800rb help

[grooveunit]

Greetings all, first post here, but I've been reading threads here for a while. I've got an old GK 800rb that I'm trying to fix, and I need some guidance.

History- Amp is 800rb rev. A, circa 1987, in regular use until 2010, probably 6500 gigs on it. No sound one night, owner replaced with a PF500, but wants his old GK back in service.

Schematic- http://music-electronics-forum.com/a...00rb_82_91.pdf

Findings- Amp blowing fuses as soon as powered on, no obvious crispy items inside. Supply board and preamp seem OK, HF (100w) amp works when LF (300w) supply is disconnected. Found 2 shorted power transistors, Q15 and Q23 (i think, might have been Q21) and shorted driver, Q14. Installed known good replacement in Q14, replaced power transistors. D7 through D12 check OK, and I checked R44 through R56 as suggested in an earlier thread; http://music-electronics-forum.com/t16074/

The 5W emitter resistors are all drifted high of .33 ohm, most read .6-.7 ohm, one is as high as .9. Amp is drawing serious current with no input signal and no load, 60W lightbulb limiter is bright. There's a few mv DC on the speaker jack. Bias trimpot R38 has no visible effect.

As I mentioned, the HF portion of the power amp works fine, as does the preamp section. I'm looking at U1, a LF353M dual op amp, which I think is part of the negative feedback loop? The chip seems to work, as both sections of the power amp share this chip. Could one side of the op amp have shorted, leaving the other intact?

If I can get this thing sorted with your help, I plan to recap it and hopefully Dave gets another few thousand gigs out of it. Dave's a good guy, BTW.

[Enzo]

Set your meter to ohms, and touch the probes together. WHat resistance do you see? Those 0.33 ohm resistors are wire wound, so I;d suspect the higher readings were from the added resistance of your meter probes. Yes/no?

COUld one side of an op amp fail but niot the other? Of course.


If it blows fuses, but not when the 300w amp is disconnected, thent he power supply is probably OK. In the 300w amp, check all the outputs Q17-24.

Q14 is the bias transistor, not a driver. Check it and Q13. And the resistors with them. The drawing shows two R38s, one is the trimmer. But whatever they are, check R38, R39, and the other R38. If Q14 doesn't conducts - whether its own fault or the fault of the other parts just mentioned, then the outputs will all turn on hard, and blow fuses.

If an amp puts DC on the output, if there is no load on it, it won;t draw current. If the amp draws hard even without a load, then the two sides are both conducting at once. or something is shorted. Have you checked to see if either power rail is shorted to ground in that amp? D11,D12 would RARELY short, but check them. Remember it is the circuit that matte3rs, so you can pull the transistors and test them, but that won;t reveal a short across circuit board traces.

[grooveunit]

Ok, I should have been more clear. I've already tested all the transistors out of circuit, Q14 was shorted, as were two power transistors. All power transistors Q17 through Q24 have been replaced. when I removed the shorted Q14, the amp stopped drawing current. When I replaced Q14 with a good transistor, the amp started drawing excessive current again.

If a power rail was shorted, the amp would still draw with Q14 removed, yes?

I had noticed that the diagram showed 2 R38s, but had sort of forgotten. I was speaking of the trimmer pot to the base of Q13.

D11 and D12 have both been checked with the diode check function on my DMM; is there a more rigorous check that I should perform?

[Enzo]

Nah, if D11,12 are not shorted, they are probably OK.

Yes, a power rail short to ground wouldn't care about Q14 at all.

But I would not expect a shorted Q14 to cause a problem, other than some crossover distortion. Q17,Q21 are your drivers, and shorting their bases together should turn them all the way cold. Q14 is connected between their bases.

Q15,Q16 are limiters, disable them by lifting D8,D9 from the board at one end. Probably not involved, but never hurts to get them out of the way.

Q14 is an MPSA56, yes? And Q13 an MPSA06. When you replaced Q14 did you replace Q13 too? An open Q13 or nonfunctioning one for whatever reason would leave Q14 not conducting, and that would max output currents.

Check out D10 and the pair of 22 ohms.

[grooveunit]

I checked Q13, Q15, and Q16 out of circuit, and they check OK. D10 is OK, and I checked all the resistors in that neighborhood, the 22s, and the 470s. This is why I'm a little baffled- I'm no amp guru, but I've checked pretty much everything to the right of Q14 on the diagram. Maybe I'm overlooking something- I'll recheck all the values and note them on a fresh copy of the diagram. This is partly why I was looking at U1- I suppose that for about a dollar I can pop the existing one out, and install a socket and a new chip.

I'm also ordering a handful of MPSA56s and MPSA06s, I'll just replace them all to be sure.

Thanks for the help thus far, I think I'm on the right track, just a little hung up. All will become clear with time, I expect.

If I get the 800RB working, I get to borrow the PF500.

[grooveunit]

Ok, here's where I'm at- the schematic had my +85 and -85 wire colors backward, so I had the rails hooked up wrong. I never ran the amp without the limiter in this state, and once I figured out my mistake, the amp stopped drawing big current. The amp now works, sort of, but it's weak and fizzy, and acts like its got a noise gate set too high. I've pulled all the output trannys and rechecked them, they're checking OK. I replaced Q13-Q16 again, just to be safe, and it's still not right. All of the resistors I've checked are in spec, so what am I missing? Did those little guys at c12 and c13 get fried with the reverse polarity? Oh, I replaced U1 as well.

Forgot to mention- I'm seeing some weird voltage fluctuation on one rail, +85 (I think-not in front of the amp right now), like a 25v swing, rhythmic every second or so. Tried replacing the bridge rectifier, no change.

[grooveunit]

OK, all the small caps are checking OK, although the 27pf at c10 is reading about 45-50 on my meter. C3 from the (working fine) 100W side reads the same, so I'm not sweating it.
The fluctuation in the power supply seems to be gone, nice steady +82 and -82. 60v rails are reading 64-65; is it possible that a few volts are bleeding over from the 85v side???? I've got about 10mv DC on the speaker jack with no load, too much?

300w side still working poorly, very fizzy and distorted, and I have to whack the strings pretty good in order to get sound. It's acting like an amp with really really dirty pots- I'm starting to think I've got a resistor way out of whack.

Any thoughts?

[lowell]

I don't have a schem as I'm on the road. To me at this point it sounds like a transistor is biased a bit cold and when you whack the strings hard enough it turns said transistor ON. Wish I had a schem so we could check some voltages. Maybe check B/E voltages on output and drivers?

[olddawg]

Regarding your driver and bias transistors. Did you use an exact replacement or a sub? You might want to double check the basing of the transistors. Also, if you reversed the rails, I would worry about any IC that was powered by them .

[grooveunit]

I was also thinking the bias was cold, but the trimpot at R38 isn't helping. All transistor replacements are exact matches, ie MPSA56 and MPSA06. There's one IC at U1, and I've got a fresh one in there with no improvement. Schem is linked in the initial post up top, BTW. Thanks for the help- I'm pretty stuck here.

Want to add that the help I've received here is very, well, helpful. I'm not really an amp guy, (I'd like to be) but you guys steering me along has really been useful. It's not that I don't sit and stare at the amp with smoke coming out of my ears, I do, but I know the thing is fixable, and I just need to not spend ages running down blind alleys.

[Enzo]

Poor bias results in either exccessive current or crossover distortion. If it sounds like a noise gaste with an extremely high threshold setting, then I suspect something earlier in the circuit. My guts are suggesting somke place with an unwanted DC offset that the signal has to overcome to get by, so just the peaks make it.

If your output has settled to roughly zero volts DC (and I don;t care about a few millivolts), then pretty much everything from Q14 over towards the right is balancing out and probably works. Maybe not, but probably. You got the 1.1 and -1.1 on the ends of Q14 as noted at the bases of the drivers?

New IC? OK, but did you check the power rails to it? Got both +15 and -15? And clean?

Not what I might have expected, but the schematic calls for -8.8vDC on the output pin of the IC, is it there? Several revisions all show it, so it doesn;t appear to be a typo.

How are you injecting the signal into this? Right to the power amp itself? Through the preamp? If preamp, are you biamping or full range to both amps?


IIf the output stage is OK (Q14 and rightward), it is up to Q12 to drive it back and forth. At idle there is 1vDC across its 100 ohm emitter resistor, is that there? That is a measure of the current through the part. And Q11 drives its base from the IC output. Looks like +83.4v on the base of Q12/collector Q11. Got that? Base Q11 grounded, is it? emitter Q11 -0.6v, close? You got a bunch of hte small transistors, did Q11 get replaced? Donl;t know how happy it would be to have seen that reversed power supply.

[grooveunit]

Enzo, I've been putting guitar signal in through the FX Return jack, jumpered over to the 300w input pin on the connector. 15v rails look good, as far as I can tell with my DMM. I'll check Q11 again when next I see the amp. I'm gonna pull some of the emitter resistors on the output trannys and recheck them; remember how they were reading high? You suggested it was resistance in my probes, but the DMM shows zero ohms when I short my probes, and seems pretty accurate on a 5W wirewound that I use to drain caps. If the ERs were high, wouldn't that run the OTs pretty cold?

[grooveunit]

OK, I've got some voltages to share. First, I had 82V on the rails yesterday, today it's 79. Whatever.

Got +14.2 and -14.8 on the 15V rails, which doesn't seem right.

Pin 7 on U1 has -7.8V.

Q12 has +78.2E, +77.6B, and +0.4C.

Q11 shows +77.6C, -.5E, and the base is ground.

Base of Q21 is -.2V, base of Q17 is +.4V. These are way too low, even given the 79 on the rails. R40, R41, R39 and R38 (3.3k) check OK.

Q14 shows -.2v on the base.

There's like no voltage on the trimpot at R38, all the resistors around R38 check OK.

So, there's no bias volts for Q13, which explains why the amp sounds cold and gate-like and clipped.

Please help me figger out why this is so. Thanks again.

I marked up my readings on the schem below.

[Enzo]

OK, this is a guitar amp, not rocket science, so power supply voltages only need to be close. 14.8 and 14.2 is fine. 82 to 79, means your mains voltage shifted about 5 volts. The 82v rails are not regulated, so they will always track the voltage at the wall outlet you plug into.

Your circled voltages are low because the bias is turned way cold. Does turning the bias adjuster change the circled voltages or not?

If R38 and R39 measure OK, and so does the trim pot? If all that is OK, then power off, a resistance reading between the two circled voltage points ought to then read about 5.3k, right? Does it? Then measure resistance from the base of Q13 to the base of Q21, whiloe turning the trim pot end to end. Does that resistance move between 1k and 2k as you turn?

[grooveunit]

I think maybe I didn't explain well, again. The trimpot R38 has no effect on voltages, I tries that ages ago. I have checked the trimpot both in and out of circuit, it's OK. The problem, as I see it is that the trimpot doesn't have any voltage to send as bias. There's like a tenth of a volt on one side, I'm trying to figure out why. I've got 39 volts after R41, R40 drops that to 0.2, and then R39 drops it further to 0.04? I'm missing something.