Harman Kardon HK770 DC offset

[Michael Maurice]

I'm repairing an HK770 power amp.

Both sides are exhibiting a DC offset at the speaker terminals of around 5V. That couldn't be adjusted with the presets.

The two dual transistors wired as long tail pairs were replaced with standard transistors but the fault persists.

Both channels exhibit the same problem.

I was thinking of increasing R404 to around 10K to give it more range also reduce R408 and R410 to say 27K.

It could be the other transistors, but many are no longer available.

Any suggestions?

[Radio Wrangler]

Check whether the voltages on the two zeners are symmetrical D404 and D402

It may be that one of the electrolytics C410 C408 could have gone leaky.

If the feedback lines from the right hand off the page stuff are what I'm guessing, the feedback in this amplifier is unusual in that it sets the amp up for full gain right down to DC. There is no C in series with R502, unlike most amplifiers. R462 is 8.8k and with R502 at 470 Ohms gives a DC gain of about 18, and this will amplify any DC offsets in the differential pairs. You may really need matched diff pairs in thermal contact in this amp to get a good low and stable DC residual on the output.

David

[thermionic]

.......If both channels are exhibiting the same problem, then I would be investigating the power supply first....

Are the rails correct? Is the smoothing ok? Does it have a regulated supply? Is this working as it’s supposed to do?

[Michael Maurice]

There are two separate power suppiies one for each channel. The only thing in common is the mains cable and switch. Each power supply has its own mains fuse. It is essentially two separate power amplifiers in the same box.

[vidjoman]

Have you got a load on the amplifier? A dummy load will do - in place of a speaker. A digital meter will usually give a reading, which may be several volts, without the load being present.

[thermionic]

Hi Michael.

I wonder if it is oscillating? Have you tried scoping the outputs with dummy loads connected, as Vidjoman has raised?

As both power amp sections have their own rectification & smoothing, I’m looking for a common denominator somewhere!

Is the ‘speaker protection circuits working? Is there a delay before the speaker relays click in?
Good luck & keep us informed please.


SimonT.

[Radio Wrangler]

The two channels are completely sepparate electrically, apart from the mains cable. But one thing they do share is history.

If there is a component from a dodgy batch, there'll be another in the other channel. If someone did a repair perhaps they 'updated' both channels to keep them the same?

Step one is to check that both channels have equal and opposite power rail voltages, then check those zener regulated sub supplies that feed the input stages and the offset pots.

Remember that this amplifier is critical on the matching of input transistors, it's an unusual one in that its gain is not rolled off towards DC and that any offset voltage gets the benefit of full amplifier gain before it appears at the output. Hitting it with some heat from a hair dryer to see if the output DC changes much can yield clues. If so then a component by component check with freezer spray can narrow things down. Checking all the capacitors for leaks and measuring all resistors will eliminate some possibilities, then you're down to looking for leaky transistors.

This is one of theose things where you have to go looking for trouble.

David

[dave cox]

I would also check the input electrolytic, in case the previous stage has some DC offset and leakage is offsetting the input stage(s). Disconnecting it would be one way of finding out !

Does the offset adjustment otherwise work ?
If so, how much adjustment range does it provide ?

dc