Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Gibson GA-77 "Fidelity" pot value?

9/5/2000 5:00 PM
Jim S.
Gibson GA-77 "Fidelity" pot value?
I'm almost done restoring a client's Gibson GA-77 amp. This is a terrific late-50's tweed amp that is patterned very closely off of a Fender tweed Pro. It is in great condition and has an original cone Jensen P15N speaker. Main differences from a Fender Pro: V3 is a 12AU7 instead of a 12AX7; no local neg. feedback look around the V2 direct coupled cathode follower stages; bass/treble tone circuit is different; second channel has a tweed Deluxe-style tone control circuit instead of a fixed bright cap across the volume pot.  
Here's the question: During pay testing, I noticed that the Fidelity control (which is really just a presence control) has no effect, no matter what its setting. I double-checked my wiring and later spoke with the client who confirmed that this had always been a "do-nothing" knob. I measured that pot (which appears to be original). It appears to be 50K audio; at max CW setting, there's a residual resistance of around 600 ohms between the two terminals in use. The schematic that came with the amp's instruction manual says "50 lin.". The GA-77 schematic in the Groove Tubes Amp book (which is a completely different drawing for the exact same circuit) says "50". This presence circuit is identical to the style Fender uses for a cathodyne PI tweed amp, except for the pot value. Fender uses 5K linear.  
(The presence circuit is connected to the junction of the 100K -fb resistor and the 1K5 cathode resistor of the pre-PI driver stage. The pot acts as a variable resistor connected to a 0.1uF cap that is connected to ground. As the pot is turned CW, the resistance is lowered, causing two things to occur at once: (1) the higher frequencies contained in the -fb signal are shunted to ground; (2) the 1K5 cathode resistor becomes more bypassed by the .1uF cap, increasing high-frequency gain in that stage.)  
50K seems way to big, but 50 seems awfully small. Maybe they left the "K" off the schematic? Or maybe the wrong pot was installed at the factory? Would the fact that the driver stage and cathodyne PI stage use a 12AU7 instead of a 12AX7 have anything to do with the departure from the Fender pot value?  
I'd appreciate hearing from anybody who has any insight into this amp and/or presence controls in this type of circuit. Thanks.
And now, a word from our sponsors:

9/5/2000 6:13 PM
Jim S.

Here's some additional info on this cool amp:  
- The model name for the GA-77 is "Vanguard".  
- Tweed is well-lacquered, and is of a lighter color than Fender tweed.  
- Rectifier tube is a 5V4G, instead of typical 5U4G or 5AR4. (This particular amp though had a non-original Bugle Boy "Made in Holland" 5AR4 installed.)  
- This amp is a lot harder to service than Fenders. The original main filter caps are a pair of dual-section radials (w/3 leads coming out one end) mounted to a bracket. All the uF size coupling caps are mounted UNDERNEATH the tag board. They have no markings except color bands. They were leaky, so I assume they have paper dielectric. I replaced them all with SBE 6PS series orange drops. I had to remove all the wiring from the tag board and remove the board from the chassis in order to do this. Very time-consuming.  
- Cathode bypass caps were a pair of 20uF (rather than Fender 25uF) e-caps, also mounted underneath the tag board. I replaced these with Atoms of the exact same value, but mounted the new ones on top. (Hopefully, nobody will need to remove the tag board EVER again.)  
- Lead dress is not as good as vintage Fender, but not terrible. Hookup wire is plastic coated stranded, not cloth-covered solid.  
- I don't know what these amps are worth on the vintage market, but they could be fantastic bargains, especially for someone with the ability (and patience) to do the electronic restoration themselves.  
- I replaced the main filter caps with individual axial caps. I replaced the mounting bracket with a terminal strip and put two new caps on there. The other two new caps were mounted onto the tag board.
9/6/2000 4:47 AM
Peter S

It's been nearly forever since I've seen one of these, but I seem to remember it being a 5k pot. I don't see why subbing a 12AU7 would significanlty change the operation of the circuit, at least not enough to require an increase in pot value by 45k. 50ohms is so low that I don't see how it could work that way. Maybe you can try some resistors values and see what kind of response you get from them.  
Peter S
9/6/2000 3:28 PM
Jim S.

"I don't see why subbing a 12AU7 would significantly change the operation of the circuit"
I was wondering if part of the problem, in addition to the pot value being too large, is the 100K -fb resistor coming off the speaker output. Are you saying that this resistor value is appropriate, now matter what the gain factor is of the pre-driver and PI stages? I was wondering if a larger -fb signal is required to compensate for the reduced gain of the 12AU7.  
"Maybe you can try some resistors values and see what kind of response you get from them."
Yes, I'm planning to remove the two wires from that pot and hook them up to my resistor substition and see what the useable range of resistances are. I'll let you know what I find out.
9/7/2000 3:19 PM
Jim S.

I found that nothing seemed to make a difference in sound. Even shorting out that pot did not add any "presence" to the guitar tone. Anyway, I ran out of time and will not be looking into it any further unless the customer asks me to and is willing to spend more money.  
I DID notice that the 5K (linear) presence control on my homebrew tweed 3x10 Bandmaster is very subtle, and hardly does anything between 1 and 9.
9/8/2000 2:45 AM
Peter S

Sparks a IS a very subtle, almost do nothing control, but I have no clue why it doesn't make ANY change at all. Maybe try lowering the value of the feedback resistor?  
Peter S
9/8/2000 8:55 PM
Jim S.

I did, by clipping the leads of my resistor substitution box on each side of the 100K fb resistor - but this caused some HF oscillation (due to the long leads?) after decreasing the net R to around 50K or less.  
In any case, time ran out and the amp is back in the client's hands. I told him he might experiment with subbing the 12AU7 in V3 with a 12AT7 or a 12AX7 to get more gain/distortion out of the amp. The clean tones are really nice, though.

  Page 1 of 1