Gibson GA-95 RVT

[Travst]

I'd put a new socket in V5 after one of the old tabs broke off, so I have a clean slate to work with. I pulled the wires to p6, p7, and p8 and tested.

V5-1 274v
V5-3 6.7v

V4-1 228
V4-6 225

I also swapped out both 6EU7 tubes before testing. Not much difference in voltages and no volume gain.

[Phil_S]

I think 274/6.7 are very good readings. The schematic says 242/8.2. Higher plate voltage should result in lower cathode voltage. I think that one is as good as its going to get. Socket replacement turned out to be a good idea.

So, now we probably work on what makes V4 plate voltage so high. It is still 2x what it should be. I will see if I can get some readings from my GA20-RVT. This requires pulling the chassis, which is a bit of a PITA. I may be a day or three getting this info.

The problem is bilateral. I don't have any "light bulb" level ideas on this one. The only other common part is the B+ rail. Maybe the cathode grounds are bad?

Any change in how it sounds with the new socket?

[Travst]

There's really no change in tone. The effects channel now sounds pretty good (but in need of some tone tweaking) and the volume is several times greater than when we started this marathon.

Can I reattach the trem wires?

[Phil_S]

I'd leave trem and reverb off line while diagnostics are going on. I'm hoping someone with stronger chops than I have will come along with some new insight.

I'm curious about current draw of the power tubes. Can you measure the ohms on each half of the primary side of the OT? Pull the power tubes. Find the CT. Measure ohms from the plate pin to the CT. Get one reading for each half. We know the plate voltage. We can calculate static plate dissipation. Maybe we just have cold bias and need to lower the bias voltage. I"m thinking that -44V is on the cold side.

[Firestorm]

Travst --

What tube are you running in V4? I don't think we'll see close to the schematic numbers there without a proprerly functioning 12AU7 (which we don't know if we have). No other noval tube I can think of will pass enough current to pull that plate down or push that cathode up to the schematic's numbers. A 12AT7 will pass only about a third as much juice; 12AX7 is tiny.

I know V4 is common (sort of) to both channels, but I think the problem must be somewhere else. Maybe Phil has something with output tube bias. Certainly worth checking.

[Travst]

Thanks Phil.

[Travst]

I have a 12au7 in v4.

[Phil_S]

Maybe Phil has something with output tube bias. Certainly worth checking.

LOL, for a guy who doesn't know that much it sometimes comes as a surprise when someone else actually agrees with me. Looking at the schematic this morning, and assuming bias voltage is contributing to the problem, I'm thinking we should covert this one to adjustable bias voltage. I see R53 is 33K to ground. I'm thinking a 25KB pot in series with a 10K or 15K resistor will do the job. I'll go out on a limb here and suggest that bais voltage needs to be in the mid to upper 30's, not 43 or 44 like we are seeing. We'll know more when Travst gives us the DCR of the primary side of the OT.

[Travst]

I think adding a bias pot is a great idea regardless. I'll locate a pot and plan to drop it in. Hopefully I can find a suitable pot locally, as I don't have anything close to that value on hand. Can you describe the hookup of the pot in more detail? I'm guessing that one outer lug would go to R53 on the ground side, center lug to R57/58 cup, and the other outer leg to the - side of C31.

I'm primed to get this thing fixed over the weekend.

[Phil_S]

Connect one outer lug and the wiper. I like to tie the other outer lug to the wiper, effectively making the pot into a variable resistor. I'm not good at circuit design. I suggest you wait for someone to come along and say that 25K + 15K is the right thing to do.

The basic logic is that the fixed resistor provides a "floor" so it never goes to zero when you rotate the pot. The question I have is about range and the best choice for the fixed R. The answer is related to choosing the best floor. It may be better to have a 10K or 5K pot and a larger R. If you need to mail order, I'd look at Apex jr. and Small Bear electronics. You should get a proper trim pot or a locking pot that mounts through the chassis.

No hurry on this. You can just change the fixed R and it will work fine. You can try a 27K, but don't do anything until we know your static plate dissipation. Please get us the DCR on the OT and I'll show you the calculation.

[Travst]

I'll start with measuring the DCR and taking a shot at calculating the dissipation. I need to try to apply as much of the math as possible, and have some good reference materials.

[Firestorm]

If you want the bias to be externally adjustable, you should eventually add the 1R current sense resistors and pin jacks so you can tweak it while it's still in the cabinet. You can still get the screwdriver adjustable pots Fender used, but I think only available in 10KL. 10K pot with a 22K series resistor should work, giving a range of 22K-32K. 25KL pot with 15K works too, with a range of 15K-40K. With a standard carbon pot, the higher you go in value, the "jumpier" the voltage will be (tiny little tweaks can make a lot of difference). As Phil says, wire the pot as a rheostat with the wiper to either lug (depends on which direction you want to be "more" and which way "less"). Just be sure not to take the bias voltage off the wiper; if the wiper opens up, you want the pot to fail to higher negative voltage, not to zero negative voltage.

It's going to be a little bit of a pain working with the bias supply. The caps you've labelled C31? and C32? are ginormous. The schematics don't show a voltage spec so somebody erred on the side of caution. 75V or 100V would have been plenty. I can't even see the diode. Also can't tell if the caps are in right (anode to ground).

[Firestorm]

I'll start with measuring the DCR and taking a shot at calculating the dissipation. I need to try to apply as much of the math as possible, and have some good reference materials.

I take it we're reluctant to measure the current directly. If your meter's good enough, shunting the OT works fine. Ken's preferred method.

[Travst]

I'll start with measuring the DCR and taking a shot at calculating the dissipation. I need to try to apply as much of the math as possible, and have some good reference materials.

I take it we're reluctant to measure the current directly. If your meter's good enough, shunting the OT works fine. Ken's preferred method.

I'm not afeered of nuthin'. Looking over the shunt method, it seems very straightforward. Whether my MM is up to the task is another question.

I'm using this (probably time to upgrade, as I'm using it a lot now):

http://www.radioshack.com/product/index ... Id=2103176

[Firestorm]

Don't know about that model specifically, but I have the 46-range, logging 22-812 and it works nearly as well as my two Flukes for measuring current that way. Can't hurt to try it (IF YOU'RE CAREFUL) and you can validate the results by doing the math.