How much power does Standby use?

Prompted by this, typically flippant, article I thought I'd undertake a little of my own research into whether turning the TV off at the plug makes any difference.

I armed myself with a digital power meter (but see below!) and went to work measuring the standby power of things about the house. If you are interested in replicating results, then bear in mind that many devices take a little while to "settle down" once turned on (my DVD recorder, for example, spins for disk for five minutes, and then properly sends itself to sleep). Also, I'm guessing there is some margin of error. However, let's slay one myth: my Motorola mobile phone charger uses no power at all (well, less than one Watt) if its plugged in, but no phone is attached.

My understanding, which Wikipedia does not entirely backup, is that many modern power supplies are "electronic" or "switched-mode", which use very little power when a device is not attached to them. You can spot these, superior, power supplies: they are typically (much) smaller than the traditional "power-bricks". You can physically tell if a power-supply is consuming energy by itself: those which do will become warm to the touch, even when apparently doing nothing.

Inefficient power supplies are just one cause of standby power usage. This Wikipedia article does a better job than I can of explaining why.

The other piece of the puzzle is how much electricity costs. This seems to be a bit of minefield: head over to uswitch to see. As I see it, the confusion arises because of upfront fees (a standing annual charge you may, or may not, pay to your supplier) and variable rates: typically a higher rate for the first N kilowatt hours (kWh) and then a lower rate, or in rare cases, the opposite (which is more sensible, from an energy conservation standpoint). To perform the calculation, I will use the cheaper rate as offered by British Gas (August 2008). So I'll go with 11 pence per kWh. A look at uswitch suggests that this is not untypical. This value has the nice effect that using 1 Watt of power constantly for a year will cost 1 pound!

ApplianceStandby powerAnnual saving (pounds)Kettle metric
TV, DVD Recorder etc.25 Watts20.081743
Desktop PC7-8 Watts6.93601
Digital Radio7 Watts6.46561
Battery charger3 Watts2.83246
Mini Hi-Fi8 Watts00

The "annual saving" is based upon my rough estimate of how long the appliance is turned off at the plug each day, multiplied by 365. The "kettle" rating is how many kettles boils that much energy gives you (one-litre, assuming 10 to 100 degrees, or 377K Watt Seconds). The zeros are for applicances which I can't be bothered to turn off (possible not rationally!)

An of quoted "low-hanging fruit" in energy saving is the low energy lightbulb. Let's look at a lounge light: a 100 Watt incandescent bulb replaced with a 20 Watt bulb, and assumed 6 hours a day usage, average, a year (this is, admittedly, a guess). This gives a saving of 19.27 pounds. Scaled up, changing lightbulbs is clearly a good way to save money, but one blub is comparable to the energy saved by turning off my (admitted old) TV.

Energy efficient bulbs currently cost 50 pence a piece in my local Morrisons: it seems very silly not to use them. The Wikipedia article linked above gives a long list of problems. I would cite lack of compatibility with dimmers as a problem (I've seen consumer dimmable bulbs, but they plug into a normal socket, and you can select the light level: they are not compatible with dimmer switches). They do take a while to warm up, although I don't see this as a huge problem, except in obviously specialised applications. I have noticed that the colour quality varies a lot: I prefer a whiter light (which seems brighter to the eye), and I'd recommend shopping around a bit. Spiral shapes are also more asthetically pleasing.

The good news is that most electronic products are getting much more efficient. For example, the analogous LG Mini Hi-Fi to my aging model is rated at 0.8W in standby mode. If you worry about such things, its the cheaper electronics which you need to watch: my new wireless phone appears not to be very efficient.

More on power meters

Recently, I was puzzled by the difference between the "watt" and "volt-amp" readings from my cheap power meter: my A-Level physics suggested to me that they should be the same. A trip to Wikipedia points out the difference; it's down to tricky AC effects, summed up in the Power factor. The physics is explained as the difference between real and apparent power. Real power is energy used per time unit; "reactive power" occurs when the circuit contains inductance and capacitance: an idealised capacitor will store energy during the positive phase of AC power, and then release it during the negative phase, meaning that no work is done; energy is simply moved about.

So, real power is where work is done, and this is what you get billed for. Reactive power is of interest to your power company, as this energy moving around has to flow through wires which have a (small) resistance, and hence get hot (as do transformers and so forth). A kettle, being a heating element, is a purely resistive load. Modern electronics use switched mode power supplies, and these are heavily reactive loads; as are motors in pumps for fridges and so forth. My power meter, for a kettle, displays almost identical readings for Watts and Volt-Amps, but for my laptop, the Volt-Amp reading is nearly twice that of the Watts reading!

So, upshot is that really I care about real power. To separate real power from reactive power, you need to study the way current and voltage vary over the cycle: as AC cycles at 50Hz in the UK, you need to sample these pretty often to get a good picture of what's going on. The contention is that cheap power meters don't really do this; in particular, they can underestimate the real power (the "watts" reading) somewhat. There's even a Wikipedia talk page argument about this! The main evidence comes from this photo where a Kill-A-Watt device shows an AC/DC transformer using less power than the DC side is loaded with! The whole thread is interesting, but sadly peters out before a conclusion is reached (user KVar reports no problems with his test of an LCD monitor and a cheap power supply-- one which doesn't have electronics to reduce the reactive load).

Not sure what to conclude here. Certainly my Kill-A-Watt seems to under-estimate the power usage of CF lightbulbs (assuming the ratings on them is correct).

Furthermore, ruminating on such things for too long leads you to thoughts such as: The electric meter in my basement is actually electronic, pretty small, and presumably doesn't cost that much, given that the power company installs them. I presume it works in a fairly similar way to the Kill-A-Watt devices, and hence cannot be that much more accurate. And this thing is used to decide how much I pay!

Head back to Adventures.