Gibson GA-95 RVT

[Phil_S]

Down at 60Hz the response of the OT may be falling off a bit, which would make the impedance ratio appear high.

Martin: I wasn't aware that it's a factor in a voltage test for turns ratio. Did I miss class on that day?

[Firestorm]

I measure 257mv at the two OT secondaries after applying 6.9v at pin 3 of V7 and V8.

That's close to what you had before. I think what we were looking to do is run the tranny test in the other direction: put the 6.9 in on the output side and measure the 200 volts or so that should come out the primary. The assumption is that the meter is likely to be more accurate measuring the higher voltages.

EDIT: Martin is right; I think we usually test these at 1kHz, but even 60Hz should show something really out of whack.

A couple other things to try after that. We don't know what shape V5 is in. What kind of output would you get with a good tube (12AT7 or 12AX7 since you have no more 12AU7s) in V4 AND a good tube (12AT7) in V5. We're running out of other things to test in the basic circuitry.

Also, I think you should try to isolate the tremolo thingy from the rest of the reverb channel. The light-dependent resistor inside it isn't really a resistor, but a silicon cell that happens to have high resistance in the dark and low resistance in the light. Sometimes these things fail to a ridiculously small resistance which will load down your signal (on the reverb channel). If you can locate C35 (a .022 between the thingy and R39, a 470K) and remove it, that will take the LDR out of the circuit.

[Travst]

I'm getting 93v when I attach 6.9v to the output wires that I disconnected from the speaker jack. The speakers are a 4.9 ohm load. Note that I'm using a heavy 20' cable for lack of something better, so I assume that'd increase the resistance.

I held up on the 'thingy' test due to this reading. I know where the opto-whatchamacallit is on the board.

[martin manning]

The voltage ratio is the OT's response at the test frequency. If we found that the calculated primary impedance came out to be 2000 ohms at 1kHz, then the 2880 result at 60Hz would say that the response has dropped about 1.6dB relative to the response at 1kHz.

Travst, do you still have 6.9V from your test transformer when it is hooked to the amp's OT secondary? You need to hook it up and then measure the voltage on both sides of the OT to get an accurate result.

[Travst]

grrr... I used the wrong lead on the first test. Ok... now I tested the voltage at 6.9v and put the transformer leads to the OT secondaries. I measured 187v at pin 3 on V7 and V8. I even tested again to be sure. I also pulled C35 and that had no effect on the volume.

[Firestorm]

Travst, do you still have 6.9V from your test transformer when it is hooked to the amp's OT secondary? You need to hook it up and then measure the voltage on both sides of the OT to get an accurate result.

I usually do this with a signal generator and two meters so I can be certain of my input voltage. Transformers usually do pull down the applied voltage (some more than others). Some even "see" the second meter and cause another (minor) change in voltage.

The best you can do is as Martin says: connect your "6.9"VAC source and measure it first to see how close it remains to 6.9. Then meter the other side.

I suppose it wouldn't hurt to lift the center tap connection and test each half of the primary as well, to see if some of the windings have shorted together.

[Firestorm]

You're testing faster than I can type. Well, that is very close to what you had before (from the other direction).

What tubes are installed in V4 and V5 when you are testing output "loudness?" New? Good? Unknown?

[Travst]

New tubes that I tested in other amps.

With new cap in C35:

V5-1 274v
V5-3 6.4v
V5-6 147v

V4-1 222v
V4-6 221v

[Firestorm]

12AX7? Or...?

Sorry, I didn't mean put a new cap in C35, I meant put no cap in C35. That disconnects the trem gizmo from the reverb channel. But since you have low output on both channels ...

I don't suppose you have spare 6EU7s around? Signal starts out at V1 and I'm not sure we know how that tube is functioning.

[Travst]

12AX7? Or...?

Sorry, I didn't mean put a new cap in C35, I meant put no cap in C35. That disconnects the trem gizmo from the reverb channel. But since you have low output on both channels ...

I don't suppose you have spare 6EU7s around? Signal starts out at V1 and I'm not sure we know how that tube is functioning.

12AU7 in V5. I did take the cap completely out, but it made no difference. And yes, I do have spare 6EU7s. I'll try them tonight. As I recall, I tried different tubes in every socket, but... anyone reading this thread knows my memory can be faulty. I now realize the importance of keeping a log of all the changes rather than my just jotting down random notes to remember things. New habit!

[Firestorm]

Maybe update the voltage chart, too, noting what type of tube is in V4 and V5.

We know (or think we know):

1) The OT seems okay. It has a primary impedance of 2938; a reasonable number. You should measure the resistance of the primary from center tap to each end.
2) Output tubes: (I think you said they were new). Voltages have always been good.
3) Interstage transformer works and has a ratio of 1: 2.4. This the same as Phil's and close to the 1:3 ratio of aftermarket interstage transformers.
4) V5a (pins 1,2,3) seems okay. Cathode voltage is low compared to the schematic, but this could be the tube itself. Supposed to be a 12AU7. You could substutute a 12AT7 for listening tests. A 12AT7 does not draw nearly as much current as a 12AU7 so the cathode voltage with a 12AT7 will be low.
5) V4. Supposed to be a 12AU7, but cathode voltages have not been close with existing samples. A good 12AT7 here produced similar voltages on each triode. Again a 12AU7 draws a lot more current than a 12AT7 so the measurements won't match the schematic, but the stage seems to work.
6) V1. 6EU7. Voltages have been good, but we don't know overall health of the tube. You have other tube samples for testing.

Output power is still very low, yet each stage seems functional, so we are losing signal somewhere. Assuming tube swaps don't improve anything, and since it affects both channels, V5a is the first place to look. I'd suggest cleaning up the wiring on that socket to make sure there are no "whisker" connections and to see that nothing is connected there that isn't mean to be. V5b is kind of a mess because the trem oscillator is built right on it, so it might be good to remove things on that triode. Make a little drawing of how it's wired.

The step after that would be to take schematic in hand and trace the signal path on each channel looking for improper connections. It would have to be something affecting both channels.

If need be, we could make a little homebrew signal tracer to find out where gain is being lost.

[Phil_S]

The step after that would be to take schematic in hand and trace the signal path on each channel looking for improper connections. It would have to be something affecting both channels.

The good news is that the normal and reverb channels mix at C25. This means the parts of the amp in common with both channels are that half of V5 and the power section.

I have pretty much ignored the output section because all voltages seem to be where they should be. Assuming that's the case (and maybe it isn't), that leaves only those few parts affecting the half of V5 that serves as the driver tube for the i/s tranny. This isn't much: C25, R41, R42, C26, the tube socket, and the tube. R41, R42, and C26 should all ground at cup #33, if the same as my amp.

[Travst]

Firestorm, what do you want to see in the drawing of the tube?

I have all the stuff to build a signal tracer Tom (Structo) and I discussed this a while back and I picked up the necessary parts.

[Firestorm]

Firestorm, what do you want to see in the drawing of the tube?

I'm just suggesting that you disconnect everything from V5b (the trem stuff) to make sure the connections to V5a are nice and clean with no accidental connections and so you can test without its influence (if there is any). The drawing is just so you can put 'em back.

Phil's right that according to the schematic the first common point for the channels (aside from the PS) is in front of C25, so V5a is first up. But someone has been inside this amp before and you never know what someone may have tried to do to "fix" it.

I just had a 1960 Champ in that had blown its OT. Someone replaced it with whatever they had lying around, which turned out to be an 18V power transformer! Sounded like crap.

[Travst]

Got it. I have some little flags I mark and put on wires as I disconnect them. I'll clean up V5 and post a new voltage charge with the tubes indicated.