1959/60 era Tweed Gibson GA-8T Discoverer

Alas, when I took the before photos, I was very sick (in the early stages of shock) but didn't realize it. As a result, many of the before photos were not usable.

Control Panel (would be on top)

This amp had been worked on. The PS cap (probably the usual Gibson "cigar" cap) had been replaced with a metal can cap inside the chassis (a short waiting to happen, in my book). The power tube cathode resistor bypass cap had also been replaced. A power resistor had burned, possibly burning the power transformer, and another power resistor (a wirewound sand resistor) had broken and possibly burned. Dick (the owner) was concerned about the power transformer as well as the resistors.

A Discoverer's Guts Laid Bare

This amp uses a 12AX7, two 6BM8s, a 5Y3, and has a lacquered tweed cabinet and a 10" speaker. The two 6BM8s were biased for a little over 8 watts static dissipation each for a total of 8 to 10 watts output. 8 watts is a bit over the stated rating for these tubes, but the specs were conservative. The tubes showed no sign of strain at all, no matter how loud and hard I played.

The power transformer turned out to be fine other than having scorched paper. I replaced the two resistors in the power supply, hooked the amp to a 12" Vintage 30 clone, and fired it up. It sounded very lackluster; the tone was mediocre, and the attack was very mushy at all volumes. The tremolo worked, but the range was fairly narrow, biased toward fast. With the frequency knob at the half way point, the tremolo was speeding into blurriness, barely noticeable as an effect at all. There was also precious little low end. As it arrived at my shop, it sounded like a really bad, 1960s solid state amp.

After replacing all caps except the can cap in the PS and the power tube cathode resistor bypass cap the tone was much better. I used larger caps than Gibson in a couple of places, notably replacing the 500pF coupling cap with a .022uF. I also used a .068 instead of a .022 as the first coupler, and went with .047s feeding the power stage grids. But the attack still suffered, and there wasn't as much sustain as there should be. I replaced the PS can cap with individual Xicons. I stayed with a 20uF first stage to keep the sag characteristics I hoped were there, but used 47uFs for the other two caps, to reduce hum. Suddenly the little monster came alive!

On to the tremolo. I put .047s in place of the first two .022s (closest to the plate), and the tremolo improved quite a bit. I tried an .047 in the third position as well, but now it dropped off at both ends of the frequency range. A fresh .022 in the third position was just right. Now the tremolo has a much wider frequency range, and never fades out.

How does it sound? The clean tone is excellent. It'll do anything from jazz to clean Hendrix to steel guitar. The crunch is as good as any tweed Deluxe, but with a flavor all its own. This is a very sweet sounding amp!

Power Supply Section: You can see the broken sand resistor below the pilot lamp. The death cap is still in place. The big. black cap toward the left feeds the tremolo waveform to the controls. The brown zip cord isn't the power cord, it's the cord to the tremolo footswitch. The black cord coming in from the lower right is the power cord. Whoever replaced it didn't bother with a grommet, and the cord was already being scraped up by the chassis; eventually this would have shorted power to the chassis. Not good!	Power supply closeup: The power supply components are all flying lead (classic Gibson). The 1W resistor poking out from under the broken sand resistor is the one that was partially burnt, and which scorched the power tranny paper. You can see the knot in the power cord for strain relief. You can also see the brown chicken head knobs. These are some of the best looking chicken head knobs ever made, IMO.
Another power supply view: There's not much new to see here except that the cap coupling the trem circuit to the rest of the amp has no bolt holding its strap in place.	Power supply cap: The PS cap is supported at its base end by a strap covered with electrical tape to hold it in place. Unfortunately the rest of the can is bare metal, grounded through that strap and a ground wire. Most likely this bounced against one of the terminal tags just visible below it in the photo, causing a short which burned the PS resistor. I should have mentioned that the amp came with a 4A fuse installed!
Knob closeup: I wish I had a better photo of these knobs. I took this off an otherwise too bright photo of the control panel.	Output transformers: The smaller transformer is the original. Gibson consistently used undersized transformers on their guitar amps. The new tranny is a Heyboer transformer designed for a Marshall 18 watt clone project elsewhere on the web. Heyboer does an excellent job. This Heyboer came with taps for 4 ohm, 8 ohm or 16 ohm speakers. The primary impedance was just a bit higher than that of the Gibson transformer. It adds a lot of clarity, more clean headroom, and more bass.
Component board: Here's the component board with all new caps. The trem coupling cap (the biggest, fattest, yellow cap) doesn't really need a strap. I therefore used the old trem cap strap and mount point to hold the new power supply caps, and just left the trem coupling cap floating on its leads. Any jolt bad enough to tear the cap loose will do much more serious damage elsewhere. As you can guess, it gets a bit tight with the soldering iron in some spots.	New PS caps: Here you can see how I mounted the new PS caps. I placed the strap around the 22uF cap, and used two cable ties to join the three caps into one bundle. I soldered the negative leads together. I used heat shrink tubing on the positive leads. I have used this setup in a number of amps with great success.
Complete guts: Here's the whole ball of wax. You can just see the new 2W PS resistor sticking out from under the left side of the new 5W resistor in the lower right part of the chassis.	Tone control caps: This photo mainly shows the new tone control caps. I didn't have a .005uF cap, so I used two .022s in parallel to get close. The brilliance cap to bypass the volume control is a 390pF silver mica.
Complete amp: The plate on the left with the two jacks is the new speaker jack plate. The customer requested jacks for the 8 ohm and 16 ohm outputs. He'll bolt the plate into his cabinet somewhere. I provided 18" of braided cable from the OT terminal strip to the bracket, using cable ties at each end and a cable tie holder on the bracket for strength. I prefer braiding over twisting; it stays like I want it and it looks cool.	Speaker Jack Bracket: Closeup of the speaker jack bracket. Since with multiple taps you can't easily use shorting jacks to protect the output section in case the speaker gets unplugged (or blows), I added a 200 ohm 5 watt resistor across one of the outputs. This will protect the amp for the short time it should take to realize there's no sound coming out and turn the amp off. I used heat shrink tubing on the resistor leads, and tied the cable down for safety.

Schematics