Measuring power consumption of immersion heater?

[tom.harrigan]

Are you suggesting it somehow compensates for the power factor of the supply automatically?

T

[Timothy Murphy]

This doesn't seem to affect power consumption in the slightest - it remains just under 30 watts (with the TV off), though the light on the Sky box turns from green to red.

[Andrew Gabriel]

Just checked a 2 year old Panasonic LCD, and it's 0.5W standby. It doesn't have a TCO'03 sticker on it, but this brings it within the TCO'03 requirements (1W max standby) and thus able to be sold in some countries where that's a requirement.

That seems an awful lot, or the power meter is not handling loads with power factor < 1 correctly.

[Rumble]

Is this a trick question? The supply doesn't have a power factor; only the load has a power factor.

[tom.harrigan]

It wasn't supposed to be a trick question. The supply has a power factor by definition. The question is whether it equals unity or not. Are you implying that a domestic supply is guaranteed to have a power factor of 1?

T

[John Rumm]

Its an odd question granted, although you could argue that the supply may have a distorted voltage waveform as a result of local high current non-power factor corrected loads. This (and other things) will introduce harmonic noise to the supply. The result would be lowering of the effective power factor seen at the load.

[Andrew Gabriel]

In which case you'll have to give your definition, because it is meaningless with the standard definition of power factor.

[tom.harrigan]

Well, it's not _my_ definition it's _the_ definition. Perhaps you could Google for one?

T

[Andrew Gabriel]

I am very familiar with the standard definition which applies to loads, having designed and built a true power meter myself some 25 years ago, and having explained power factor many times here over the years.

You'll have to point out a definition of power factor of a supply, as that's not covered by the standard definition. Google draws a blank too. The only related thing I could think of would be a measure of the worse power factor load a particular supply can drive, but even that's meaningless as supplies are rated in [k]VA and in many cases can handle worst case loads with a PF of zero.

[dennis]

They are charging me per kwHr so they had better have the correct power factor.

[tom.harrigan]

If they are charging per kWh, power factor is irrelevant.

T

[Adrian C]

Tom, the issue is *how* the meter makes a measurement that can be accurately displayed as "real power" - Watts - what the domestic customer is billed for.

The device can measure voltage of the supply. The device can measure current drawn.

That's VA.

To arrive at Watts, it has to calculate (guess, fiddle) what the power factor is and this is where the over reading inaccuracies result, especially with low current devices, and cheapo gadgets of this sort.

Power factor of the supply? What's that then?

[John Evans]

?????

Not so!

The numerical power factor eg 1, 0.9 etc - is a measure of the phase angle between the LOAD current and the applied voltage. It is also the measurement of the ratio of real power and apparent power or the ratio of watts (W) to voltamperes (VA).

Look at Google:

power companies worry about the overall power factor of the supply for obvious reasons.

[tom.harrigan]

I'm clearly trying to discover whether the power factor I measure is affected in any way by the electricity network. If you put a purely resistive load across the supply coming into your house, are you guaranteed to measure no phase difference between the voltage and the current? If the phase difference is zero, then the power factor of your supply is 1.

There's a whole load of stuff on the web about the lengths the National Grid go to power factor correct their supply. Perhaps you should ring them to tell them they are wasting their time?

T

[dennis]

If the load is resistive there is never any phase difference between current and voltage it is governed by ohms law..

[tom.harrigan]

It's something the National Grid are obliged to keep within certain limits: 0.85 lagging - 0.95 leading.

T

[tom.harrigan]

....

You have just contradicted yourself.

T

[John Rumm]

IIUC, it is not effected by the network as such, although it may be affected by other users of it.

With purely resistive loads then it is not an issue. For loads with a reactive component then the reactive component will typically be the major influence on the power factor. However the quality of the waveform that you are supplied with can further influence it.

They go to some effort to mitigate the effects their users have on the supply. A big industrial user (or the cumulative effect of many smaller ones) pulling large currents from the supply that are not phase aligned with the voltage, can end up distorting the supply waveform. This leaves the waveform non sinusoidal and hence introduces other frequency components into it. These will interact with the reactive elements of any load differently than would a plain 50Hz supply.

[John Rumm]

Well sort of...

The concept of power factor only really has any meaning when you know the characteristics of the load. With a resistive one, the power factor is 1 regardless of the supply.

What the grid are obliged to do is provide a stable supply, and ensure that the required true power output of the generating plant is available to the grid. Obviously this would not be the case if capacity on the network is going to be absorbed by large reactive currents flowing around it. They also need to ensure local voltage stabilisation.

The limits you specify above are limits for the *load*. i.e. they must meet certain performance criteria when the connected load lies within those limits. With applied loads outside of these limits they are not required to achieve the same standards.

A distribution system itself will also have a number of reactive components within it, like the cables (transmission line effects), and transformers. The effects of these will also need to be compensated for.

[John Rumm]

I think he was just being a bit sloppy with terminology. However it is worth remembering that the distribution grid is also part of the load.