The best thing to do with paper is to leave it on a shelf, or if you are not going to keep it then do not use it in the first place.

If it goes into land fill then, provided that you don’t let the anaerobic bugs at it (thus creating methane, which is Very Bad), again we have an excellent form of carbon sequestration. What we want to be doing is growing trees (made of CO2 and H2O), chopping them down, and then leaving them to sit there forever – underground or on my book shelf, does not matter really!

Recycling paper is very energy intensive, and I remain convinced that it is not a large amount more to extract it from the trees in the first place, but I would be delighted if someone could prove to me otherwise. My belief is, I admit, based on anecdotal descriptions of the processes involved, but my sources include the CEO of a large recycled paper supplier.

As for burning, if you use gasification then the efficiencies are much higher (well over 50%), and some serious green folks with money to invest are looking to build biomass-fired power plants. That biomass is mostly fast-growing trees. May as well be paper!

Peter’s estimation was that non-specialist kit (which costs around £400) might be 10-20% inaccurate when measuring low-wattage AC. Even then though, a measurement of 0.5W is still definitely below one Watt for the standby figure. The innacuracies are not an order of magnitude, and I don’t believe this detracts from my fundamental point;

My view is that technology should be the solution wherever possible, not changing behaviors and causing people minor inconvenience and irritation. Low standby power is one example, as is using the auto-hibernate option rather switching PCs off. Additionally, in our country where we need heating for a large part of the year, the wasted energy just warms the home and marginally reduces fuel bills.

Also, the level of standby power in most modern equipment (which even with possible inefficiencies is still well below 1 Watt) makes it an inconsequential thing to worry about, and has the real danger of making people think they are doing some real good when they are not.

Therefore, I believe we should encourage people to do things that have a real effect (using public transport, turning off lights etc) rather than fretting about the trivial.

Obviously if you are a super-informed individual who cares about every bit of your carbon footprint then more power to your elbow! When trying to make changes in behavior of the less-well informed, though, we should focus on the big issues.

Basic multimeters cannot measure voltage and current at the same time, which means there is no way to determine the power factor, and so the real power cannot be calculated (wattage is volts times amps times power factor). There is at least one multimeter that has a plug-in attachment so that it can make these measurements. And safely. But it isn’t cheap. And it’s not much better than a power monitor or clamp meter. None of them have enough sensitivity for small wattages.

And the worry with cheap meters is always how they determine an RMS value for the current. Many have simple average-responding circuits that only work with currents that follow a pure sine wave. Give them a waveform with harmonics and the readings will be incorrect.

The figure we use for losses is 15.5%, which comes from industry data (1.5% high-voltage grid, 6% distribution and 8% production). If you have reason to believe this is wrong, it is important that it’s publicised.

Just as an example of what can go wrong if we’re not careful, suppose a meter shows 2W for the standby power for some appliance. If the meter lacks sensitivity, or has not been calibrated for non-sinusoidal currents, the true wattage could be a lot higher. It might even fluctuate over time and our reading might not be representative. But let’s call it 3W. If the site power factor (after attempted correction) is 0.6 (just to make the maths easy), then 3W puts a load on the supply of 5VA. Taking losses into account, that’s equivalent to 6VA at the generators. If the appliance is left on standby for a year, the generators would average 24 kg of CO2. But if all appliances were no longer left on standby, the emissions would tend to be reduced at the highest rate (by burning less coal) and the saving for our example appliance could be as much as 47 kg of CO2. Not 8 kg, which is the figure we would get if we based the calculation solely on the 2W displayed on the meter.

BTW I appreciate that AC power is a complex subject. If I’ve used any terminology that you’re not familiar with, I’d be happy to explain it.

Anyway, you are quite wrong that equipment to measure millwatts is expensive!! Any standard multimeter will do it. The only trouble is that they don’t usually have a handy way to do it safely (I would not recommend my method).

Equally, transmission losses are only a few percent at most. We looked into that when determining our “real” data centre footprint, and worrying about variations during the day due to different sorts of power station being online is taking it to an extreme! Again, a quick sanity-check reveals that cannot be a significant factor in the grander scheme.

Given that good test equipment costs £hundreds, I wouldn’t expect a £10 meter to be very accurate at small wattages, especially with the non-sinusoidal loads presented by switched-mode power supplies. The meter has to detect a few milliamps and it has to determine the power factor from harmonics in the waveform, which is a lot to ask at the price.

You mentioned putting “a clamp meter around the power cable” to measure the standby power consumption of PCs. I hope you’re aware of the need to separate the live and neutral wires and put the clamp meter around just one of them. By clamping the whole flex, you’d be measuring the difference between the live and neutral, and getting a false reading close to 0.

In any case, when calculating CO2 emissions, what matters is not the real power of the appliance but the apparent power (normally expressed in VA rather than W) “seen” by the electricity generators after power factor correction. This is not trivial to determine, but will be higher than the wattage shown by the meter. A full analysis would take into account transmission losses, and the fact that CO2 per VA varies throughout the day because of the different mixes of power plants that are brought on line. CO2 calculations can only really give ballpark figures. That doesn’t stop web sites such as http://www.energysavingtrust.org.uk quoting figures with 3 or 4-digit precision!

A quick trawl of manufacturers’ data for a random selection of current models gives a wide range of standby powers: 0.15W (for a Philips set), 0.19W (Sony), 0.7W (Toshiba), 1.2W (JVC) and 6W (Beko). Which demonstrates the differences between leading-edge makes, good makes and cheap makes. Since it takes a while for leading-edge technology to filter through to other manufacturers and to consumers, I suspect no one knows what the average actually is for the TVs currently in British homes. But I guess that 2-4W figure sounds plausible. Dunno really.

One point I would make is that it’s not just an LED that is powered in standby: it’s also the circuits that detect “turn on” signals from the remote control or connected equipment. And in many TVs this is now configurable; it is one way that manufacturers are obtaining ultra-low consumptions, but it means that many consumers will not be operating their TVs at the lowest possible settings. e.g. My mate’s new Sony has a 0.19W standby. But configure it so that a HDMI signal can turn it on and the standby consumption trebles. It also has a “quick start” mode which uses 16W for the first 2 hours in standby, giving an average during a day of over 2W.

The one DVD recorder that I’m aware of uses either 2W or 15W depending on which standby option is selected. In general, I don’t think it’s TVs that are the worst offenders when standing by. The boxes underneath them use more (partly because they need to power circuits to loop through the RF and SCART signals). In our house, the boxes use 13W between them in standby. The main culprit is the Freeview receiver, at 7W. How representative that is, I have no idea. But I suspect people who unplug their TVs and leave their digital receivers on standby have been misinformed. I’d be interested to know whether you have any better figures than mine for the under-TV boxes.

As for what a modern TV (or stereo) consumes on standby, I have looked at both the manufacturers’ data and measured it myself, and it is indeed very small (only a few hundred mili-Watts). The main difference, I believe, is in improvements in switch-mode power supplies (the bit that converts 200+V AC into low-voltage DC) which used to be very inefficient and had a constant power draw of a few Watts even when the devices they were powering were using very little, for example just keeping the LED glowing.

I did not use a clamp meter actually – I used one of the “plug in and measure” types, a Prodigit Electronics 2000MU, reviewed here. They are supposed to be pretty accurate! If you are getting some different figures I’d love to hear them though – I have only tried it on kit from big-name companies like Panasonic, Phillips & Sony.

Can you tell us how you came up with the 0.5W figure for TVs: is this from manufacturers’ data or did you measure some samples yourself with your clamp meter?

What make and model clamp meter are you using?

Can you confirm that you’re just putting it around the device’s own mains flex without modifying the flex in any way?

As for hibernate, I have not come across a case of it not working in the last few years…. admittedly, I have not tried using Vista in a serious manner personally (I gave up Windows some time ago) and it would not surprise me if it failed to hibernate reliably