Bad ADC chip?

I was going through an old audio interface to test it for sell-off and noticed I had some noisy channels.  Digging in I find that on certain ranges of input to the ADC, that the noise happens.  For instance, loop DAC to ADC, output sine wave at -54dbFS = clean, -53dbFS = noisy, -52dbFS = clean.  I can shift that window around by cooling the chip (which runs pretty warm to touch).

This leads me to believe that the ADC chip is bad.  The chip in question is the AK5394VS.

you always could fry an egg on those darn things.

check the clocks going in, as well as reference voltages.

Not sure what to look for on the clock signal...But the reference voltages and power rails are all clean.

$25 a chip....I guess I'll try one and see how it goes.  I've never re-worked a 28-pin SOIC, but I ain't skeeered. 

TheJames said:

Not sure what to look for on the clock signal...But the reference voltages and power rails are all clean.

Click to expand...

Verify 12.288 MHz or 24.576 MHz for the MCLK, 48 kHz for the LRCLK and 3.02 MHz for the BCLK.

You need a fast 'scope.

Just thinking out loud and I'm by no means a digital circuits/theory guy...

I can check the clocking, but since it's a multi-channel converter, I would assume the clock is the same across the ADC chips.  There would logically be some sort of clock distribution, to all of the ADC chips, but the base clock would be the same for all.  Since the problem only occurs with two of the ADC chips, and only one side of one of those chips, I would assume the clock is fine, but something specific to those two chips is at fault. 

If the digital clock frequency were unstable, when the input is a pure sine wave the observed signal (as reconverted through DAC) would be a sine with a wavering pitch similar to wow/flutter on an analog deck.  Perhaps more specifically, that would be the case with a varying clock.  A noisy clock would potentially generate random pops/clicks.

The noise that's experienced is like digital clicking (hash) that only occurs with certain input levels to the ADC.  If you change the physical temperature of the ADC chip ( like with a cold spray), you can shift the input level that the digital hash occurs at.  A sine wave input at say -20dbFS is perfectly fine, yet the same sine wave at -21dbFS has hash and the hash goes away at -22dbFS.

My gut says ADC chip.  I'm open to input though.  I've got a 30Mhz scope, I just don't know if I can find an "easy" clock point to stick the probe.

The converter in question is a MOTU HD192.  It uses six AK5394VS chips. The two that I'm having trouble with are in the middle of the board (inputs 5, 7 and 8).

I think the Assumption that the clock is the same across the chips is a "should" but not a "shall" condition....

Technician rule of thumb is to probe the actual clock pins* on each chip to verify...

(*the ACTUAL PINS, and not a circuit node nearby that may be a direct connect to the pin)

The fact that cooling the chip is interesting; used to use a lot of freeze spray and chiller stations... http://temptronic.com/MobileTemp/MobileTemp12.htm

I hope nothing has come unsoldered with the heat

It may be some sort of heat induced jitter internally to the chip where thresholds drift badly.  Or coupled with an ESD event, there may be something vastly out of tolerance inside the chip...  Used have many issues with ESD, self heating and the like for analog chips we used to design in the 2GHz region...

A replacement of the chips may be in order, unless the clock distribution is not sending the chip a proper clock...

Another thought might be electrolytics on any reference voltage pins and noisy voltages... probing each pin and comparing to known good channels might be an approach...

Check it out... curious indeed... 

Check the components (and solder connections) around the ADC... a bad or poorly connected decoupling cap might be enough to alter timing/logic thresholds and introduce noise... Of course damaged chip is always a real possibility, but with  expensive parts it's worth checking the cheap parts first.

JR

Digidesign/Avid 192 interfaces also use this converter.  I've had noise problems similar to the one you describe, as well as more general snaps and pops (similar to the noise you get when the clock is incorrect).  In each case, replacing the chip solved the problem.  I had to invest in some SMT rework tools, but they have more than paid for themselves at this point in saved downtime and repair costs.

TheJames said:

  I've got a 30Mhz scope, I just don't know if I can find an "easy" clock point to stick the probe.

Click to expand...

You won't see a 24.576 MHz clock with that. You need something much faster.

-a

Well...Hot air rework station gets here on Tuesday, and replacement ADC chips and some caps show up on Wednesday.

I guess we'll see if that straightens things out.

Verdict:

Bad ADC chip.

Replaced the ADC chip for channels 7 and 8, they're great now.  Sadly, it appears other channels are acting up the same way.

Another order to Digikey and I should be all good on this guy.

(Anyone want a modded MOTU192 with fresh ADC chips?)

Sometimes it is the obvious problem... Some ICs are less reliable than others.. I recall one old codec that was very temperature sensitive.

JR