Measuring power usage

[DaveOC]

Hi, I wanted to check my electric meter for accuracy and was told a clamp meter would be best. Can anyone tell me how to do this or is there a better way?
Thanks

 

[ericmark]

The old and new both are clamp on meters, left one was cheap bought it in Hong Kong before take over, so shows it's age, and it will show 0.01 amp, one to right less than a week old will show 0.001 amp and also works with DC.

Clearly would not have got second if first could do it all. But they show current in real time, so one can assume power factor of unity and 230 volt to work out watts, but it will not show the same as the electric meter, as they are measuring different things, the clamp on measures power the electric meter measures energy.

So a watt is a joule per second, and the joule is a very small unit, so since 3600 seconds in an hour, the are 3600 joules in a watt.hour. So we use kW.h with the electric meter but kW with the clamp on (after some maths) so they don't measure the same thing. May seem odd but there is no time in a kW/h (the two times cancel each other) but there is time in a kW.

A device like this will measure nearly kW/h I say nearly as it has no power factor correction and assumes 230 volt. I have the plug in 13 amp version and would assume similar so you will get some thing like this I am using it to see if beer is warm enough, but in many ways the plug in monitors are better, what seems odd is in many ways the non computer connected device tells you more.

However to show you that picture I had to re-boot the device, don't know when it stopped, but clearly you could not use that to show errors with an electric meter, it would need to be a calibrated data logger.

 

[Taylortwocities]

A simple clamp meter will only show you the instantaneous power (kw). So you can see the amount go up when you turn on the kettle.
But to check the accuracy of the electric meter, as mentioned above, you need a calibrated data logger. And you need one that is going to display the same units as the lekky meter (kWh) over a set period of time. Then you can compare apples with apples.
There are many devices on the market, if I were doing this, I would buy a Shelly EM.

Shelly EM - Shelly Cloud

shelly.cloud

 

[ericmark]

if I were doing this, I would buy a Shelly EM

If I was doing it I would hire the data logger. Remember you will need the calibration certificate.

 

[Taylortwocities]

If I was doing it I would hire the data logger. Remember you will need the calibration certificate.

Hmm
Depends on how deep your pockets are

Hire of Fluke 1730 for a week would be about £130. You can buy a Shelly EM and a clamp for half that and you have it FOREVER!

 

[JohnW2]

So a watt is a joule per second, and the joule is a very small unit, so since 360 seconds in an hour, the are 360 joules in a watt/hour.

Actually 3,600 seconds in an hour (that was presumably a typo on your part). More importantly, not "watt/hour" but, rather watt.hour (i.e. watts x hours)

So we use kW/h with the electric meter but kW with the clamp on (after some maths) so they don't measure the same thing.

As above, the meter measures kilowatt.hours (kWh), not kW/h.

May seem odd but there is no time in a kW/h (the two times cancel each other) but there is time in a kW.

Coupled with the above, that seems unnecessarily confusing. As you are aware, it's really very simple ...

With constant power ....
P Watts consumed for T seconds equates to (P x T) Joules of energy expended
... or, in more practical units ...
P kW consumed for T hours equates to (P x T) kWh of energy expended

[ with 1 kWh being equal to 3,600,000 Joules ]

I would think that " kW/h " would be a useless, and essentially meaningless, measure of anything. If you had, say, a 2 kW load running for 50 hours, that would equate to 40 kW/h - but, as far as I can see, that information would be of little use to man or beast!

Kind Regards, John

 

[JohnW2]

... you could not use that to show errors with an electric meter, it would need to be a calibrated data logger.

But to check the accuracy of the electric meter, as mentioned above, you need a calibrated data logger. ... There are many devices on the market, if I were doing this, I would buy a Shelly EM. ...

If I was doing it I would hire the data logger. Remember you will need the calibration certificate.

I don't think that 'calibrated' (or 'calibration certificate') are really relevant. If one wants to check the accuracy of a supplier's meter, all one really needs is a device which one is fairly confident will gives reasonably accurate readings.

If such a device appears to show that the supplier's meter is giving significantly erroneous readings, and reports that to them, the first thing they will do is install a 'check meter' of their own, and all subsequent 'decisions' (and 'actions') will be based upon what their 'check meter' reveals.

I find it hard to believe that they would ever make decisions and/or 'take any action' on the basis of a consumer's measurements, even if they were derived from a device which came with a 'calibration certificate'.

Kind Regards, John

 

[EFLImpudence]

[ with 1 kWh being equal to 3,600,000 Joules ]

I would think that " kW/h " would be a useless, and essentially meaningless, measure of anything. If you had, say, a 2 kW load running for 50 hours, that would equate to 40 kW/h - but, as far as I can see, that information would be of little use to man or beast!

Struggling with the last sentence.

Wouldn't 100kWh in 50hours be equivalent to 2kWh/h?

 

[JohnW2]

Struggling with the last sentence. Wouldn't 100kWh in 50hours be equivalent to 2kWh/h?

It would - but eric was not talking about kWh/h - he was talking (writing) about kW/h.

Kind Regards, John

 

[ericmark]

If one wants to check the accuracy of a supplier's meter, all one really needs is a device which one is fairly confident will gives reasonably accurate readings.

I see your point, if the meter has stopped, and you plug in a 3 kW heater for 20 minutes and it shows 1 kW.h then not far out.

As to using a clamp on to see it really is 3 kW that's icing on the cake. But you need to be able to switch off all loads.

When I got my first plug in tester I tried all sorts of electrical equipment to see what they used. The Sky box was a surprise, but most items were as expected. The other one which caught me out was fluorescent lamps, they were 110 volt 58 watt, so quick look assumed ½ amp each, so 16 amp supply should run 32, well to be on safe side, put just 25 on a string. The 16 amp MCB tripped, and I found over 20 amp draw, cure was to use 127 volt tapping not the 110 volt one.

But if it is found to be faulty, how far back will they go?

 

[JohnW2]

I see your point, if the meter has stopped, and you plug in a 3 kW heater for 20 minutes and it shows 1 kW.h then not far out.

I'm not sure what you mean by "the meter has stopped". If you mean that it has 'stopped working' (recording new usage) then one presumbly would not need to make any measurements to confirm that there was a problem!

If the meter was still 'working' and showed something like 1 kWh usage when a 3 kW heater was run for 20 mins (or better, if it showed roughly the right change in readings when the heater as run for longer periods than that), then one could conclude that the meter was "not far out". However, as you go on to say, that would require all other loads to be switched off, which may well be a problem.

However, I was thinking more about using some sort of clamp meter (on tails) over an appreciable period of time (maybe a day or three, or even more). Even given the need to 'guess' voltage and PF (perhaps aided by measuring voltage from time to time) that ought to show whether the meter readings were in the right ballpark.

As to using a clamp on to see it really is 3 kW that's icing on the cake. But you need to be able to switch off all loads.

One could use some sort of cheapo plug-in monitor to confirm that the heater was actually using 3 kW. However, as you say, that would only be useful if one could switch off all other loads.

.... The other one which caught me out was fluorescent lamps, they were 110 volt 58 watt, so quick look assumed ½ amp each, so 16 amp supply should run 32, well to be on safe side, put just 25 on a string. The 16 amp MCB tripped, and I found over 20 amp draw, cure was to use 127 volt tapping not the 110 volt one.

Don't forget that MCBs are current-operated. Even if each of the lamps were only drawing, say, 58W, it could have been appreciably more than 58 VA, hence potentially a fair bit more current than you were expecting.

But if it is found to be faulty, how far back will they go?

That probably depends upon the direction of the error and, if the meter has been over-reading, how much fighting the consumer is prepared to do! (I imagine that only 'particularly honest' consumers would report a meter which they believed was consistently under-recording their usage!).

If the supplier wanted to investigate the matter conscientiously, I suppose they could look back at the history of meter readings over a long period of time. If the meter had 'become faulty' at some point in time (rather than having 'always been inaccurate'), then one would expect to see an abrupt change in the rate of apparent energy usage - in which case one could retrospectively re-estimate usage subsequent to that change on the basis of usage prior to the change.

If examination of historic records suggested that the meter had 'always been inaccurate', then there's nothing much they could do other than guess/estimate what the consumption 'should have been' - but that would be a bit iffy!

Kind Regards, John

 

[pcaouolte]

The Shelly EM recommended by taylortwocities above is adequate to test the accuracy of the meter. It uses a current clamp on the live feed between the meter and the CU. Take a meter reading, connect the current clamp of the Shelly EM. Carry on using electricity as normal for a period of time (perhaps a day or a week). Read the meter again. If the kWh recorded by the Shelly EM (visible in the Shelly app) is the same as the difference between the two meter readings then the meter is ok. The Shelly EM is pretty accurate as it is connected to the mains supply to read voltage and current simultaneously.

 

[JohnW2]

The Shelly EM recommended by taylortwocities above is adequate to test the accuracy of the meter.

Indeed - I would say 'more than adequate'. Any sort of clamp-based device, even if without any wired connection (hence requiring guessing/estimation of voltage and PF) is likely to be adequate for identifying any significant inaccuracies of the supplier's meter, particularly if used over a reasonable period of time (ideally at least a few days).

In my, albeit limited, experience, in the great majority of cases in which a consumer suspects the supplier's meter, there's actually nothing wrong with the meter, the problem being their underestimation of their electricity usage! However, on the few occasions in which the meter is apparently inaccurate, the inaccuracy is (again, in my limited experience) usually fairly 'gross', and easily detected by any clamp-based device, as above.

It may be time to again wheel out my illustration of how measurements from my OWL monitor (clamp, but no direct mains connection) have compared with figures based on my supplier's meter over the past 5+ years. The system uses 'constant' (reviewed annually, but rarely changed appreciably) estimates of average voltage and PF and, as can be seen the difference between my and my supplier's figures are as close to zero as makes no difference - in fact, currently a cumulative difference of only about 20 kWh in about 24,000 total kWh since April 2017 (about a 0.08% difference) ...

Kind Regards, John