Now accurate are domestic electricity meters typically?

[Alan]

After a high electricity bill I've been trying to find out why. I've been checking my meter readings each day and so far (after 3 days) the usage is around 23kwh / day peak rate and 8kwh / day low rate (Economy 7). This seems very high to me. My house is a 3 bed semi, heated by gas (water also). Washing machine & dishwasher are always run at night in the E7 period. The only form of electrical heating is a 500w "frost watcher" in the [insulated] garage - I will fit my plug-in power watt / usage meter on this tonight to see how much it's on, but seems very little this time of year. No immersion heater. Most lights are compact flourescents, apart from ones which are on dimmers and a few low voltage fittings. (CFL's are in all the regulally used lights left on for long periods though). No electric shower or cooker. Oven *is* electric but used rarely, and only about 2kw IIRC.

There are 3 computers running 24/7, when I last checked the power used by these was about 60w each - will re-check. Will reduce this to 2 computers soon. Other stuff on 24/7 is a fish tank, again when I last checked total wattage was pretty low, and the heater doesn't come on much due to being in a warm room etc. Lights on for about 6 hours/day (fluroescents and Metal Halides). Will fit watt meter here after it's stint in the garage.

My equipment / power usuage hasn't really changed in the last few years, but looking at an electricity bill from 2001 the used units were way below what it is now.

I have an ammeter connected to a current transformer on the incomming mains supply, and even on a typical evening with lights on etc it reads about 4 amps, and during the day the meter is usually resting on it's end stop (0-80A scale).

So, I'm wondering of the accuracy of the electricity meter which was replaced about a year ago. Any tests I can do to confirm? I will have a better idea in about a week of power usage of equipment but to me I just can't see where a total of around 31 kwh per day is going!

Thanks in advance,

Alan.

[Wanderer]

You don't say - is it a solid-state meter, or an old-style with a rotating disc? If it's disc type, they are moderately easy to check with a bit of simple arithmetic, the name plate will have a figure for the number of revolutions per kwh.

Switch *everything* off at each individual appliance - not just flipping off the consumer unit - and see if the disc is still revolving. If it is, then either there's something still on that you've forgotten about, or the meter is faulty.

If the meter stops with everything switched off then switch on one single appliance with a bit of load, like a 3kw fire for a few minutes, and take an *accurate* timing for, say, twenty revolutions of the disc. You should then be able to work out by simple arithmetic whether the meter is running fast or not.

Say the name plate gives 80 revs per kwh. You have 3kw of load, so in 60 minutes you'd expect 240 revs. Simple ratio will tell you how many kwh's your 20 revs represents.

If it's a modern solid-state meter, sorry, but you've got to rely on the company testing it.

[Ben Blaukopf]

Turn off the power at the CU, and see if the meter records any usage. Turn on the CU and turn off all circuits, and check that the meter still isn't recording usage. Turn on one circuit at a time. Etc - you get the idea.

Ben

[Alan]

Sorry, modern solid-state meter :-(

[Christian McArdle]

It certainly sounds buggered. Alternatively the chavs next door are stealing it, or there is something plugged in that you don't know about or forgot about, like a fan heater in the loft or something.

Christian.

[Lee]

I'd recheck the PCs, unless they are old/low powered models, my guess would be that they are using a lot more than 60W each ;) My desktop uses between 150 and 300W depending on what it's doing. Whereas the laptop I use mostly now, is between 40 and 90W depending on whether I'm playing 3d games or not.

Also when my BiL had a 6ft marine tank it was costing him nearly UKP10/wk, lots and lots of low wattage stuff soon adds up when it comes to fish tanks - although the lights are usually the biggest load :) :)

You can roughly check the electronic meters by switching on a known load and counting the led flashes over a timed period.

Lee

[Lee]

Oops, missed a "just" there, but you know what I meant :)

Lee

[fred]

In article , Alan writes

The principle is exactly the same, measure the exact time between 2 readings ie. start when the smallest digit clicks over & then stop again when the smallest digit ticks over some (much higher) reading later.

My electronic meter has a 0.01kWhr resolution, so the worst measurement reading error you can have if measuring over ~1kWhr is 1% and if you start & stop exactly when the dials (or lcd) clicks over then it will be better than that.

The load you are using (1kw fire or 3kw fan heater) will certainly not be an accurate load, just measured one here (1kW) and it was 1088W so 8.8% error. That will be the limiting factor of your accuracy.

So you should be able to check accuracy within about 10% quite easily which may suit you as it appears you are looking for gross errors rather than the last 0.1%.

[Alan]

It is a 6' marine tank, but it's been running 5 years and it's only recently my electricity bill has gone up. Most pumps on the tank are about 5w each, and only 4 of them in total. I'll check load though but don't expect it to be much.

[Alan]

I see a weekend of testing coming up. Just found my power meter, blew the dust off and replaced the batteries. Ready for business now!

[Mike Harrison]

Fridge? Freezer..?

[Lee]

Ok, if you setup has stayed stable then you can obviously discount it :) Of course what happened in the case I mentioned was him was adding a UV sterilizer one month, then a couple of metal halide lights etc.

Lee

[Owain]

"Alan" wrote | After a high electricity bill I've been trying to find out why. | I've been checking my meter readings each day and so far (after | 3 days) the usage is around 23kwh / day peak rate and 8kwh / | day low rate (Economy 7). This seems very high to me.

If you were using electric storage heating, my first suspicion would be the timeswitch was erroenously putting the heavy off-peak load on in peak hours. Even though you are using plug-in timers for the washing m/c etc, it is still possible the timeswitch is switching your meter over to cheap rate at the incorrect times.

With that ratio of day:night units I wonder if E7 is the right tariff for you. Its daytime rates cost more.

A 500 W "frost watcher" will use 12 kWh per day unless its thermostat reduces down, which it may not in this weather, compare that to your 30 kWh a day and you can see it isn't much.

| I have an ammeter connected to a current transformer on the incomming | mains supply, and even on a typical evening with lights on etc it | reads about 4 amps, and during the day the meter is usually resting | on it's end stop (0-80A scale).

Switch off everything on your consumer unit and see if the current drops to nil.

Switch everything in the house off, and then put the circuits back on at the CU one by one, and watch for any increase in load.

Owain

[Andy Wade]

Doesn't it have a flashing LED? Mine does - it gives 1000 flashes per kWh, so the power being used (in watts) can be calculated as 60 * flashes_per_minute or 3600 / seconds_per_flash.

AFAIK all domestic electricity meters are accuracy Class 2, which means ±2% basic accuracy. AIUI this basic accuracy applies for a resistive load (unity power factor, no harmonics) for currents between the marked upper and lower figures (typically 20 - 80 A). For currents below the lower figure and for poor power factors (whether due to phase shift or waveform distortion) the accuracy is allowed to deteriorate somewhat.

[Ian Stirling]

Some models can be significantly inaccurate. I've measured 250% inaccuracy on some models of plug-in power meter for some loads.

[Ian Stirling]

This may be inaccurate. How are you measuring the 150-300W? My PC also read 150W+ on a plug-in power meter, but a proper power reading (using a scope) revealed that the plug-in power meter was overreading by a factor of 2.

300W is very, very unlikely, unless you have a multiprocessor PC with terabytes of storage, or you use a projector for your monitor.

[tony sayer]

Remind me which way that is:)(.....

[nightjar]

Custom PC recently ran an article on power supplies. They estimated a fairly modest desktop machine as needing 280W without reserves, 295W with headroom. Admittedly, you probably would not often need to use the DVD drive (25.5W) and the CD-RW drive (19.88W) simultaneously when also running everything else at full chat. However, they were the only significant power users that you might not want to use together. My CPU, graphics card and memory would need about 60W more than the ones they used in the example, so it is not that hard to top 300W on a home machine working flat out.

Colin Bignell

[Pete C]

Hi,

What is the frost watcher used for specifically?

cheers, Pete.

[Lee]

Err, yes that may be a bit misleading since it includes the monitor, Sorry about that. I measured the voltage across a shunt resistor with a true-RMS Fluke DMM. I *think* I did the maths properly. Hmm...maybe I should measure it again.

For the laptop I just measured the DC input current - can't get that wrong ;) :)

Lee