View Full Version : Power Supply Voltage fluctuation

24-06-2006, 11:57 PM
I have an intel motherboard running xp pro, P4, 1GB Ram. The motherboard has a utility which monitors temperature and voltage. The +12V keeps fluctuating up to 12.313V. So I reckon its time I got a new psu cos this could damage my computer. Anyone out there thinks thats this is normal or has a better idea about doing something about the fluctuation than I would love to hear about it.


25-06-2006, 05:50 AM
I just looked at my SmartGuardian, and it shows a slight fluctuation too...but I have to believe that it's WNL.

I imagine there are loads and unloads going on all the time on that buss, and that the actual voltages are not outside normal parameters anyway.

My rate on the 12v buss is 12.19 to 12.29, which has been the case since I've owned this unit..over 3 years now.

25-06-2006, 09:25 AM
AFAIK most things in the computer that use raw 12v are motors and would tolerate small variations. The important things like the CPU use the 12v after it has been through a precision voltage converter down to something under 2v. Voltages reported by the motherboard aren't that accurate anyway.

25-06-2006, 11:27 AM
I have a ASUS MB and the 12v, 5v legs fluctuate slightly due to load like someone above mentioned. The 3.3 and Vcore are steady. This is normal so I wouldn't worry. My 2cents+gst

25-06-2006, 11:32 AM
<........due to load like someone above mentioned.....>

Wow! I've graduated to "someone" now!

Graham L
25-06-2006, 02:43 PM
The 5V and 3.3V are fairly critical. They are usually specified to be +/-5%
and the other supplies are allowed to be +/- 10% or 15%. COU coltages are usually governed by motherboard regulators. 12 volt is iued only for motors in disks and fans (and serial ports). None of these loads needs close regulation.

I'm not sure about ATX supplies, but on PC and AT supplies only the 5V is regulated. The levels of the other supplies are determined by the number of turns each has on the transformer. The 5V might be a centre-tapped 6 turn winding, whereas the 12V might have a centre-tapped 10 turn winding. The voltage drop in the diodes and chokes will cause the 12V to drop with increasing load. The switchmode regulator chip gets its feedback from the 5V supply, so if its load varies, the 12V might go up and down because of that too.

Worry if your 12V goes to 0V.

Who says the measurements aren't accurate? I put in a programme to show the voltages, and it gave me about 15 digit values. They were totally meaningless, but they were to 15 places. :D

25-06-2006, 02:58 PM
the sensors are very accurate. however the fluctuation is the thing to keep an eye on. they normall fluctuate a little bit. to much is a sign of a power supply failing or struggling to supply. one thing to keep an eye on is sudden dips in power. also the 5v is often tied to the 12v rail, if the 5v dips it causes the 12v to go high.

what sort of PSU have you got? if its a cheap pos then it would be worth changing it.

25-06-2006, 03:54 PM
Its a 400W ATX psu, i bought it new with the case a few months ago. I wasnt too sure wheather it was normal for it to fluctuate like that.

Graham L
25-06-2006, 03:58 PM
The sensors are not accurate. They might be "precise" because they give "digital" numbers. But they are not as accurate as a $5 multimeter. They don't have to be, and it would cost money.

It's normal for the 12V to vary. That does no harm.

25-06-2006, 06:03 PM
yes its normall but it depends what normall for your pc. sorry but its sometimes hard to tell if its just normall fluctuations or ones casued by a fault. i've had faulty PSU's that looked like normall......untill i put a good one in and saw the difference.

whats the brand of the PSU ?

also whats the specs of the pc? the psu might be a bit light.

25-06-2006, 06:26 PM
Cant acutally say what the brand is...
TUV Product Services...S/N 50968763. (if this helps)

PC is P4 2.6 with HT, 1GB Ram, 128DDR Geforce card, 80GB HHD, DVD Writer...

I reckon the psu is good enough for this.... nuthing much else on the pc..

25-06-2006, 07:44 PM
i don't think TUV is a brand....mind you it wouldn't suprise me if it was !

what brand case?

its proberly a cheapy but its fairly decent size so it should last ok.

25-06-2006, 08:56 PM
12.3v is fine, as long as it stays at 12.3v (or somewhere near it) if it was going all over the place all the time I would be suspicious

25-06-2006, 09:06 PM
12.3v is fine, as long as it stays at 12.3v (or somewhere near it) if it was going all over the place all the time I would be suspicious thats is just it.......what do you call "somewhere near it" ?? i've had plenty of faulty PSU's that only show +/-0.5v.

the only way to tell is to install a known good psu and compare the two.

i normally find its not so much the volt difference or the drop in volts but the way it drops ie small but sudden drops for no reason and most important a change in sounds.

edit: sorry to try and expain it better..... just because it says its 12v dosn't mean its good. unforunatly its not as easy as just measing volts to determin if its good or bad. power factor and electical noise are also big factors exspecialy when caps have blown.

25-06-2006, 09:49 PM
According to the official ATX PSU Specifications:

3.2. DC Output
3.2.1. DC Voltage Regulation
The DC output voltages shall remain within the regulation ranges shown in Table 2 when
measured at the load end of the output connectors under all line, load, and environmental
conditions. The voltage regulation limits shall be maintained under continuous operation
for any steady state temperature and operating conditions specified in Section 5.
Table 2. DC Output Voltage Regulation
Output Range Min. Nom. Max. Unit
+12V1DC (1) 5% +11.40 +12.00 +12.60 Volts
+12V2DC (3) 5% +11.40 +12.00 +12.60 Volts
+5VDC 5% +4.75 +5.00 +5.25 Volts
+3.3VDC (2) 5% +3.14 +3.30 +3.47 Volts
-12VDC 10% -10.80 -12.00 -13.20 Volts
+5VSB 5% +4.75 +5.00 +5.25 Volts
(1) At +12 VDC peak loading, regulation at the +12 VDC output can go to 10%.
(2) Voltage tolerance is required at main connector and S-ATA connector (if used).
(3) Minimum voltage during peak is greater than 11.0 VDC

However, like you say - this still does not determine if the PSU is faulty