Searching \ for '[EE] ATX-PSU question' 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/index.htm?key=atx+psu+question
Search entire site for: 'ATX-PSU question'.

Exact match. Not showing close matches.
PICList Thread
'[EE] ATX-PSU question'
2008\01\08@190230 by Rikard Bosnjakovic

picon face
I have two working ATX-PSU (different manufacturers) that I was about
to modify a bit, since I needed the 12V outputs for a project. Before
doing any modifications, I tested both PSU:s like this: put a load (10
ohm 5W resistor) across +5 and GND, short green cable (PS_ON) to
ground and hooked the psu into the mains. Worked perfect for both
PSU:s, I got my 12 volts.

Then I started the modding by desoldering all cables except for the
green one (which I cut and soldered to ground), soldered the resistor
between +5 and ground and finally one piece of cable to +12V and
ground respectively.

Now, the first psu works just like it did when I tested it prior to
desoldering, but the second doesn't. What happens is that the psu
shuts off after an instant of a second. The 12V-rail goes up to 12V
and then the psu shuts off. The 5V-rail goes up to 0.9V (or similiar).
I suspected that one of the cables was some sort of feedback-cable to
the psu, but I cannot find anyone except the gray one (PWR_OK) here:
http://pinouts.ru/Power/atx_v2_pinout.shtml - but this seems to be a
feedback to the PC and not the psu itself.

Anyone got a clue of what I have done wrong? As I said, the test
before the cable-desoldering went fine, so I must have botched it
somewhere.


--
- Rikard - http://bos.hack.org/cv/

2008\01\09@053835 by Tamas Rudnai

face picon face
Hi,

Usally I put 100 ohm to the +5 as 50mA just as fine. When a stable +12 is
more important than maybe it's better to remove the +5 parts and put a
voltage divider from +12 to the feedback line as +12 depends on +5 feedback.
Also it worth to check the diodes as some cheap PSU claims 13A @12V while
uses PR3003 (3A diodes). At least they put it onto the cooler :-). If that
the case it may worth to check the ring if they were using thick enough
coils.

Most ATX PSU has a similar circuit of this:
http://www.pavouk.org/hw/en_atxps.html

Not all of them though, for example Dell does not use TL494 nor TL431 -
there is a DNA1001D and DNA1002C if I could recall well.

But this thing you've made supposed to work just as fine. Are you sure there
is no short circuit made when desoldered the cables? Did you check your
resistor if that's ok? (You might not even need that) Did you check VSB,
3.3V and the negative rails?

Tamas


On Jan 9, 2008 12:02 AM, Rikard Bosnjakovic <spam_OUTrikard.bosnjakovicTakeThisOuTspamgmail.com>
wrote:

{Quote hidden}

> -

2008\01\09@060654 by Alan B. Pearce

face picon face
> Anyone got a clue of what I have done wrong? As I said, the test
> before the cable-desoldering went fine, so I must have botched it
> somewhere.

By tying the line low it may think you are wanting to power off, as in
holding the power on button down for several seconds on a PC. Quite why the
other one doesn't behave in the same manner is anyones guess, but it may
want to see the button go open after power on before looking for power off.

2008\01\09@064020 by Steven Howes

flavicon
face
> By tying the line low it may think you are wanting to power off, as in
> holding the power on button down for several seconds on a PC. Quite  
> why the
> other one doesn't behave in the same manner is anyones guess, but it  
> may
> want to see the button go open after power on before looking for  
> power off.


Nope, its the motherboard that senses the 'holding'. It has to be  
taken low and stay low to remain active.

More... (looser matching)
- Last day of these posts
- In 2008 , 2009 only
- Today
- New search...