I normally wouldn't ask such a question in here but I know there's a lot of good electrical people here who'll hopefully know the answer.
I'm trying to work out exactly how much it costs in electricity to have 2 PCs running all the time - a webserver and my development machine.
Both are AMD Athlons with 350W power supplies running the motherboard, half a gig of ram, 2 SCSI disks and 1 or 2 case fans for cooling. I know I'm not running the PSUs to full capacity so is there a way of working out the consumption?
Oh, the webserver's monitor is pretty much always off and when I'm not using the development machine the monitor is either off or in standby mode, dunno if that makes a difference :)
On occasions when I want to "size" consumption of computer equipment (typically for working out what capacity UPS is required), I tend to use a clamp meter and a specially made up short extension lead. The lead has a couple of meters of the outer insulation stripped off. Hook whatever you want to measure up to the lead, and then clip the clamp meter round the live conductor. That should give you the current draw. Multiply by
240 to get a VA rating, and finally (in the case of computer equipment) multiply the result by an estimated power factor of 0.8 to give watts.
For small loads like PCs I tend to coil up a section of live conductor and put the clamp meter round ten turns of it, then divide the result read from the meter by ten. This gives a more accurate result.
A few months ago I replaced my ancient 75 MHz Pentium system with a shiny new 2.8 GHz P4 - Dell Dimension 4600 with 250 W PSU. With the machine running continuously there was a noticeable increase in electricity consumption. Out of interest I took the following measurements:
Power Current Earth Equipment /W /A VA PF /uA
-------------------------------- ----- ------- ----- ---- ----- Old sys unit (P75) idle 30 0.185 44.4 0.68 30 Old sys unit (P75) CPU 100% 39 0.233 55.9 0.70 30 Iiyama 17in. CRT monitor 90 0.566 135.8 0.66 60
New sys unit (P4/2.8GHz) idle 100 0.505 121.2 0.83 30 New sys unit (P4/2.8GHz) CPU 100% 155 0.800 192.0 0.81 30 Dell 18 in. TFT monitor 42 0.268 64.3 0.65 40
----- Power measurements taken with a Feedback Instruments wattmeter.
Current figures are true RMS, taken with a Fluke 87.
VA figures are apparent power, calculated assuming Uo = 240 V (i.e. the approximate actual supply voltage, rather than the nominal 230 V).
'PF' column is power factor (W / VA).
'Earth' column shows protective conductor current (in microamps) taken on a Fluke 77 (mean responding). Note that the measured figures are far less than is commonly assumed in threads about spurious RCD trips.
Figures for the monitors were with 'typical screens', not max. brightness & contrast.
'CPU 100%' figures taken with the SETI at home client running.
Now THAT looks like a handy device! My whole reason for asking the question in the first place was the higher electricity costs associated with a much bigger house than the last one despite us using pretty much the same stuff here. The only things we could come up with were the extra PCs being used, but looking at Dave Baker's post below they can't be the whole reason.....nice to be able to measure pretty much everything else.
As little as that? Hmm, that averages out at less than £25 per quarter then, which doesn't explain the high bills we're getting....I wonder if the old washer and dishwasher were using more power than the ones we left at the other house......hopefully that's academic since we've got AAA rated stuff now as well as a B rated fridge/freezer. If the bills don't come down in price I'll be stumped!
You could happily put allthat kit *except* the laser printer on an 800VA UPS, yes. *Don't* put laser printers (inkjets are fine) on "domestic sized" UPSen - as the figures show, when the heater for the laser printer kicks in (to fuse the ink to the paper) it draws a much bigger peak than most under-desk UPSes will cope with; you'll end up shutting down everything on the UPS, and maybe blowing its own overcurrent protection (most usually a fuse).
UPS sizing is a victim of market-speak specmanship, BTW: an "800VA" UPS will typically supply that power for only 5 minutes or so once disconnected from mains input. (Check out individual manufacturers' websites for full details). If you want, say, 15 minutes' runtime, you need to "oversize" the UPS, i.e. get one with bigger batteries. Some mid-range and upwards models allow you to add extra batteries to increase the time of independent working, rather than the peak output. The excuse for the "5 minute" rating is that that's long enough for the automated orderly shutdown procedure, or to provide continuity until a dedicated backup generator kicks in. For a single webserver, running headless or with the screen pulling very little power in its standby mode, you'd probably get
30+ minutes of autonomous running out of an 800VA UPS; but again, you'd need to check manufacturer's specs before you buy. UPSes made by APC - and for all I know those made by other makers too - come with a "calibration" function where you can run the UPS down to "near-empty" with the particular load you run, allowing the monitoring software to know how many minutes the batteries are good for with your particular load.
You shouldn't run CRTs or laser printers on small domestic UPSs. Or at least according to the doco that came with mine. The inrush currents at switch-on are excessive.
On an associated subject I had to measure the quiescent power/current consumption of the whole house recently, with "everything off" - which still leaves 2 fridges, a freezer, various tv/ video, computer / hub, modem etc etc on standby and the result was some
400 watts, averaged over some 6 hours of sleep time - meter readings at midnight and
06.00 hrs on several days. I couldn't believe it so I bought and fitted temporarily an additional electricity meter which confirmed it - the elec. board changed the meter about six months ago and that also gives the same results.....
I am now setting about fitting the second meter into each of the MCB circuits to find out where all this is being used - its about 9 units a day which is some 40% of our daily consumption of about 23 units/day
Be aware that if using the calibration function the UPS may well shutdown at the end of the cycle even though it has mains input. This may be related the the various settings probably the interrelated, shutdown on low battery, restart time or restart after x percentage recharge. BTDTGTTS. B-)
I fail to understand your surprise. You have two fridges and a freezer as part of your "everything off" load, right? Even the "A"-rated fridges eat about 1kWh per day (the Euro Standard rating system seems to be to give kWh per year, with common figures around 300-400). So that's 1kWh in 25 hours (we round upwards for ease of calculation ;-) meaning 40W for a single, energy-efficient fridge. Multiply by three and fudge in a bit more for the age of your fridges and we've got 200W at least. Add on your kit on standby, maybe a CH boiler and pump, any "security" lights being turned on by foxes at 3 a.m., and a 400W draw seems unremarkable.
Rather than messing about with:
why don't you just unplug one of your fridges/freezers each night for a few nights and watch the effect on your overall consumption? (They have good enough insulation that 6 hours without power won't turn them into a soggy salmonella-breeding mess... at least one hopes so!) And repeat that procedure for your other circuits - if you're that keen, you can work at the whole-MCB-circuit level first, then delve deeper into individual bits of kit which are on the higher-background-consumption circuits. But it does seem to me that the figures you give are nothing remarkable. If you want to cut down consumption, look into consolidating your chilled food storage, maybe into a new energy-efficient thing if your existing ones are old (you get better efficiency overall from one fuller fridge than two mostly empty ones!), use compact-flourescent bulbs, turn anything which you can (teli, 'puter - it's got a battery for its realtime clock, dammit) right off rather than leaving it on standby. To draw in another thread, a UPS can be a useful single-point of power control for all your computer kit (if need be you plug in a couple of 4-way extension strips into it); or at less cost, wire some
4-way extension strips into one of those 4-way plugs, or do *one* step of daisy-chaining from a "master" strip into subsidiary ones, to give you a single place where you can turn off *all* your 'puter kit. But do
*measure* where your "background" power is going before putting in lots of effort into a minor contributor; I suspect the bulk of your baseload consumption is from those fridges and freezers.
400W for 6hrs is 2.4kW/hrs, ah I guess your doing a x4 to get the "whole house off" consumption for a 24hr period.
We use about that amount per day, cook by 'lectric but have rather too many incandescant lights on from 0700 to 0100ish which costs approx =A330/qtr... Must replace some with CFLs I reckon they'll pay for themselves within 2 quarters if not 1.
Base load when I very quickly checked it one evening was about 700W (+/- 200W) thats with around 400W of lighting on. Similar stuff on to yourself so 400W average base load probably isn't two far from the mark.
And don't forget that if you have TRVs on your radiators, the hot air produced by your computers reduces the amount of heat needed by the radiator. My computer room is toasty warm even in winter with the radiator on the "frost" (*) setting.
I have, OS/2 or Linux. Just haven't got round to finishing the client that will talk to OS/2 box for the linux boxen. With only one cut every year on average it's hardly worth the agro and the Linux box appears to survive not being shut down properly anyway.
One day the UPS comms will be plugged into a linux box and then I might get the others to shutdown properly as well. Of course there is the added problem that even if shutdown they remain on (old machines) so I need a hardware solution for that and the newer machines don't power back up when the power comes back without a physical prod of the power button. Yes, I have tried the various options in the BIOS to no effect.
Yeah, I'd only be powering the 2 base units, one 19" monitor and the router; the rest can die. All I'm bothered about is shutting the webserver down if the brownout's going to be longer than 5 or so minutes 'cos if it IS that long then it wasn't my fault :o)
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.