Re: Gibson circuit mods

On Thu, 30 Oct 2008 08:55:56 -0400, "Phil S."
<psymonds_no_spam@xxxxxxxxxxx> wrote:


"RS" <RS@xxxxxxxxxx> wrote in message
news:0qaig41mfojb5ub5fmfr67fl84cc607lh2@xxxxxxxxxx

...The treble circuit works by shunting highs to ground through
the cap. I was pointing out that it would be more efficient to do
that right from the plate rather than after the .003 (prev .0047)

The .01 would indeed connect to the plate, as you have it.
The other side of the .01 connects to the treble pot. "
The 'low' side of the treble pot goes to ground.

The bass cutoff cap (.003) would still connect to the plate, just
like the previous .0047 did.
The rest of the bass cutoff circuit would be configured like it is
now.

I think I've got it now. Treble and bass caps both connect to the plate.
Ground the treble pot.

Sounds like you've got it. If you get stuck, I'll draw you an ascii
schematic.

You don't have the schematic for my GA20-RVT (Gibson issued 2 or 3
iterations of this amp). They actually rearranged the tonestack on both
channels. I'm not so good at describing, I'll try to post it so you can see
it. The bass is hooked to the plate. Volume and treble are wired in
parallel after the bass. The volume wiper feeds the next grid.

Interesting circuit. Slightly better in some respects; not so in
others. What's labeled as the 'bass' control is configured more like
a mid control. Not sure why they'd use the .01 cap to ground, as a
resistor in that spot would be a better way to do bass rolloff.

And I'd still go with treble up front if you're going to use that type
of circuit. Easier to scale impedances. That was not done optimally
in the Gibson circuits.

That bridged T is something like a Fender-type tone control with the
bass and treble up full. You should be able to see the similarity.
But the different values place the mid notch at a different spot.

Fender's mid notch is around 300 hz. Just looking at the Gibson notch
circuits, it appears to be around 500hz to 600hz, with a shallow
slope, but it will have a sharp notch right at the resonance point.
That's caused by opposite phase shifts between signal going through
the top C/R (330p/220k) and the bottom R/C (220k/.0047). Highs go
through the 300p cap at the top. Lows go through the resistors at the
bottom, and are phase-shifted by the .0047 cap. They add together
again at the other side. At the resonant frequency, the signals cancel
when they recombine. That's why I said it would sound 'phasey'--it's
similar to the circuit used in some phase shifters.

This amp can be rather ice pick sounding. I think maybe this is the cause?
The notch is at the "wrong" Hz?

Definitely could be a big contributor to that. Try lifting the mid
cut cap and you'll see what I mean.

If you think that the I-S transformer is part of the cause, you could
try 'damping' it with a paralleled resistor to see what happens.

That's a simplified explanation for what is actually a network. You'd
be able to simulate that with Duncan Monroe's tone circuit program.

I have that. If I understood better about frequency/rolloff etc., I might
actually be able to make sense of it.

There are ways of shortcutting that to understand it better, but it's
hard to explain in print.

If you do this amp right, you could get a great tweed-ish blues/jazz
sounding amp. Should be very organic and responsive.

Yes, that would be a clear improvement.

The circuit is based on variable bass cut and treble cut rather than
the usual (Fender) bass boost and treble boost. Of course all things
are relative re the terms 'boost' and 'cut' but the former is no doubt
why Gibson used separate mid cut circuits (the bridged T's). The mids
are not cut by the main tone circuit.

Many of the tone controls of the Tweed era don't notch mids like the
Fender/Marshall controls do. In general, this is why they're favored
for that warm balls-out Chicago blues sound. If you keep the mid
notch circuit and make it variable, you could get the best of both
worlds.

If you're building a new amp, why not go with turret board or
something?

Yes, but now you've put a second project on the board. Build one as a
prototype, then mod the existing one. The prototype will essentially be ch2
of the GA20-RVT without the optoisolator in the tremolo. That opto is an
incandescent made of unobtainium. I can just follow the roachless GA15-RVT
design.

I still think it would be easier to light an LED than a neon bulb, if
that's where you're going with this. Less problems with glitches when
it fires. More practical to be 'tube-purist' where it counts. Someone
may even make a modular trem circuit that could be used. I don't use
tremolo much, but I've used LED modulators in a lot of circuits
(phasers, flangers), and had no problem with them.

Also re relays: You sounded like you thought it would be tough to work
with them. Not at all. Kinda like a flashlight. You hook a voltage to
the coil and the relay closes. Nothing could be simpler.

.

Relevant Pages