Gibson GA-95 RVT

[Phil_S]

Sure you've got a low voltage a/c source. Pull all the tubes and install some gator clip jumpers on the filament supply. That will give you 6.3VAC to test T3. You have 3 wires to lift on the tx -- 2 primary and the CT on the secondary. Pull the power tubes and find the output reading at the grid pins.

T3 on my amp measures as follows:
Primary DCR (red to blue) is 724 ohms.
Secondary DCR is 3436 ohms (green to green/white), 1834 ohms (green/white to yellow CT), 1603 ohms (green to yellow CT).

Are you sure that you measured correctly? If you measured without lifting the leads from the circuit board, you are measuring the circuit, not the tx. You've got to lift the wires. I'm having some difficulty imagining your tx is very different from mine. Maybe the secondary is different, but the amp circuit to the primary is nearly identical, so I'd expect the same primary. I think it is reasonably safe to assume, if you get ohm readings similar to mine, you can rule out the tx as part of the problem.

You need a lantern battery to run the super secret tx test! The little batteries don't have enough oomph.

On the other hand, you say the amp is working. Maybe we don't worry about it?

Good luck.

[Structo]

John I didn't really read that article about testing the tranny but a tranny needs to see alternating current or pulsating DC to work.
It's the change in voltage or amplitude that is "transformed" across the core from the primary to secondary winding.
If the secondary has more windings then the primary you will get more out than you put in and if the secondary has fewer windings than the primary, you will get less than you put in.

That's where the terms "step up" and step down" came from.

To me it sounds like T3 is probably OK because of the voltages you measured there.
The fact that you get -43 volts on the secondaries for your bias voltage is exactly what the schematic calls for.

Phil has given you a lot of good information and with his help since he knows these particular amps, you should be able to get it going in no time.

[Travst]

I'm going to take the time this week to go through all the rest of the suggestions. It keeps getting better and better, but there's just more than a single problem to work through. I've gotten a lot of valuable exerience on this amp and I appreciate all the pointers. I guess I'll tackle a Dumble next... can't be more complicated than this over-engineered Gibson.

[Phil_S]

No doubt about it. This is a difficult amp to work on. The board is a rat's nest. The schematic is fuzzy. It is not a particularly good implementation of a mediocre design. All that said, there is a pile of potential in the amp. Get it running right and you will be able to get it into a zone you'll like. The reverb and tremolo circuits are first rate.

[Travst]

After a long interlude in which I built another amp and a guitar, I'm back to working on this amp. I picked up a 12/6v transformer and am able to supply the low voltage needed to test T3. My understanding is that I clip the two transformer wires to each of the two primaries and measure the current at the center tap (wire going to 33K resistor R48) to measure the turns ratio. I'll be doing that this afternoon.

I noted that the wiring in the amp differed somewhat from the schematic, at least to my eyes. First, the secondary going to V7 shows the lead going to Pin 5, while the wire was actually connected at Pin 1. Second, the lead going to R46 and then V10 Pin 1 was actually connected to V9, Pin 1.

Are these difference cause by changes in the design for which I have no schematic? Or, did someone mis-wire T3? I'm curious as to your thoughts on this.

[Phil_S]

No, you need to lift the CT and measure voltage on the outer legs. You are not measuring current (Amps). Here's a bit of reading on the topic:
http://www.radioremembered.org/outimp.htm
http://sarris.info/main/calculating-out ... -impedance

You may ask, what does calculating impedance have to do with this? Actually, not that much. What you are trying to verify is whether the windings are intact and not shorted. To do this, you want to confirm the turns ratio. If this is a PT, then the turns ratio will tell you what to expect when you apply line voltage to the primary. If this is a OT, then you do want to go through the calculation of the impedance ratio to be sure it makes sense as a match to the power tubes.

Edit: I finally found this amp is equivalent to the EA300-RVT and have a schematic. Pin 1 is not used on the 6L6 power tubes V6-V9. The build uses this as a simple tie point because the pin is unused.

I see you are testing the interstage PP transformer. If this is the same as on the GA20-RVT or the GA15-RVT, the turns ratio on this should be about 1:2.5. So, 6V in on the primary will generate about 15V out on the power tube side. You may want to run the test in the other direction, 6.3V in on the power tube side will give you about 2.5V out on the primary. This will be less stress on the transformer, which is rather small.

Good luck with it.
Phil

[Travst]

Thanks Phil. I haven't addressed the OT yet, but will do that next. I've lifted the leads on the interstage transformer and will be testing it today as well.

[Travst]

Here's what I found:

T3 - I fed 6.9v in from the other direction as suggested. I calculate the turns ration at 1:2.55, which is consistent with the readings on Phil's amp. So, I soldered everything back into position and tested the OT.

OT - I fed 6.9v in from the other side and got a reading of .24v. 6.9/.24 yields a turns ratio of 28.75:1.

For the next step, I squared 28.75 to calculate 826.56:1 as the impedance ratio.

Does that sound right? It seems to approximate the addition of the voltages going into the OT.

During the downtime, I also checked out the rest of the components and replaced a couple more resistors and corrected a cracked solder joint. After testing, there was no improvement.

[troobka]

In your opening you mention replacing 470k resistors on the 6l6s. Did you mean 470 ohm? I think that would make quite a difference on in volume at the output stage.
All the best,
Mark

[Phil_S]

Ok, so let's figure .24v isn't so accurate for a moment. I'd run the OT test in the other direction so I could get voltages that are more reliably read by your meter. Anyhow, working with what you got...

The schematic shows two 8 ohm speakers wired in parallel. Please verify this, as it is crucial to reaching the correct conclusion. 8||8 = 4 ohms. If your turns are 28.75:1, that says the primary is 3300K. That is certainly within reason for a quad of 6L6 at 440V (also from the schematic).

I think your OT is OK and it is time to look elsewhere.

Have you got the schematic for the Epiphone EA300-RVT? Look at Schematic Heaven.

Can we get a voltage chart for all tube pins? You can omit the heater readings (too much clutter to sort through), but to check to see that there is proper heater voltage at all sockets. We are lucky the schematic indicates voltages. See if you can identify where the voltage isn't right.

[Travst]

In your opening you mention replacing 470k resistors on the 6l6s. Did you mean 470 ohm? I think that would make quite a difference on in volume at the output stage.
All the best,
Mark

Typo... or brain meltdown. They are 470 ohm resistors. Good catch!

Phil, I'll run the voltages again in the next few days and update the numbers.

[Travst]

Here's the latest chart:

[Phil_S]

Please ignore most of what I said about bias voltage in my earlier PM. Bias voltage is very good at -44 on pin 5. I don't know what that ~-4.5 is on pins 2 and 7, but I'm going to ignore it for now. Pin 1 is indicated as the "shell," meaning it isn't internally connected, and on a glass tube it probably isn't connected at all. making it available as a tie point. It appears to have been used as a tie point for the bias voltage circuit, so I'm going to assume that's OK.

Pins 2 and 7 on V6-V9 are heater voltage. You measure that by putting one probe on pin 2 and the other on pin 7. Set the meter for VAC. You should see about 6.3VAC. This seems to true for the other tubes, as well. For the 6EU7's heaters on on pins 1 and 2. For the 12AU7's heaters are on pin 9 and on pins 4+5 which are jumpered together.

I think I see a few typos in your voltage chart. Those tiny numbers are hard to see. For expected voltage from the schematic:
V4 pin 6, 94V
V5 pin 3, 8.2V
V5 pin 6, 94V
V5 pin 8, 1.65V

What pops out at me from the voltage chart is that V4 has plate voltage that is ~2x what it's supposed to be (pins 1 and 6) and cathode voltage (pins 3 and 8 ) is ~40% of what it's supposed to be. I suggest you swap V4 for one of the other 12AU7's (V2, V5) to see if this problem follows the tube or remains with the socket.

Let us know.

[Travst]

This is my latest voltage worksheet. I'm a bit confused on C30 and C31, but I think I have the other caps right. There are a few things in this thread I haven't tried and I'll back up and begin working on them.

[Phil_S]

I have attempted to improve your voltage chart a little. I hope I did that and didn't make it worse I've indicated certain things, like the heater pins, which you don't need to read or report, and where the pin is not used by the tube, but may be used as a connection point by the circuit builder.

I don't understand some of your cap notations at the bottom and left you a few comments in the worksheet. I don't see R50 (22K) as a match to the schematic; same for R47. These are both at the lower right, part of the tremolo circuit. I've indicated what I think you may be looking at, but you'll need to say what for sure.

I suggest one overriding problem is the voltages on V4 are all wrong. Until this is corrected, the rest of the amp is not going to operate well. The circuit is interactive, meaning that fixing V4 will likely change all the rest of the preamp tube readings to some extent. Unfortunately, I am an amateur, not a professional tech, so I may be near the edge of my capability. I'll be candid with you when I'm not sure of something.

I don't have a clear theory on why the voltage is so high on V4. The good news is that V4A and V4B each perform the same function; which is a simple final gain stage, one for the reverb channel, and one for the normal channel. They are biased alike, so the results should be nearly the same for both sections, which they are.

I suggest tracing the plate voltage supply backwards from the tubes and record what you find at each spot.
1) There is a 100K plate load resistor for each section (pins 1 and 6). Trace these back to where they are joined together. What is the voltage there?
2) That junction should feed a wire(or simply join) to the junction of C29A and R43 (4.7K). I'd look for the 4.7K on the board and a wire to C29A, probably a red wire. What is the voltage to ground on each side of R43?
3) Follow C29A/R43 back to C29B/R44 (4K 5W -- a physically large resistor). What is the voltage on each side of R44?
4) R44 is connected to V10. This is a 0A2 voltage regulator tube and this might be a source of the problem. See if you can get a voltage reading on the plate of this tube. It could be either pin 1 or pin 5.
5) R44 forms a junction with R55 (4K 10W, a really large resistor) probably on a terminal strip, not the board. Find this junction. The other side of R55 is connected to C30B. You report 474V there, which is easily within the margin of expectation for an expected value of 448V. This is particularly true because line voltage is higher today then when the amp was built.

I'd also like to know the voltage reading at the junction of the two 100K plate load resistors for V1 (pins 6 and 7). This should be the same as #1, above. Let's check and make sure.

We are looking for the spot where voltage is supposed to drop but does not. That would be at across R55 (4K 10W). Measure the voltage drop across R55 by placing one probe on each side of the resistor (no ground reference).

I don't remember if you supplied a photo of the board. If you are having trouble finding stuff, a hi res picture will help. If that board is like my Gibson, the solder cups are numbered, making references fairly easy (famous last words.)