Gibson GA-95 RVT

[Phil_S]

There is an insulating layer between the board (below the cups) and the chassis. It makes it difficult to screw up but not impossible. I don't think this is an issue.

[Structo]

Hi Phil, you guys are devoted I'll give you that.
Weird that the output is low. It's will probably surprise you at how simple the fix was when you find it.

One thing I can mention is, it is not unheard of to have a factory wiring mistake or omission. Or even some weird component failure that is not obvious.
The amp is just passed from person to person, never getting fixed.
This time hopefully you can find the problem.

Yeah, with eyelet boards you have to watch how much solder you feed into it when soldering.
You would be surprised after tweaking an amp for a while how ugly it gets underneath.
You can have balls of solder 1/4" thick hanging off the bottom.

I have a little dentist type mirror that is good for inspecting underneath without removing the board.

I remember those solder cups, I'm glad they ditched those and went with eyelets.

[Travst]

The amp is just passed from person to person, never getting fixed.

Tom, that's exactly what happened here. I had a cheap guitar for sale and this guy wanted to trade the amp for it. He owns a boutique pedal company and, I would assume from his stories and having seen his workshop, knows something about electronics. His take was that it was a simple fix. Here's what I found right off the bat:

[img:500:375]http://farm5.static.flickr.com/4093/481 ... 92d4a8.jpg[/img]

Note the tube that was broken at the base... four mis-matched power tubes. Either he hadn't even looked at the amp, or he misrepresented it. He stated that he'd bought it that way, so at least two owners passed it along without fixing it. I'm not worried as I've enjoyed the project and don't have much money in it.

Hm... I need to test those remaining tubes.

[Phil_S]

Hi Phil, you guys are devoted I'll give you that.
Weird that the output is low. It's will probably surprise you at how simple the fix was when you find it.

Yeah, even tho' I'm a rank amateur, I know my way around this one. As I said, I cut my teeth on one that's almost identical. You are undoubtedly right about the fix. If I had it here, it would be done already. The goal is not to throw him a fish, it is to teach some fishing along the way. Otherwise, he'd just take it to a tech...

The good news is that he's passing some decent signal from ch 2 to the finals, which suggests to me the PA section works OK. Voltages in the PA are reasonable, too. Most of all, the dreaded malfunction of the i/s transformer seems to be off the problem list.

From here, it's just a matter of sticking to it, working systematically until it gets right, one channel at a time. If we had a 'scope, we could probably find what's eating the signal a bit quicker, but we'll find it with a volt/ohm meter eventually. I'm also pretty sure we'll turn left when we should turn right, more than one time. That is the nature of the work at this stage.

So, have you got any suggestions? The main problem appears to be plate voltage 2x what it should be on V4, which is the final gain stage before the i/s driver tube. He says Rk checks out, and Ra is new. The B+ rail is a bit high, but still appropriate. Also, we have the correct voltage drop across the last dropping resistor in the B+ ladder. On the Epi schematic, that's R43. Since V1 voltages are reasonable in respect to the schematic, the problem is most likely between V1 and V4 (these are the two gain stage tubes, split between the two channels, V2 and V3 run the reverb and tremolo).

[Travst]

I'm just grateful for the help!

Keep in mind that there are two 6EU7s in the preamp.

[Travst]

I scrubbed down the board around all the solder joints in question with no effect. Just for grins, I reattached R14 to see if anything would change. The voltage dropped even further. I did my best to peer under the board. It appears it does have the cups. I guess it makes sense to start checking voltages around that area to see if I can try to narrow things down.

[Phil_S]

The reason for lifting R14 is to simplify the diagnostics. There are two channels in the preamp tht mix at C25. Lifting R14 or R30 will take one channel off line. This reduces the number of variables. I figure, channel 2 is working, so let's see if we can finish that off by getting V4 voltages around where they belong.

All voltages are running high, so I am guessing we'd like to see a drop of about 130v from C29A to the plate of V4 pin 1. You've got 207V at the plate and 263 at C29A. I think we are looking for 263-130 or around 133V at the plate. When plate voltage drops, cathode volage will rise, probably to around 4V. Those are the targets.

Rather than go all over the lot, let's try to focus on the area that is close to being right.

Back to C25, which presents as a short. I guess you should pull that from the board and test it with the ohm meter. You only need lift one side. It should read infinite. If it reads zero, it's a short and needs replacement. If it meters OK, get ready to reinstall, but hold up a minute.

With one leg out of the circuit, test each cup (#13, #14) for DC voltage. The means measure from the cup to ground, not cup to cup. You should see voltage at #13 but not at #14. Please verify that #14 is the one that connects to R41 and V5 pin 2 -- if that tunrs out to be #13, adjust accordingly. If you have voltage on #14, that's a problem to address. It also means the cap isn't the culprit and you can return it to the circuit, but not before you do one more thing, below.

While you are in there, we also know voltages aren't quite right with V5 pins 1 and 3. This is the driver for the interstage transformer. Plate voltage might be OK at 286, but cathode voltage at 1.9 is way low. We are looking for 8V at the cathode. My amp operates the same way. A quick thing to do is isolate that section of V5 by liting C25! You've already got it lifted --either end, doesn't matter. With C25 out of circuit, what is the voltage on V5, pins 1 and 3?

Since we've removed the whole preamp by lifting C25, it should move towards the schematic values. You already verified R41 and R42. If there is no meaningful movement of the cathode voltage, lift one end of the bypass cap C26 and check voltage readings again for V5, 1 & 3. IF lifting the cap fixes, it, the cap was leaky and you should replace it. You can leave it out of circuit until you find something appropriate to put in there. It calls for a 2uf cap. I would not beef that up for fear of stressing the interstage transformer. I promise that you'll be very sorry if it blows.

OK, let's see what that gets us. If no progress, we will start moving backwards towards V1.

[Travst]

Alrighty...

C25 is in good shape. Cup 13 reads 128v, while cup 14 reads 0v with C25 lifted. V5-p1 reads 281v and V5-p3 reads 1.9v.

I pulled C26, which is passing current through. I'll have to find a 2uf cap to replace it, as I don't have one in stock. Note that the Gibson schematic calls for a 5uf. When measuring p1 and p3 at V5 with C26 lifted, there was no change.

I checked with the Shack, and they have 2.2uf caps in stock. I'll hit them on the way home tomorrow.

[Phil_S]

I like 128V. That side is doing what it should. 281v is OK, too. I see the Apollo calls for 5u. You should keep the 5u. Remeber that is a bypass cap. We made this mistake before. If you are measuring voltage in circuit, then you are metering the resistor, not the cap. If the voltage changes when you pull one end, the cap is bad. If no change in cathode voltage without the cap, assume the cap is good.

Let me sleep on the rest.

[Phil_S]

C25 is in good shape. Cup 13 reads 128v, while cup 14 reads 0v with C25 lifted. V5-p1 reads 281v and V5-p3 reads 1.9v.

So far, this is good.

I pulled C26, which is passing current through. I'll have to find a 2uf cap to replace it, as I don't have one in stock. Note that the Gibson schematic calls for a 5uf. When measuring p1 and p3 at V5 with C26 lifted, there was no change.

You must mean voltage. Voltage and current are not interchangeable terms, though they do go together. I think your conclusion is not correct. As I said last night, no change in cathode voltage means the cap is not shorting. You will get better tone sticking with a 5uf cap. I see it on the Gibson schematic. I would just reinstall what's there.

Would you mind retracing some details for me? With c25 lifted, you report 128v at #13. Earlier, you reported 263v at #50 and 207v at V4 pin 1. There is a 68K (R30) between V4 pin 1 and #13. I am having some difficulty understanding this much of a voltage drop across R30. With c25 still lifted, please post voltages for the following:
#13
#14
#31
#32
#50
the cup where the two 100K plate load resistor for V1 meet, middle row, maybe third cup from the right edge (input side) of the board, don't know the cup #

V1 pin 6
V1 pin 4
V1 pin 7
V1 pin 9

V4 pin 1
V4 pin 3
V4 pin 6
V4 pin 8

V5 pin 1
V5 pin 3

Now, reconnect c25 and get a second set of voltage readings with it connected. Is R14 (330K) lifted or in circuit -- just tell me if it's in or out of circuit -- no need to reconnect, but keep it constant for the two sets of measurements.

I should tell you what I'm thinking, and also let you know that, at this point a fair amount of guessing is in order. Maybe one or a combination of the of the following is the cause of the problem:
-resistor of the wrong value (4.7K or 47K where you need 470K, that sort of thing)
-resistor mis-wired
-bleed resistor not properly grounded (R41?)
-unintended connection(s), look for whiskers under a bright light and magnifying glass.

If this doesn't disclose anything, I'll suggest we pull channel 2 off line and work with only channel 1. Channel 1 is much simpler. Just wanted to let you know I'm not out of ideas just yet. Fixing my Minuteman required a similar strategy and I did get it fixed.

[Structo]

John, I apologize if you have done this but, have you measured the grid voltage on the 6L6?

I see on the print it notes -43v.

Something I find strange is that it gets the plate and screen voltage from the same node.

Not sure I have ever seen that before.

I'd be tempted to take the plate voltage from the node before the choke.
Then take the screen voltage after the choke.

[Travst]

Thanks, Phil. I'm stopping to pick up some caps tonight and will get the testing done. Good point about current vs. voltage.

[Phil_S]

John, I apologize if you have done this but, have you measured the grid voltage on the 6L6?

I see on the print it notes -43v.

Something I find strange is that it gets the plate and screen voltage from the same node.

Tom, this is a Gibson "special". It is what it is. Both the Apollo (Gibson) and EA-300 (Epi) schematics for this model show the plate and screen from the same node. Given the comparatively low current draw on the screens, even with a 470z stopper, it seems to me that screen voltage will be above plate voltage. Plate voltage is specified at 440V, the node voltage is spec'd at 448V. There is no screen voltage spec, but with 448 from the node....

John gave us a voltage chart a while back. Plate were 461, 461, 461, 459. Screens (same order V6-V9) were 483, 466, 466, 463. Node voltage was 474, so I am assuming 483 is a mistake of some sort. In know, weird....anyway, this does not seem like the time to address it. Maybe later when the preamp is right. It's a simple thing to move the supply wire. I wouldn't be surprised if this is an error in the schematic and they built it as drawn.

[Firestorm]

Fender did this for a couple of years, too. Seems economically odd because it requires a monster choke to cope with the plate current. But clearly the designers of the day were following someone's lead.

[Travst]

With R14 out of the circuit, and after replacing C26 with a 4.7uf cap:

First value is with C25 lifted, 2nd is C25 in circuit

#13 - 203v : 201v
#14 - .5mv : oscillates, can't get a stable read
#31 - 260v : 257v
#32 - 205v : 203v
#50 - 263v : 261v

V1 pin 6 - 187v : 187v
V1 pin 4 - 1.1v : 1.1v
V1 pin 7 - 175v : 175v
V1 pin 9 - 1.2v : 1.2v

V4 pin 1 - 205v : 203v
V4 pin 3 - 2v : 2v
V4 pin 6 - 260v : 260v
V4 pin 8 - 4.5v : 4.5v

V5 pin 1 - 281v : 280v
V5 pin 3 - 1.9v : 1.9v