Acoustic 270 -- Weird voltages in power amp

[disorder]

I have an Acoustic 270 on the bench that powers up but won't put out sound. Its hanging up on the 80V rail... where I'm supposed to be getting the half supply ~40V I'm reading the full 80V. Both the 40408 and the 2N2484 were pulled and tested fine. Surrounding caps and resistors measured ok as well.

I started to test around the 2N4248 pair (can anyone explain what these two transistors do!?) and got some strange readings. Picture of readings marked on the schematic...

http://img21.imageshack.us/img21/9321/270badvolt.jpg

Notice how the emitters are tied together and should have 11.6V resting on them but have 1.3V and both bases should be at 11V but in my case one is at 10.6V and the other 0.69V. It seems like this should be impossible??? Both of these transistors tested fine on a multimeter and my Peak transistor meter. Any one have any sort of technical insight or just an idea? Thanks in advance!

Heres the schem if anyone needs it...
http://music-electronics-forum.com/t25117/

[R.G.]

I can't believe they really did that.

Q401 is biased by divider chain R402/403/404. I bet there's an alignment section that tells you to diddle R404 to make base of Q401 be 11V. That accounts for the voltage on base of Q401. Q402 is raised to its nominal voltage by feedback from the output through R414.

If the base of Q402 was to be pulled ... way... down somehow, it would pull both emitters down, and the base-emitter junction of Q401 would remain high as long as it didn't reverse breakover. 10.6 is a little high for that, but it could be.

If it were me, I'd suspect that R412/413/415 or C406 was shorted or otherwise pulled to ground, forcing base of Q402 low.

Using a 40V transistor there and not otherwise providing for making is safe from an 87V power supply is putting an awful lot of trust in the design.

Just after I hit the 'send' button I thought that you may as well check for shorted outputs while you're at it. I'm not sure what would happen if the output was shorted to ground through the 270K R414, exactly. But it could be similar.

[Jazz P Bass]

I have an Acoustic 270 on the bench that powers up but won't put out sound. Its hanging up on the 80V rail... where I'm supposed to be getting the half supply ~40V I'm reading the full 80V. Both the 40408 and the 2N2484 were pulled and tested fine. Surrounding caps and resistors measured ok as well.
I started to test around the 2N4248 pair (can anyone explain what these two transistors do!?) and got some strange readings. Picture of readings marked on the schematic..
http://img21.imageshack.us/img21/9321/270badvolt.jpg

Notice how the emitters are tied together and should have 11.6V resting on them but have 1.3V and both bases should be at 11V but in my case one is at 10.6V and the other 0.69V. It seems like this should be impossible??? Both of these transistors tested fine on a multimeter and my Peak transistor meter. Any one have any sort of technical insight or just an idea? Thanks in advance!

Heres the schem if anyone needs it...
http://music-electronics-forum.com/t25117/

First off:
The power supply is single sided.
There is only the 80Vdc rail.
There is a capacitor before the output jack to the speaker.

The two transistors are called a differential pair.
Link:Differential amplifier - Wikipedia, the free encyclopedia

I would agree that you should test for shorted transistors.
Test the other voltage test points.
This may help narrow down what is shorted.

[g1]

"Its hanging up on the 80V rail... where I'm supposed to be getting the half supply ~40V I'm reading the full 80V". Haven't looked at the schematic yet but this same problem appeared in a recent post because there was no bleed resistor for the output cap. when speaker was disconnected. If it doesn't have one, tie a 1k across the output and see what it does to your center voltage.

Edit: Ooops, the problem in the other amp was that the center voltage was appearing on the speaker side of the output cap. when the speaker was disconnected. My mistake, sorry for the confusion.

[disorder]

Yeah I guess you can say this is a two-fold problem. One issue exists with those voltage readings at the differential pair and also at the readings at the 40408 and the 2N2484 where they should be ~40V but I'm still getting the 80V. I'll be back at it tonight so hopefully something turns up. Thanks for the help so far!

[gbono]

I'm assuming that the preamp board is functioning normally and you get the appropriate signal into the power amp board?

Your differential amplifier on the output board is not happy - I'm suspecting that there is an open somewhere in the output transistors/drivers.

[Enzo]

What he meant was, I bet, was that the output point - where the speaker cap connects - was stuck on the 80v rail voltage. That point should sit at half B+, so 40v, but it is at 80.

I suspect the whole output is being told to swing up to +80. We'd be blowing fuses is there was contention at the output. This is a common, ordinary solid state amp section. Doesn;t really matter if it is split supplies connected to a speaker, or single supply with a cap for DC blocking. Q405,410 are the drivers for the outputs. Q404 is the bias transistor that keeps the driver bases a steady 2v apart. Those bases can be pushed up and down in voltage, but Q404 will always keep them 2v apart. Q403 is the voltage amp, and its job is the yank the output stage up and down to make music. When Q403 conducts hard, it pulls Q404 down towards ground, and all the outputs follow along. If Q403 turns off, then Q404 is free to be pulled positive by R416,417.

SO first, check Q403. Look at its collector. I bet you have closer to 80v than the 41v it wants. What is at the base of Q403?

Like all solid state amps, this is very heavily fed back. The typical differntial pair tries to do its job. Q401 is the input, and it sends teh signal along to Q403. Q402 monitors the output, and tries to correct any differences betwen the output and the input. Usually works well. But if there is a problem, say DC on the output, then that diffy pair will try as hard as it can to move the output the other way to compensate. But what if the feedback path is broken? Make sure R414, which samples the output back to the base of Q402, is not open. And as I like to point out a lot, the connectiions and traces between parts are just as important as the parts. An open trace between resistors is just the same as an open resistor to the circuit.

Look at R412,413,414,415. They form a voltage divider. R414 normally connects to the 40v at the output. R413,415 are in parallel making about 2.3k, call it 2k, so the total then is about 372k to ground. (270k+2k+100k) SO solving the voltage divider: 40v x 102k/372k = 11v. Just what the schematic tells us to expect.

But you have 0.6v there at Q402. SO with 80v on the output, instead of seeing like 22v on the base of Q402, we have that low voltage. SO if R415 is not open, then I wonder if maybe C406 is shorted across R412. It is one of those little 2uf caps, I am always suspicious of those. ANd removing thus the 100k from the voltage divider and starting with 80v on the output (And just for science, doe to the low number I will make the 2k back into the 2.35k it should be), I get: 80v x 2.35k/272.35k = 0.69v. How does that compare to what you found?

Also possible Q402 is shorted base-collector, leaving R409 1k as the path to ground. The solving: 80v x 1k/271k = 0.29v. That doesn't match the facts, but then there is a lot of fudging in my thought experiment. And the transistor copuld be real leaky instead of sshorted. In any case, those areas are worth checking.

But my advice is to focus on the big picture. I suspect those diffy transistors are working and doing their job, and the voltages around them will fall into place one the real fault is corrected.

[disorder]

Enzo, thanks for the reply. First thing I've come across is that the feedback resistor to the diff pair (R414) is open... will replace and report back

[Enzo]

Well, that is enough to do it.

[disorder]

That fixed it, amp sounds good again. I'm getting the appropriate 45V at the output now, however the diff pair isn't quite where it should be. I am getting 13.9V at the joined emitters... with this voltage here I have R404 adjusted such that 13.3V appears on the base of the first PNP in the pair, the second PNP rests at 13.3V as well.

What is causing the slightly higher voltage here? R411 (100k) and R410 (56K) all measure correct as do R408, R407, and R409. C405 was replaced with a brand new 10u 100V cap.

[Enzo]

Don't worry about it. SS amps are one big loop. Some transistor somewher conducts a trifle more current, and another one elsewhere reduces current to compendate. SO you have about 0.6v B-E on the transistors. SOUnds fine to me. You have the output centered where it ought to be, and the amp sounds OK. Stop fixing it.

[disorder]

sounds good to me. thanks again for the help!