Searching \ for '[EE]: PC Power Supply... revisited' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power/actodc.htm?key=power
Search entire site for: 'PC Power Supply... revisited'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: PC Power Supply... revisited'
2000\06\07@023259 by Sebastian Garcia

flavicon
face
Hi all,

Just wanna know if anyone can send me a schematic of the typical switching
mode power supply of the AT PC, or point me to a web link where I can find
it.

Got one of these ( "Achieve ML200C" ) with a problem...don't starts (either
with load) . It seems that one of the MJE13005 's at the input is dead. Any
hint will be welcome !

TIA,

S.-

2000\06\07@024141 by Dale Botkin

flavicon
face
There is a system board detect pin called PWRGOOD (or similar) that is used
to prevent the power supply from coming up if there is no load on it.  It's
been a while, but I think it needs to be connected to +5V.  Don't hold me to
it, but I think that's what it needs.  Geez, who would have thought I'd ever
need those old PC & XT system board schematics I tossed a few years ago?

Anyway, this may help if it's a NEW power supply that looks dead.  If it's
an old one that actually IS dead, forget I spoke.

Dale

> Got one of these ( "Achieve ML200C" ) with a problem...don't
> starts (either
> with load) . It seems that one of the MJE13005 's at the input is
> dead. Any
> hint will be welcome !

2000\06\07@025636 by David Duffy

flavicon
face
<x-flowed>Dale wrote:
>There is a system board detect pin called PWRGOOD (or similar) that is used
>to prevent the power supply from coming up if there is no load on it.  It's
>been a while, but I think it needs to be connected to +5V.  Don't hold me to
>it, but I think that's what it needs.  Geez, who would have thought I'd ever
>need those old PC & XT system board schematics I tossed a few years ago?

As far as I can remember, the PWRGOOD signal is an output from the PSU
to indicate "power is good" to the main board. (+5 when all is ok)
Can't say I've ever checked to see if any main boards actually use it though!
One common fault with AT PSU's is the 220K .25W resistors on the push-pull
transistors go high or O/C - I think they are a part of the kick start
mechanism.
Regards...

___________________________________________
David Duffy        Audio Visual Devices P/L
U8, 9-11 Trade St, Cleveland 4163 Australia
Ph: +61 7 38210362   Fax: +61 7 38210281
New Web: http://www.audiovisualdevices.com.au
New email: spam_OUTavdTakeThisOuTspamaudiovisualdevices.com.au
___________________________________________

</x-flowed>

2000\06\07@031535 by Mark Willis

flavicon
face
Dale, IIRC, !PwrBad or PWRGood, is an output from the Power Supply - it
emits +5V normally when the power supply is working right, if AC power
drops out it's designed to hold the CPU in reset etc.;  Sorta like a
!MClr only for rather a higher power processor!

A friend was making a Linux Beowulf cluster (diskless) out of a bunch of
old identical 386's & NICs with custom boot ROMs he had, sharing one
power supply to run several momboards per custom case, he may have built
a little buffer circuit he built that made sure each momboard had a good
fanout left from that pin.  Ought to call him up and see what he's up to
nowadays <G>

Sebastian, I'll look but the chances of my power supply schematics
matching theirs aren't perfect.

 Mark

Dale Botkin wrote:
> There is a system board detect pin called PWRGOOD (or similar) that is used
> to prevent the power supply from coming up if there is no load on it.  <snip>
> Dale
>
> > Got one of these ( "Achieve ML200C" ) with a problem...don't
> > starts (either
> > with load) . It seems that one of the MJE13005 's at the input is
> > dead. Any
> > hint will be welcome !

--
I re-ship for small US & overseas businesses, world-wide.
(For private individuals at cost; ask.)

2000\06\07@090250 by John C. Frenzel

flavicon
face
> Hi all,
>
> Just wanna know if anyone can send me a schematic of the typical switching
> mode power supply of the AT PC, or point me to a web link where I can find
> it.
>
> Got one of these ( "Achieve ML200C" ) with a problem...don't starts
(either
> with load) . It seems that one of the MJE13005 's at the input is dead.
Any
> hint will be welcome !
>
> TIA,
>
> S.-
>
Alot of times its just the fan that is dead.  The supply is designed so that
if the fan dies, the supply shuts down.  Try replacing the fan to see if
that's it.
John

2000\06\07@165247 by Sebastian Garcia

flavicon
face
John,

-----Original Message-----
From: John C. Frenzel <.....johnfrenzelKILLspamspam@spam@HOTPOP.COM>


|Alot of times its just the fan that is dead.  The supply is designed so
|that
|if the fan dies, the supply shuts down.  Try replacing the fan to see if
|that's it.
|John

Thanks for your help, but the fan is OK. I've checked this.

Regards,

S.-

2000\06\07@165251 by Sebastian Garcia

flavicon
face
Dave,

-----Original Message-----
From: David Duffy <piclistspamKILLspamAUDIOVISUALDEVICES.COM.AU>


|One common fault with AT PSU's is the 220K .25W resistors on the push-pull
|transistors go high or O/C

What is O/C ? (Cutoff...?)

I see two 150K resistors, think in the place of the 220K 's . I will work
around this.

Thanks for the hint,

S.-

2000\06\07@180201 by Sebastian Garcia

flavicon
face
Well people, I found that one of the NPN MJE13005 BJT's is dead. It's
compatible with ECG51 and now I'm looking for a replacement.

Closed thread.

Sorry if I bored a dozen guys with this. I know that it's more easy to buy a
new one but, What happen with the technical spirit?  ;)

Best Regards,

S.-

2000\06\07@234217 by paulb

flavicon
face
John C. Frenzel wrote:

> Alot of times its just the fan that is dead.  The supply is designed
> so that if the fan dies, the supply shuts down.

 Interesting theory, but I've never seen any evidence of it.  That
would require a fan with (rotation) monitor device.  While these *are*
used in top-end processor fans, it's a bit of an ask in power supplies.
(OK, so by definition I don't see top-end machines!)

James Paul wrote:

>  In my experience, I've found that it's usually less expensive to buy
> a new PS than to try to repair the old one.

 Dead right.  What I however enjoy, emphasis on that, is collecting
especially "shop rejects" - the ones that bounced on the way out.  These
often have easy and non-cascading faults.

 The two faults I have described here before are the 1 ohm power
resistor from the factory that *should* by rights have caught fire, as
all specimens behaved as two amp fast-blow fuses, and the one where the
switch wires were connected with wrong colours, so the front panel
switch became a mains crowbar.

 Do *NOT* accept unbranded power resistors (3W, 5W, 10W) from a
supplier.
--
 Cheers,
       Paul B.

2000\06\08@002854 by John C. Frenzel

flavicon
face
> John C. Frenzel wrote:
>
> > Alot of times its just the fan that is dead.  The supply is designed
> > so that if the fan dies, the supply shuts down.
>
>   Interesting theory, but I've never seen any evidence of it.  That
> would require a fan with (rotation) monitor device.  While these *are*
> used in top-end processor fans, it's a bit of an ask in power supplies.
> (OK, so by definition I don't see top-end machines!)
>
Paul,
It's more than some half-witted theory.  I have returned to use more than a
half dozen PC power supplies by just replacing the fan.  You are right, one
way of accomplishing shutdown on failure is to mechanically monitor
rotation, and you are right, these supplies don't. They monitor current
load. No load, no output.  Generic PC power supply fans are always 12v, not
mains driven for this very reason.  If you don't believe me, open a supply,
and unplug, or desolder the fan connector.  Let me know what you find.
John

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...