Sansui G-3500 No Output

[sol7]

Hello. I just recently picked up a G-3500 receiver. I tested it before purchase and there was no output to the speakers. Complete silence. I bought it on the notion that it was blown fuses. I get it home and pull the cover. Yup, blown fuses. Ran to RS and picked up some, stuck them in and powered on. About 10 secs later I see a nice flash. Both fuses to the amp section blew.

What's the easiest route to diagnosing the fault? I can't check voltages since the fuses blow. I don't see any burnt components. No bulging/leaky caps. Power from the transformer is good. All lights work, Tuner works (as far as meter deflection and stereo light are concerned). It's not a high value receiver, so taking it somewhere is not economically viable.

Worse case, I part it out?

 

[nosirrah]

To start with get the service manual...try Hifiengine, free there.
Pull the output transistors, Google "transistor testing" and test them, they really cannot be tested in circuit.
Post your results...

 

[sol7]

Ok, I pulled one output transistor. 2SA771. Using a DMM. Got these reading:

B-C 1.6ohm
B-E 1.5ohm
C-E .1ohm

Analog Meter:

B-C 2ohm
B-E 2ohm
C-E .5ohm

Readings are the same when switching leads. If I'm doing this right, that transistor is bad. Right?

 

[avionic]

Ok, I pulled one output transistor. 2SA771. Using a DMM. Got these reading:

B-C 1.6ohm
B-E 1.5ohm
C-E .1ohm

Analog Meter:

B-C 2ohm
B-E 2ohm
C-E .5ohm

Readings are the same when switching leads. If I'm doing this right, that transistor is bad. Right?

Bigtime shorted:yes:

Better check the rest of them.

 

[sol7]

What would cause them to crap out? Pushing speakers below the rated ohms?

Google only seems to show one reseller of that transistor. eBay shows some listings. All seem to be generic. Mouser and DigiKey show nothing. Any other sources that might have them or do I go generic?

 

[kevzep]

What would cause them to crap out? Pushing speakers below the rated ohms?

Google only seems to show one reseller of that transistor. eBay shows some listings. All seem to be generic. Mouser and DigiKey show nothing. Any other sources that might have them or do I go generic?

Speaker outputs probably shorted out at some stage......or a driver transistor may have died and shorted out taking the outputs with it......could be any number of things.......

There will be modern equivalents you can use, the original transistors will be long out of production......

What are the transistors? 2SA??? 2SC????

 

[sol7]

Tested one 2sa771 and 2sc1986 on system B side. Both bad. I'm sure the other channel will be the same since fuses for both channels are blowing.

Pulled TR16/18 so far.

 

[nosirrah]

There will be equivalents you can use, but it would be good to know what was the problem....
It's possible that new output transistors will fix it, but if they don't, and the original issue still exists, then the new ones will go as soon as you power up.
If they are inexpensive I would be tempted to try it.
I'm thinking that someone hooked it up wrong {to speakers} and shorted it out, the fuses didn't save the outputs, because both channels went.
But I am no expert, there are some good techs here, they may chime in on this...

 

[Hyperion]

I would remove and check the driver transistors, if any show suspect/different readings replace them. Also check R73/R75 and R74/R76, and R78/R80 and R77/R79.

Put the amp back together again MINUS the output transistors and check the ability to set the DC offset. Measure this at the junction of R71/R69 and R70/R72, if this works OK, fit the outputs and bring it up via a Dim Bulb Tester (or variac) - preferably one channel at a time, measuring bias current on each channel as you do this.

If this tests OK, take out the DBT and try it, again measuring bias current, and ability to set it - again preferably one channel at a time. Check/set the DC offset, and if both bias readings and both offset readings are stable then you should be good to give it a listening test.

Maybe a bit overcautious for some - but this is what I would do.

 

[sol7]

So are the fuses blowing because of the shorted outputs? If so, what if I power up with new fuses and minus the output transistors. Would I be able to verify the voltages on the schematic or are they dependent on the outputs?

My thing is, I don't want to put $50 in parts into a $40 piece, figuratively speaking.

 

[Hyperion]

Apologies - what I should have clarified was...

"if this works OK, fit the New outputs and bring it up via a Dim Bulb Tester (or variac) - preferably one channel at a time, measuring bias current on each channel as you do this"

And, yes if you remove the old output transistors, this will almost certainly stop the fuses blowing (unless you have a PSU fault) - so, with new fuses you will then be able to measure the voltages.

 

[sol7]

Apologies - what I should have clarified was...

"if this works OK, fit the New outputs and bring it up via a Dim Bulb Tester (or variac) - preferably one channel at a time, measuring bias current on each channel as you do this"

And, yes if you remove the old output transistors, this will almost certainly stop the fuses blowing (unless you have a PSU fault) - so, with new fuses you will then be able to measure the voltages.

I figured you meant new outputs.

Later today, I will try power without the outputs and see were I stand.

 

[sol7]

Ok, in the schematics referenced above, it only shows voltages for system B. Will they be the same as system A? Or only system B is tested and if all is well, leave it at that?

I pulled A & B outputs. Only 1 side was bad. With outputs removed, I put in some fuses and powered on to check voltages against reference. They were close. What is the tolerance or do they have to be dead on?

Also, I notice TR08 was very hot. I pulled it and TR07. Both tested good. Pulled TR14 and that tested good. And when poking around I noticed the voltage on IC02 bottom leg (as you look at schematic going to R54) was around 10v, while IC01 was around .6v. Pulled IC02 and it tested good. Haven't pulled IC01 yet.

 

[sol7]

Pulled IC01 and it tested good.

 

[dr*audio]

Test TR10. It is the bias transistor, located on or near the heatsink. Then test all the transistors on that channel.
If everything looks ok, with the output transistors not installed, turn it on and check the voltage on the output of the bad channel. It should be close to 0V, + or - 50mV max. If it reads higher, shut it down immediately and start checking transistors again, you missed something. Also check R70, R72, R74, R76, R78, R80.

 

[sregor]

First - as far as I can tell, your left channel is probably OK, so concentrate on the right channel or you'll end up making things worse before its over. (Assuming even transistor numbers is right channel) Also, pay attention to plus and minus when giving voltages. If components are getting hot, then do not leave on for long times when testing voltages (turn on - measure voltages - turn off. As far as how exact voltages it - it depends - a lot of times it is the difference between two voltages that is important - especially where base and emitter is involved.
10 volts on base of ic02 is definitely bad. Even if it tests good it may be bad, but other voltages should be way off also. If its + 10, then the other half should be comletely turned on, and the other side is turned on max, and you should have about a volt and a half or so less than the negative voltage, turning on tr08 hard and causing it to get hot, This is trying to pull the output negative, but something is pulling it positive. Most likely tr12 is shorted or leaky collector to emitter. BTW, thanks for posting the output segment of the amp - it makes assistance easier.

 

[sregor]

Information heads up - the fuses are not for each channel, but for the power supply, so one channel shorted outputs will cause both fuses to go.

One more warning on q10 (bias transistor. Watch out for ksc945 if it looks like it had been replaced. ksc945 and 2sa945 are electrically the same, but basing is different. ksc945 has base in the middle. If you are purchasing, look for ksc945C this one has collector in the middle, same as 2sc945.
Good luck with it. My money is on bad tr12 or 14 or ic02. Good luck.

 

[sol7]

Well, it looks like TR16, TR18, and TR12 are bad.

R77, R78, R79, R80 read about .5-.6 ohm and schematic shows .33. They are beyond 20%. Should I replace them? Haven't tested R69-72 yet.

I have been using this to test the transistors. Didn't realize I had it until the other day:

 

[sol7]

Disregard about R77-80. Bad connection on meter. R69-72 read good.

So where can I get replacements for the bad transistors? Digikey and Mouser show nothing. Can they cross reference if I call them? Or is there somewhere online where I can find cross references?

I did find this place for TR12:
http://www.goldmine-elec-products.com/prodinfo.asp?number=A10408

and this for TR16/18:
http://www.electronica-usa.com/Merchant2/merchant.mvc?Screen=SFNT&Store_Code=EU

I'd like to order from one place. Plus the first reference has min $10 order.

 

[sregor]

Good replacement for 2sd438 would be 2sd667 from elctronica (2sb560 use 2sb647) Same case and almost the same specs - same or better.