Why not to trust mobo voltage sensors:

TheSmJTheSmJ
edited January 2004 in Hardware
Check this out: http://www.antec-inc.com/pdf/article/mobo_voltage_test.html#

I know it's from Antec and all (who would be a bit bias) but I checked my voltage readings on my own system (in my sig) with my digital VOM, and the VOM's ratings were showing the correct output, while the mobo's were showing a couple tenths of a volt off.

I only bring this up because I've seen a lot of you knock good PSUs (Like the TruePower) for giving low output. Truth is, you have to use a decent (~$15) VOM to know for sure.

Comments

  • madmatmadmat
    edited December 2003
    I agree with you SmJ, I saw the same thing with my Digi-Doc5 after swapping from my ASUS A7V333 to my Soyo the bios and MBM5 voltages were no longer in line with the Digi-Doc.
    The ASUS was usually right on with it.
  • Geeky1Geeky1 University of the Pacific (Stockton, CA, USA)
    edited December 2003
    Yeah... this is one of the reasons why the vast majority of power supply reviews are totally worthless. Who cares what the voltages in MBM5 or whatever are? Those voltages mean jack squat. The only way to get accurate data is to use a multimeter, as you said, and furthermore, any power supply review that uses a computer to test the power supply is invalid ANYHOW. The only way to get valid results is to hook a power supply up to a load of a fixed, known value, and test the voltages with a multimeter.
  • SputnikSputnik Worcester, MA
    edited December 2003
    good, i's starting to get a little worried about the 11.73V, 4.89, and 3.25v i's getting. figured it was alright (even though out of spec) since this things as stable as a rock
  • muddocktormuddocktor
    edited December 2003
    Geeky1 wrote:
    Yeah... this is one of the reasons why the vast majority of power supply reviews are totally worthless. Who cares what the voltages in MBM5 or whatever are? Those voltages mean jack squat. The only way to get accurate data is to use a multimeter, as you said, and furthermore, any power supply review that uses a computer to test the power supply is invalid ANYHOW. The only way to get valid results is to hook a power supply up to a load of a fixed, known value, and test the voltages with a multimeter.

    That's why I only read psu reviews that use a known load and are reading actual voltages with good equipment, like Anandtech tests with(even Tom's tests psu's well).

    Another point that affects voltage readings in bios is the resistances encountered through the various connectors and circuits on the mobo and wires feeding the mobo. I'm sure that you have seen pics of melted ATX connectors on mobos or motherboards and that didn't happen for no reason.

    For a good read on causes of low rails and some ways to cure the problem I recommend reading this thread here at overclockers.com's forums. It will give you a basic understanding of some of the problems and also some solutions for fixing a low 5v rail, which isn't uncommon at all.
  • drasnordrasnor Starship Operator Hawthorne, CA Icrontian
    edited January 2004
    I had one TruePower go bad... ever. It read 4.6V on the +5V line according to my Fluke DMM, and the machine gave random errors in memory, disk, graphics, you name it. The replacement reads 4.6V in hardware display, but 4.95V on my DMM both on my dummy load (Antec ATX PSU tester) and in my system. Lesson learned.

    -drasnor :fold:
Sign In or Register to comment.