• 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 AU-217 restore

TupolevFan

Super Member
Good day everyone. A week ago I was passing by the yard sale and saw a dusty and a bit beaten up Sansui AU - 217. Owner asked 15 USD and I could not resist and picked it up. Came home powered the unit up and to my surprize relay clicked. I measured DC offset and one channel was almost perfect 3.2mV and the other was 135mV and jumping a bit around. I turned it off and disassembled. Made a list of Parts and caps for replacement + relay. Order arrived last Friday. I removed all more or less large caps from the main board and gave it a isopropil bath and some soap water with brash and then hair dried it :). The component side came out pretty clean but from the "solder" side there were some white traces from the isopropyl alcohol. Not sure what to use to remove them.
I started to replace the components(capps and resistors) and it was going OK until i lifted a trace and "string" of curses went out of my mouth. :)))) Then it happened couple of more times until i realized what is going on. I usually use a stainless steel needle to clean the whole from the solder (briefly hitting it up and sticking the needle in and pulling it out). The issue was that this time my needle was not stainless steel and the solder was sticking to it a bit better than i expected. So i was a bit upset with myself since reinforcing the trace does not look as nice as if I did not have to do it and extra operation wastes time.
I found all 220 Ohm(R39,R41, R40, R42) resistors out of spec the measurements were all over from 270 to 380 Ohm. Replaced them with 1/2 W 220 1% metal films. Also in a channel where DC offset 135mV i found that R35 was out of specs(82 Ohm was measuring 400+ Ohm). In the beginning I thought that high DC offset was most likely because of faulty differential pair 2SA798 and now after finding this resistor almost 5 times of it's value i am not so sure. Because the negative rail would be affected. I replaced all four 82 Ohms resistors. Plan to replace all small transistors with their ON Semmi equivalents. Not sure if i have to touch 2SA798 or do you guys suggest to replace it anyway? Output stage will be original as it seems like original are in good shape. replace dual diodes VD1212 with the pair of 1N4148 (if i remember the part# correctly). I almost completed to recap main board. Just was not happy with myself about lifted/damaged traces. There are R43,R44,R45,R46 6.8 Ohm resistors (they masure OK) should I replace those as well as they looked like those 220 Ohm resistors. If you could think about any other failure prone component in this amp please let me know.
Will Post some pics later
 
Last edited:
Register to hide this ad
Sounds like you did, but replace all fusable resistors and make sure all glue is gone. Glue on some diodes had killed my 217 before I restored it.
 
Looks like a more elegant solution with factory matching. Let us know how it goes. Pics will be nice once you get her fixed.
 
Looks like a more elegant solution with factory matching. Let us know how it goes. Pics will be nice once you get her fixed.
You are right about factory matching but i am not so sure if it is more elegant since legs bending would still be involved. I used this in two technics amps (SU-8600 and SU-7600) to replace 2SA798 and results were fine.

One needs to watch the gain because for 2SA798 min is 250 but for this one i think min is 200 but typical 290. the chips i installed had gains from 270 to 290.
 
Question regarding transistor subs. for a bias circuit transistor (TR11 and TR12) 2SC711 i will replace it with KSC945C Or i can do KSC1845. it seems like pinout for 711 is ecb and for 945 or 1845 is bce which leads to the flipped position. And this would not be an issue unless flat part of the to-92 would need to face a heatsink. Any issues because transistor would be facing the heatsink with the rounded side? Whould't it effect the heat transfer? Yes there is a metal clip. Any thoughts?
 
Any issues because transistor would be facing the heatsink with the rounded side? Whould't it effect the heat transfer? Yes there is a metal clip. Any thoughts?
Yes, it would affect the heat transfer.
 
Last edited:
Here some pics of main board. I almost finished the recap + replaced resistors Transistors and prepared replacement for 2SA798.
 

Attachments

  • 1.jpg
    1.jpg
    61.4 KB · Views: 114
  • 20190814_163927.jpg
    20190814_163927.jpg
    56.5 KB · Views: 114
  • 20190814_223437.jpg
    20190814_223437.jpg
    161.6 KB · Views: 121
  • 20190814_223444.jpg
    20190814_223444.jpg
    139.8 KB · Views: 116
  • 20190814_223449.jpg
    20190814_223449.jpg
    129.9 KB · Views: 113
  • 20190814_223451.jpg
    20190814_223451.jpg
    134.7 KB · Views: 112
  • 20190814_223501.jpg
    20190814_223501.jpg
    132.3 KB · Views: 113
what would be the suggestion? 1 of the transistors has very corroded legs. And also i am replacing it just to be thorough. that is why it is rebuild.
Use a suitable replacement with the same pinout, KSC945C would be ok I think, mounting a different replacement with the rounded side of the case to the heatsink is a bad idea.
 
Use a suitable replacement with the same pinout, KSC945C would be ok I think, mounting a different replacement with the rounded side of the case to the heatsink is a bad idea.
That is exactly the issue. Both KSC1845 and KSC945c have flipped pinouts compare to 2SC711
 
What i will probably do is cut and flip the traces on the board - it is the easiest solution and not so hard to do. a bit ugly but looks like i have no choice
 
That is exactly the issue. Both KSC1845 and KSC945c have flipped pinouts compare to 2SC711
Ah yes, I hadn't fully understood the problem, I see now. Personally, I would be inclined to cross the leads - even if this means extending them slightly, but whatever works for you is the one to go for. ;)
 
Cut and flip traces?? Just leave the transistor alone......it's fine......

I'm all for being through but that's just a bad idea.
 
Cut and flip traces?? Just leave the transistor alone......it's fine......

I'm all for being through but that's just a bad idea.
I understand that leaving it alone is an option for this particular case. However what would one do if this transistor needed a replacement. I also do not like bent legs or flipped traces.
 
Do you have the gain specs on the original 2SC711's? You could use a KSC3503 or KSC2690 (overkill) and mount them on their faces, you may have to drill the solder pads with a 1mm bit (from solder side) to allow for the larger leads on the to-126's. The gain on the KSC3503 may be lower than the original 2SC711's because you can only buy these in the 'D' suffix.
 
Do you have the gain specs on the original 2SC711's? You could use a KSC3503 or KSC2690 (overkill) and mount them on their faces, you may have to drill the solder pads with a 1mm bit (from solder side) to allow for the larger leads on the to-126's. The gain on the KSC3503 may be lower than the original 2SC711's because you can only buy these in the 'D' suffix.
Thanks for the suggestions.

Some notes regarding 2SA798. After experimenting with different adapters, it was still too time consuming to put all of those jumpers and bent legs and it looks a bit ugly, i decided to use easyEDA program and designed my own adapter with correct pinout and ordered 20 of them for the price of 15 USD ( this includes shipping) if i order 20 dual Nexperia chips for 0.40USD each + shipping i would get 1.50 per 2SA798 replacement
 
Back
Top Bottom