after the problem had sprung up with the filter caps. It didn't help.

[Chuck H]

This is why I didn't ask for plate voltages, suggest coupling cap leakage or question original cap position.

Take it to a tech.

[Whymeworry]

What I am planning on doing. If the tech fixes it, and can explain to me what went wrong, I will post the answer for anyone who also has this problem in the future.

[gtrplayr1976]

And that schematic indicates that the resistor values should be 2.2k and 10k.

What are your preamp tube plate voltages?

The Silverface Twin I am working on has a 100K ,and 2.2K although the schematic shows different. I've found a few inconsistencies between the schematic ,and what is actually there.

[g1]

He was talking about the resistors at the filter caps, in the dog house. There should never be a 100K there. Double check yours, if it is 100K you should have some low preamp voltages.

[g1]

Well, this is a bit of a mess. The layout (not schematic) shown here: http://music-electronics-forum.com/a...win_rev-2-.pdf shows a 10K and 100K for supply droppers, rather than a 2.2K and 10K as the schematic shows.
I believe this must be a typo, as the voltage drop across a 100K in that position would be huge.
Yet the OP stated that it made no difference to the sound whether it was 10K or 100K.
gtrplayr1976: Can you verify it is a 100K in the doghouse rather than 10K, and does it look like the factory part or has it possibly been replaced?
Anyone care to give an opinion on whether the layout has a typo or whether there are units out there with 10K and 100K supply droppers?

[bob p]

Wow, you guys who could ID the 100k resistor in the photos have really good eyes!

Steve had the right suggestion: Post a voltage table for the supply rail and all of the tube pins.

[bob p]

Well, this is a bit of a mess. The layout (not schematic) shown here: http://music-electronics-forum.com/a...win_rev-2-.pdf shows a 10K and 100K for supply droppers, rather than a 2.2K and 10K as the schematic shows.
I believe this must be a typo, as the voltage drop across a 100K in that position would be huge.

Noticing that was a very good pickup on your part.

Anyone care to give an opinion on whether the layout has a typo or whether there are units out there with 10K and 100K supply droppers?

I'll bite. That's definitely a typo. Looking at the voltage drop across the 100k resistor, it's 420 - 345 = 75V. It's spec'd as a 1W resistor. Do the math, and it's dissipating 50 mW. There's no way that can be right: A) 50 mW won't drive the downstream load, and B) they spec'd a 1W resistor for a 50 mW job instead of just using a 500 mW resistor like they did everywhere else in the amp. 100k can't be right.

Doing the math for 10k / 1 W the power dissipation is about 500mW. At least those numbers make sense.

I'm wondering if somebody just read the layout diagram and put in the parts shown on the layout, without double checking the schematic.

Yet the OP stated that it made no difference to the sound whether it was 10K or 100K.

No difference in sound can't be correct. We're talking about a 10x difference in resistance, which would cause a HUGE voltage drop. Every time that I've seen old CC resistors go bad and drift up, a 50% increase in resistance is enough to cause a significant voltage drop downstream, significant shifts in the load lines, significant loss of headroom, and crappy distortion. that's with a 50% increase. I can't imagine how bad things would be with a 10x increase, and I can't imagine that a 10x error would make no audible difference.

Can you verify it is a 100K in the doghouse rather than 10K, and does it look like the factory part or has it possibly been replaced?

The picture to me looks like a Carbon Film replacement. Original would likely have been Carbon Comp. I'm thinking we're looking at a parts error when somebody serviced the amp. Best way to get the answer would be to clamp on an ohmmeter.

[Enzo]

Schematics certainly can be wrong, but I almost never refer to layouts for component values. I always use the schematic unless there is a compelling reason to do otherwise.

75v across a 100k resistor is about 7 tenths of a milliamp. One triode would use that much, so if there is more than a single triode downstream of the 100k, it must be wrong.

[gtrplayr1976]

This is a picture of the first time I opened the dog house. I was changing filter caps. I know the owner ,and know he bought this amp new. To my knowledge has never had it to anyone for repair, or service. I will try to verify this with him. I will get a better picture this evening when I get home.

[Chuck H]

I mentioned the same in a PM to Bob. And I also noted that it IS an official printed document. So it's entirely possible that someone put the wrong value part in as per that document at some time or another. But notice also that the node of the 100k dropper says it goes to the trem controls?

[Chuck H]

That looks like brown-black-orange to me.

[gtrplayr1976]

You are 100% right Chuck. My bad. I have no idea what I was thinking. Definitely a 10K ! Apologies to all ! This is what happens when you have insomnia
I have seen other discrepancies even though this was a mistake on my part. I think I was looking at the layout. Some of the things on the AA270 ,and AA769 match more with what I have than the schematic for the MVS which is what I have. In this case I guess I will replace with like parts.