• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Very low 12v rail

D

digitalx0

Gawd
Joined
Jan 4, 2005
Messages
801
I just put the computer together that is in my sig, and in Everest I find something to be quite peculiar:

volts.jpg

The rail fluctuates between 10.5 to 11.1

Misreporting? Low, but not too low? Dangerously low? The system is running very stable with my cpu, ram, and video card oc'ed a fair amount. And yes, the rail was that low even before I overclocked. It is also odd how Everest is reporting the wrong volts to the wrong labels..
 
Have you checked the power supply with a digital multi-meter?
 
I don't own one, so no. Do you recommend I purchase one? or I may be able to find somebody with one
 
yeah definately check with a digital multimeter.

If it is not within +/- 5% (under 11.4v) you might have a problem... I'm guessing if it is actually 10.6v, there would be some stability issues.
 
It should be pretty accruate, much more accurate than any softwar.
 
mikeblas said:
How accurate is the meter that you recommended?
Click to expand...

I'm thinking it's 1% but I'd have to dig up the documentation to be certain.
 
3-1/2 digit DMMs simply don't have a big enough display to make a very accurate reading at 12 volts, and cheap meters like that one don't have a good basic accuracy to begin with.
 
As long as it gives a reading to 2 decimal places and is within 1-2% it's perfectly acceptable. I'm not spending $300 on a 5 digit Fluke for testing power supplies. A DMM that errors on the order of .12 volts per 12V is still giving a far more accurate reading than software ever will.
 
For some reason i get the feeling that DMM's resolution are going to be the new wattage/amperage <sigh>

For one it isn't neccesarily the number of digits it is the number of counts. And as I am sure mikeblas knows there are a 3 1/2 digit DMM's with enhanced resolution counts up to 6000 instead of the standard assumed value of 1999. So while it can be about display size....like many things in life...size isn't all it's cracked up to be.

Special note: I don't have the manual or specs on linked DMM above to know if it is a 1999 count or 3200 or 6000.
 
madmat said:
A DMM that errors on the order of .12 volts per 12V is still giving a far more accurate reading than software ever will.
Click to expand...
The monitor chips are more accurate than you think, even accounting for errors possible in prescaling. If you sit down and do the math, including all the error sources, you'll find that a meter you're calling "perfectly acceptable" is not as accurate as the hardware sensors you're second-guessing.

While you don't need to buy a $300 Fluke, it's foolish to buy a meter without knowing its accuracy specs. You can get a much more accurate meter for only a little bit more money.

madmat said:
It's 6000...
Click to expand...
Are you sure? This is what the website says:
Sears said:
Manual-ranging multimeter with large 2000 count LCD display,
Click to expand...

While the meter might have 1% basic accuracy, it probably has a 3- or a 5-count LSD accuracy, which throws a 12.00 volt reading off pretty badly. That's why the number of digits matters.

A 6000-count meter would be called a 3-3/4 digit DMM, not 3-1/2 digits.

I think we'll find out, though, that Everest isn't reading the sensors correctly. Before spending any money, it might be a good idea to check the reading in the BIOS, or try another program.
 
mikeblas said:
3-1/2 digit DMMs simply don't have a big enough display to make a very accurate reading at 12 volts, and cheap meters like that one don't have a good basic accuracy to begin with.
Click to expand...

Don't confuse accuracy and precision. You can have a clock without a second hand and it can be accurate but not precise.
 
dandragonrage said:
Don't confuse accuracy and precision. You can have a clock without a second hand and it can be accurate but not precise.
Click to expand...

I'm not.

If you read the specs for various meters, you'll see they usually list a basic accuracy as a percentage, then an LSB accuracy expressed as a number of counts. (The exception seems to be high-end lab meters, in the thousands of dollars; they usually specify accuracy as a percentage -- parts per million, actually -- and that's that.)

The LSB count affects a 3-1/2 digit multimeter far more than it affects a 5 digit meter: 5 counts on 3-1/2 digits in a 15 volt range is +/- 0.05 volts. 5 counts on a 5 digit meter is +/- 0.005 volts -- ten times better net accuracy.
 
You must log in or register to reply here.
Back
Top