[rescue] Decstation 5000 Power Supply

[CJ Reha]

Hi Mike,

Makes sense you're not getting anything on the secondary. If it's sensing a
fault and going into some kind of protection mode, the supply won't
oscillate at all, so you won't get anything past the transformer. Though,
reading through Rob's posts...check to see if you have proper B+ at the
main filter caps (be careful of course - the primary side is hot). Every
once in a blue moon I've seen those caps go open like that, so it's
absolutely worth checking, especially since it's already been observed
once...

That being said, output caps can and will still cause this behavior - if
they're out of spec or high ESR, the supply will find it harder and harder
to maintain proper regulation and at some point shut down for safety. Also,
capacitor electrolyte is conductive (and corrosive) - so if there's a lot
of it that managed to creep out onto the board, it might be causing leakage
problems the supply will also shut down in response to. Since your fuses
are good and the supply failed quietly like it was probably a "safe"
protection mode shutdown, I'm hoping - like with Rob's supply - a simple
recap will bring it back. I've fixed plenty that just needed that! I
wouldn't be worried about longevity. If you're thorough with cleaning and
use good quality caps, you should be golden for a while. I've fixed
hundreds of old supplies like it and I would estimate 1% or less come back
to me.

It's always difficult to trace out supplies without being able to
manipulate them in person, but it does appear the vertical subboard next to
the switching MOSFET and inrush limiters contains the driver circuitry.
Many supplies of this era still used discrete driver circuits, though the
feedback stuff was usually integrated (like your supply which uses an
LM393/LM339 comparator to sample the outputs).

Hope at least some of this helps.

Regards,
CJ

On Fri, May 31, 2024 at 10:30 PM Michael Dombrowski via rescue <
rescue at sunhelp.org> wrote:

> I spent some more time with the PSU tonight. I'm not getting voltage
> anywhere on the secondary side. I pulled out a thermal camera and the whole
> secondary side is cold, as are the mosfets. I don't see an IC to drive the
> mosfets, there is another control board with discrete SMD transistors,
> which I assume provides the same function. Would bad caps cause this
> behavior? I'm a little worried there is something wrong on the primary side.
>
> mike
>
> On Friday, May 31, 2024 at 07:24:50 PM CDT, CJ Reha via rescue <
> rescue at sunhelp.org> wrote:
>
>
> Hi Mike,
>
> Though I unfortunately can't give advice on the wiring of new power
> supplies, I can potentially be useful in bringing back the original. Much
> like basically any other switcher from the 80s or 90s, those supplies
> suffer from the output capacitors leaking. Even if there doesn't appear to
> be any leakage, pull the caps anyway - in my experience many supplies will
> have little pools of electrolyte under each cap, or the solder will sizzle
> and smell like fish even if it looks OK from the outside. Sometimes it's an
> oily "film" not easily seen unless you're really looking for it. I saw that
> in your VCFed thread another gentleman posted about his machine with the
> same problem, and he had a bunch of caps leaking on both the main board and
> the auxiliary control board...so it's likely your supply has the same
> ailment. Do check to make sure the fuse is intact - if not, there may be
> more serious issues afoot - otherwise, I think a simple recap may get you
> back on the road so to speak. What have you done on the supply to this
> point?
>
> Regards,
> CJ
>
> On Fri, May 31, 2024 at 7:54 PM Michael Dombrowski via rescue <
> rescue at sunhelp.org> wrote:
>
> As I've mentioned on this list, I recently picked up a Decstation 5000/200
> and was mid-way through Ultrix install when the system powered off and
> wouldn't come back on. I opened up the H7878A power supply, but don't see
> any obvious faults. I'm slowly working through it, but I don't have any
> experience repairing switching power supplies, so this will likely take
> some time.
>
> As a temporary measure, I ordered a beefy 5v power supply and a -12/+12v
> buck boost converter. Unfortunately, I'm not able to boot. The power rails
> all are steady, and thermal imaging does not reveal any suspiciously hot
> areas. There are two other connectors from the power supply to the logic
> board: warning and POK. I've tried various combinations tying these to
> ground or +5v without success. Does anyone know the expected behavior of
> these circuits? I can find references for older Vax-en, but not for the
> Decstation.
>
> thanks
> mike
> _______________________________________________
> rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org
>
> _______________________________________________
> rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org
> _______________________________________________
> rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sunhelp.org/pipermail/rescue_sunhelp.org/attachments/20240531/aceb9147/attachment-0001.html>