Sansui AU-X901, replacing power transistors, 2SC3519 / 2SA1386, options?

Hyperion said:
There are a couple of possibles - but really those MUR1620's are every bit as good as you need.
Click to expand...
Hyperion this MUR1620 are coming in complementary pairs (positive /negative )
I tried to find other options in pairs but I couldn’t .
The options you are referring , are complementary pairs ?
 
Sui-Yogi said:
are complementary pairs ?
Click to expand...
Yes they are, it's been a while since I researched these - IIRC, I came up with much higher current alternatives, ~20Amp (?) - something like that, and in the same package!

I believe in not using significantly over-rated components, as you can get yourself in a World of hurt (transformer-wise) if you use massively over-rated rectifiers. You want the rectifiers to go 'pop' before the transformer does.... 8A rectifiers (per leg) is plenty enough I think. ;)
 
Last edited:
Hyperion said:
Yes they are, it's been a while since I researched these - IIRC, I came up with much higher current alternatives, ~20Amp (?) - something like that, and in the same package!

I believe in not using significantly over-rated components, as you can get yourself in a World of hurt (transformer-wise) if you use massively over-rated rectifiers. You want the rectifiers to go 'pop' before the transformer does.... 8A rectifiers (per leg) is plenty enough I think. ;)
Click to expand...
Yes that is correct , 8A per side is more than enough for any Sansui I know of; I used them in a modified AU-X11 PSU with heatsink and they hardly getting warm , didn’t need a heatsink at all .

The reason I asked , maybe there was an option to avoid the Chinese element .
This is something that concerns me in the long run but might I’m overthinking it .
 
As for the power transistors: if not 3519, you can go for 3263 / and his pair, if esear available.
Dont go for Toshiba / Motorola variants - these are all 30MHz, Sanken are 60/35MHz and the circuit is optimized for such frequency (note that 4pF in the NFB phase compensation).
And above all: replace them all with the same batch, don't mix them.
My 2c‍♂️
 
The Power Supply works now after replacing the rectifiers.
Please note the Service Manual has an error on both the schematic and the board layout.
mD4, mD6 and mD5, mD7 are swapped.
It's good to use a DBT ...

Bias is good, Hot-GND is good. Cold to GND is off on the left side but good on the right. Something is still broken there.
 
Ok found a couple of issues, a bias transistor was bad and also at least two transistors on the protection board, Q32 and Q33 (this threw me off for a while ...).
Question is what to replace 2SA1115/2SA2603 with?
Voltage and current in normal cases shouldn't be high. KSA1015/KSC1815?
 
JoseHH said:
Yes, KSA1015/KSC1815 would work in my opinion. If you can find these KEC parts KTC3199/KTA1267, or KTC3198/KTA1266 or KTC3227/KTA1274 would be good alternatives. Or even KSA733C/KSC945C if you happen to have them.
Click to expand...
I replaced them all since I had enough KSA1015s and now I see they are EOL and not available. I have some KSA733s but they are also EOL so need to find another future solution.
There's still another issue on the left channel somewhere from kQ54 and forward but have been to busy with work to find it.
 
Back to this one. I had some issues with the pre-amp so I redid that one. Now all four DC offset settings for the pre-amp works.
The HOT-GND also works for the driver amp but COLD-GND does not.
Spinning my wheels on that one.
Right side is 1.8V and not adjustable, left side is 27V!
The plus/minus voltages for the kVR2 rail is fine.
The voltages over the kD51-kD54 diodes are also off on left side by an equal amount. I don't even understand how it's possible ...
See picture.
 

Attachments

  • 20260504_212616714_iOS.jpg
    20260504_212616714_iOS.jpg
    148 KB · Views: 9
Last edited:
Magnus Back said:
Back to this one. I had some issues with the pre-amp so I redid that one. Now all four DC offset settings for the pre-amp works.
The HOT-GND also works for the driver amp but COLD-GND does not.
Spinning my wheels on that one.
Right side is 1.8V and not adjustable, left side is 27V!
The plus/minus voltages for the kVR2 rail is fine.
The voltages over the kD51-kD54 diodes are also off on left side by an equal amount. I don't even understand how it's possible ...
See picture.
Click to expand...
They're all replaced, see table. And checked again (green mark) ...
It's almost as if there is something swapped but I can't find it.

1777934962391.png
 

Attachments

  • 1777933647320.png
    1777933647320.png
    25.6 KB · Views: 5
Last edited:
Ok those problems are solved.
A number of resistors were a bit out of spec, some a lot. After replacing these all DC and bias settings are now adjustable and stable.
The COLD-GND is finicky to set but stable once set.

However it still won't come out of protect and the light isn't flashing initially so something is most likely bad on the protect board.
AC 30V in works but no voltage at lQ1&lQ2. mD1, mD2, mD5 are good.
A number of broken components. mR1, mQ1, mQ2, mDZ1 found broken so far. Still not working.
 
Last edited:
Plimpington2 said:
Good lord! Did you put this table together yourself?
Click to expand...
I make an Excel for every amp I work on. Prepare it beforehand. By now I have one for a lot of models of course and just copy.
Over time I add in things I see broken so I check those on the next one etc.
 
Both Q1 and Q2 were broken as well.
Once replaced the "flashing" works but no joy of it coming out of protect.
Trying to understand how the TA7317 and circuit is supposed to work now.
It seems there is no driving voltage on pin 9, only 50mV.
DC in and current in seems fine.
 
Can't find any other broken components so I'm assuming the TA7317P is broken.
Ordered from eBay, hoping for a real one ...
 
You must log in or register to reply here.
Back
Top Bottom