• 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

AU-999 initial powerup, after restoration

Well that's a relief. All 4x 2SC898's appear to be good: I get no reading from base-->collector & base-->emitter one way, and .5v from both the other way, using the diode tester on my MM.
 
Here's another clue: when I initially powered it up with the DBT I had my bias probes hooked up to the wrong resistor, either 855 or 856. ( although now, looking at the schematic I see they mirror 853 & 854. So you should be able to use them to set bias, too? )

Anyway I disregarded this once I realized they were the wrong resistors. But the (bias??) reading climbed up really high after powerup: to over 220mV. And didn't respond when I tried adjusting the bias pots. Then dropped back down to 0.

When I realized I had the wrong resistors, I figured it was something unrelated. But maybe it is relevant?
 
I have been going thru boards testing for continuity. I doubt it's on the 1159 driver boards, as I didn't replace any bias related parts other than the 1K trimmers.

The 1277, 1232 ripple current boards underneath seem the most likely culprit, as they were a pain to solder & awkward to get to. But so far all the connections seem correct & no shorts.

Will x2 check F1230, maybe I installed a cap backwards
 
So if you swap 1159 boards there will be no changes to symptoms? In that case it could very well be the 1277, 1231 - a fault on either would affect both sides, they are not part of the L/R symmetry.
 
It is very strange that the bias on each 1159 directly affects the other, but inversely. And the bias only kicks in at the extreme end of the 1K adj pot. At max I can only get 3-4mV from both. If I back one off, the other goes up to 8-12 mV but no more.

I tested the traces on both 1159 boards: everything seemed to be intact. My MM had bad signals along the connector edges so I cleaned them with 99% iso, but it made no difference when I plugged the boards in again.

I guess I will start testing the original transistors C875E x2, A5326 x1 on the 1159 boards
 
Last edited:
bolero said:
It is very strange that the bias on each 1159 directly affects the other, but inversely. And the bias only kicks in at the extreme end of the 1K adj pot. At max I can only get 3-4mV from both. If I back one off, the other goes up to 8-12 mV but no more.
Click to expand...
It might be the action of the DBT (if you are still on DBT?), when you back one off, the voltage dropped is lower so you get more bias on the one channel?

Don't be tempted to take it off DBT just yet, because I could be wrong about the above - or something else that we don't know about might be happening. ;)
 
Oh, it's been off the DBT for a while!

I still use it to power up, after doing something. How long does it normally take the big filter caps to discharge? I get 0V across them within minutes of turning it off.

I may have found a short across cap 816 on F-1277. Small piece of solder in there.

And thanks!
 
Another strange thing: all 3x AU-999 schems I downloaded from hifiengine are chopped off at the bottom. So I can't trace out those boards. But there's no way to report messed up files over there!

*edit* ok I found the schem here on AK, but it's not as clear a scan. At least it has that section!
 
Last edited:
Cleaning up between cap 816 on F-1277 made no difference. I guess I was getting a reading from a charge stored in the cap.

I'll start measuring voltages in the circuit, now that I have a full schem. Will keep it on the DBT too.
 
I just measured all the resistors on boards F-1231, F-1277, they are ok & within tolerance. The caps are all new except for the small .01uF C817, C818 ceramics.

I'll pull the F-1159 boards again & start measuring components.
 
Ps I have not tried running a signal into the amp yet, or hooking up a load/speakers. I assumed getting a stable bias first, would be priority?

Thanks again, all!
 
bolero said:
Ps I have not tried running a signal into the amp yet, or hooking up a load/speakers. I assumed getting a stable bias first, would be priority?

Thanks again, all!
Click to expand...
Correct, my personal approach would be to get the DC conditions correct (i.e. DC Offset & Bias), before applying a signal.
 
Thanks!

Another clue: I figured I'd measure some voltage before pulling the 1159 boards out again...I get -44v directly off the negative regulator tap, which matches the schem, but only +30v from the positive tap. It should be +44v
 
Last edited:
Here's pics before & after the filter cap change , I'm pretty sure I installed the regulator properly?

Notch @Bottom left is +
Then AC bottom right & diagonal
Top right is -


999_pre_filter_swap.jpg 999_post_filter_swap.jpg
 
Last edited:
Man I don't get those 90's resolution pictures on this server, I would think that in this day and age storage would be cheap enough, but enough of my silly rants.
I can see the crowbar is still there (F1313) that's an ancient protection circuit, by this day and age it give more problems than anything IME (MAC1700 and 2 AU-999). Is that connected?
I can't see much, and I have no idea what filter caps did you use, but those wires could be connected to dummy terminal on the cap.

upload_2022-12-1_17-22-47.png
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom