Advice on Hartke 3500 bass amp repair--power amp module

[drm1]

Does someone have a minute to advise about repairing a Hartke 3500 bass amp?

1. Amp purchased used and soon started putting out distorted sound, then eventually blew its fuse. it now blows immediately on power-up.

2. I removed the power amp module, and the fuse does not blow. Rail voltages (+/- 92V) are present and clean. And there is a clean signal at the effect send, which I guess means the preamp stage is OK.

3. [Schematic of this module attached as pdf.] I don't find any shorts on the eight output transistors tested in-circuit with a voltmeter on the diode test setting. (All test essentially the same.) The 0.47ohm power resistors on each also test OK in-circuit (not shorted).

Am I on the right track? Can you suggest where I should look next, or better tests for the transistors?

Thanks for any help you can offer.

Hartke3500PowerAmpModuleSchematic.pdf

[J M Fahey]

I don't find any shorts on the eight output transistors tested in-circuit with a voltmeter on the diode test setting. (All test essentially the same.)

Are you sure? *What* exactly do you measure, on the diode scale, on each leg pair, both ways?

The 0.47ohm power resistors on each also test OK in-circuit (not shorted).

Well , they *should* test "almost" a short.
In very low resistance power resistors, *open*is bad.
What do you read on what scale?

[drm1]

Are you sure? *What* exactly do you measure, on the diode scale, on each leg pair, both ways?


Well , they *should* test "almost" a short.
In very low resistance power resistors, *open*is bad.
What do you read on what scale?

Thank you for taking the time to reply.

Transistors: Measured on the diode scale, every leg pair, both ways. All combinations measure one or two diode drops or open circuit.

Resistors: These measure 0.8 ohms on the 200ohm scale of my little digital voltmeter, and approximately 0.5 ohms on my VTVM on the 10 ohm scale.

[J M Fahey]

Well, so far so good.
Please build a series lamp bulb limiter, to be able to measure some voltages without blowing fuses.

[drm1]

Thank you again. I do have a lightbulb limiter. [I also have a Variac if that's relevant, though I understand it's not a current limiter. ]

[J M Fahey]

OK. Plug your amp into it, with a 60 or 100W bulb in series, and turn it on.
The bulb will glow quite brighta, but you should have *some* B+ and B- rails.
How much?
Also measure the "hot" speaker out pin, and last but not least, Vce and Vbe at those output transistors, and also at the drivers.
We expect around 600mV Vbe of the proper polarity and more than 3 or 4V DC Vce.
Also measure voltage across the bias transistor.

[drm1]

With the amp in series with a 60W bulb, rail voltages are +/- 61V

Hot speaker out pin is at ground.

Vce for all eight output transistors is +/- 60V
Vbe for all eight output transistors is +/- 0.4V
Both drivers (Q309 and Q310) read Vce of +/- 60V and Vbe of +/- 0.8V
Q311 and Q312 show a drop of +/- 0.8V

[Jazz P Bass]

With the amp in series with a 60W bulb, rail voltages are +/- 61V

Hot speaker out pin is at ground.

Can you please clarify this statement.
Why is it at ground?

[J M Fahey]

With the amp in series with a 60W bulb, rail voltages are +/- 61V

Good. Voltage loss is low, meaning the amp is passing "low", quite normal current, compatible with a sane amplifier.

Hot speaker out pin is at ground.

Good, so connecting a speaker there should cause no problem.
Vce for all eight output transistors is +/- 60V
All normal voltages. Those 0.8V look somewhat high but you probably measured on the 200V scale, just after measuring rail voltages. On the 2V scale you should have .65V or less.
All you post points to a healthy power amp.
Turn it on, still using the lamp limiter, and then connect the speaker (all controls on 0 and no signal).
What happens?

[drm1]

Using the lamp limiter, speaker connected, all controls on 0 and no signal:

Dimly lit series lightbulb; amp powers up fine.

[Steve Conner]

Looks like there is no obvious fault according to those tests.

Did we already check all the "systems" stuff? Heatsinks free of dust bunnies, cooling fan working, speaker system is the right impedance, no intermittent shorts in the speaker or cabling, line voltage correct?

If so then it may be an intermittent fault that makes the bias voltage run away. Bad connector, solder joint or similar. Or maybe the previous owner blew it up already, replaced the output transistors, but didn't bias the new ones properly, or didn't fit the temperature sensor properly.

You said the speaker output was at ground. The voltage measured there should be almost zero but not quite. If it was exactly 0 I would get suspicious that it was shorted. However I have seen amps with DC servos that have only 1mV offset.

[drm1]

First, thanks to everyone who has made suggestions; I am learning a lot.

Second, the output isn't quite at ground--measuring with a VTVM on a very sensitive scale shows it to be a tiny voltage above.

Third, I powered up the amp without the series limiter, first with a Variac and then without. It does not blow the fuse. I think I may have spotted the problem while reassembling: The connector to the power amp module (GND, B+, B-, AC, Out) has a scorch mark on the circuit board; I think there may have been a loose connector that shorted out one of the rails or the AC.

In any event the thing powers up. And the distortion is back but I am now almost certain that it is from a problem in the compression circuit. I will work on that (it's a voltage-controlled amplifier on an IC, controlled by a sample of the input signal). I'll start a new thread when I need help!

Thanks again for getting me this far. --DRM

[Jazz P Bass]

Easy enough to narrow down the distortion.
Insert a signal in the return jack.
That will effectively bypass the compression circuit.