• 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

Well that was weird... AU-719 problem.

Hyperion

Hyperion

Roobarb & Custard
Subscriber
I have just been bench testing an AU-719 that I reconditioned for an AK member a couple of years ago.

He complained of losing a channel and at first I couldn't find anything wrong, then I found something, if I tapped the left channel (rearmost) Driver PCB it would burst into oscillation, and go into protection mode. I tried this a couple of times, but I made a fatal mistake - I continued reproducing the fault without using a DBT in circuit, and true to form, disaster struck, 2 fuses blown, +2 OP transistors and 2 driver transistors. :( - I was pretty annoyed with myself for that, but was a bit distracted with this particular fault and forgot the basics. All the OP transistor junctions for that channel were blown short circuit, and most of the drivers.

Anyway, today I replaced the fuses & driver transistors and put new OP transistors in, powered it up on DBT, all good, removed the DBT and set DC offset and bias no problem. So I let it soak for a while, to check for DC offset and or bias drift, all good, then turned it off and left it for a while. I came back to it later, but this time I put it on DBT. Fault was still there, one tap on the driver board, and it bursts into oscillation, triggers protection, and the DBT bulb brightens considerably. I have some freeze spray and try all the usual suspects - nothing doing. Then I try prodding a few things - still nothing until I tapped C16 very lightly, (470µF 80V) - and all hell broke loose ! - so I very quickly powered off. Powered on again, and still oscillating, powered off quickly. So I thought "surely not C16" ??? - anyway, out came the driver assembly and I changed C16 - re-assembled, powered on (on DBT of course) - with my hand on the power switch, I tap the driver board and...... clean as a whistle, more tapping, no response, absolutely fine now. Removed the DBT, powered on - more tapping, still fine, conclusion? a faulty electrolytic capacitor !!! (you might think, 'solder joint'? - not a chance, the joints were spot-on).

I tested the capacitor on my Peak Tester, and also on a multimeter, and I can't get it to show any problem at all, It was a brand new capacitor fitted by me.

I can see why it caused such havoc, it's in a crucial spot, any disruption there (random charging/discharging, becoming resistive etc) and the amplifier stage will be all over the place.

I hope the description above might help someone with a similar problem. Or for other problems, just the general fault finding approach, and how getting distracted can have disastrous results. :)
 
Last edited:
Register to hide this ad
ghazzer said:
Any thoughts on the possible failure mode(s) of this cap that would trigger the oscillations?
Click to expand...

Just a guess, intermittant internal connection, maybe causing it to be open circuit, or very low value, or with a really significant series resistance. Probably desoldering it from the board has remade the internal connection for a while.
 
Last edited:
Overundr1 said:
Lead spacing perhaps stress the cap internally, strange :(
Click to expand...
Lead spacing exactly the same as the originals, @ConradH taught me the importance of this. The capacitor was/is mounted tight to the PCB, and has a rubber pad at the bottom to ease stress on the leads.

It is just possible that the issue was not the capacitor but very close by, and changing the capacitor has disturbed whatever it was sufficiently to make it go away. I really doubt that as when I touched the capacitor it really was the lightest of touches, but had the same effect as a more remote tap on the PCB. Prior to changing the suspect capacitor I inspected the solder joints, and all the others in the surrounding area, and the rest of the board with a watchmakers eyeglass.
 
Last edited:
Glad you got it sorted out! As you know--brand new components does not = perfect and up to spec--you get a "dud" every here and there. Again--congrats on your "retirement"--working harder than ever :rflmao:
 
Thanks John.
Before retirement I worked for a major aerospace corporation and failures like this were routinely sent to the Failure Analysis Lab (FAL). They had a myriad of fancy (& expensive) gear to dissect and analyze the part - mainly to see if it was a design/manufacturing problem or it may have been a secondary failure. Always interesting to read those reports, and hope that the feedback would improve products.
 
Thinking from aged memory here, if that cap had a microscopic trace break would that not kick your supply voltage to full -or + on that leg of the driver board? If so yes the drivers would quickly become very unhappy with the idea of continued existence :)
 
Long shot.. I have had 2 ghost issues which came down to the "PAD" where it joined to the trace right at the solder mask had FAILED....putting a hairline crack between the pad & trace....
 
Thanks guys, but no solder pad or broken track problems here, however work is temporarily halted as I need TO-3P insulating washers. I was replacing the OP transistors in the other channel and found a damaged one, new ones ordered and on their way. ;)
 
vigman said:
:banana:YEAH !!!!!! Welcome to the GRUMPY OL MAN CLUB !!!!!! :banana:
Click to expand...

Yeah, I think I joined a couple of years ago, I began to get feelings that I wanted to write to a lot of companies to tell them what I thought of their useless so called 'Customer Service' or whatever. Now... I have the time to do it !!! :deal: :D

This years score so far, is around £170 returned to me as compensation for useless Customer Service - they ain't seen nothin yet.... :)
 
Last edited:
Hyperion said:
Yeah, I think I joined a couple of years ago, I began to get feelings that I wanted to write to a lot of companies to tell them what I thought of their useless so called 'Customer Service' or whatever. Now... I have the time to do it !!! :deal: :D

This years score so far, is around £170 returned to me as compensation for useless Customer Service - they ain't seen nothin yet.... :)
Click to expand...

I'll have to remember that when I call it quits or at least slow down--there are so many missed opportunities to recover (financially) from poor products or customer service--I just don't take the time to pursue them. If "the hit" is for under $100--it would cost me more than that to take the time to e-mail back and forth every day for a month, or spend two hours on hold waiting to get to someone (if you can even get a real human on the phone) to deal with the issue.
 
Hyperion said:
I have just been bench testing an AU-719 that I reconditioned for an AK member a couple of years ago.

He complained of losing a channel and at first I couldn't find anything wrong, then I found something, if I tapped the left channel (rearmost) Driver PCB it would burst into oscillation, and go into protection mode. I tried this a couple of times, but I made a fatal mistake - I continued reproducing the fault without using a DBT in circuit, and true to form, disaster struck, 2 fuses blown, +2 OP transistors and 2 driver transistors. :( - I was pretty annoyed with myself for that, but was a bit distracted with this particular fault and forgot the basics. All the OP transistor junctions for that channel were blown short circuit, and most of the drivers.

Anyway, today I replaced the fuses & driver transistors and put new OP transistors in, powered it up on DBT, all good, removed the DBT and set DC offset and bias no problem. So I let it soak for a while, to check for DC offset and or bias drift, all good, then turned it off and left it for a while. I came back to it later, but this time I put it on DBT. Fault was still there, one tap on the driver board, and it bursts into oscillation, triggers protection, and the DBT bulb brightens considerably. I have some freeze spray and try all the usual suspects - nothing doing. Then I try prodding a few things - still nothing until I tapped C16 very lightly, (470µF 80V) - and all hell broke loose ! - so I very quickly powered off. Powered on again, and still oscillating, powered off quickly. So I thought "surely not C16" ??? - anyway, out came the driver assembly and I changed C16 - re-assembled, powered on (on DBT of course) - with my hand on the power switch, I tap the driver board and...... clean as a whistle, more tapping, no response, absolutely fine now. Removed the DBT, powered on - more tapping, still fine, conclusion? a faulty electrolytic capacitor !!! (you might think, 'solder joint'? - not a chance, the joints were spot-on).

I tested the capacitor on my Peak Tester, and also on a multimeter, and I can't get it to show any problem at all, It was a brand new capacitor fitted by me.

I can see why it caused such havoc, it's in a crucial spot, any disruption there (random charging/discharging, becoming resistive etc) and the amplifier stage will be all over the place.

I hope the description above might help someone with a similar problem. Or for other problems, just the general fault finding approach, and how getting distracted can have disastrous results. :)
Click to expand...
I also use the tapping method with good results usually. .
 
Hyperion said:
Lead spacing exactly the same as the originals, @ConradH taught me the importance of this. The capacitor was/is mounted tight to the PCB, and has a rubber pad at the bottom to ease stress on the leads.

It is just possible that the issue was not the capacitor but very close by, and changing the capacitor has disturbed whatever it was sufficiently to make it go away. I really doubt that as when I touched the capacitor it really was the lightest of touches, but had the same effect as a more remote tap on the PCB. Prior to changing the suspect capacitor I inspected the solder joints, and all the others in the surrounding area, and the rest of the board with a watchmakers eyeglass.
Click to expand...
Plastic sleeve 180° out .
 
You must log in or register to reply here.
Back
Top Bottom