Trying to bring Sansui AU-517 back to life

Dahond

New Member
Hi,

I'm from Belgium. I've been reading through this forum for a while trying to revive my sansui AU-517.
I've already found some great tips here but now it seems I'm stuck.

The amp was given to me a few years ago, including a sansui turntable and sansui speakers.
I worked great for a few weeks when one day, playing a record, it started smoking and wouldn't come out of protection.
2 weeks ago I've finally had some time to start repairing it.
It was a shorted diode on the powerboard and blown fuses in the left channel.
Fixed that easily, started it back up but it wouldn't come out of protection...

Long story short:
I'm not a technician, I just took the schematic and started measuring and shorted an output transistor doing that :-/
I've changed the output and driver transistors, did a recap of the of the driverboards and the powerboard.
Changed all transistors in the protection circuit, including TR09 on the driver board. Changed out some shorted resistors.
Every component I now measure seems to be within specs.

So here's the problem:

If I use headphones (64 Ohm) it works and sounds great although sometimes at low volume a channel turns scratchy and lowers in volume. Most of the time the right channel, sometimes the left. This can be fixed by turning up the volume one step and then I can lower it again, sounding great again.
If I use speakers (8 Ohm) it immediately clicks back in protection mode. I've tested different speakers.
If I disable the protection circuit it sounds great with my speakers although sometimes the scratchy lower volume random channel thing happens too with speakers but it is fixed with the volume one step up and back down again.
It is also fixed by simply touching the signal coming from the driver board to the protection relay with the probe of my multimeter.
If I listen to the preamp on the preamp out on the back it sounds great, no scratching or lower volumes.

I've read some stuff about the fuse resistors and the diode VD1212 but I didn't replace them because they measured ok.
Could they start misbehaving under load?

I don't know where to go from here. I'm starting to think I should just disable the protection circuit.

Guess I'm looking for some direction on what to check/measure/...
Any help would be very much appreciated. I quite love this amp.

EDIT:
Bias = +-20mV left and right
DC offset = +-1mV left and right
 
Last edited:
oh man, do I love my au-517 !! I think I would prefer it and it's build quality over the AU-719
 
and sorry should read more of the OP but I'm distracted with the saints game :)

I'm sure this is a common fault you are aware of, but use typical RCA jacks and hardwire the preout to main amp input on the rear of the amplifier panel. Lost of corrosion and loos of contact occure frequently in this particular area.....jiggle the switch an lost channels seems to come back to life. I just got tired of wondering everytime I noticed something was off and just wired the preouts into the main-ins via RCA audio jacks.....haven't had an issue yet :rockon: or since then...)knock on wood)
 
I expereinced all the issues you were describing and vanished after another karma member suggested this fix
 
Thanks for the reply.
I bridged the switch and even tried the poweramp with a different preamp. It still goes in protection with speakers.

I can't find the cause of why it works with headphones but not with speakers. 64 ohm <-> 8 ohm.
 
Thanks for the reply.
I bridged the switch and even tried the poweramp with a different preamp. It still goes in protection with speakers.

I can't find the cause of why it works with headphones but not with speakers. 64 ohm <-> 8 ohm.
Because there is an imbalance somewhere caused when there is a load on the output. When it is idling, or when you are running your HiZ Headphones, there is no load on the amp.
As soon as there is load, then something is restricting current somewhere, it could be an emitter resistor, or an emitter resistor on the drivers, base resistors on the outputs or drivers.
You will have to go through and meticulously measure everything and make sure its in spec.
 
I've read some stuff about the fuse resistors and the diode VD1212 but I didn't replace them because they measured ok. Could they start misbehaving under load?
I strongly suggest you DO replace the fuse resistors, they are so flakey they may have failed since you checked them.

Not quite so important with the VD1212's because they fail less frequently, and the faults they cause are usually more obvious. People change them because they fail just often enough to be a bloody nuisance. ;)
 
Last edited:
hope you get her fixed....I'm never going to get rid of my 517....no matter whatever preamp I seem to purchase to in order to take the preamp of the 517 out of a bedroom or den system it turns out that she continues to best all the preamps I try and substitute.....and the amp section is worth it's weight in gold too
 
I strongly suggest you DO replace the fuse resistors, they are so flakey they may have failed since you checked them.

Not quite so important with the VD1212's because they fail less frequently, and the faults they cause are usually more obvious. People change them because they fail just often enough to be a bloody nuisance. ;)

Yes, and this is what I feel the issue might be, if these resistors have drifted even slightly, when you try to pull some current through them, they go even higher Ω and put the system out of balance, boom, lots of DC offset and the protection kicks in...
 
I really would like to get this fixed. I really like the sound through the sansui SP3500 speakers.

Don't know how to quote so I'll @ for reply :)

@slimecity : the problem occurs in both channels

@Hyperion : The fuse resistors seem to be changed already, they look like ordinary brown colored resistors. R23 and R25 on F2722 and F2721.

@kevzep :
I've been measuring voltages since your reply. Was afraid to do so since the first time I tried I shorted the output transistors. This is what I found:

Voltages at the zenerdiodes on F2721 and F2722: ZD01 and ZD02 are 24,3v and -20,5v while they should be 22,5v and -22,5v on the right channel.
Same kind of readings on the left channel.
I suppose this is not ok.
R04 and R10 measure within tolerance so I think the zenerdiodes are bad.

Are they 1W 22v? Is this info enough for my local component supplier?

Many thanks already :))
 
Have the output transistor emitter resistors been measured and checked, I ask, as you say you blew some output transistors.
 
Yes. Output resistors R44 and R45 have been replaced together wit the output transistors.
I've checked every resistor on the boards (maybe even four times :)) and replaced everything that seemed dodgy.

Do you think the difference in voltage at the zenerdiodes could activate protection circuit?
 
Do you think the difference in voltage at the zenerdiodes could activate protection circuit?
Yes, that will play havoc with the DC conditions of the amplifier, you need to investigate why both channels and all 4 zeners are not functioning correctly, something tells me there is an elephant in the room. :idea:
 
Not sure what issues would be caused by the VD1212 if they acted up - how did you measure these?

I have read enough about these to outright replace them in every sansui I've had - its simple to do.
 
I used the diode test on my multimeter for the VD1212. As I said I'm not a skilled technician so everytime I suspect a component I search the web how to check it.
Before I started ressurecting this amp I hardly knew anything about transistors and diodes. I only knew how to follow schematics building guitar effects.
I've learned much this last few weeks.

@Hyperion
The supplyvoltage on the zeners is 50,8v and -50,8v. Service manual says it should be 51v.
Could the zeners have drifted due to age?
Or do you think the elephant in the room is something drawing too much current taking the zeners of by +2 volts and the zeners are probably still within spec?
I'm starting to think this goes beyond my ability to comprehend electronics.
 
Could the zeners have drifted due to age?
I would find that hard to believe, all you can do is experimentally change them, and see what happens. You say you have checked everything else and that ~50V is plenty enough to get the right voltages. You are correct, 22V 1W is the right component spec' - don't be tempted to put in a higher (or lower) wattage component as this will have a detrimental effect.

By 'elephant in the room' I was thinking of something like the mains voltage selector set incorrectly? or more likely some wrong value, or wrong orientation replacement components installed. It wouldn't be the first time someone had misread the resistor colour code and installed resistors wrong in value by a factor of 10 for example.

Personally I would like to see a picture of one or both of your driver boards, good sized clear pictures if you can. ;)
 
Last edited:
So.
I've measured the voltages again this morning after the last reply.
+ and - 50,8v supply
Zenerdiodes
Left channel
ZD01 21,82v
ZD02 -23,16v

Right channel
ZD01 22,45v
ZD02 -23,09v

They changed a little. Didn't do anything to it :-/

Right and left:
20181011_103541.jpg 20181011_103522.jpg
 
Back
Top Bottom