Digi 192 I/O troubleshooting thread

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

johnR

Well-known member
Joined
Feb 20, 2009
Messages
156
(This is a re-post from a thread originally in The Brewery. Hopefully someone will find it useful.)

I recently repaired a 192 I/O A-D extension card that had noisy inputs. After being on for a while a frying noise would start, along with random pops and bangs. It turned out to be the J506 constant current diodes on the other side of the board from the converter chips. There are two sets of these diodes, and it was the set nearest the converters (4 diodes per channel). Those AK5394 converter chips run hot and the heat had damaged the diodes. I strongly suspect this is a common problem with these cards.

Unfortunately the J506 is obsolete. I substituted 1N5301 constant current diodes, which are available from Mouser. They're expensive, but cheaper than repair by Avid. I didn't have a schematic, but I tracked down the noisy diodes with a can of freezer spray. The 1N5301 is in a DO-35 case instead of a 2 pin TO-92, but it fits and it does the job.
 
I mis-read what John had posted and only replaced two diodes per channel -- the closest two to the converters on the other side of the card.
This did save one channel though. I will go back in and replace the remaining diodes on the bad channels.

On a previous repair, I definitely had bad burned surface mount caps which were also replaced.

Before (removed first diode):
192_8.JPG



After replacing half of what I should have:
192_13.JPG



I will investigate the caps again as well...
 
Avid Fixed small SMT caps  on the card.

Burnt caps include C175,C215,C312,C170.  All on underside of I/O card with AK539 converters.

Note These burn caps were black charcoal and burnt the mounting pad right off the card but a small pass thru hole was soldered with small wire from cap.

The Caps that were replaced (not burn) look like the same SMT ceramic caps.  They are on C261,259,C282,C300,  also C258 C260, c281,c299  C180,C181,C179,C178 C223,C224,C222,C221.  The are part Of AK 539 chip Circuit . I cant tell the values on caps.
 

Attachments

  • Replaced caps.jpg
    Replaced caps.jpg
    90.5 KB
Hi, just curious how do you track a noisy diode with a freezer can? This may be useful knowledge to apply on other builds.
 
i think he is talking about a tech version of the "fire and ice" style  of kung fu.

-- a heat gun and a can of freeze spray are the weapons to wield against your foe.

often times you can isolate a noisy component by shooting the suspects,
one at a time, with  a freeze spray, most of the time the noise will fade when
you hit the offender.

as we know, heat related problems can be hard to re-create out of the rack, and
especially with the top off.
here we pull out  the heat gun and warm up areas of suspect.
now the heat intolerant components will begin to talk...
and bad solder joints and poor connections begin to reveal themselves as their
surroundings expand away from them.





hope that was right
-grasshopper
 
QUEEF BAG said:
i think he is talking about a tech version of the "fire and ice" style  of kung fu.

-- a heat gun and a can of freeze spray are the weapons to wield against your foe.
Yes. Obviously you have to be careful not to overheat things. If it's already running hot you probably don't want to heat it further.  When using freezer spray be careful not to use so much that the resulting condensation causes a short circuit. Don't use freezer on high voltage circuitry.

In this case the noise increased dramatically for a few seconds and then disappeared when one of the suspect diodes was sprayed with freezer. As it warmed up again there were some loud pops and bangs and the noise returned to its previous level.

Never do this with your monitors turned up loud.
 

Latest posts

Back
Top