• 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

Sansui BA-3000 runs super hot when bias set correctly

dr*audio said:
If you connected your meter to the bias test points and you used it's mA connections, it's no wonder the driver board heated up. The current connections on a meter have a shunt resistor across them. If it's the Amp connections, rather than mA, the shunt resistor is a very low value, so you effectively connected a very low resistance across the emitter resistors in the output stage, hence the damage. Surprised the fuse in your meter didn't blow.
Click to expand...
ok, good to know. other than just now, I never tried mA actually. The right card blew up because I turned the bias too high. :)
 
I edited an earlier post, so you might not have seen the edit.

Adjust your DC-offset as per the manual ...

richardpp said:
Mutli-turn trimmer allows a finer adjustment but you're close enough. Connect up F-2497 and adjust the dc offset at the speaker terminals as per the service manual's instructions

Then measure the voltage across pins 2 and 4 and adjust VR02 as per the manual instructions for setting bias current.
Click to expand...
 
Tina Louise said:
actually, I just realized I had the F-2497s plugged into the wrong sides (back F-2497 is supposed to be the right). bias is looking better now. I'm at 14.5mV now and slightly more than half way.

I looked at my AU-20000 and compared.

Holy crap, it came that way...
Click to expand...

replying to myself...

looking at the circuit, I don't think this should make a difference. I think it might point to a problem with the back output module?

regardless, need to pick it up again tomorrow.
 
The back heatsink is usually the left channel, otherwise the speakers will be reversed.
 
Last edited:
richardpp said:
The back heatsink is usually the left channel, otherwise the speakers will be reversed.
Click to expand...
Ok, so I think all of this is going to point to that module being at least partially bad. I did all the work to put molex connectors on these connections specifically to rule out bad connections on the board and to debug the channels easier and then started having all these other issues.

And then I have to look at my Au-20000 for left and right being switched. Lol.
 
Tina Louise said:
0.0mV with VR02 same as before (1/4 for counter clockwise).
1.4mV with VR02 half way.
scared to go higher right now.
Click to expand...
If these are your readings through the back (left) channel it's likely there's a problem with the output transistors or their installation, or the F-2497 board itself. The downfall of the board is, being double-sided, the pass through connections can go intermittent.
 
Last edited:
Tina Louise said:
And then I have to look at my Au-20000 for left and right being switched. Lol.
Click to expand...
I checked your photos, just swap the F-2510 connector boards in your AU-20000 and you'll be fine. The BA-3000 was wired correctly when you got it so just keep your new connectors as they were.
 
Tina Louise said:
As an aside, just confirming that all the PNP output transistors are together on one side of each output heatsink and NPN are all together on the other side? the output transistors were mis-matched when I got them, but I still think the PNP and NPNs were all together (and the manual says so too).
Click to expand...
That's correct, the PNP transistors all face the back of the amp and the NPN transistors all face to the front.
 
Tina Louise said:
Ok, so I think all of this is going to point to that module being at least partially bad
Click to expand...

If that's the case, first check F-2497 left channel for:

+68Vdc on board connection 25
-68Vdc on board connection 26

Edit: 25 and 26 are the connections from the main filter capacitors. If the voltages are good, next I would turn the power off and check for continuity between 25 and each of the NPN transistor cases and between 26 and each of the PNP transitor cases.
 
Last edited:
richardpp said:
If that's the case, first check F-2497 left channel for:

+68Vdc on board connection 25
-68Vdc on board connection 26

Edit: 25 and 26 are the connections from the main filter capacitors. If the voltages are good, next I would turn the power off and check for continuity between 25 and each of the NPN transistor cases and between 26 and each of the PNP transitor cases.
Click to expand...
+65/-65V on my receiver, but yeah, I have voltages.

As an aside, I put inline fuses between 25 and 26 and the main capacitors. @Hyperion used 5amp fuses. Do you concur? I have 2amp fuses in there now because I figured I'd start small.

25->NPN continuity.
26->PNP continuity.
that's cool. that's testing for any breaks in the F-2497 boards?
 
richardpp said:
If these are your readings through the back (left) channel it's likely there's a problem with the output transistors or their installation, or the F-2497 board itself. The downfall of the board is, being double-sided, the pass through connections can go intermittent.
Click to expand...

so the back brick clearly has something wrong at this point....

"left" F-2507 board
left slot
front brick
bias = 10.2mV with VR02 only turned slight past 1/2

"left" F-2507 board
left slot
back brick
bias = 3.7mV with VR02 same as above

"right" F-2507 board
left slot
front brick
bias = 11.3 mV with VR02 only turned slight past 1/2

"right" F-2507 board
left slot
back brick
bias = 0.7mV with VR02 same as above

My new molex connectors make this nice and easy.

When I was thinking about the bricks potentially being a problem a week ago or so, I was trying to figure out where the pass-throughs were on the F-2497 boards from photos on the forum, but couldn't figure it out. This was before I realized I had the F-2497 board and not the earlier design which confused the heck out of me.

So, I think what I should do is take the back brick out, check all transistors again for shorts, check the resistors, check for any breaks in the board... anything else? The thermistor is just a standalone loop (and I've already tested that through the wires going to the protection board).

Based on what other people have gone through taking the brick apart, this is going to take a while.

Any other tests I missed before I get into this? (I forgot the voltages on the right f-2507 doing that now)

Thanks again for all your help.
 
Last edited:
I didn't match hFE when putting the main transistors in, btw. (I bought them all new at the same time). my multimeter has hFE on it, but couldn't figure out what to connect to what. I can probably test it by clipping lead wires on the transistors and plugging it into my small transistor tester (which I used for gain matching on smaller transistors). but how is it done with a multimeter without the socket for transistor testing? searching on google shows people making a bread board as part of the testing in which case I'll just clip lead wires on and use my small tester.

What am I gain matching though? all PNP? all NPN? all PNP and NPN together? PNP and its matched NPN (TR01 with TR03 e.g.)?
 
Tina Louise said:
+65/-65V on my receiver, but yeah, I have voltages.

As an aside, I put inline fuses between 25 and 26 and the main capacitors. @Hyperion used 5amp fuses. Do you concur? I have 2amp fuses in there now because I figured I'd start small.

25->NPN continuity.
26->PNP continuity.
that's cool. that's testing for any breaks in the F-2497 boards?
Click to expand...
Do I concur with Hyperion? Absolutely :bowdown:

2 amp fuse rating is good for testing the amp at idle

Correct, continuity between the board edge connections and the transistor cases is testing for any interruption in the supply voltage to the transistor collectors.
 
Tina Louise said:
so the back brick clearly has something wrong at this point....

"left" F-2507 board
left slot
front brick
bias = 10.2mV with VR02 only turned slight past 1/2

"left" F-2507 board
left slot
back brick
bias = 3.7mV with VR02 same as above

"right" F-2507 board
left slot
front brick
bias = 11.3 mV with VR02 only turned slight past 1/2

"right" F-2507 board
left slot
back brick
bias = 0.7mV with VR02 same as above
Click to expand...
Yes, I would agree that there is clearly something wrong within the rear heatsink assembly.

But besides the problems with the bias readings when connected to left channel heatsink, overall your readings appear lower than expected on the right channel. Adding to this, rail voltage is lower than expected. The question is whether this can be all tied to a single fault.

tbc
 
Tina Louise said:
So, I think what I should do is take the back brick out, check all transistors again for shorts, check the resistors, check for any breaks in the board... anything else? The thermistor is just a standalone loop (and I've already tested that through the wires going to the protection board).

Based on what other people have gone through taking the brick apart, this is going to take a while.

Any other tests I missed before I get into this? (I forgot the voltages on the right f-2507 doing that now)
Click to expand...
I wouldn't rush into dismantling the left heatsink. My first inclination would be to isolate it from the power supply by removing the in-line fuses. Then I would retest your supply voltages at the main filter capacitor terminals.

Have you carried out the power supply check #61
 
You must log in or register to reply here.
Back
Top Bottom