• Please note that Audiokarma will be offline briefly for updates early on Monday morning, September 28th.

Please Help...Did Some Damage to Sansui AU-999 Left Output Stage

Zekeman

Zekeman

AK Subscriber
Subscriber
I was following the instructions for replacing the "flying-saucer" transistors in the "AU-999 Redux" thread here

http://www.audiokarma.org/forums/showthread.php?t=144862

...When I made a big mistake and installed the wrong transistors in four locations on the left channel driver board. The first picture shows what I originally installed and then corrected. I had already recapped this board as well prior to powering it up with the incorrect trannies...

When I had the incorrect transistors installed I powered it up, got a hum, the main fuse blew, and a bit of smoke rose up the driver board past the right side of the big Elna cap in the first picture. So I shut it down cursing at myself.

I replaced the trannies as well as the two diodes at teh top of the board and then went ahead and replaced the 8002 trannie on the little board F-1231 under the chassis. I also recapped it (see second pic).

Then I checked all the resistors on the driver board and all checked out with positive resistance (I checked them while mounted in the board). I pulled the output trannies and they both checked out, meaning they did not read open/shorted (last picture).

I installed new micas and heat sink grease (there was no heat sink grease at all).

I powered it up and I get undistorted sound out of both channels but the left channel heats up like a Weber grille within a few seconds so something is definitely wrong.

Could the Bias be whacked out due to installing new transistors on the driver board?

I am not sure where to turn next.
 
Last edited:
Register to hide this ad
ALWAYS power up with a dim-bulb tester after major work. Always.

No magic fix here. Start checking transistors and resistors. And until you can get the current flow under control, power the unit up with a dim-bulb tester & 100W bulb.

I don't see you checking to see if the C-E is shorted on the outputs. And hopefully you didn't kill the TO-66 drivers, 'cause those are gonna suck to find if you did.
 
Zekeman said:
<snip>

I powered it up and I get undistorted sound out of both channels but the left channel heats up like a Weber grille within a few seconds so something is definitely wrong.

Could the Bias be whacked out due to installing new transistors on the driver board?

<snip>
Click to expand...



Yes.....

Did you locate WHERE the smoke you saw came from?

I suggest you check the VR803 pot out of circuit to see if that got scorched internally.

Check your bias servo circuit transistor, TR809,
and recheck all the resistors around it now that power has been re-applied:
R835, R833, R831, R829 etc
I have been fooled in the past in thinking resistors are ok, yet when lifting one leg got a zapped reading, something in the circuit spoofed the measurement. YMMV....

Read the collector and emitter voltages at TR809 on the cool side, and then on the hot side, see if you can adjust them to the same (down probably on the hot side) voltage.


The good news is with relatively clear sound, almost everything in there is behaving itself - I HOPE, because second harmonic distortion does hide in the music...

edit: he PM'd me calling for help...

edit2: the to-66's are TR815 & TR817 ??
 
Last edited:
OK...Attached first pic shows that the big output trannies, TR819, TR821 tested OK.

Mark...the second pic shows graphic of whet I believe was the path of the visible smoke...When this happens it was a bit of a freak-out so that's why I say "I believe..."

I then pulled TR815, the little and it tested OK as an NPN.

Tested TR 817 (a PNP) and get what is in the fourth pic: so it tests OK.

Mark Question: your quote reads "...Read the collector and emitter voltages at TR809 on the cool side, and then on the hot side, see if you can adjust them to the same (down probably on the hot side) voltage...." ....what does Cool and Hot mean?

======================================================

Resistor test Results:

TR809 B-E=0.349, B-C=0.370

thirdpic shows board with one leg of about 17 resistors lifted...all test OK.

I also pulled both pots and they both adjust resistance...no open result (although I am not 100% sure if I am testing correctly: positive lead on center pin and neg on one...test...change negative to other pin and get opposite result).


All transistors on this board test OK (In board test).

======================================================

Question: The two caps were replaced before the initial smoke...Should they be replaced as well?
 
Last edited:
...Update...Pulled one leg of all resistors and they are within spec. Measured all resistors on the small board underneath as well as the cement emitter resistors and they are OK. I am going to build a dim bulb tester tomorrow and try it...only I am not quite sure what I am looking for other than a lightbulb that burns at full brightness (indicating a short). I believe I need to measure the bias as all trannies in that board have been replaced and I am hoping this is the source of my heat problem.

Advice?
 
Last edited:
Bright bulb can mean a shorted transistor, or that the bias is cranked way up. If you power up with the bulb tester and don't like what you see, measure the bias and turn it down to minimum. If you can't, then obviously your problems are more serious than high bias.
 
OK but question. I tested all the transistors on the driver board as well as the little board under the chassis with the emitter resistors on it, and none are showing as shorted. Could any circuit boards other than these driver boards be in question? ...anything earlier in the signal process?
 
Last edited:
Built dim bulb tester with 100 watt bulb.


The output trannies get warm pretty quickly with the dim bulb tester attached but never too hot to touch. I plugged it in directly and the trannies get too hot to touch within 5-10 seconds....as well as the cement resistors underneath (see pic)

I replaced the bias trimpot. I am measuring a couple volts (volts not millivolts across the emitter resistor (second pic).

I replaced TR801 and 803 with fresh 795's....same result.

Both aluminum electrolytics on the driver were replaced. I have not replaced the small flat round capacitors...Could these be the problem? (see third pic)

Also, the serial number on my unit is 210030467 and it does not have a protection board near the middle of the underside of the chassis (see pic)...and the large green one in the fourth pic.

Any suggestions on what to test next?
 
Last edited:
Gee...I wish there was some way to narrow the problem down to 'driver boards' vs 'something on the heatsink or other component'....

Hmmmmm.....whatta you think? :scratch2:
 
So I pulled the left driver board and powered it up with the dim bulb tester; bulb goes out completely and right channel plays fine.

I put the right channel driver board in the left's slot....Dim bulb goes out, channel plays with no heat.

Desoldered and removed TR809 (NPN) off of left driver board and get the following readings:
...with leads positioned as follows: positive on base, negative on emitter = OL
positive on base, negative on collector = 0.472
negative on base, positive on emitter = OL
negative on base, positive on collector = 0.550

...so being an NPN it looks like this transistor appears to be bad as I believe I should have gotten a positive reading in the second measurement, correct? Then why does the smoking of the transistor result in the positive reading in the fourth reading?
 
Last edited:
Those readings don't make any sense. What's the number on the transistor? Seems more like you have mis-identified the leads.

This is the bias servo transistor, and it should be remote-mounted.
 
That's the one...the dangler. 2SC281...I just went down and confirmed the measurements. I marked the leads before I pulled it...
 
Last edited:
Just givin' you my impression...you have no shorted leads, and two good diode drops. Looks a lot like misidentified leads to me. You can stuff pretty much any NPN in there as a quick test.
 
The schematic corresponds to the board and the installation of the component....
 
Last edited:
When you have been junction / conduction testing the transistors, have you exercised the principle of:

"When in doubt, take it out" ??

and removed the transistor from the circuit? To eliminate any possible interference?
 
markthefixer said:
When you have been junction / conduction testing the transistors, have you exercised the principle of:

"When in doubt, take it out" ??

and removed the transistor from the circuit? To eliminate any possible interference?
Click to expand...

Well...being a novice here, no I have not, except for the outputs on the heatsinks...but I did pull the TR809 tow nights ago (which gave me the strange measurement results).

I did receive some additional 694b NPN's tonight, stuck one into the TR809 spot and powered it with the dim bulb: bulb lights and goes out...left channel played OK. Ordered some new dangling transistors. It appears that TR809 was it...! I am still confused about the bizarre readings/leads though (two good two "OL").
 
Last edited:
I received some fresh 2SC281's from Acme. Now it runs cool. Adjusted bias to 13mV.:banana::music::thmbsp:

I installed big new 10,000uf 50V Nichicon ps caps and replaced all the FS's on the right channel driver.

So it now has all new caps and transistors.

Thanks Again gents!
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom