Harman Kardon Restoration issue

tutu75

Member
2013-05-28 1:08 am
Hello,

I'm the proud owner of a 1973 HK630, which works pretty well 90% of the time. However I want to get new speakers and I don't them to be destroyed by this amplifier.

The issue: After powering up the amplifier, I can see the speaker cones going all the way out, then all the way in, and finally stabilizing in the middle position. This takes about 5 seconds and happens even if the volume control is set to the minimum.

I have no idea where this problem might come from? Is this normal behaviour to output a quasi direct current for a few seconds after starting up? Is there something I should investigate in the electronics to pinpoint the issue? I do have the full schematics from the service manual, a meter and a scope, but not much knowledge in amplifier designs.

Thanks for any input you could provide.
David
 

kevinkr

Administrator
Paid Member
No IMHO this would not be considered normal behavior in most amplifiers.. I've seen a few direct coupled designs without relays that do develop appreciable offsets at power up, but nothing like what you describe.. More likely there are some capacitance time constants that are out of whack due to deteriorating capacitors, it could also be the case that some are just leaky when voltage is first applied.

The place to start is to replace all of the coupling capacitors in the audio path, and for insurance it is usually a good idea to do all of the power supply capacitors.

If you are really ambitious replacing thermal grease and mica insulators is not a bad idea either..
 

tutu75

Member
2013-05-28 1:08 am
Thanks for your answer!

I'll try to order a few capacitors and start replacing them, that should keep me busy for a while. I'll post again to let people here know how it turns out. I guess I should try to order a few different cap to hear differences in sound, I'm not trying to stick to the original 1973 sound though.

I just like to have that nice amp working forever ;)
Thanks
 

tutu75

Member
2013-05-28 1:08 am
Check the DC power supplies...B3, B4, B5, B6 for ripple and voltage.
Check/replace C807 and all caps in the low voltage dc supply stage.

Dan.

Download the Harman Kardon HK-630 service manual for free - Hifi Manuals

Thanks, I've opened up the case and check for voltage and ripple.
B3 B4 B5 and B6 have current voltage and no observable ripples as far as I can measure.

When I power up, B3, B4, B5 and B6 and nicely going from 0V to their correct voltage. They also smoothly discharge when I power down.

The time it takes for charging is about the same time I see the speaker outputs going full power in quasi DC.

B1 and B2 have correct voltage and about 0.2V of ripples.

This amplifier have been obviously restored previously as many of the caps looks pretty new Elna ones and I see some marks of soldering. However C807 is not one of these elna ones and looks to be the same brand as the others that looks original.

Maybe I'll try to trace "backwards" where the DC at startup is coming from, by starting from the speaker outputs and trying to measure that variation backwards to the pre-amp inputs?

Do just a few at a time and check the end result.. Sounds like you are on top of it.

Will do, I just start by ordering most of them that look to be the original ones and will change a few at a time. I added C807 to the list.

Thanks you both for your help, it got me started on this project :)
I'll post my findings
 

tutu75

Member
2013-05-28 1:08 am
QA few more observations:

The very slow current pushing my speakers all the way in and out when the amplifier is powered up is generated in the Tone Amp circuit board.

The power amp board has two coupling capacitors C601 and C621 which are letting this very slow alternative current go through the power amp board and the output signal follows exactly what the Tone Amp circuit outputs.

This very slow oscillation is generated around the TR503, TR504 and TR553, TR554 transistors and components.

This very slow oscillation is not measurable on the negative pole of C511 and C561.

TR501, TR502 and TR551, TR552 are producing the same slow oscillation.

As kevinkr suggested
More likely there are some capacitance time constants that are out of whack due to deteriorating capacitors, it could also be the case that some are just leaky when voltage is first applied.
it looks like a time constant issue. When the power B3 rise up slowly nothing happens, then then B3 reached 21V (or a little before) the couples of transistors on the Tone Amp board are going alive and oscillate very slowly for 5 seconds.

The measured error around TR503, TR504 and TR553, TR554 looks like a step response of a 2nd order filter with a big overshot followed by two unwanted oscillations before settling down.

Not knowing exactly what to do I tried to change R801 by a 44k value then a 11k to see what happens when B3 voltage rises twice as fast or twice as slow.
The results on the unwanted oscillation was worse in both cases.
 
I guess that if both channels show the same problem then it might either be a power supply issue, or may be it's just the nature of the circuit. It's extremely hard that both channels develop the same issue at the same time.
If that's the only problem, if it's such, you can add a speaker time delay/DC protector board and forget about it.
 
QA few more observations:

The very slow current pushing my speakers all the way in and out when the amplifier is powered up is generated in the Tone Amp circuit board.

The power amp board has two coupling capacitors C601 and C621 which are letting this very slow alternative current go through the power amp board and the output signal follows exactly what the Tone Amp circuit outputs.

This very slow oscillation is generated around the TR503, TR504 and TR553, TR554 transistors and components.

This very slow oscillation is not measurable on the negative pole of C511 and C561.

TR501, TR502 and TR551, TR552 are producing the same slow oscillation.

As kevinkr suggested it looks like a time constant issue. When the power B3 rise up slowly nothing happens, then then B3 reached 21V (or a little before) the couples of transistors on the Tone Amp board are going alive and oscillate very slowly for 5 seconds.

The measured error around TR503, TR504 and TR553, TR554 looks like a step response of a 2nd order filter with a big overshot followed by two unwanted oscillations before settling down.

Not knowing exactly what to do I tried to change R801 by a 44k value then a 11k to see what happens when B3 voltage rises twice as fast or twice as slow.
The results on the unwanted oscillation was worse in both cases.
Ok, you have done quite some investigation. I wanted to know the time up period of the B3 supply....it looks like they have built in a time constant there.
I would perhaps not trust the power supply caps that have been replaced, and renew regardless...little cost.


I notice that this receiver has pre-out/main-in connection on the rear panel.
Did you confirm if the mis-operation does not occur with this link removed ?....if the 'fault' is still there measure/replace C603/C623 NFB bypass caps.
These caps also apply a strong voicing to amps, try bypassing with solid caps also..... try higher values like 0.22uF and up as low values can cause ringing of the parallel electro.

If it is so that the oscillation is due to the tone section, I would be prepared to replace or at least measure all the caps in the tone section.

I doubt this slow oscillation is completely normal, usually a quiet quick turn on thump/click on these kind of old amps/recievers.

Dan.
 

hifly

Member
2011-07-03 4:39 pm
Hi David, I bougt a H/K 630 in early 1974 and still have it.
The phenomenon you discribed was there since new. The speaker cones moved very slowly but they did not bottom out or make any sound so I never saw it as any issue.
Today the amp sits retired and maybe waiting for a recap, because I feel a bit emotional about it after decades of trouble free service.
 
These caps are the ones I've replaced so far:
db8b2ccc8f7d329ded2e08797a1362d0.png
 

hifly

Member
2011-07-03 4:39 pm
Hi hifly,
Can you remember if there also were a pop from the speakers when you turned the receiver on when it was new?

If I remember correctly, yes there was. It also took some maybe four seconds to power up making a brief mechanical buzz therein.

Although I could, it may not be wise to actually see what it does because the amp has been unused for a couple of years now and all caps are original.

That is a well built and good looking amp.