Converter broken - where to look?

My coveted but ancient Troisi 8 channel AD converter just broke. The converter's output, from one second to the next, dropped in volume and lost all the low end. This shows in the noise which has a scooped high end and no bass. What you put in comes out of the digital end, just a a much lower volume and sounding high pass (or maybe even band pass filtered).

I've checked all the voltages in the analog section and on the converter chips and they are all appear to be fine. The symptoms are exactly the same across all 8 channels. Any idea what this could be and where to look? Thanks!

Are the voltages the same during the issue???

The issue is permanent. The voltages are in the expected range, but I don't have technical documentation.

Since the problem is identical on all 8 channels my expectation was a power supply issue. Which would make sense, because of the unit is running very hot all the time.  It seems unlikely a power surge destroyed all of the input op amps in a similar way, doesn't it?

There's a small circuit section that seems to handle the LED indicating switched input sensitivity that appears to be not working, since the LEDs don't light up. Pushing an input sensitivity button I can hear the relais switch though. In that section there's a tantalum cap with no voltage between it. But it might just be signal filtering.

I checked all of the other tantalums and they all have the same ballpark voltages between them.

What kind of problem would the symptoms be compatible with?

I know I wish I could see some converter schematics.....

I have an old Mytek waiting for one day to be figured out,,,, It did have a bad relay but it was more of an intermittent thing with the different input selections....so they do go bad.....
I think there's some kind of ground issue with mine and or something bad with the clock crystal.... 
There are a few of those weird 3 leg filters that look like caps too that I suspect....I've been tempted to buy another just to figure it out......Ok digression....

Yes I suspected power supply because of the entirety of your issue too..... I wonder if it could still be suspect.....  I've never starved anything to the point where I've ever heard weird stuff so I'm more or less asking.....

That input sensitivity sounds suspect but I wouldn't know why the filtering thing would happen but, maybe it could with these things...

Totally way out of my element as with many things so sorry I can't be more helpful...

Hopefully you get it sorted......  I know there's a few guys on here that seem pretty knowledgeable about these things so maybe you'll get lucky......


edit///I guess I'll ask the obvious....... any changes to your setup prior to this happening???

Thanks Scott. I will recap it (not many electrolytics in there anyway), but the way the problem started - out of the blue after it was on for a few hours (no setup changes either),  so this doesn't suggest a recap will fix it.

I took out all four input op amps (AD797) for a pair of inputs and it made zero difference to the way the channels responded relative to the ones with op amps soldered in. Put in NE5534N - no difference either. The analog input stage does not appear to have anything to do with it.  I also do not think there's something wrong with the AD converter chips. My next supsicion would have been reference voltages, but those are fine, too.

The clock and the AES/EBU drivers are fine, otherwise there would be no stable digital signal coming out. Would four AD converter chips receiving an overvoltage display the exact same symptoms? Seems a little unlikely to me.

The unit seems to be build of 100% standard, fully labelled through-hole components. Nothing appears to be burned, so in theory by replacing the defective component it should easily be fixed...

Good luck with the Mytek, too!

I guess you could always look at the signal and see where it takes a dive????

Lucky you with the labeled components.....I guess it's top secret stuff going on in the Mytek with the scraped and sanded op amps ,no board values and cut traces with some jumpers.... It's an older one so maybe it was still in prototype..... I had another with the exact same thing....

Good luck! Please let us know how you get along.....

Check the obvious suspects that are cheap/easy to fix, like bad solder connections.

Lower level could be caused by a compromised upper address line but this is just a wild guess...

JR 

A bad solder connection seems unlikely given how the problem appeared. Where would the upper address line be located on this board?  8)

Here's a foto of the PCB:

http://www.dropbox.com/s/bk5ojh1aaef2emb/IMG_20180612_164803.jpg?dl=0

As you can see it's a tidy design. Any obvious culprit from looking at it? The only thing that looks visibly decayed are the EMI-filters.  But they measure fine (no short across the capacitive part).

Yes, I'm lucky to own this unit. It's a high quality, very well engineered no-nonsense piece of gear, the kind that's rare today. I like this 20 year old 20 bit converter better than the modern ones I have.

The only thing suspicious I could so far find ist that the chassis ground is connected to the 110V ground connector, but not to the circuit's grounds.

The led you mentioned shouldn't have anything to do with this should it??

I'm guessing you've chop- sticked everything  to death already???

Wow that is an old board! Very cool...

What is that at R40???

https://s20.postimg.cc/x15npxka5/IMG_20180612_164803.jpg

a hair...;p


never mind....looks like a mark...

scott2000 said:

What is that at R40???

https://s20.postimg.cc/x15npxka5/IMG_20180612_164803.jpg

a hair...;p


never mind....looks like a mark...

Click to expand...

Paint. No idea why it's there, but it's non-conducting. 

scott2000 said:

The led you mentioned shouldn't have anything to do with this should it??

I'm guessing you've chop- sticked everything  to death already???

Wow that is an old board! Very cool...

Click to expand...

I will get out the scope and trace the signal next.  Nothing else to try.

I know those transformer secondaries can get hokey too. I've seen the solid core crack.....Wouldn't make sense seeing as how your voltages are all good .....

Good luck!

scott2000 said:

I know those transformer secondaries can get hokey too. I've seen the solid core crack.....Wouldn't make sense seeing as how your voltages are all good .....

Good luck!

Click to expand...

Yes, with a broken transformer there shouldn't be any sound at all coming from the unit...

OK guys, I found the problem: I shouldn't be messing with electronics late at night. I somehow had a digital highpass filter switched on in the DAW by accident. The converter is working all right.  :

Sorry for bothering you and thank you for your time.

OK obvious suspect is try reseating the convertors in their sockets. Over time socket (mechanical) connections can become oxidized and simply reseating the ICs can improve marginal connections.

JR

Thanks John, but I found "where to look" (see my above post). 

living sounds said:

Thanks John, but I found "where to look" (see my above post). 

Click to expand...

ah.. pilot error

JR

Awesome! Can't tell you how many times it's been that component between the monitor and chair for me!