3.3v, 5v, 12v

[tterris]

i just bought the fsp-group ax500a blue storm psu because of the nice dual 12v lines of 15 amps a piece, but failed to realize that there are only 28 amps on the 5v line and 30 amps on the 3.3v line..

so what peripherals do the 3.3 and 5v and 12v lines power?

 

[Ice Czar]

a few minutes playing with this will save me several paragraphs
http://takaman.jp/D/index.html?english

make sure to select different mobos (P3, P4, AMD with and with out +12V connector)
and various CPUs and note how the rails that power them alter

then try dfifferent components

older boards relied on the combined +3.3V\+5V rail , newer boards and more and more componnts are biased towards the +12V rail
its likely youve made a good choice

full instructons on figuring your load are here > http://hardforum.com/showthread.php?t=843254

 

[tterris]

cool... if the calculator wasn't made for 4 year old computers

 

[Ice Czar]

processors are relatively up to date
and its simple to add the values for other components if you know what they are
drives, fans and most cards and such arent any great difference
its primarilly the Video cards it lacks
its still a few light years ahead of most PSU "calculators"
which simply give you watts, which means next to nothing

 

[category_five]

how does he get the values for the different componants. Also anyone have word as to the power draw on the newer video cards? I'm assuming >80w

 

[Bullitt]

I thought it was around 70ish watts for the 6800Ultra and 88 for the 800XT under gaming load.

Either way, your 80w figure is accurate.

 

[burningrave101]

I thought it was around 70ish watts for the 6800Ultra and 88 for the 800XT under gaming load.

Either way, your 80w figure is accurate.

The 6800U uses more power then the X800XT. The X800XT consumes around 65w while the 6800U consumes around 110w of power.

 

[tops2]

The 6800U uses more power then the X800XT. The X800XT consumes around 65w while the 6800U consumes around 110w of power.

burningrave101, do u know how much current approximately the 6800 series draw from each voltage rail?

 

[Ice Czar]

http://hardforum.com/showthread.php?t=844691

specifically
Power Consumption of Contemporary Graphics Accelerators: ATi
Power Consumption of Contemporary Graphics Accelerators: Nvidia

however

http://hardforum.com/showthread.php?t=846983&highlight=shunt
(wherein we discuss the accuracy of the xbit tests compared to BillR's own tests)

any idea how those test proceedures came up with such divergent results?

Measuring 12V on a computer is *HARD*.

12V hops all over the place... whenever a drive access occurs and a head flies across a hard disk surface, a load pulse occurs on the 12V line. whenever a fan does a 90 degree rotation a 12V pulse occurs, and when CPU's switch between idle and running states the 12V load goes all over the place at KHz rates... Sometimes when the CPU goes into idle, it's possible for the VRM to actually put 12V current *back into* the power supply!

Most DMM's are peak detecting, so they'll read artificially high. And on top of that, most DMM's have an error specification which is usally something like 5% of full scale, which means that if you're measuring on a 10A scale, 2A can measure anywhere from 1.5A to 2.5A. And finally, you need your shunt to be non-inductive, because if it's a wirewound resistor you'll be getting extra voltage on it just due to the inductance combined with the current changes...

To do things right you really need a calibrated, true RMS multimeter and a calibrated, non-inductive current shunt. At my work, we use calibrated current shunts read with a calibrated millivoltmeter for this sort of work...