800rb help

[lowell]

I remember a similar issue with a GK. Aren't there 1 or 2 transistors related to the bias circuit? I think you said you checked and replaced them but not sure.

[grooveunit]

I remember a similar issue with a GK. Aren't there 1 or 2 transistors related to the bias circuit? I think you said you checked and replaced them but not sure.

All the transistors on the 300 watt side are new. I think enzo said that Q13 and 14 are bias, and 15 and 16 are limiters. 17 and 21 are the drivers.

[Enzo]

The resistors are OK, the pot is OK, but is the CIRCUIT OK? That is why I suggested measuring resistance between the two circled bases. It is not enough that the resistors all be OK, they also must have intact connectione between them. A cracked circuit trace can leave them open as far as the circuit is concerned. Same thing with measuring between those two bases. It may seem the same to simply measure the parts themselves, but if we measure from the actual transistor base to base, we are also checking the integrity of the copper tracework. It may all be fine, but we need to know for sure.

But read on...

Look at the circuit as a whole. From -85v rail through R41, R41,R39,R38,R38,Q12,R37, to +85v. You have 1.4v across the 100 ohm R37. That is 14ma, 0.014A. Ignore for now any side paths. Current through a series circuit is constant. SO predict the voltage drop across 2.7k R41. 14ma through 2700 ohms = 0.014 x 2700 = 37.8v. You are getting about 39v across each of them. Not far off.

Now that much current through R38,38,39 ought to drop about 74 volts!!! So normally most of it goes through Q14. You get a whopping 0.6v across it. Q14 must be on hard or shorted to shunt that current past those resistors. Now look at your voltage readings. You have -0.2v at the Q14 base and -0.2v at Q14 collector. The base is shorted to the collector, or at least there is zero voltage drop. Short base to collector turns a transistor into a diode. And a diode drop is darn close to that 0.6v you have across Q14.

Is Q14 shorted? MAybe, check. Maybe C12 is shorted, lift it out of circuit to see if it brings us back. Or Q13 is shorted E-C.

Other possibilities might be a solder bridge we have not noticed on the board. And then the possibility the parts are wrong. Q13 is an MPSA06 and Q14 is an MPSA56. Are they right? ANd my pictorial shows Q13 as closer to the power transistors than Q14, so are they in the right holes?

The pictorial also calls the trimmer R38, and resistor R38 as "R38A." That might explain the schematic with two R38s.

You are worried about getting some voltage to the base of Q13. ANy voltage there would be the result of voltage division along the R38,38,39 chain, which in turn is part of the larger chain I described above. NO voltages drop across those three resistances because for whatever reason, Q14 won't let it.

[grooveunit]

Ok, with everyone's help, especially Enzo, I was able to stop going in circles and figure out what was wrong. I knew it was just some stupid thing, probably staring me in the face. I'm almost ashamed to admit what it was, but in the event that someone reads this thread and avoids my mistake, here goes.

Enzo's suggestion to lift c12 out of the circuit was the winner. I had been looking right past c12 on the schematic, and never really bothered to pinpoint it on the board. Looking for it today, and what I should have seen was a .01uF cap. There's no cap there. Finally occurs to me to locate c5, the equivalent cap on the 100w side, and what do I see? That funny resistor that I thought was R39, a 1k, that didn't ohm out right, that I replaced with another 1k.

Google auto-complete tells me that I'm not the only person to search "capacitors that look like resistors", in fact, I only got to the "t" in "that" before I saw the term in the drop-down list. I guess I'm not the only guy that's done this, but boy do I feel like a knob. It just never occurred to me that that little pink resistor wasn't a resistor.

Pulled the 1k out, and the amp is fine. I need to find an equivalent cap to install- the schem shows .01uF, but the color code on the one I pulled indicates .02uF. Are these axial-lead ceramics still available, or has technology moved on since 1987?

Thanks again for the help, everyone, especially Enzo. You most of all got me thinking the right way, when I was ready to give up. I would've felt like a jackass sending this thing to GK with a frigging resistor installed where a cap should go.

BTW, I recapped the power supply board, and upped the values on the filter caps a bit, and replaced the electrolytics on the power amp board as well. I'll recap the preamp tomorrow, and this thing will hopefully be good for another 20 years. Dave says thanks.

[Enzo]

And you won't be the last to make that mistake.

Always a red flag when that "resistor" on the board has C123 printed next to it.

They certainly are still available, but the type hardly matters, any 0.01uf cap should work. Even the voltage doesn't matter much. Normally Q14 has a couple volts across it. If it gets up to 3v, your output would start cooking. SO unless the circuit has burnt up, that cap will never see more than 3v. So 50v ceramic or mylar should work if you lack the original type.

[Steve Conner]

Ah, the old "capacitors that look like resistors" trick.

We recently had a contract manufacturer make us some complicated boards stuffed with tiny surface mount parts. They couldn't get any of them to work, so two got shipped to me for troubleshooting. Turns out that they got their components mixed up and installed a capacitor where there should have been an inductor. The components are 0402 size and none of them has any markings, so it took quite a bit of detective work to figure this out.

Oddly, in the demo board that we based our circuit on, this inductor was installed in a spot labelled "C21" on the schematic. It showed an 8.2 nanohenry capacitor. But we redrew the schematics and ours were correct.

Anyway, I hope that the guy who loaded the pick-and-place machine feels like a jackass for wasting a whole day of my time.

[grooveunit]

Always a red flag when that "resistor" on the board has C123 printed next to it.

Oh, but to have labels on the board. The only markings on the board in this amp are numbers for the transistors, and I have no pictorial or layout diagram. I've been running circuit traces and keeping track in my head, which, it turns out, is not ideal.

[Enzo]

AH. My schematic set has layouts as well as schematics. COntact G-K and ask for the full set.

[grooveunit]

GK has been as helpful as they could be, which is to say not super-helpful, mostly given the age of the amp. The lady (Lani) that helped me told me that their older computer files are incompatible with their newer computers, blah blah blah. They could not spec the cap to replace c12 from the old parts list, but told me that the newer amp uses a 50v 10% ceramic disc at c12. I grabbed c5 from the HF amp and installed it at c12, Dave never uses the HF anyway. I'll throw a Rat Shack disc in the HF, and button it back up. Enzo, does your layout show the connections for the yellow/orange/green/brown from the preamp board? I took pics when I started, but they suck. Orange and yellow go to the direct out, but I don't know which is which. I think Green and brown went to the send/return jacks.

[Enzo]

I don't know.

Look through this stuff.