• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Kenwood KR-6050 5v DC on left channel with subsonic filter

spaghetti

spaghetti

Active Member
After sitting neglected for a few months, it's time to try and fix this nice receiver that I bought during the spring and left sitting until now.
It's basically working, but after some warmup minutes, when I use the subsonic filter switch the receiver goes in protection. What happens is that DC offset on the left channel gets as high as 5V, so the protection circuit is fine. This happens only when speakers (or dummy loads) are connected and volume is below a certain level (let's say about 1 o'clock). When the subsonic switch is off, or I am listening with headphones, things keep normal and the receiver keeps working fine.

This might be some kind of design flaw or common failure since I was able to track another unit with the same issue at this link: http://www.oaktreevintage.com/Kenwood_Stereo_Parts_Unit.htm (page is long, you can search and find "protection relay tripping" to get to the relevant part).

So, I am looking for some ideas on where I could start my troubleshooting, I can upload the schematics if that could help.

Thanks for help,

Andrea
 
Register to hide this ad
Sounds like multiple problems. For one, the protection should trip whether listening to speakers or through headphones. It's voltage detection at the amp output, which will change very little from one to the other. So the change from headphones to speakers baffles me a bit. Should be FUBAR for both, not one or the other, unless the overload detection is getting tripped. Tried to follow that on the scat, but the screwed-up way it's drawn makes it tough to see how the detection is done.

Subsonic filter simply adds a basic R/C network right before the first power amp stage. I think it strange that adding a capacitor to the signal path causes an increase in offset. What I suspect is that your proper DC offset is being shown when the subsonic filter is engaged, which would mean that the C7 2.2uf 50V bipolar cap is bad, and adding the C19 subsonic cap filters it out. But if the offset was adjusted when the subsonic was not engaged, now it goes nuts when the C19 cap is added to the signal path.

Bottom line is that it appears you may have a number of 'past due' caps in that preamp, mainly those between the volume control and the JFET power amp input stage.
 
EchoWars said:
Sounds like multiple problems. For one, the protection should trip whether listening to speakers or through headphones.
Click to expand...

I'm sorry my description was not clear: if, after turning on the receiver, I only use the headphones then the protection circuit will never cut the music. In this case, the DC offset when operating the subsonic filter rises only to 0.6V, which seems to be not enough to trigger IC1 offset voltage protection. On the other hand the issue happens when speakers are connected for 5~10 minutes, so I think it somehow has to do with certain circuits warming up or working intensively. Yesterday I also checked the outputs bias, both at power-on and when the subsonic filter starts to trigger protection, but the measurement was consistent with the expected values on the service manual so it looks like a dead end.

Should be FUBAR for both, not one or the other, unless the overload detection is getting tripped.
Click to expand...

Can you confirm that the overload detection actually used in this amp? I looked at the schematics, pin 4 and 6 (thermal detection, overload protection) are all connected to ground (pin 2), so I guessed that those functionalities were disabled, but since I don't actually know how that functionalities should be wired, that was pure guessing.

Bottom line is that it appears you may have a number of 'past due' caps in that preamp, mainly those between the volume control and the JFET power amp input stage.
Click to expand...

Thank you very much for the analysis, I checked a few small elect. caps (1 and 10uF) and they were close to their spec limit, but found no crazy numbers. I will try to make some more measurements and tests, I will come back if I find something that I can't make sense of.
 
Don't bother measuring the caps...it won't tell you what you need to know. Replace them.

Kenwood has drawn the scat in a different manner than they previously did, but after studying I think you may be correct in that they did not utilize the thermal and overload protection on the IC.
 
@Watthour thank you for the schematic, it's much more convenient than the service manual!

Yesterday I decided to isolate the preamp from the amp, removing R39 and R40.
I noticed that after a while on the base of Q9 some voltage started to grow (at some point it was ~200mV and still growing), while the base of Q10 on the other channel remained ~0V.

I thought that this voltage could be the reason for the malfunction, and since there was no other component that could generate the voltage I swapped Q9 and Q10 and now the issue moved to the other channel, so I think I have found the culprit.

Now I have to source a replacement, 2SK150A seems to be unobtanium nowadays, while UPA68H can be easily sourced on ebay, but there's good chances they may be fake. I will visit the local shop later this week, but I am not very positive about finding a replacement there.
 
Since the circuit operates around 21 Vdss it may be safe to use a matched pair of 30V N-FETs bonded together to replace the SIP package.

Another choice would be to use 30V pr 50V SMTs converted to through-hole configuration, matched and bonded together, of course. I'm fighting one of those now, thus my thread here:

http://audiokarma.org/forums/index.php?threads/smt-adapters.735171/#post-9954297

There are a few through-hole 50V N-FETs still in the marketplace, but they are NOT inexpensive at $7-12 apiece, or more. Moreover, most of those are intended for RF applications and have maximum drain currents down into the low mA or even µA range.

Of course, there are inexpensive and plentiful 50V versions available but only in SMT devices, like the 2SK880s or similar: http://www.mouser.com/ProductDetail...0HnGdrLjpqW5Yfjz4xG4m%2b1PWVu9TxxD0QTKJmRiQ==
 
This morning I went to the shop, of course I was not able to find the exact replacement. On the other hand the old man gave me some stuff, I hope among these options there is something that will work:

1) a couple of 2N3819 (they're max 25V, it should be ok anyway since there should be 21V in the circuit)
2) a couple of 2N3822, max 50V like the 2SK150A but metal case and a bit more larger than 2N3819
3) a double jFET U406, which is marketed as "switch", so I don't know if it may be really suitable for this application

Is there something I should prefer to use?

Thanks,
Andrea
 
The 25V devices are just a little too close to the operating voltage for comfort. Any excursion in supply voltage might be a problem. The physical size of the metal cans should net pose a problem.
 
I decided to buy also one cheap chinese uPA68H on ebay and wait for it to arrive, I will decide eventually what to use.
 
You must log in or register to reply here.
Back
Top Bottom