Is the power factor of some energy saver lamps so low that they actually consume more VA than incandescent lamps of equivalent light output? If a 100W incandescent is replaced by about a 20W CFL, as recommended by lamp manufacturers, that would imply PFs for the new lamps below 0.2. Admittedly we may wish to use higher wattages than this to get more light, perhaps to allow for some new lamps' greater fall of efficiency with aging.

At present the new lamp types are being sold to the public who buy them on the assumption that they can be used as plug-in replacements. Here in UK consumers have little choice, as the sale of GLS tungsten lamps is now banned, except for low wattages.

A quick look at Google suggests that fluorescent types might have PFs more like 0.5, but some LEDs may be very bad indeed. Are we heading for trouble with these latest types? It would be interesting to hear from anyone with real knowledge of the subject.

 

I moved your post because it was a topic hijack and interesting enough to get some commentary.

 

Typically, to lower costs they use a diode bridge and small filter capacitor. This makes the power factor from 0.4-0.5, which is poor, but a poor power factor doesn't really matter for the average consumer who only pays for unity power factor. It might be a concern for a company which is charged based on power factor, in which case active PFC lamps also exist.

The replacement of a 100W bulb with a 20W CFL does not indicate the power factor is 0.2. CFLs are about 4-5x more efficent than incandescents. On a grand scale however, they are only 8% efficient compared to an ideal light source.

I'm not a major fan of the CFLs: they do save energy (this is why I use them), but I really think we should be putting more into LEDs as they are more likely to be the most common future lighting technology.

 

Here is an article on the comparison. http://open4energy.com/forum/home/lig/energy_saving_light_bulbs_incandescent_cfl_0912142219


I have never heard of "different" types of PF loss, Given the chart that was supplied, and that at least for consumers here, who are not charged for PF, It doesn't apply. The losses appear to be insignificant compared to the overall savings.

 

Here is a thought... given my limited understanding of PF and CFL's, power factor losses are largely due to inductive reactance. (Industrial locations) If the power factor losses in CFL's are capacitative, won't there be a improvement in overall PF from the use of CFL's?

 

Typically, to lower costs they use a diode bridge and small filter capacitor. This makes the power factor from 0.4-0.5, which is poor, but a poor power factor doesn't really matter for the average consumer who only pays for unity power factor. It might be a concern for a company which is charged based on power factor, in which case active PFC lamps also exist.

The replacement of a 100W bulb with a 20W CFL does not indicate the power factor is 0.2. CFLs are about 4-5x more efficent than incandescents. On a grand scale however, they are only 8% efficient compared to an ideal light source.

I'm not a major fan of the CFLs: they do save energy (this is why I use them), but I really think we should be putting more into LEDs as they are more likely to be the most common future lighting technology.

My point was originally made in reply to a post on another thread, which seemed to imply that new lamps might take more current than their predecessors. This seemed unlikely, so I tried to estimate what the PF would then need to be, expecting to get a ridiculous answer. Assuming the tungsten lamps to have unity PF, but five times lower efficiency, the new lamps would have to have (less than) 0.2PF to consume more current when replacing like for like light output.

What surprised me was that on searching on the internet, I found some claims that LED lights could have as low as 0.15 PF. A this level, even allowing for their greater efficiency, they really might take more current than the lamps they replaced. That does sound worrying, perhaps in terms of harmonic generation as much as anything else.

 

My point was originally made in reply to a post on another thread, which seemed to imply that new lamps might take more current than their predecessors. This seemed unlikely, so I tried to estimate what the PF would then need to be, expecting to get a ridiculous answer. Assuming the tungsten lamps to have unity PF, but five times lower efficiency, the new lamps would have to have 0.2PF to consume less current when replacing like for like light output.

What surprised me was that on searching on the internet, I found some claims that LED lights could have as low as 0.15 PF. A this level, even allowing for their greater efficiency, they really might take more current than the lamps they replaced. That does sound worrying, perhaps in terms of harmonic generation as much as anything else.

Yes, they may take in more current. But AC is an alternating waveform and although they make draw in more current at the highest point of the waveform on average the power is lower. Lower PF does mean higher losses in wiring, which is why bigger organisations are billed on it, because it costs the power company more to provide power at a lower PF.

Many of the cheaper LED lights use just a capacitive power supply, or even a resistive supply! These are horribly inefficient - PF is also pretty bad. Look for ones with an active PFC current-mode supply - but you pay the price.

 

Oh dear, I seem to be having some trouble getting my point over here. Yes, clearly we must accept that "energy saving" lamps consume less power, if they come anywhere near living up to their manufacturers' claims.

The question I wanted to address was simply whether or not the RMS current drawn by the new lamps was likely to exceed that drawn by their tungsten equivalents. It was suggested that transformers, cable etc. could require uprating to support the new lamps.

I was originally very sceptical about this, but power factors as low as 0.15 (if accurate) seem to imply that this could be the case. If the lamps have narrow conduction angles, the peak currents could also be considerably more than (√2)*RMS, but that's a subject in itself.

Let us hope that we can forget about the really cheap and nasty designs as being too inefficient or unreliable to be widely used, but what about the more common switched mode systems? My gut feel is that there ought to be some regulations imposing minimum values- maybe I'm out of date here, and this is already the case in some countries?

 

Oh dear, I seem to be having some trouble getting my point over here. Yes, clearly we must accept that "energy saving" lamps consume less power, if they come anywhere near living up to their manufacturers' claims.

They do, just watch your electric bill go down.

The question I wanted to address was simply whether or not the RMS current drawn by the new lamps was likely to exceed that drawn by their tungsten equivalents. It was suggested that transformers, cable etc. could require uprating to support the new lamps.

I was originally very sceptical about this, but power factors as low as 0.15 (if accurate) seem to imply that this could be the case. If the lamps have narrow conduction angles, the peak currents could also be considerably more than (√2)*RMS, but that's a subject in itself.

RMS current should be lower in energy saving lamps. However, as you mention peak current may be much higher. It is peak current that matters most, due to I^2 * R. If a CFL draws say 1A pulses 10% of the time and the wiring has a resistance of about 1 ohm then the power wasted is 1W/10 = 100mW, but if it draws 100mA 100% of the time the (*exactly the same amount of power output on the lamp but in a different form*) power wasted is only 10mW. This is only an example, but it shows how peak current is important. So yes, wiring may need to be upgraded and in very extreme cases the wiring may end up wasting more than is saved.

Let us hope that we can forget about the really cheap and nasty designs as being too inefficient or unreliable to be widely used, but what about the more common switched mode systems? My gut feel is that there ought to be some regulations imposing minimum values- maybe I'm out of date here, and this is already the case in some countries?

There is little regulation on PFC. For power supplies, to claim Energy Star 80 PLUS there must be some kind of PFC, usually active. And it must be active PFC, not just switch mode - technically speaking, a CFL ballast *is* a switch mode converter.

 

How about some actual data... I plugged in the CFL that has been in my reading lamp next to my recliner for a little over a year and used a Kill-a-Watt to measure its characteristics:

At 118.3 VAC line voltage:
0.32 A
22 W, 38 VA
Power factor = 0.59

75 W GE incandescent bulb:
0.62 A
72 W, 73 VA
Power factor = 0.98