SF Twin Reverb - Vibrato Channel Breakup

[nickb]

If your probe kills the burst, why not add a little cap and/or resistor there to mimic the probe and kill the burst?

If blasting the reverb driver into cutoff sounds crap, how about back to back clipping diodes (value?) across the input so that it cannot be driven into cutoff?

Somehow that seems like cheating.

It may come to that yet. I'm going to back out my changes, that is then my fallback position, and then figure out the feedback path for the oscillation. With that nailed the reverb issue will be as the standard amp. The owner obviously likes to use these settings and the sound he gets from them so he should be happy.

I'm am greatly relieved to hear I'm not the only one to experience the rerverb overload issue. The oscillation may be unique to this example.

Thanks for all the suggestions. I'm still listening.

[cjenrick]

check grounds from pots, had same problem with same amp, turned out to be nuts holding down pots, in the V channel

amp probably did not oscillate at factory so something has drifted.

miles of coax in that thing.

[nickb]

check grounds from pots, had same problem with same amp, turned out to be nuts holding down pots, in the V channel

amp probably did not oscillate at factory so something has drifted.

miles of coax in that thing.

I did see your earlier post suggesting this. It's a good idea and on my to try list.

[The Dude]

I have a couple of questions. In post #1 you said the amp was stock excepting the reverb transformer. Was it replaced with an exact factory replacement or something else? Also, when the reverb is turned down does the oscillation stop?

[nickb]

I have a couple of questions. In post #1 you said the amp was stock excepting the reverb transformer. Was it replaced with an exact factory replacement or something else? Also, when the reverb is turned down does the oscillation stop?

Its a Hammond 1750A 22.8K / 8 ohm. It may be the thing that pushed it over the edge. The precipice should not have been so close, IMHO.

The reverb level control makes no difference. It's only if you have enough volts on the grid of the reverb driver that you can make it happen.

[The Dude]

Its a Hammond 1750A 22.8K / 8 ohm. It may be the thing that pushed it over the edge. The precipice should not have been so close, IMHO ......

Agreed, but I think that's what's going on here. IMO, the non-musical distortion is typical of these amps. I never liked it. But, I've not encountered oscillation issues. The only thing different seems to be the transformer. I see this as part of that repair. Outside of trying a stock transformer (which would be silly since there's a "working" one there), I think the best option is to lower the drive to the transformer or at least the high frequency content of that drive.

[The Dude]

FWIW: I always thought putting a Master Volume in a Twin is akin to putting wheelie bars on a Cadillac.

[nickb]

Agreed, but I think that's what's going on here. IMO, the non-musical distortion is typical of these amps. I never liked it. But, I've not encountered oscillation issues. The only thing different seems to be the transformer. I see this as part of that repair. Outside of trying a stock transformer (which would be silly since there's a "working" one there), I think the best option is to lower the drive to the transformer or at least the high frequency content of that drive.

I hear ya... and that was exactly what I did. The worries are that this owner, I guess, actually uses & likes (or hates his neighbors) these settings that cause this problem and it will reduce the huge reverb so I'm about to back that out.

Timmyp's clipper diodes idea above has real merit as done properly the 12AT7 will still be conducting and that will prevent inductive voltage spikes.

FWIW: I always thought putting a Master Volume in a Twin was akin to putting wheelie bars on a Cadillac.

Ah, the marketing dept. effect.

[cjenrick]

try switching the phase on the sec of the rev xfmr, i got a better overdrive sound by doing that,

[66 Kicks]

Its a Hammond 1750A 22.8K / 8 ohm. It may be the thing that pushed it over the edge. The precipice should not have been so close, IMHO.
The reverb level control makes no difference. It's only if you have enough volts on the grid of the reverb driver that you can make it happen.

Maybe a resistor in series with the primary?

[nickb]

try switching the phase on the sec of the rev xfmr, i got a better overdrive sound by doing that,

Tried that see this post. But I find the same as you, if I have the output leads swapped the drive is better. It doesn't affect the problem.

I really need to get back to the bench and do some non-desktop debugging. The problem is we have building work going on right now and i can't get to it.


Pete: I think I might know why a grid stopper on the reverb drive doesn't help. It's in cutoff when this happens.

[Familyortiz]

I can't see a reference designator for these 2 vibrato tubes, but I'd replace them with a known good tube as a simple test.

[pdf64]

[ATTACH=CONFIG]44576[/ATTACH]
I can't see a reference designator for these 2 vibrato tubes, but I'd replace them with a known good tube as a simple test.

Do you mean 'reverb' rather than 'vibrato'?

[Familyortiz]

Thanks for the correction. Here is the vibrato section.

[nickb]

Yes, all this is familiar to me, the preamp overdrive on these things is pretty horrid, and some control settings can send things crazy.
I've tried the various mods mentioned to try and improve things a bit, but one of my friends actually likes the pull boost feature, so I couldn't change it too much.
I think that V3 grid stoppers and making the V3 cathode resistor 1k5 (bypassed) in series with 820 (un-bypassed) would be a good starting point if I get another one in.

Builders have gone so got back to the bench. Hooray! We have a solution.

I backed out my changes aside from the 1.5K ohm cathode resistor in V3. Oddly, the problem was much more persistent for no obvious reason. I could now get it to happen with the lead from the reverb output to the MV switch removed so ruling out that direct path.

The solder bridges from the pot brass strip to chassis all looked doubtful and indeed one of the original problems with this amp was an open chassis ground so I re-soldered those. Didn't help.

Measured oscillation at 172MHz (I have a 500MHz scope). Moving wires around did affect it but not by much. Added very aggressive 56K grid stoppers to V2B and V4B. That helped a little. Added 47K stopper to V3 and that did it, so destroying my notion of it being in cut off when it happened. The really annoying thing is that was the very first thing I tried except I used 5.6k which I felt was entirely suitable to kill oscillations at 100's of MHz. Wrong.

So, Pete wins the prize. It's hidden in an undisclosed location but with your prodigious deductive powers, you'll have no problem finding it. You're welcome.


So, no takers for the bonus points in post 1?