11,19V on the +12V rail - Is this too low ?

Jerg

n00b
Joined
Mar 3, 2006
Messages
9
Hi everyone,

Lately, I have been getting some BSOD. Minidump file is telling me its hardware and so far I have not figured it out.

Since I wanted to monitor my system, I installed Speedfan and I noticed that my +12V is going from 11,25V to 11,19V and Im not running my cpu at 100%, almost IDLE at the moment.

Is this normal ?
Can I check the voltage with another software ?
Could this be causing me the BSOD ?

Thanks for any input you may have..

BTW, my power supply is a PC Power&Cooling Silencer 750W.
 
Buy a digital multimeter and check it yourself.

Seriously I know this has been said 100000 times but software voltage programs are notoriously worthless.
 
As the above poster said, the only way to reliably measure voltages is with a digital multimeter. However, the ATX specification is +/- 5%, which corresponds to a range of between 11.4V and 12.6V. If your voltage is actually 11.19V, then it is too low. However, check with a multimeter before coming to any conclusions.
 
Assuming I get a multimeter, how do I measure the +12V ?
Grab a molex connector, set it to voltmeter mode, stick the positive probe into the yellow wire and the negative probe into one of the black wires.
 
Measure by sticking the probes into the motherboard 24 or 20 pin connector under load. Molex with nothing attached to it only sees voltage at the power supply's rail.

Falling out of ATX specs can happen if the wire from PSU to connector is too thin for the length.
 
Measure by sticking the probes into the motherboard 24 or 20 pin connector under load. Molex with nothing attached to it only sees voltage at the power supply's rail.
The Silencer 750W is a single-rail PSU. Voltage measured from a molex connector will be the same as voltage measured from the ATX connector, and it is much easier to measure from an unused connector as well.
 
The Silencer 750W is a single-rail PSU. Voltage measured from a molex connector will be the same as voltage measured from the ATX connector, and it is much easier to measure from an unused connector as well.

No it won't. They both go back to the rail. Unloaded molex connector will provide voltage at the rail. If you measure at the ATX connector, that includes voltage drop in the connector and wires. Remember that on-board sensing is downstream of the connector.

If you were to measure the voltage between the yellow wire on an unused Molex (nothing connected in that string) and a yellow wire on the ATX connector with the DVM in 2v range, you will read a small voltage. That is the voltage drop in the ATX wiring. To get a even accurate value, you'll have to measure voltage between black & yellow at ATX, then between black & yellow at unused Molex and take the difference. Voltage drop occurs in return path too, but you can't simply multiply the result from the former test by two, because the resistance in 12v wiring and ground wiring is not the same, simply because there are more ground wires going to ATX connectors than yellow wires.
 
No it won't. They both go back to the rail. Unloaded molex connector will provide voltage at the rail. If you measure at the ATX connector, that includes voltage drop in the connector and wires. Remember that on-board sensing is downstream of the connector.

If you were to measure the voltage between the yellow wire on an unused Molex (nothing connected in that string) and a yellow wire on the ATX connector with the DVM in 2v range, you will read a small voltage. That is the voltage drop in the ATX wiring. To get a even accurate value, you'll have to measure voltage between black & yellow at ATX, then between black & yellow at unused Molex and take the difference. Voltage drop occurs in return path too, but you can't simply multiply the result from the former test by two, because the resistance in 12v wiring and ground wiring is not the same, simply because there are more ground wires going to ATX connectors than yellow wires.
Fair enough. I thought you were referring to voltage differences between rails with your previous post.
 
Jerg said:
I installed Speedfan and I noticed that my +12V is going from 11,25V to 11,19V and Im not running my cpu at 100%, almost IDLE at the moment.

Is this normal ?
For SpeedFan to be inaccurate? Yes. When I used it with an old ECS K7VTA3 v. 8, it said the +12V rail alternated between 6V and 8V about twice a second. Obviously it wasn't. Another monitoring program read 10.6xV, which also had to be wrong because the hard drive wouldn't spin at such a low voltage (I tested).

As the experts have recommended, get a digital multimeter -- very handy tool, not just for checking computers but also batteries (mobo, flashlight, car), wiring, automotive charging systems,...
 
The programs can read something obviously wrong and you know it's wrong like a ladder missing an entire rung, but accuracy is not known unless you compare against a calibrated meter.
 
Back
Top