Protek 3023B Power Supply with bad display - fixed!

[Robomeds]

So I recently came across a Protek 3032B power supply.  These are reasonably nice bench units with two 30V, 3.2A outputs.  Each output has a volt and amp meter.  This particular supply was marked as defective with no display on the right side.  Well I took it home and decided to take a look.  The inside of the system is rather interesting.  I haven't really looked at one of these 2 channel supplies before.  It is 100% dual construction starting with the secondaries on the transformer.  The left and right power taps are separate windings on the transformer.  I guess it makes sense but I simply had never thought about it before. 

Anyway, as I was saying the right side didn't display showed nothing.  With a meter I found the voltage and current adjust still seemed to work just fine.  The voltage meters are kind of interesting as well.  It simply has 4 basic multimeter chips, one for each amp and/or current display.  At first I thought something might be wrong with the board with the displays.  Nope, I checked that by swapping the left and right displays.  The problem turned out to be the painfully simple 12V AC to 5V DC part of the amp board.  The bridge rectifier was bad.  I'm not sure I've ever heard of (not that I have extensive experience) a bridge rectifier failing.  I didn't have the correct replacement part so I had to cobble one together out of 4 diodes.  The 5 V linear regulator seemed good and the system is now up and running.

Free power supply... just needs rectifier

[SeanB]

Bridge rectifiers do fail especially if you have underspecced it or stressed the leads during assembly. If you have a peak current that is over the max limit, even though the average current is very low, it can eventually fail. If lucky it fails open circuit on one diode then the other diode goes open, in worst case a diode fails short circuit and burns the transformer out with the DC current through the core.

I had bad experience with diodes in power supplies failing short circuit and cooking the whole thing into a toasted blob.

[Robomeds]

I'm starting to think I might have spoke too quickly. 

So I put in the new rectifier.  Mind you this rectifier only powers the right side voltage and current meters.  It doesn't interface with the voltage or current controls nor the main output so even if the main outs were overdrawn I can't figure out how this secondary supply would be over drawn. 

In any case, that side seems to operate correctly but now the meter display on the left side has gone out.  The left side was also doing something odd.  When I would turn the voltage up the voltage would jump from about 6 to 10 volts as if one of the relays that switches the transformer windings didn't click.  Additionally, the voltage display on the front panel wouldn't follow the outputs.  So it would say 10V even though a multimeter would measure 12+.  Once I got over 14 or so volts it seemed like the display and multimeter would again agree.  Also, it seemed like the left and right outputs could affect one another.  So a high current load (say 2.5A) on one would cause a small drop in voltage on the other (perhaps as much as 1V).  In theory these are isolated outputs but perhaps at high loads one output affects the transformer windings of the other.  Then again, perhaps I have a failing transformer...

Suggestions?
Thanks!

[ftiong]

My Protek 3003B is having the same issue with the meter as well as a jumpy output when adjusting the nobs.  I wonder if you have any fix for it yet?  Thank you for this post.  I am going to at least fix the meter.