AU-517 - one noisy channel

[Hyperion]

A story, once upon a time there was an old amplifier...

An AU-517 which I have been reconditioning for an AK member, this is the first time I have reconditioned an amplifier without any intermediate listening tests to see if everything is as it should be. Of course I used a DBT after every major section of work, took voltage readings, checked things with an oscilloscope and so on. This approach was chosen because of the rather poor job that had been made of mounting new binding posts. The posts were a bit of a liability if I used them to hook up my speakers, as they were spinning because they couldn't be tightened up enough - the rest I'll leave to your imagination. (I guess I could have used the old spring clip terminals for the 'B' output - but who does that? )

So, today I completed remounting the new binding posts and was able for the first time to listen to the fruits of my labours. At first, very nice - then a rustling and popping sound from the right speaker. Off to the bench it goes and the oscilloscope shows the issue very nicely. Sudden short jumps in DC offset, - like a square wave almost, and some very unstable looking spikes, all in all a very jittery output. The usual elimination is performed (connected-separated switch ) and it is the power amp section that is the source of the noise. So I am thinking I really hope it isn't the 2SK97... and I resolve to hit the easy targets first, namely the transistors. I change the two 2SA750's (TR03/TR04) first, one had an hfe of 360 and the other one 512 - I change them for a much better matched pair of KSA992's - hook it up again (still on the bench) no cure. So I try giving the power amp heatsink some sharp taps with the blunt end of a screwdriver, and also the PCB itself, just a glimmer of response - something happening there, too faint to be useful. Then I put my finger on each one of the transistors in turn, giving a little push to each one, and BINGO ! a big reaction from TR02 (2SC1400) I stop, and it quietens down a bit, another wiggle and it goes crazy, I inspect the solder joints which are OK.

So I change TR02 and TR01 for a gain matched pair of KSC1845's - hook it all up again - jittery output is gone... I left it on the bench for a while to warm up, beautifully quiet now. A reset of DC Offset and Bias, and connect the amplifier back into my system, very nice sounding amp, and both channels quiet when they should be. The faulty transistor is another one of those with 'socketed legs' - one or more of the legs moving independently from the rest of the transistor.

So that is the tale of a very merry old amplifier, whose resistors are now 'resistoring', transistors are 'transistoring' and capacitors... well they're doing their thing too. - I am sure they will all live happily ever after together.

The End.

 

[Multimode]

Nice bedtime story!

You say you didn't carry out any intermediate listening tests, but did you do any before/after listening tests? If so, what was the difference?

The difference with the AU-919 that you refurbished for me is definitely 'chalk and cheese'.

 

[Hyperion]

I didn't listen to it at all before, but usually the sound might be described as 'dull, lifeless, lacking in presence, with poor dynamics'. My own AU-717 was like this, the better appointed relative of the AU-517, it was my first recondition and the first listen was pretty a emotional one, it sounded absolutely breathtaking to me.

Glad you liked the bedtime story, it's been a bit quiet on here lately so I thought I would have a little fun, and be informative (troubleshooting method) at the same time.

 

[Multimode]

... usually the sound might be described as 'dull, lifeless, lacking in presence, with poor dynamics'...

That describes very well my 919 which, you will recall, you didn't listen to prior to refurbishment as you didn't want to risk the possibility of black flag capacitor failure killing the output transistors.

I wonder how many other people drifted away from hi-fi listening due to a gradual and hence unnoticed deterioration in the sound quality. I use every excuse to get some listening time in now, like never before!

 

[Hyperion]

a gradual and hence unnoticed deterioration in the sound quality

This is quite often cited as the reason why people don't think there is anything wrong with their un-restored masterpiece. And rather perversely why the same people bemoan having been persuaded to have a restore/recap done saying that the sound quality post-restore has been decimated, most criticism levelled at lack of bass IIRC. I'll leave that one there as that's a whole different discussion, causing emotions to run high, very high in some cases.

 

[mikeyax]

This is quite often cited as the reason why people don't think there is anything wrong with their un-restored masterpiece. And rather perversely why the same people bemoan having been persuaded to have a restore/recap done saying that the sound quality post-restore has been decimated, most criticism levelled at lack of bass IIRC. I'll leave that one there as that's a whole different discussion, causing emotions to run high, very high in some cases.

Mine now sounds most excellent, thanks to you.

 

[Hyperion]

My pleasure Sir.

 

[Overundr1]

Funny it was TR02, had that one bite me too. In my case it was a solder break under the solder pad which was invisible even under magnification. If TR02 is not doing its job offset will suddenly swing raising the thoughts that the SK97 has gone belly up. From now on whenever I get any of that series in will remind myself to de-solder that transistor as a matter of course on both amp cards and very very carefully inspect the solder pad for cracks. Why that one is a nuisance evades me, however your findings are the second in recent chat lore. Well done sir.
Forgot to mention it was CDfixer that finally found it, much appreciated help from one of the grand master techs in the forum.
-Lee

 

[LBPete]

John, do I have that serial number?

- Pete

 

[Hyperion]

Almost certainly you don't have it, but it doesn't belong to me - it belongs to another AK member, I'll make a note of the S/No and ask the owner if he minds me passing it on to you. I can't think of any reason why he might not want you to have it, but it is not my place to say.

 

[ConradH]

All "electronic" problems are, at some level, ultimately mechanical.

 

[Hyperion]

Funny it was TR02

It is a little out-in-the-open on that board, and it wouldn't take much of an impact on the top of the transistor to crack the tracks underneath, which I know is not unknown in this game. I did inspect the joints and track very carefully, (watchmaker's eyeglass ) no sign of bad joint or cracked track. the only conclusion I could draw was bad transistor.

 

[LBPete]

it belongs to another AK member, I'll make a note of the S/No and ask the owner if he minds me passing it on to you. I can't think of any reason why he might not want you to have it, but it is not my place to say.

I understand, people may have reasons they don't want it public. At the same time it is tied to an AK username so it is record of ownership at that point in time.

- Pete

 

[ConradH]

I'm always reluctant to put serial numbers in public, but if I do I blank the last two digits- 123456xx.

 

[LBPete]

Conrad, the username is not published together with the serial number in the public report. The xx would be somewhat problematic. The database is set up not to accept duplicate serial numbers. The xx would create the possibility of 99 duplicate numbers.

- Pete

 

[Hyperion]

The owner AK member HCeline came and collected his amplifier today, after an audition he seemed very pleased and couldn't wait to get it home.