Mesa Boogie - DC Offset Protection Circuit

[ServiceBob]

bb750 dc oFFSET.JPG

I have an MB Big Block 750 that has stopped producing sound. I have traced the signal through the preamp and the power block and have determined it enters the relay immediately before the speaker jacks but never exits. I'm not so good with solid-state or logic circuits, and I'll admit I can't figure out exactly what's happening here.
Of course, the PC boards are installed in such a jam-packed fashion that I can't determine whether or not the relay coils are even getting any juice. So, I don't know if it's a bad relay (which I honestly doubt) or whether it's not closing. Could someone explain the actual aim of the circuit, how it works, and how I can troubleshoot it?
Finally, is it actually necessary, and would it be dangerous to simply short the relay switch contacts?
Thanks in advance!

Oh, and Merry Christmas!

[sluckey]

KEEP IN MIND THAT THERE MAY REALLY BE A DC VOLTAGE PRESENT ON THE SPEAKERS!!! In that case this circuit is probably operating as it should.

First look at the Relay Driver circuit. When power is initially turned on, C63 begins charging through R103. It would take over 30 seconds to full charge to 5 volts, but that will never happen. As soon as the cap voltage reaches about .6V, Q8 will turn on and energize the relay which connects the speakers to the amp. This is the normal operating condition.

Now look at the DC Offset detector. The two op amps are being used as comparators (no feedback loop). The top comparator has a -3V reference voltage applied to the inverting input. As long as the inverting input is greater than the non-inverting input, the output of the top comparator will be equal to the positive supply voltage. This keeps D16 reverse biased, effectively disconnecting from the relay driver circuit. Similarly, the bottom comparator has a +3V reference voltage applied to the non-inverting input. As long as the non-inverting input is larger than the inverting input, the output of the bottom comparator will be equal to the positive supply voltage. This keeps D17 reverse biased, effectively disconnecting from the relay driver circuit.

These two comparators work together to form a +/-3V voltage detector window. As long as the speaker sample voltage applied to the other inputs of the comparators stays in the window, everything stays happy and the relay driver will keep the relay energized. But if any dc voltage on the speaker should exceed +3V or -3V, the appropriate comparator will change states and it's output will become equal to the negative supply voltage, causing D16 or D17 to become forward biased. This will apply the negative voltage to the base of Q8, turning it off, and de-energizing the relay.

Hopefully this will help you troubleshoot this circuit. Divide and conquer by disconnecting D16 and D17. Then see if the relay driver works. I suggest unplugging the speakers in the event there is a real dc voltage on the speakers that is causing this circuit to work as designed. Of course if you discover there really is a dc voltage on the speakers you will need to troubleshoot the power amp.

[ServiceBob]

Thanks, Sluckey!

That was exactly what I was looking for. You are not only gifted in your knowledge, but also in your ability to share it. And we are all lucky for your willingness to share it.

I'll post the results of my troubleshooting!

Merry Christmas!

[ServiceBob]

UPDATE.

I verified that there is no significant DC voltage on the speaker output. ( <0.2V)
Then I lifted a lead on D16 and D17. No Change.
I take that to mean I should focus my attention on the Delay/Driver portion of the circuit. Right?

In checking the voltage on the positive side of C63 I watched it rise through +0.6V and make it all the way to 5V. I also measured the potential across the emitter and collector of Q8, which is also ground, I expected to see it drop from 5V to 0V when (and if) the transistor turned on. There was never a change.

At this point, I think I have a bad transistor (Q8). To verify that is the case couldn't I short the collector to ground to switch the relay on? Isn't the transistor, when closed, providing a path to ground for the transistor side of the relay?

Thanks again.

[martin manning]

Correct, and that would mean that the transistor has failed open. Try shorting its collector to ground. That should energize the relay.

PS If this is the problem, short the collector lead to ground using your DMM in current measuring mode to see what the current through the relay is. I'm thinking you can substitute a common-as-dirt 2N3904 NPN for that obsolete Darlington transistor.

[ServiceBob]

Thank you!

I've ordered a new Q8. I'll post again when it's fixed.