Half Of A Gibson GA40 Les Paul Amp Is ...

"Let a Gibson amp be like a box of chocolates ...." Orville 'Forrest Gump' Gibson. My GA20 works pretty well, especially channel 2 with its 5879 which sets it and its big brother, the GA40 Les Paul apart from nearly every amp of the period (read more about the GA40 here). The public domain schematic indicates 12AY7, 5879, 6SQ7, and a 12AX7 in the preamp. However, the first GA20T I bought is fitted with 3 6EU7s and the critical 5879 in the preamp (thanks Orville); not the lineup listed above and the gap between the preamp and power tubes doesn't allow enough clearance for the bell of an alnico speaker (see lower half of pic below).




Because I prefer alnico speakers and wanted the same tube line up in Channel 2 as in the GA40, I bought another GA20T configured like the amp in the upper half of the picture above, including the 12AY7 in Channel 1 and a 12AU7 driver/ phase inverter and with the wider speaker gap. Fine ... except someone painted it medium gray/tan :evil: ... although it's complete with all five original Gibson ox-blood chickenheads:




Included a Jensen P12R however the glue holding the cone in place failed; it's at a speaker shop in West Palm along with a 10" Jensen 'Viking' that our BBer RWood laid on me to address the shallow Silvertone 1481 cabinet:




Upper left before stripping, lower left an idea of what it looked like before it was painted. Upper and lower right, after stripping with Strip-Ez and a coating or two of Easy-Off oven cleaner. Web buzz is that the Easy-Off is the way to go ... and it would have been for conventional tolex. On this amp, the vinyl topcoat on the tolex was gone; maybe taken off to enhance adhesion of the paint. Anyway, the stripped tolex is now nothing but cotton - no vinyl - the texture of fine burlap. Some of the tweed-colored weave still has some dye in it but the Strip-Ez doesn't distinguish between paint pigment and fabric dye.




Progress on the new filter caps; same approach as the other GA20 - stabilizing wiring strips, dry fitting, and with soldered-in supply leads:



Filter caps in, primary and screen caps grounded separately from preamp filter cap IAW good practice. New cord at right:




Control panel: except for some crud just to the right of the on/off switch, really in pretty good shape; the silk-screening about 80% intact:



A little Forrest Gump twist on things; GA20T #2 has its signal, tremolo, and one of its cathode caps mounted on the underside of the circuit board (keep thieves from snipping them out ??? - seems Mesa Boogies are done the same way ... )the 5 pots and 4 jacks in the control panel will have to be loosened, and dropped out so the circuit board can be flipped over :evil: :evil: to get at the cathode and other caps. Replacing the cathode cap isn't elective but the others might be except that if I go to the trouble to replace the cathode cap ... and one of the others lets go ....





Why all the fuss? It's the 5879 ... Read more about it here.

John, I lost count. Don't mean anything by it, not trying to be funny, but I honestly don't know which amp is which anymore. Not that it matters thàt much, I enjoy reading your threads about them anyway - I'd be pretty happy just sitting in the same room with you and just looking over your shoulder.

Hi Walter; thank you. When posting about them, even I get confused which is which; I have two GA20Ts .... I guess one is an earlier model and the other a later one with a higher serial #. Unlike the extraordinary work that Hans has done tracking Guild serial #s, I know of no source; web or hardcopy, that has tried to sort out Gibson serial numbers.

Anyway, the first one I bought and threaded here is the later model; 6EU7s in the preamp but no space for a speaker bell. The second one (the more recent above), I believe is the earlier model; room for a speaker bell but with different tubes in the preamp; otherwise they're the same. This latter one came with paint on the tolex and caps under the board.

The 6EU7 model/1st one runs fine, the other is still in parts. When done; the 'tone victor' gets the spoils. The cabinet that housed the 6EU7 chassis, for an old Gibson, is in pretty good condition and the tone champ gets the better cabinet ... the other gets ... I believe the Flemish expression is 'hind tit' :wink: It must be close to mid-night there; have an Absinthe on me and a very happy holiday! John

Oh, no, more like eight in the PM! Thanks for the explanation. And very happy holidays to you and yours as well.

Progress pics; coupling caps. The top panel is the resistor side of the circuit board looking straight down. The center is the board after the controls and jacks have been pulled out of the faceplate and the board flipped over and part way out of the chassis. The lower panel shows the new parts in relation to the old ones; all the caps with the highest voltage ratings have been replaced; these caps connect the consecutive amplification stages by providing a signal path and, at the same time, block DC from appearing at the input of the next stage. The large tan cylinder in the middle panel is one of the electrolytic cathode caps for the 5879; it had to be replaced regardless and no point in not doing the couplings caps at the same time. Notice the bulky bumblebee caps.




In the next generation of this circuit shown below (my other GA20T; bought first but later model), the bumblebees have been replace by a collection of silver micas (tan and maroon disks), paktrons (yellow), and polyester chocolate drop (lumpy brown) caps. The Sprague Orange Drop at the far left is a replacement and the tan cylinder at the far right is the original 6V6 cathode resistor bypass cap; it and the coupling caps will be replaced in the near future.

Progress Post; GA20T 1 of 2 ... the first one; the 6EU7 model. Some visual inspiration Link; matches the other GA20T with 12AY7 / 5879 / 6SQ7 / 12AX7 preamp not the one discussed below. I updated the power supply caps and it hummed like nobody's business. I took it to the local amp shop and Bill Potter straightened the problem out; a ground loop and a known design deficiency in this amp. From a reply in another forum ... I lost track of the URL, paraphrased:

".... it wasn't the CT for the heater, but was the CT for the power transformer secondary that was grounded incorrectly ... the red wire off the PT was grounded to one of the power tube sockets pins ... causing the hum to be picked up by the grid-leak biased 12AY7 in the preamp because the preamp wasn't grounded correctly either ... moving the PT CT to the main chassis ground point, and re-routing the preamp ground to ground in the correct order reduced the hum level to just about zero..." which is what Bill Potter did.




I changed out all the coupling and cathode resistor bypass caps, original circuit board on top, finished on the bottom;




Earlier in this post, I was crabbing about the arrangement of the tubes preventing the bell of an alnico magnet from fitting between the preamp and output tubes. I already had a re-coned P12R, recently bought a pair of P12Qs in unknown condition but with heavier magnets and larger voice coils, and generally prefer alnico speakers anyway. The only solution was to cut a new baffle that moved the center of the speaker an inch and a half or so to the right (viewed from the rear) and down a little but without cutting the lower baffleboard batten; measuring indicated that, even if it had been cut and the speaker lowered until the lip was resting on the bottom of the cabinet, there still wouldn't be enough room for a bell. I can live without the bell ....

New baffleboard process below: upper left, all the dimensions needed drawn onto the masonite template ... having screwed one of these up before ... and the new baffle, 9-ply 1/2" birch plywood on the left. Upper right; stainless steel machine screws, flat and cut washers and a nut for the baffle mounting studs; the aftermarket doesn't support a reverse-threaded Ferlingheimer screw of the type shown in the lower right of the picture long enough to accomodate a thicker baffleboard. 'Ferlingheimer' screw is how they are referred to on the south Island of New Zealand where I've been smuggling them to the leading audio amateur in Greymouth ...



Above, lower left, two coats of flat black spray paint and lower right the speaker fastened to the baffleboard. Gotta go through some trouble with the baffle mounting hardware; if not and if someone were to try to remove the baffle, the mounting screws will turn in the same direction as the inside nut and the outcome is FUBAR. I re-cycled the original grillcloth and used it as well as the original skirted nuts. Pleasant surprise after stripping the grillcloth; somewhere along the line, Gibson stopped using the 'winged threaded nail' with its impossible-to-remove little flat nut and started using Ferlingheimers for baffle and speaker mounting. Rescuing the grillcloth produced an unlooked-for benefit.

In the picture below; on top is this amp as delivered with center of the speaker hole in the middle of the baffle and a replacement alnico speaker. As delivered and because of the interference, getting the preamp/phase inverter tube out of the preamp risks bending the crap out of the pins; it just looks like it was engineered to have a shallower, ceramic speaker. On the bottom with the center of the speaker hole moved to the right and down, you can see how the P12R's magnet clears the tubes. I have a Weber 12A125 P12Q equivalent here and measured the diameter of the speaker bell. My concern was that the heavier magnet would also be larger; the P12R magnet approximately 3" square but the measurements of the Weber suggest that the magnet on the in-bound P12Q shouldn't be much bigger ... or so I hope.




I got interested in the GA20T in part because of Mad Dog's glowing description of his GA40LP and it's 5879 pentodes in the preamps of both channels and twin 6V6s. Shown below is a comparison between the preamps of the GA20T and GA40 and the most noticeable distinction is the 12AY7 in the preamp of the GA20 instead of a 5879. When the amp came back hum-free from the shop and with its no-name speaker and polyester chocolate drop coupling caps, tired-out cathode caps, the tone was ... non-descript; channel 1 was thin, bright, and with dried-out bass. Channel 2 with the magic 5879 was also sort of disappointing; bright, tiring, nothing exciting.



My Gibson standard is somewhere between Bluesdan's GA18 and my GA8; the top line of the circuit - 12AY7 shown above (6EU7 in this amp) and 12AX7 driver (6EU7) in this amp needs to get where that amp is now and I got part way there by adding a cathode resistor bypass cap to V1 (6EU7, cap not shown) ... getting closer; bigger, pushier, more bass, more forward ... more better gooder. Not a GA18 but getting there but what I wanted was for channel two of the GA20T to match channel 1 of the GA40. In another forum, I asked what the purpose or function was of the resistor/capacitor array shown in the red boxes above; 35-odd peeks, no replies.

Superficially, the only difference between channel 2 in the GA20 and channel 1 in the GA40 was the filter? resistors and capacitors in the green box below. I added a switch and a capacitor and bypassed that entire array as shown below. The upper red circle is the switch shown schematically; the red circle on the right is the switch sitting in the hole formerly occupied by the Polarity switch. I picked up the signal as it was entering the green box, interrupted it with the switch, added a capacitor, and terminated it on the channel 2 volume switch where it normally would have appeared after exiting that array. The two red arrows indicate the points on the circuit board where the signal was picked up (left) and tied off (right). The added cap appears as a yellow cylinder in the upper center of the pic of the circuit board just below the indicating arrow. When the switch is off/normal, the signal runs through the filter?/array. When the switch is on/bypass, the signal follows the magenta line, goes around the green box, and shows up on the volume control.




The result is dramatic; not yet to the hot, swirling, smoking thing but that 5879 sounds like a 12AX7 on musical steroids. There's a little hum because the signal is carrying DC all the way to the cap and the wire is unshielded and passing the power switch and transformer and the volume controls ... but that's easy enough to fix but I'm getting pumped about my Half GA40!

Left to do; re-work the bypass switch - not just the hum but determine whether the outbound signal is a mix of bypassed and filtered signal, replace the preamp cathode resistors which I was too lazy to do today, re-finish the mahogany footswitch ... these things are just killer .... (footswitch on Bluesdan's GA18)




... and keep looking for some fresher 6EU7s; this amp was delivered with its original Gibson-logoed 6EU7s. I have fresh ones in the GA30 ... and they work fine and replace the knobs; this amp was sold with ox-blood chickenheads and its knobs are currently on the GA30. I really won't know what this amp can do until the speaker breaks in and I dress up the bypass mod but if I got channel 1 to come close to a GA18 and channel 2 to resemble channel 1 in a GA40 ... I'd have to think of this as a badly under-rated amp. CJ

What about a push pull pot instead of the switch? Short signal run, and "stealth", important with amps as well as Buick Skylarks with 2" duals and open element air cleaners. :wink:





"Mine was Silver Green"

Thanks Steve; I'll look around ... most of those pull on/push offs pots are the 10K used in Fenders for boosting mids ... but stealth is good; sweet ride too!

Default said:

What about a push pull pot instead of the switch? Short signal run, and "stealth", important with amps as well as Buick Skylarks with 2" duals and open element air cleaners.

Click to expand...

capnjuan said:

most of those pull on/push offs pots are the 10K used in Fenders for boosting mids ... but stealth is good; sweet ride too!

Click to expand...

... but I do have a 1meg audio pot w/ SP switch-on contacts on it. I could stealth it in as the channel 1 volume control. If played through channel 2, the channel 1 volume control could be turned right just enough to operate the bypass. If played through channel 1, whether the filter in channel 2 were bypassed wouldn't make any difference ... Eureka and Myreka ... :mrgreen:

I love reading this stuff...........keep it coming! 8)
Crystall ball sees a GA77 Vanguard in the Caps future..... :wink:

Hi Dan and thanks; dunno just yet about the GA77 .... because this is what's driving the car right now:

Talking about his GA40LP in the Tonquest thread to shamelessly and irresponsibly enflame CaptJuan said:

The GA40 has a dark, hot sound, swirly and alive sound. West coast hot, jazzy too. Not much clean ... The amp itself is compression rich, so I can do without early alnico breakup. The clean you do get is so rich, dense, and the overdriven tone is not easy to describe. VG called it "chocolate thunder" ... funny but somehow appropriate.

Click to expand...

As the expression goes; "I gotta get me some of that!" :wink: These GA20Ts are dangling the possibility, except for a 12" and not a 15" speaker, of getting GA40LP-like performance out of channel 2 and something at least similar to the GA18 out of channel 1.

... and these re-coned Jensen P12Qs / Weber 12A125s showed up today ...




One of which is going in this 6EU7 GA20T. Yes ... I'm being greedy here ... going for the inside-the-park home run but I think I have 3rd base for sure. Besides, as often as I screw up and take my gear to the amp shop, that GA77 doesn't have a chance to get too far off the radar screen. John

Ok ... this is about it for the 1st of the two 2 GA20Ts (the 6EU7 model); on the left, with a re-coned Jensen P12Q / Weber 12A125, a set of fairly well-matched Rowe-branded Sylvania 6EU7s, an RCA 5879, matched CBS/Hytron 6V6s, and a military-grade 6106/5Y3 rectifier. On the right, with its new baffleboard and original grillcloth; I have the badge - it'll go on when I refinish the footswitch.




Stealth Steve had the right heading on this; hide the switch under the veil of a volume control which I did. Hoffman Amps sells a 1 meg audio taper pot with a set of pull-on/push-off contacts on the back that I substituted for the channel 1 volume control; the position of the switch contacts has no significance on channel 1.

When the switch is pulled on, the R/C filter array is bypassed and the 5879 output goes directly to the set of contacts on the back of the channel 1 volume control indicated below by the magenta arrow. The yellow .01 cap in the magenta box blocks DC from appearing ultimately on the channel 2 volume control and the yellow wire connects the cap and the switch contacts. The yellow wire continues right and feeds the channel 2 volume control.




The two blue boxes are the 25uf/25V cathode resistor bypass caps that I added and were discussed in LTG's Voicing Amps thread some time ago. The bypass caps change the bias of the tube and its frequency response. In this case, they act like a low-pass filter allowing lower frequencies to pass while blocking some of the higher.

The effect in this amp is to provided a fuller-bodied tone; somewhat more bass but a generally thicker, more compressed signal - they tilt the amp towards a little darker from, at least in my opinion, the overly bright tone without them. If, downstream, someone were to prefer a brighter tone, these caps can be easily snipped out and, for that matter, the bypass mod could be removed by snipping the coupling cap at the circuit board and at the switch.

The amp has speaker with a new cone and rather than try to describe its tone and what it can and can't do, I'm going wait a while for the cone to break in. However channel 2 with its 5879 is unlike any 12A_7 channel on the planet; firey hot, pulsing, edgy, cleanish, loads of gain and with a smoldering, undulating quality in the sustain. All I need to do now is go buy a GA40LP for nearly 3 times what I have in this or a Victoria Electro King for nearly 6 times as much to see just how close I got to the target! :wink:

When this amp was at the shop getting the grounding straightened out, the tech offered to draw a schematic; there just isn't one out there in Webdom for this weird version. When he gets done, I'll forward it and pics of this amp to Mr. Wally Marx whose Gibson book arrived today ... he missed this version ... and a few others ... see comments in that thread. CJ

Hmmm......GA18 on ch1 and a GA40LP on ch2.............sounds like the ultimate hybrid!

capnjuan said:

However channel 2 with its 5879 is unlike any 12A_7 channel on the planet; firey hot, pulsing, edgy, cleanish, loads of gain and with a smoldering, undulating quality in the sustain.

Click to expand...

:shock:

Back in a few, I gotta take a cold shower...

BluesDan said:

Hmmm......GA18 on ch1 and a GA40LP on ch2 .............sounds like the ultimate hybrid!

Click to expand...

Hi Dan; yeah, a man's gotta have dreams; ... open for the Rolling Stones at Shea Stadium ... break par from the tips ... have Catherine Zeta-Jones return his calls ... and I have another shot at the brass ring. The other GA20T - with the tube lineup that matches the public schematic - I think will already be a little closer to the 'hybrid' mark. I'm pumped! :wink: J

Default said:

capnjuan said:

However channel 2 with its 5879 is unlike any 12A_7 channel on the planet; firey hot, pulsing, edgy, cleanish, loads of gain and with a smoldering, undulating quality in the sustain.

Click to expand...

:shock: Back in a few, I gotta take a cold shower...

Click to expand...

:lol: Victoria, Divided by 13, and others are producing amps built around this tube ... not sure if anything else back then used these ... but it's a 9-pin tube and with a little fiddling to a 12A_7 preamp socket ...

As they used to say on the Bob Newhart show ... this is my other brother Darryl (not you Hatz :wink: ) ... the other GA20T... it arrived at Juanzampz in an old coat of light gray paint:




Upper left with its coat of paint; the other three are after successive coats of Strip-Eze and Easy-Off oven cleaner. Unfortunately, the vinyl surface of the tolex had either been scuffed to help the paint adhere or was melted by the Strip-Eze. As a result, the Strip-Eze pulled off not just the paint and its pigment but the pigment in the cotton tweed backing fabric as well. What was left was the color of linen and as rough as #60 sandpaper.




I sprayed it once with clear varnish, coated it in diluted amber shellac, and a final coat of gloss spar varnish ... and what it looks like now. On the right and from left to right; 6087/5y3 rectifier, twin Ratheon JAN 6V6s and an odd-assortment of preamp tubes including an RCA 5879 next to the last, and a re-coned Jensen P12Q:




Side views show spotchy, inconsistent surface damage and uneven absorption of the varnish and shellac:




Closeup of some of the more uniform surface;




In the upper half of the pic below; the circuit board was flipped (again :evil: ) becase first go-around, I only had one of the three .047 tremolo oscillator caps; the trem didn't fire up, and so back again to replace the two bumblees next to the yellow Mallory in the box.



The lower half of the pic above shows the filter bypass mod in place; the cap going left/right in the lower, right-hand corner is picking up the signal directly from the plate of the 5879 and blocking off the DC. The cap is connected via the yellow wire, routed under the board to keep it from flopping around, to the wiper of the channel 2 volume control pot. When it gets here and as above, I'll switch out the channel 1 volume pot for one with a set of pull-on/push-off contacts. Before someone decided it 'wasn't stock', they ought to hear what they were giving up by reverting to stock by un-doing the bypass.


Simplified drawing of the bypass mod; that set of caps and resistors just chokes it do death ...




And here it is ... if I was Scratch, I'd probably call it Buckaroo on account of its color. The tolex is fully adhered, the corners are tabbed down; it's as tight as it can get. It's one of those things ... it's just how it came out ... this is not the amp for a girlie-man ..




Each of these two amps came with an opportunity in their layouts; a blowup of the channel 1 preamp below. Notice that each jack leads to a tube half as opposed to being summed at the input of the first half. In the GA18 for example, all the inputs lead to V1A then to volume and tone controls, and then to the other side of the tube, V2B. Since the player can only plug into one jack, the other 1/2 of the tube does nothing.




Bluesdan asked if I was going to finish this as a GA18 in channel 1 and a GA40 in channel 2 ... the answer is not just that configuration but because each input is tied to it's own tube 1/2, one input can be voiced like a traditional GA18 with a 1.2K cathode resistor and 25uf/25V bypass cap and the other 1/2 of the tube will also get a 1.2K cathode resistor and maybe a 'Marshall' .68uf bypass cap ... thick, rich, rounded, and compressed in jack 1 and something with more pronounced high frequency, brighter, more intense in the other jack. Did not figure on this but it has presented an opportunity that I'm going to try to exploit ... three separate, very diferent voices from the same amp.

Not quite finished with a few items; re-finish one footswitch, re-re-finish the other because ... I scratched it requiring a do-over. Both get their badges replaced, both will get the channel 1 preamp voicing, and this one probably needs another tube ... getting a ringing/whistling on top of the notes ... could be something microphonic, who knows but the 5879 ... that's another story; gainy, warm, driving, full frequency ... more depth that a 12A_7 ... more!

Gee, I'm 50/50 on suggestions.
+1 on the switched pot
-100 on refinishing the tweed. :evil:

Default said:

-100 on refinishing the tweed. :evil:

Click to expand...

One of those deals where you start without knowing where you'll end up. Not being defensive but it actually looks better in person; except for the torn corners, it looks like it was professionally recovered with something somebody spilled stuff on. Wanna take a shot at this? May seem familiar ..... referring to the pic of the 12AY7 above;

1. Is there an argument for replacing the input caps with resistors?
2. Value for the grid resistors?
3. Would .47uf as a cathode resistor bypass cap do enough to justify its presence?

What year is this one again?