Assuming that the batteries are fully charged and all that, how much extra power does a UPS draw from the grid when simply plugged in? Some amount of energy must go towards the monitoring hardware and all that, right?

Interesting question. I can only answer for the UPS unit I have on hand. Will take some measurements in the next little while and post the results.

IIRC, ~90% efficiency is fairly normal for a UPS running normally.

The extra power goes to regulating power, keeping the battery charged, etc.

Even with high efficiency, manufacturing and disposing of the lead acid battery inside can't be good for the environment. I've even heard from people who use UPS's that the battery doesn't last very long.

@ drees -- I guess the question for me is how much "extra" power is drawn when whatever's plugged into the UPS is off. I'm asking partly because of the UN's report from several years ago that devices on "standby" are drawing a substantial chunk of our energy resources.

I've had one UPS that was a dud but several that have lasted (so far) for years. It's good to get one with a replaceable battery, pain as the actual replacement process might be.

I'm not sure if I'd agree that it's so "substantial". There's a report in the latest issue of Home Power magazine that examines the issue of "standby" energy loss.

Some quick numbers:
Avg US home has 40-60W of continuous phantom loads = 1.2 KWH/day.
Avg US home uses 36 KWH/day; 4% is wasted on standby power.

Based on the above, the article states that the standby waste from 122 million US homes is enough energy to run all Australian homes. (Well, that's mostly a numbers game -- the standby waste from all the Australian homes is enough to run all the homes in Brunei.)

Residential electricity consumption in the US is around 36% of the total. So if EVERY home reduced its consumption by 4%, the net result to total electricity consumption would be a drop of something like 1.3%. That sounds pretty good, but remember, everyone has to do it.

Also, electricity is only one form of energy used by the average household. A single car represents at least 50% of a typical household's total energy consumption. If there's more than one car, you can be assured that the energy consumption of those cars is double that of the electricity in the home.

Finally, if you consider energy consumption in general, electricity consumption is substantially lower (as a percentage of total energy consumption) than direct burning of fossil fuel for transportation -- cars, planes, ships, trains.

I guess what I am saying is that the typical 40-60W waster per household we're talking about really represents a drop in the bucket.

But it's true that every drop counts, especially when there are so many leaks.

Are you positive about that? The source below indicates that the electricity/power sector uses about twice as much primary energy as transportation:

steve koonin seminar

pages 8 and 52 in particular.

Your link looks authoritative, Jaganath, and very interesting. Thanks for that!

I should have been more precise -- "US home electricity consumption is substantially lower..." The data in your linked presentation does not split electricity consumption into home/commerce/industry. Homes (in the US) account for 36% of total electricity consumption. Also, your dataset appears to be referring to the world while I am referring specifically to the US data (which I've been studying a bit). Hence my comment above.

I have an older APC UPS, model BE350C. I measured it once a while back and if memory serves, it uses about 6 watts, definetly less than 10, when its on but nothing is plugged into it.

This is an interesting question, as I am looking to get a UPS to run all my AV gear through, and I was wondering about the phantom load I would be adding. I'll be watching this with interest for Mike's measurements.

On a related note, I'm pleased that my Denon receiver, LG LCD TV, and DVD player all standby at unmeasurable loads on the Kill-a-Watt. Too bad I can't say the same for the PC - I think the USB for the wireless KB/Mouse pulls most of that.

OK, got around to the measurements.

This is a CyberPower 525SL. A few years old. Just checked to see if it was working by unplugging from the AC. It works. Computer kept running, the UPS started beeping to warn of AC power loss.

Anyway, when it's plugged into AC and nothing is plugged into it, it draws 6~7W. I think it's not a bad price to pay for some security from AC instability... but considering how infrequent they are in Vancouver, perhaps I should turn it off & just use it as a power bar.

My current PC, with a Seasonic PSU of one kind or another, draws less than 1W when turned off but plugged in.
In standby, it appears to draw 1-2W.

In normal usage (idle, like right now), it draws ~95W.
In multi-task madness and high load, I've seen peaks of ~150W or more, but that's rare. Sustained long term high load is maybe 130-135W.
This is with an A64-x2-4800, 2GB RAM, 2 HDDs, x1800GTO (a more powerful vidcard than really desired, but it was the only thing available in the lab at the time the PC was assembled.)

Thanks, everyone. I got our UPS's back when the utility company was killing our power unpredictably at a rate of about 2/month (on average). That seems to have stopped. Also, my partner is now on a laptop, so her UPS is completely unnecessary. I think I'm going to switch that thing over to the power-triggered switch, assuming that it's sensitive to a laptop's variable power draw...

An interesting feature of my UPS (APC Back-UPS CS 500 USB) is that I can use their software to see how much power my PC is consuming.

I did some testing with my X2 3800+, and C&Q made a big difference. With C&Q disabled it'd draw 130W constantly when doing nothing, with C&Q enabled it went down to 98W when idle.

I also tried underclocking my gfx card (7900GS) in 2D mode to less than half the original speed, but that didn't make any differnce whatsoever.

I have 3 harddrives in this machine, as soon as they are turned off consumption goes down to around 90W.

This thing is a real lifesaver for me. We have some pretty crappy power over here with lots of fluctuation and lots of outages. We have power outages about once a month, but when they do come it usually happens about 20 times during that day. Without it I wouldn't be able to get any work done on those kind of days

Try measuring it again with a load. That will give the efficiency of the inverter and rectifier in normal mode. Note that most modern UPSes use a "high rail" design to maximize efficiency. In fact, it's not uncommon to attain efficiencies of 90% or above. Note that the high/low DC/DC converter often operates in both directions.
Sine wave inverters are somewhat less efficient and are actually unnecessary for PCs. The best is actually a slightly filtered square wave at about 170v peak.

I also conducted some quick tests on the efficiency of my 1000VA rackmount UPS a little while ago, check out this thread.

They just don't make any sense in the first place. I had one, and it broke. I now just use a regular power strip, and I find that it works better, as it doesn't cut out to battery when the voltage drops or anything crazy like that. The PSU just deals with it.

Plus, laptops have a UPS built-in.

i'll be getting a power meter soon too and will let you know how much my UPS draws one thing i didnt see, which i will test also, is how much more it draws when charging, so i'll unplug it for a while then plug it in again and watch the watts

Probably a relative thing. Lead not ideal, but the batteries do last a long time (I have had UPSs for several years, yet to have a battery go out on me.)

The lead batteries are readily recyleable.
As compared to alkaline batteries - which are hard to find recycling options for (our city has had a high level of recycling for a long time, and they still haven't found a way to recycle alkaline batteries.) Multiply that by all the
plethora of devices (clocks, tape players, radios, thermometers, keyboards, cameras, flashes, PDAs, remotes, ...) using AA, AAA, etc. batteries.

Compared to all the car batteries out there, or even worse the use of lead in fuel, UPSs are probably a relatively small contributor to lead use/exposure.