Page 1 of 1

Low 5v rail.

Posted: Fri Jan 23, 2004 4:11 pm
by Mobius
My ancient A7V133 is reporting the following voltages:
<pre>
VCore......1.75v
+3.3.......3.49
+5.........4.48
+12........12.40
-12........-12.84
-5.........-5.51
</pre>
These seem kida poor to me. The 5 volt rail in particular is identified by the BIOS as being undervolted - since it is over 10% low.

I'm gonna clean and re-seat the power connector on the mobo, and I have recently de-gunged the interior of the PSU - so it's not overheating or anything.

Any ideas what symptoms I could expect with a low 5v rail? This morning I had repeated crashes at boot time. However, System Restore seems to have fixed the issues.

I'm running the 133 RAM at 145MHz (always have!) and hence the Thunderbird runs at 1.45GHz instead of 1.0. (multiplier 10x)

PSU is Enermax 350Watt. It's about 2 years old. Probably due a new one I guess - if nothing raises the voltages.

Posted: Fri Jan 23, 2004 5:13 pm
by Jeff250
Another possible culprit is the mobo, so you may want to rule that out before purchasing a new PSU.

Posted: Sat Jan 24, 2004 12:29 am
by MD-2389
Have you verified the reading with a multimeter? Try disconnecting everything from the power supply, short the appropriate pins on the mobo connector and check it with a multimeter that way? It could easily be a faulty reading by the BIOS.

Posted: Sat Jan 24, 2004 8:53 am
by De Rigueur
Maybe a cicada got in there. Image

I was under the impression that voltage levels could be fine tuned with adjustment screws.

Posted: Sun Jan 25, 2004 7:17 pm
by AceCombat
is this a ATX or a AT PSU mobi?


<BLOCKQUOTE><font size="1" face="Arial">quote:</font><HR><font face="Arial" size="3">Originally posted by MD-2389:
Have you verified the reading with a multimeter? Try disconnecting everything from the power supply, short the appropriate pins on the mobo connector and check it with a multimeter that way? It could easily be a faulty reading by the BIOS.</font><HR></BLOCKQUOTE>


if you do that to a ATX PSU and do not have any load attatched to it, you can possibly cook the PSU and wind up with having to buy a new one regardless.


here is the ATX pinout:
Image


and here is the warning:

Quoted from www.compute-aid.com

At any event, do not try to manually jump-start the power supply without attaching motherboard. Since the power supply is expecting certain sensing circuit feedback to regulate the output voltage, munaully starting it without attaching it to the motherboard could cause damage to the power supply.

How do I test if the power supply is bad?

ATX power connector is layout like above. 5VSB constantly provides 5V power to the connector through pin 9. If you have a voltmeter, while the power supply fan does not turn on, you could identify which pin is pin 9. Normally there is a clip on the connector between pin 15 and 16. If you can identify pin 14, which is power-on pin, you can using a piece of wire short the pin 14 and 15. If you power supply has power connected to it and power switch on the power supply is on, you will see the fan is turnning by now. Otherwise, your power supply may have problem.

WarningThis test is for those who have extensive electronics training. Do not try this if you do not have extensive electronics training. Make sure there is enough load connected to the power supply before testing ( at least motherboard and one hard drive)!

PS: EDIT: ITEMS FOUND Image

Posted: Mon Jan 26, 2004 3:55 pm
by MD-2389
<BLOCKQUOTE><font size="1" face="Arial">quote:</font><HR><font face="Arial" size="3">Originally posted by AceCombat:
if you do that to a ATX PSU and do not have any load attatched to it, you can possibly cook the PSU and wind up with having to buy a new one regardless.</font><HR></BLOCKQUOTE>

Umm....no. It will run just fine like that Ace. Why do you think koolance gives you a jumper cable and instructions on where to put it to test the system if it would fry the power supply? Image This is exactly how those ATX power supply testers work Ace.

Posted: Mon Jan 26, 2004 6:48 pm
by Vindicator
Youre forgetting, of course, that the Koolance itself puts a load on the PSU and therefore fulfills that requirement. AFAIK Ace is right, its always a good idea to have a load of some sort on a PSU during testing.

Posted: Mon Jan 26, 2004 7:42 pm
by Honest Bob
AT power supplys died in the middle of the P1/P2 erra. Not many PII system had AT. I beleive I came across one last week though.

Posted: Tue Jan 27, 2004 11:33 am
by AceCombat
<BLOCKQUOTE><font size="1" face="Arial">quote:</font><HR><font face="Arial" size="3">Originally posted by Vindicator:
Youre forgetting, of course, that the Koolance itself puts a load on the PSU and therefore fulfills that requirement. AFAIK Ace is right, its always a good idea to have a load of some sort on a PSU during testing.</font><HR></BLOCKQUOTE>

thank you Vindy, the koolance tester does put a load on it.


<BLOCKQUOTE><font size="1" face="Arial">quote:</font><HR><font face="Arial" size="3">Originally posted by Honest Bob:
AT power supplys died in the middle of the P1/P2 erra. Not many PII system had AT. I beleive I came across one last week though. </font><HR></BLOCKQUOTE>

ive come across alot of PSU AT/ATX hybrids, one even powered a P-4, it had both AT and ATX style connectors plus the AUX12 connector for the P-4