Re: CBM 710: Replacing Power Supply - Help Needed

Den Fri, 13 Apr 2018 23:29:47 +0200 skrev Francesco Messineo
<francesco.messineo@gmail.com>:
> On Fri, Apr 13, 2018 at 11:04 PM, Michał Pleban <lists@michau.name>
> wrote:
> > Hello!
> >
> > Mia Magnusson wrote:
> >
> >> Or you could repair the broken PSU. Unless a TDA1060 / NE 5560 or
> >> some replacement for the BUV48A/BUV47A power transistor is hard to
> >> get, there aren't any parts that's likely to be break and hard to
> >> get a hold of.
> >> I'd start with checking the electrolytic capacitors for short
> >> circuits.
> >
> > The problem is, I replaced all these capacitors a month ago, so
> > it's not likely that one of them failed.
> 
> well, that can't be excluded anyway. I would not replace a non
> defective capacitor (I assume they weren't all faulty, were they?)
> Some switching mode regulators depend on the right range of the output
> capacitors ESR to be stable, though I think those old designs are not
> affected by this "problem".

Also unless components were of a good quality brand and bought from a
reliable source, the new capacitors could be bad. Either counterfeits
or just that they had been stored so long that they should had been
reformed before usage.

> > Nothing else in the PSU looks obviously broken or burnt, so I really
> > have no pointers where to look. I have never repaired a PSU before.
> 
> first things: AC main bridge, the primary-side active devices (most of
> the times, two mosfets or two BJTs), the secondary side reference and
> feedback (usually an optocoupler) and the secondary side rectifiers.
> You can usually power the output with the same voltage supposed to be
> generated, with a current limited supply, AC disconnected, and look at
> the feedback signals, the voltage reference comparator and the PWM
> regulator in a safe way.
> Most of the times I get lucky by checking all discrete (diodes, BJTs)
> semiconductors with the diode check position on the multimeter and ESR
> of all electrolytics.

In this specific PSU there is a 50/60Hz standard transformer, bridge
and capacitor providing the initial power to start the switch mode
specific IC. But the switch mode transformer itself also powers the IC,
so maybe the standard transformer (only 2.2VA) can't sustain the
continous power. Atleast some activity should be possible to see.

The +5V and +12V rails only go to a feedback look, injecting power to
those outputs would probably not start the PSU.

Btw this one is old enough to use a small transformer instead of an
opto coupler, and the switch mode regulator IC is connected to the
secondary side instead of the primary side. Back in the days it was
almost only some rare computer PSU's (usually for far larger computers)
and TVs/monitors that used switch mode power supplies, and at least the
TV's never used opto couplers at that time.

Repeating the link to the schematic that were in the start of this
thread:
http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/b/cbm700-ceag-PSU.jpg

-- 
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.