Gibson GA35RVT - fizzy overdrive, weird compression

[afinitemind]

Hi everybody!


I inherited a fun project: an old Gibson Lancer GA35RVT. Looks like I'm not the first person on these forums to dig in to one of these amps! This one came to me with a power issue where it'd turn on and sound fine and then the B+ would slowly start to drop until the amp lost all power. Traced that issue down to a shorted reverb transformer. The amp came back to life once I'd taken the RT out of circuit. It's already got JJ 7591s and newer sovtek 12ax7s, plus the 6eu7 sockets have been converted for 12ax7s.


This particular specimen sounds ok, but when I turn it up into overdrive territory it is VERY fizzy and not particularly musical. Also, when I hit a big staccato chord, I can hear the amp compress and it just sounds weird. As in, the background hiss goes silent and then slowly fades back in. Like maybe I'm hearing the power supply caps refill? I've been reading other people's threads about the Lancer and following as many links as possible (like the super helpful youtube videos). These things happen the same on both channels, so I'm thinking power amp or PI. Anyone have initial thoughts on where I should dig in to try to suss out these problems?


Now, just to make things interesting, someone else has really gone to town on this amp. The filter caps all look newer, rectifier diodes are 1N5408, all of the electrolytics on board are new, but best of all: the ENTIRE circuit has been transplanted from Gibson's original eyelet board rat's nest to a much more spacious piece of tag board. Best I can tell, looks like they were kinda following this adjusted schematic: http://infernalracketstudios.com/Gibson%20GA-35%20RVT%20Repair%20&%20Mods%20(5).pdf


Guess it'd also be nice to replace that reverb xformer but I'm not sure how to figure out what exactly I'd need.


I'll post specific voltages later, but on my initial check, things looked mostly ok if a little higher than the schematic reference.

[PRR]

Welcome.

> B+ would slowly start to drop until the amp lost all power. Traced that issue down to a shorted reverb transformer.

That doesn't quite fit.

> overdrive territory it is VERY fizzy...big staccato chord, I can hear the amp compress and it just sounds weird. As in, the background hiss goes silent and then slowly fades back in. Like maybe I'm hearing the power supply caps refill?

I'm suspecting the RT is a side-show, that your real problem is a weak power supply.

> someone else has really gone to town on this amp.

Maybe best intentions, but how was the execution? Perfect? Or 99.44% right?

Have you taken any voltage readings while playing? (Use clip leads! Don't get yourself killed!!) B+ to ground at C24B (1st cap), C24A (2nd cap), also C25 (bias). Steady, saggy, going high?

R408 is wired dubiously in the plan. I have to ponder that. Meanwhile what happens at the CT of T3 when you beat the amp?

[afinitemind]

Hi PRR, thanks for the response!


Best intentions for sure. The execution looks decent if a little messy. I cleaned it up as much as seemed necessary. Schematic-wise, a few minor details changed...no 'heater bias' section, 2x 10W 500ohm resistors in parallel (probably sag resistors?) after the rectifier, that R40B "bias" pot is 25k with the surrounding resistors changed to accommodate the different values. Also, there is no C25 in the bias supply, but C23 is 100uf 150V.


I just took some measurements while playing...clip leads for safety! And the power supply is definitely struggling when I slam the amp. These measurements were observed while hitting staccato chords: before > during > after the chord


C24B (1st cap): 447.5V > ~415 > ~456
C24A (2nd cap): 422.9V > ~358 > ~443
^^these went back to the 'before' voltage after a second or 2...seems to correspond pretty well to the noise fading back in after the chords.

C23: -26.3V > -38.5 > -26.3
T3 CT: -21.6V > -41.1 > -21.6

[PRR]

> T3 CT: -21.6V > -41.1 > -21.6

That is the key to unhappiness.

That point should be nearly solid.

The fact it goes 2X the idle value shows it is going into extreme cut-off, "blocking distortion".

I have to re-hose a tractor while the sun shines. Maybe later I can look at the plan.

Is there a plan (link) for the un-modified factory original?

[afinitemind]

Here's the only original I can find: http://archive.gibson.com/Files/schematics/GA-35RVT.jpg

I've read online (might've been on this forum, actually) that there were multiple schematics used, but I can't find another plan. This amp appears to be slightly different from that schematic but who knows what it looked like before the tagboard & circuit mods were installed. The bigger differences are in the EQ section anyway, so most likely not related to the PI / bias supply issues.

Thanks for being willing to help, PRR! I appreciate you taking the time.

[sluckey]

There are a couple schematic files for that amp right here in Hoffman's schematic library.

[PRR]

Just to keep you busy: Jumper bias pot from 20u cap to driver CT end. That added several K-ohms probably is not helping. But I don't think it hurts much.

You *may* just be playing it harder than it was supposed to go. The bias shift shows you are pushing the 7591 into grid current. 7591 are not rated to pass grid current. Lowe-gain tubes like 6L6 can take small grid current (a few mA) safely. The hotted-up 7591 probably can't be pushed to such current safely. With that bias supply, 0.4mA average grid current would give your observed ~~20V to ~~ 40V shift. This could be over 1mA on peaks.

> 2nd cap 423V > ~358

18% sag from idle to FULL power is no big deal. A bit high, but I think you are pushed to total over-drive, past any "sine test" condition.

Does anybody have a clue how an original amp played with heavy strums?

[afinitemind]

Life got in the way! I'm traveling for the holiday week, but I'll try jumpering the bias section when I'm home on Sunday.

[afinitemind]

Alright, tried the suggested jumper, same result.


Not sure if it's worth noting, but I did observe that the bias voltage jumps a lot less (-21.7 -> -23.5) when I set the channel volume for cleaner operation. This is true with or without the jumper installed.