'Surprisingly' (to me) low PF of gadgets

[JohnW2]

This is an observation, rather than a question.

In the course of playing with the 'wireless PIR' I recently wrote about (see here ), and the 'dusk to dawn' photocell unit I am using with it, I was initially a little taken by surprise.

My house-wide 'energy monitoring' (clip on, hence VA, not W) indicated that the receiver part of the wireless PIR and the photocell unit were each taking about 30 VA continuously whilst powered, which did not sound too good.

However, furrher investigation indicated that both had PFs of around 0.1, so were in fact each 'consuming' only about 3W.

Looking at the devices, the photocell unit certainly, and the PIR receiver probably, use a capacitor as the voltage-dropping element to provide power for the electronics, the relays that both contain are primarily inductive and the electronics presumably take very little (so not much resistive element to the load), so I imagine that both devices look like predominantly reactive loads.

What I observed is therefore reasonably explicable, but I have to say that it hadn't previously occurred to me that the PFs of such devices would be so low.

Kind Regards, John

 

[AdrianUK]

I have, over the last few years, observed the PF of my house slowly changing as the filament lamp & transformer fed loads disappear and more and more SMPS, or variants of it, powered devices appear.

I have my own 'whole house' energy monitoring device, which is a measurement centre (a ISKRA MC750 to be precise) which does take a voltage reference and is therefore able to correctly distinguish between apparent and real power. The device also logs V, I, F, S, W & P at 1 second intervals.

At the time of writing, the house demand is 4.8A, 246V, 968W, - 622VAr, 1230 VA with a PF of 0.84 CAP. (These figures don't exactly work out because I can't capture them all at the same instant). Looking back at the log, its rare that the PF is inductive and the only time the PF comes close to unity is when the immersion heater cuts in and the 3kW resistive load swamps everything else!

Also of interest is the THD, quiet high at 42% with 14% 3rd order & 29% 5th order (The MC750 does FFT upto 63rd harmonic on both V & I) . I think the high THD is down to 'cheap' power supplies, maybe in LED lamps(?)

The current waveform is ..... interesting!

 

[JohnW2]

I have, over the last few years, observed the PF of my house slowly changing as the filament lamp & transformer fed loads disappear and more and more SMPS, or variants of it, powered devices appear. ... Looking back at the log, its rare that the PF is inductive and the only time the PF comes close to unity is when the immersion heater cuts in and the 3kW resistive load swamps everything else!

Interesting. Whilst there certainly is an increasing number of loads which are far from resistive, as your immersion indicates, most of the largest loads, which are therefore going to dominate the overall picture, are almost entirely resistive (immersions, heaters, cookers {other than induction hobs}, showers etc.).

As a result, although I have countless LEDs and things powered by SMPSUs/similar, my overall/average PF remains very close to unity. My monitoring system cannot measure it directly, but comparing my figures (kVAh, and even that assuming constant voltage) with my electricity meter (kWh) over any period of time always indicates a PF very close to 0.98 - and the meter itself, which provides a figure for kvarh, always gives a figure very similar to that.

Also of interest is the THD, quiet high at 42% with 14% 3rd order & 29% 5th order (The MC750 does FFT upto 63rd harmonic on both V & I) . I think the high THD is down to 'cheap' power supplies, maybe in LED lamps(?) .... The current waveform is ..... interesting!

As you say, very 'interesting'. I again assume that this waveform represents a time when there were no major resistive loads?

Kind Regards, John

 

[ericmark]

I have two plug in units, only the one that shows PF but never seen it worse than 0.8.

 

[JohnW2]

I have two plug in units, only the one that shows PF but never seen it worse than 0.8.

Have you tried it with small devices which use cheap SMPSUs or 'worse' (cheap LEDs etc.)?

I've just tried a couple of meters on a range of cheapo LED lamps/bulbs, and got PFs ranging for 0.12 to 0.72.

Kind Regards, John

 

[AdrianUK]

Yes, the current waveform was, pretty much, the same time as the powers I quoted. The loads which were on included 3 PCs & LCD monitors, 42" LCD TV, various LED lights including 7 x 5W LED candle on a dimmer (and dimmed to around 30% ish). The immersion heater, obviously, hadn't cut in so pretty much all of the load was 'electronic'.

At the moment, its crept up a little, PF is .92 CAP with 1.11kw & 1.21 kVA, I'm not quite sure what has come on-line... maybe the central heating (pump adding and inductive element to cancel out some of the C, maybe)....

 

[JohnW2]

At the moment, its crept up a little, PF is .92 CAP with 1.11kw & 1.21 kVA, I'm not quite sure what has come on-line... maybe the central heating (pump adding and inductive element to cancel out some of the C, maybe)....

Interesting. Compared with early earlier waveform, it's now much less 'symmetrical' (positive and negative half cycles now differ appreciably).

Kind Regards, John

 

[Colin Brenton]

Interesting thread! What does "CAP" stand for in relation to power factor? That's a new one on me...

Thanks

 

[JohnW2]

Interesting thread! What does "CAP" stand for in relation to power factor? That's a new one on me...

I presume it means that the reactive component of the load is net capacative (i.e. current 'leads' voltage by the phase angle in question). With net inductive reactance ("IND" I guess ), current lags behind voltage.

Loads such as we are discussing (LED lamps/bulbs, and anything electronic using an SMPSU or similar) will be predominantly capacitative, whilst ones with significant motors etc. will be predominantly inductive. If both are present, they will partially 'cancel', thereby bringing PF closer to unity.

Kind Regards, John

 

[ban-all-sheds]

I presume it means that the reactive component of the load is net capacative (i.e. current 'leads' voltage by the phase angle in question). With net inductive reactance ("IND" I guess ), current lags behind voltage.

CIVIL.

 

[Simon35]

CIVIL.

Not come across that before, and it took me a moment to work it out. Very good.

 

[JohnW2]

Not come across that before, and it took me a moment to work it out. Very good.

I had come across it, but the way I was taught to remember was that with "L" (as in inductance), current "Lags". That works because one always talks/thinks about current leading or lagging voltage, rather than vice versa.

Kind Regards, John

 

[ban-all-sheds]

I had come across it, but the way I was taught to remember was that with "L" (as in inductance), current "Lags".

How were you taught to not remember that with "L" (as in inductance), current "Leads"?

 

[JohnW2]

How were you taught to not remember that with "L" (as in inductance), current "Leads"?

Good question, and I knew someone would ask that ( suppose I should have pre-empted the question!), and I don't know the answer - but for some reason it seems to have worked for me. Maybe there were additional bits to what I was taught, or maybe I've remembered it slightly wrong. If the layer of dust is not too thick, I'll see if I can find my old notes!

Whatever, it has somehow got indelibly engraved on my mind that "capacitive current leads".

Kind Regards, John

 

[ban-all-sheds]

Oh well - CIVIL works for me....