Gibson GA42RVT revised schematic

[EFK]

Don't laugh as my bad drawing! I have been spending a lot of time playing with this amp and have (and have posted in other posts over the past 6-12 months) a schematic that Gibson sent me. Well I can tell you for sure their schematic is completely wrong - maybe it's for a different amp or something but it certainly is not for this one. I've been tracing out the preamp and the PI sections to the best of my ability; I'm a real novice but I do have these sections correct I believe. The reverb and tremolo sections are a bit beyond me. What I would like to do is reduce bass - this amp has a HUGE amount of bass and even with the bass knobs down at 0-2, it is too much bass with paf-type pickups. I would very much appreciate anyone's input as to suggested mods or improvements. In particular: the tone stack in channel 2 looks kind of odd, at least compared to the 'usual' Fender or Marshall TMB stacks. I think there are a lot of 'dark' values throughout this circuit? The 100K treble pot seems of a particularly low value? Where could some of the best changes be implemented to reduce the overwhelming bass?

[octal]

You might try reducing the value of the cathode bypass caps in the preamp (the 22uf caps) down significantly; something in the range of .68 to 4.7 uf would be a big help cutting a bit of flab.

You can also trim the values of the coupling caps going to the output tubes down. They're .1uf now- you could go down to .022 or so.

There's more to tone shaping than the tone stacks themselves!

Good luck,

Nathan

I think you may have made some errors in the preamp schematic. I would examine exactly where the 15k 5W resistors go to...as drawn, they appear to be in the signal path, rather than in the power supply circuit where they almost certainly should be.

[EFK]

I think you are correct about the 15k/5W although something funny is going on in there. I took a look at some other 'standard' Fender/Marshall schematics and had something of a "!!!" moment so I think I'll go poke around in there again. I'm sure I've got it right up until that point, however. Where I was having a hard time was following the signal out of the tone stacks because from there it gets into V3 and V4 (reverb and trem) territory and I'm completely lost.

How much of a drop is necessary, down from 22uf on the bypass caps, until a noticeable difference can be heard? I think the plexi .68 would be fairly dramatic difference, maybe too much. Should I leave the second stage 22uf alone? The coupling caps are also a good idea. Definitely will try that one.

Two immediate questions: following the first stage, why are the 470 resistors joined in series directly to the .022 coupling caps? Does this serve a purpose? This seems different from say a JTM45 where the resistors follow the volume pot. What does lowering this 470 value to say the JTM45 270 do? It looks like plexis went to 470 but they also follow the volume pots rather than preceding them. And, why does the second gain stage into the tone stack on the first channel get a .022 coupling cap but the second channel does not? What effect is this having?

Thanks for the help - trying to learn on this quite a bit!

[octal]

Re: dropping the cathode bypass caps. These are typically WAY oversized in guitar amps. IMHO, there's no reason for a guitar amp to have frequency response below 80Hz. I believe the "Valve Wizard's" site has a chart showing freq response versus size of cathode bypass caps. There will be a steeper slope to the low frequency roll-off if you alter both the first and second gain stage bypass caps.

Re: 470K resistors after the first gain stage. These are being used as signal voltage dividers with the 1meg resistance of the volume pots. You could certainly try bypassing one of them with a paralleled 470pf cap (try different values) to create a treble peaking circuit, ala Marshall.

That's all I have time to explain at the moment.

Nathan

[corworld]

This amp has the Vox Top Boost Tone stack like the GA77. Very cool. I was working on my AC15 type amp to reduce the bass last night and what I found was you could swap the .022 caps before the PI to a .01 cap. The .1uf caps after the PI are really to much. Try a .047 here instead. I used a .047 and it still has lots of bass with a humbucker but the muddiness is not there.

[TheTinMan]

FWIW here are a few reference materials which are useful regarding guitar frequencies and cap/resistor combinations:

Valve Wizard Keep in mind his chart is ONLY for 100k plate resistor and 1.5k cathode resistor

Piano & Guitar frequencies

I've attached a cap-resistance table that came from Mission Amps (that's what my notes say anyway)

Then there's an old-fashioned but useful table called a "frequency reactance chart" which I've attached. You find the cutoff frequency you want along the x-axis (bottom), find its intersection with the resistance on the y-axis (horizontal lines), and the diagonal line leads you to the appropriate capacitance. All of the lines are logarithmic so it goes 10uf, 20uf... 100uf (for example). It seems really cumbersome at first but is great once you get the hang of it IMHO.

You will see from all of this that the standard 1.5k/22uf combination has a cutoff frequency WAY below anything we need coming out of a guitar.

Hope this helps,

Chip

[EFK]

2 things:

(1) Top boost? IS this a top boost? I thought it looks like a Bandaxahl/James? Huh?

(2) I have changed a whole lot back and forth with this circuit. Most of the cap swaps have resulted in a thinner, weaker sound that does not sound too good. So, how to clear things up without weakening it too much? The real problem is not so much the excess bass in each channel individually, but when one uses the jump input which is basically both channels in parallel. Really sings and screams, but the low notes do get muddy. My solution has been to leave most all of the stock values intact EXCEPT I ditched the 22uf cathode bypass caps on channel 1 and replaced them with 1 uf (a bit too much) and settled on 2.2 uf. This sounds good to me.

This really is an interesting amp. Depending upon tube and speaker choice you can go a lot of ways with it. Has a tweedy, semi-marshally tone but very chimey on top yet overall a bit dark. Kind of like some older Gibson amps I guess! I've went more brit with it, KT66 and Private Jack/Weber Blue Dog mix. Can easily cop a lot of that Beano/Cream/Black Cat Bones sound. I added a PPMIV which sounds better than an attenuator, but to really appreciate this amp (if you are going for an overdriven tone) it absolutely has to be wound up and smoking (figuratively, but it does get VERY hot). The cleans are absolutely beautiful, I just don't use them all that much. Ultimately is also likes very low output pickups - in fact, oddly enough to have been made by Gibson, it would probably be best with single coils or filtertrons. With HB pickups and mahogany, I think you need to keep them really low output to get the best out of this amp.

Sometime I would really like to talk to Pyotr Belov about where he was trying to go with this amp and where Gibson wanted him to go with it.

[corworld]

Yes, you are right, it is a Bandaxahl. I wasn't looking at the values, just a quick glance at the layout of the schemo.

I built a GA40 and had a similar experience. For tweed blues sounds it was great. The cleans were nice but dark and HB pickups put it into a beeno-type sound. It may be a bit of a one-trick pony, but it is an interesting amp to mess around with.

[EFK]

Perfect description: That's exactly what it is, a one trick pony. The stock speakers suck, but depending upon speakers you can push it either towards dark American tweedy or dark Brit blues tone. This seems to be what it does best.

[tekhed7]

To help eliminate the mud & make it more responsive overall... once you've "lightened" the preamp, you might also want to switch out that .022 uF cap running the sound into the driver tube (pin 2 of V5A) with a .001 uF {or even a 500pf} Hi-V disc cap... like you see in olde Fenders. You can leave the other one going to pin 7 at .022 uF. This will take the lowest octave out right at the driver stage. As was mentioned, no reason to have response in a guitar amp extend below 70Hz or so.