PC component power consumption: idle vs. load?

[Maverick2002]

Not sure if this qualifies as "highly technical", but I was wondering, what kind of power consumption difference could you expect from running components idle and stressing them?

CPU is easy to figure out; calculators are all over the place. But what about CD-type drives, HDDs, PCI cards, motherboards, memory, etc?

 

[Mday]

the specifications should be listed in the manual, assume the min req is the idle load, and the max req to be the non-idle load.

and no, this is not what this forum was intended for.

 

[Maverick2002]

Well, I don't know where to post this then; people in other forums don't reply. Anyway, I was allowed to take about 10x old HDDs and 10x old CD-drives from work for testing, but something doesn't add up. Most of the drives are rated:

5v - .41A
12v - .21A

That's between 2-3watts per drive. Am I missing something? Seems incredibly low. As for the CD-drives, most are:

5v - 1A
12v - .8A

That's between 5-10watts per drive? I could have sworn I read somewhere these things use like 30watts.

So does voltage jump and current drop (for HDDs and CD drives) whenever you go from idle to load?

 

[Leo V]

I believe you can use a device called Kill-a-Watt to measure your system's power draw (or anything else that plugs into an outlet). To measure individual components' power draw, you could remove the component from the computer and measure the differences (idle/load). Of course, you would need to adjust the numbers for the PSU's efficiency factor, i.e. with a 65% efficient PSU your component itself only accounts for 65% of the difference in power draw, the rest is PSU heat.

Oh, and here is a typical components max power usage chart.

 

[Mark R]

As for the CD-drives, most are:
5v - 1A
12v - .8A

The drive has 2 seperate supplies - a 5 V and 12 V. The max power on the 5 V is 1 A (5 W), and on the 12 V is (0.8 A) 10 W. Total 15 W.

Usually, the 5 V supply is used for the drive electronics - whereas the 12 V supply is used for the motor. So power consumption on the 5 V supply is likely to remain relatively constant, whereas ther 12 V supply load can cary greatly.

I've tried to measure the difference in power consumption of different components by measuring the total power consumption of the computer.

I used different software to tax different parts - e.g. prime 95 to test CPU, Quake to test CPU + 3D, Defrag to test drives. The result was that only the CPU seems to make a measurable difference to power consumption.

I used an Athlon 1000 + GF 2 system + DVD + 2x HDDs running Win 98. At idle (sitting at windows) power consumption is 90 W - when running Quake or Prime 95 or Seti@Home power consumption rises to 120 W - no difference between the different programs. Defrag + Prime 95 (to maintain CPU usage at 100 %) also used approx 120 W. Essentially the different contributions of the HDs and graphics card are negligable.

It's worth pointing out that the max ratings of the CD-ROM drive only apply as the drive is accelerating - once it has reached operating speed, power consumption drops to a much lower level.

 

[Maverick2002]

So what's the best way to load 10 CD drives (they range between 4x-48x) then? I was thinking of popping an audio CD in each one (that way I don't have to get controller cards but only hook them up to the molex). Any way to measure the power consumption then? And is there an estimated usage of the 12v once it reaches full speed?