• 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

9090 Static / Current at Idle into Protection SOLVED

Johnny_Law

Johnny_Law

o.O
Subscriber
EDIT: Solution was to replace TR01-TR04 (2SC1708) on driver board with KSC1845. Each measured OK using multimeter but would heat up during operation and introduce much noise into the signal path.

After extensive searching and not finding this exact issue ... here goes.

I have a 9090 that goes in and out of protection. Sometimes it'll last for a few minutes out of protection, but will go back in. While it's out of protection, the VU meters jump around, and it produces static through speakers.

The power supply and driver board both have new caps, and all transistors have been bench tested good. All output transistors have also been pulled and tested good. All fuse resistors on the driver board are either new or tested good.

With the driver board pulled, it doesn't go into protection.

Notes so far:

The measured DC offset before it goes into protection jumps all around and maxes out around 4.5V.

I have cleaned the trim pots on the driver board, and de-oxited the pots, with no effect.

I observed no insulators in place between the output transistors and the heat sink - since the body itself is the collector, shouldn't there be insulators in place?

And to help explain, here's a video I took where you can see the VU meter bouncing around then going into protection. Plugged in through a DBT. It was originally doing this on the right channel, but today apparently it decided to do it through the left.

Thanks for any thoughts ...

 
Last edited:
Register to hide this ad
Do you get the static thru both speakers - or just one.

Does the problem stop when you have a particular driver board pulled - or only if both are pulled?
 
I'd try to take out the F07/F08 fuses (same ones you measure bias current on) off the PS board, clean the holders and fuses, make sure they make good firm contact. I'd also thoroughly clean the driver amp socket pins at the PS board and the respective contacts on the driver amp board.
Johnny_Law said:
and all transistors have been bench tested good
Click to expand...
They can test good and still be crap, notably the Toshiba 2SC1628 (TR9/TR10) and the Toshiba 2SA818 (TR5/TR6). I would just swap them for ZTX694B and KSA1381ES respectively (watch out for the pinout). I would also replace all the rectifier diodes for 1N4148. There's no insulator sheets on the power transistors. The bias/offset trimmers can get tricky even after cleaning. It really depends on how oxidized they are and they do get oxidized. It's best to replace them all together. Bourns have nice alternatives that everyone here uses, whether multi or single turn trimpots.

Johnny_Law said:
The measured DC offset before it goes into protection jumps all around and maxes out around 4.5V.
Click to expand...
That's 4500mV and highly unlikely at speaker terminals. IIRC, the protection relay would kick in at between 350mV - 600mV. Are you sure you're meter is set to the right scale?
 
To me this seems to be a classic example of bad differential amp transistors (noisy). this problem is usually not identified by testing the transistors in or out of circuit as they will usually show good. They at times can be identified using chiller, and heat to create the problem or temporarily fix the problem. I would suspect transistors TR02 and TR04, as causing the problem most identified in the right channel, but would also replace TR01 and TR03 as well.
I also agree with Tnsilver about those problem 2SC1628 and 2SA818 transistors.
 
slimecity said:
Do you get the static thru both speakers - or just one.

Does the problem stop when you have a particular driver board pulled - or only if both are pulled?
Click to expand...

Static through both speakers - and it's just a single driver board (F-2436) so it's both.

I did take an earlier video, it didn't go into protection in the video, but you can see the other channel is bouncing. I didn't swap anything between channels in the meantime, so it seems random which channel bounces.


tnsilver said:
I'd try to take out the F07/F08 fuses (same ones you measure bias current on) off the PS board, clean the holders and fuses, make sure they make good firm contact. I'd also thoroughly clean the driver amp socket pins at the PS board and the respective contacts on the driver amp board.
They can test good and still be crap, notably the Toshiba 2SC1628 (TR9/TR10) and the Toshiba 2SA818 (TR5/TR6). I would just swap them for ZTX694B and KSA1381ES respectively (watch out for the pinout). I would also replace all the rectifier diodes for 1N4148. There's no insulator sheets on the power transistors. The bias/offset trimmers can get tricky even after cleaning. It really depends on how oxidized they are and they do get oxidized. It's best to replace them all together. Bourns have nice alternatives that everyone here uses, whether multi or single turn trimpots.

That's 4500mV and highly unlikely at speaker terminals. IIRC, the protection relay would kick in at between 350mV - 600mV. Are you sure you're meter is set to the right scale?
Click to expand...

Thanks, I did clean the fuse holders but will do again. I am about to place a Mouser order so I'll add TR5/TR6 and TR9/TR10 replacements to the shopping list, as well as the rectifier diodes. Hopefully can get away with the pots but thanks for the heads up if they prove to have gotten too corroded.

Tom B said:
To me this seems to be a classic example of bad differential amp transistors (noisy). this problem is usually not identified by testing the transistors in or out of circuit as they will usually show good. They at times can be identified using chiller, and heat to create the problem or temporarily fix the problem. I would suspect transistors TR02 and TR04, as causing the problem most identified in the right channel, but would also replace TR01 and TR03 as well.
I also agree with Tnsilver about those problem 2SC1628 and 2SA818 transistors.
Click to expand...

Ah, thanks, yes I've seen a YouTube video where the transistors are frozen and behave themselves. I do have an aerosol can of duster so I could try this myself I suppose. I'll add TR01 - TR04 to the list as well.

 
avionic said:
Those are thicker than ordinary sil-pads..
Click to expand...
However at pressures of 50 psi/square inch or greater those Aavid pads have pretty impressive thermal resistance figures. I wonder what torque value equates 50 psi /square inch to a TO-3 transistor.:idea:
 
All better. It was TR01-TR04 on the driver board. Replaced with KSC1845 and quiet as a mouse now. Thanks all!

(The output transistors are insulted, just copious amount of thermal compound.)

This was a good learning experience for me - even if transistors measure OK doing the diode test with a multimeter, they may still be bad, and a clue is that they heat up.
 
Last edited:
This was a good learning experience for me - even if transistors measure OK doing the diode test with a multimeter, they may still be bad, and a clue is that they heat up.
Click to expand...

Yes, here too. I had the left channel of my AU-317 "untamed" (DC offset), then tested the two 2SA906 and they measured fine. Then saw tnsilver comment above and replaced them with a couple of 2SA999 and now things are good again.

o7q35e.jpg

 
You must log in or register to reply here.
Back
Top Bottom