Accuracy of power load on a UPS

ICE_9 · Post by **ICE_9** » Mon Dec 10, 2007 8:05 am

Does anyone have a comparison reading on a load that is measured on both a UPS with an LED readout and a Kill-a-watt? I am trying to determine the accuracy of the UPS's readout. Any info would be helpful. Hopefully this will lead into a power usage topic in this forum.

Post by **toTOW** » Mon Dec 10, 2007 11:33 am

It depends on your UPS.

Since PSU have active PFC, we can assume that apparent power (in VA) equals real power (in W). So when plugged to that main power, the value you read is real power (this value can be given by a kill-a-watt).

If your UPS is an offline one, meaning that when power is aviable, it will give power from the main, it will display a value where apparent and real power are the same.

But if you have a line interactive or online UPS, or if your offline UPS is giving power from batteries (power outage), real power is different from apparent power, because the power coming from batteries is not sinusoidal.

Most UPS have monitoring softwares. Mine is looking like this :

"Niveau de sortie" means output level ... it's a percentage of the maximum output power which is 600 VA for this UPS (offline).

7im · Post by **7im** » Mon Dec 10, 2007 4:56 pm

If it's an APC model, I know the UPS to be accurate. No experience with other brands.

Vester · Post by **Vester** » Mon Dec 10, 2007 5:52 pm

I have a model P4400 Kill A Watt meter. I used it to measure the power factor on two computers and a PS3 at the wall outlet. The PS3 system measured an unexpectedly high 0.99 PF (TV off), the HP Pavilion Elite m9047c measured 0.61, and the Dell Dimension 8200 (with a PC Power and Cooling power supply) has a PF of 0.65. All three have UPS Back-UPS NS units.

Post by **toTOW** » Mon Dec 10, 2007 5:58 pm

Bad PF for your PCs

... do PSUs have active PFC

Vester · Post by **Vester** » Mon Dec 10, 2007 8:20 pm

I don't know about the HP power supply, but the PC Power and Cooling unit has a PF of 0.99 per the website information. I haven't pulled the printer, scanner, or speakers on the Dell to see if there is any improvement. Disconnecting the monitor power and speakers power on the HP made no difference. Anything under 0.90 is really poor for systems IMHO. I may do some more experimenting when I have time.

Post by **toTOW** » Mon Dec 10, 2007 9:46 pm

Yes it's strange ...

What are your power consumption values for your machines

Vester · Post by **Vester** » Tue Dec 11, 2007 2:06 am

Dell Dimension Intel P4 @ 2.8 GHz, printer, scanner, speakers, LCD monitor (in standy): 171 watts. With 19 inch LCD monitor on: 194 watts.

HP Intel Q6600 @ 2.4 GHz: 224 watts w/22 inch LCD monitor on and 196 with monitor off.

PS3 (80 GB drive) with TV off (unplugged): 234 watts.

Nonymoussurfer · Post by **Nonymoussurfer** » Tue Dec 11, 2007 2:20 am

Vester wrote:I have a model P4400 Kill A Watt meter. I used it to measure the power factor on two computers and a PS3 at the wall outlet. The PS3 system measured an unexpectedly high 0.99 PF (TV off), the HP Pavilion Elite m9047c measured 0.61, and the Dell Dimension 8200 (with a PC Power and Cooling power supply) has a PF of 0.65. All three have UPS Back-UPS NS units.

Are you sure you aren't looking @ line freq on the kill-a-watt? i believe the same button is used to switch between PF & Hz. If so, 61 Hz wouldn't be too bad I don't think, 65 Hz is probably farther out than out to be but still within 10%... (I made the assumption you are in U.S.)

Have you cross referenced the W & VA figures, to ensure that a 0.6 PF makes sense?

Vester · Post by **Vester** » Tue Dec 11, 2007 3:53 am

Hertz are shown in format xx.x and PF is shown as 0.xx. I double-checked...again. Thanks for the suggestion, Nonymoussurfer.

I'm not surprised about the HP, but the expensive, quiet power supply in the Dell is a shocker. I've read that some computers have Power Factors of 0.4 or lower. (Can't seem to find it on Google at the moment.)

My AC voltage is 120.3 to 120.5 at 59.9 to 60.0 Hz.

Post by **toTOW** » Tue Dec 11, 2007 11:09 am

Your values in Watts seem correct ... maybe a little high for your Q6600, but if the case is full of drives it could be normal. The differences could be explained by your AC voltage (120V - 60 Hz) where mine in France is 230V - 50 Hz