Power Factor Discussion

STEVETERRY

Well-Known Member
This thread began as a side conversation in the thread: Experience with Philips Ambient LED in the Home. I have split it out into it's own thread to make it easier to search for in the future.
Gaff


Like many of you I have been playing around with LED bulbs in the home. While I have had good success in a few applications like under counter task lighting, I had not found anything that I really liked for replacing a plain old incandescent A lamp.

Well - that has changed. at LDI last year Philips was showing a bulb which looked funny, ( it has a very YELLOW appearance) but seemed like it might bear further investigation. I never saw one in the local stores, but I ran into it on amazon the other day.

Amazon.com: PHILIPS Endura LED 12.5 Watt A-Shape A19 LED Dimmable Light Bulb ~ 60 watt incandescent equivalent: Home Improvement

I got two and tried them in a ceiling unit, and I have to say that I am very happy. The color temperature is a very warm 2700K. Skin tones look good to my eye. It seems that it takes a fraction of a second longer to turn on that the old incandescent unit did, but it gets to full bright immediately.

I am a happy camper.



If you are looking for an LED to replace your standard 60 to 75 watt A lamp - you should give these a look.


It's interesting to me that no post in this thread mentions power factor. In their effort to get the cost of LED bulbs down, manufacturers have sacrificed power factor. Many of these LED bulbs have very poor PF--sometimes as low as 0.4. This has the utilities quite scared, and their method of controlling this will be:

A. Change the standards to require higher power factors. This is already happening, led by the Canadians in North America. This will make LED bulbs cost more.
B. Charge you a hefty premium for lower power factors.

So, when selecting an LED bulb, I think you need to include power factor in the payback equation.


ST
 
Last edited by a moderator:
re: Experience with Philips Ambient LED in the home

And it isn't just a simple power factor issue, with the switch mode power supplies, the neutral current problems we are accustomed to with theatre dimmers could become mainstream, causing transformers to overheat and the neutrals having to be increased in size.
 
re: Experience with Philips Ambient LED in the home

And it isn't just a simple power factor issue, with the switch mode power supplies, the neutral current problems we are accustomed to with theatre dimmers could become mainstream, causing transformers to overheat and the neutrals having to be increased in size.

Most homes don't have enough lighting to make this a concern. It is a factor in larger buildings that electrical engineers incorporate into the design.

Both compact fluorescent and LED radiate a fair amount of RF from their power supplies. The FCC seems oblivious to it, but various radio systems are affected. Ham radio is suffering from increasing noise levels on many bands. AM broadcast is sinking due to a sea of noise, and FM broadcast is next.
 
re: Experience with Philips Ambient LED in the home

And it isn't just a simple power factor issue, with the switch mode power supplies, the neutral current problems we are accustomed to with theatre dimmers could become mainstream, causing transformers to overheat and the neutrals having to be increased in size.

Yes, indeed! And unlike phase control dimmers that generate the most harmonics at half power, switch mode supplies in ballasts and LED drivers generate harmonics at full power.

ST
 
Re: Experience with Philips Ambient LED in the home

It's far from being just a lighting issue with switch mode style microwave ovens and inverter air conditioners, all computer supplies, amplifiers, the list is endless, maybe we'll soon get a rebate for incandescent lamps.
 
Re: Experience with Philips Ambient LED in the home

It's far from being just a lighting issue with switch mode style microwave ovens and inverter air conditioners, all computer supplies, amplifiers, the list is endless, maybe we'll soon get a rebate for incandescent lamps.

Agreed!
With all the emphasis on energy savings, the new appliance norm may fall to .5pf ! (How long before a washing machine motor becomes a small efficient 3 phase motor driven by a small inverter/driver circuit?) On the other hand, newer guidelines push for "home run" neutrals on all circuits on the home. (and away from the old x/3 with a common.) When only two wires are involved, both would carry the same current.

So, will there be push-back? Well, if the government continues along its same course, they may simply mandate that power companies accept this new norm in order to promote the efficiencies. After all, isn't that how this whole CLF/LED thing started?
 
JD - IIRC, Energy Star guidelines are 0.5pf or greater for CFL lamps and resi ballasts. Can't recall anything regarding LED's.
 
Re: Experience with Philips Ambient LED in the home

JD - IIRC, Energy Star guidelines are 0.5pf or greater for CFL lamps and resi ballasts. Can't recall anything regarding LED's.

Can't imagine LEDs are any better! Welcome to the .5pf world!

Odd thing is, there are regulations on permanent indoor fixtures, at least in commercial buildings, that mandate 0.9pf or better. Can't remember the specifics, but if I remember right, this is not something through the NEC. Ran into this issue a few years back with some outdoor HID roof fixtures. Although these were "Outside" (which would normally be exempt) they were attached to the structure, so we had to buy corrected fixtures.

It will be interesting to see how this all washes out. On one hand, low pf, on the other hand, massive energy savings. Which will win?
 
Re: Experience with Philips Ambient LED in the home

Power Factor relates to the phase relationship between voltage and current, you can be chopping the guts out of a sinewave and causing all kinds of harmonics without affecting the power factor as in standard theatre dimmers.

Actually, except at 100%, phase chop dimmers have a horrible power factor.

Power Factor Correction

"Another notorious non-sinusoidal current draw is the popular phase controlled light dimmer, which uses a TRIAC or SCR to reduce the RMS voltage to the load by turning on partway through the half cycle. Not only is the current waveform highly non-sinusoidal, but it is also out of phase with the voltage supply. Hence, these loads have a non-unity power factor"

Think about a dimmer set at 33%. The full power draw is out of phase with the waveform, drawing 100% of its load in the last 33% of the AC cycle. (assumed liner curve)
 
Re: Experience with Philips Ambient LED in the home

I don't buy that line of argument, when the same light is at full it has pf 1 and as you dim it the pf goes down? I would want to see some convincing evidence to back that theory.During the time the current is flowing it is in phase with the voltage, so the quote you showed is wrong. The same quote apears on 7 different forums, so there's some plagiarism going on, but it's still wrong however many times it's copied.
 
Last edited:
Re: Experience with Philips Ambient LED in the home

I don't buy that line of argument, when the same light is at full it has pf 1 and as you dim it the pf goes down? I would want to see some convincing evidence to back that theory.During the time the current is flowing it is in phase with the voltage, so the quote you showed is wrong. The same quote apears on 7 different forums, so there's some plagiarism going on, but it's still wrong however many times it's copied.

Ah, but it isn't! At peak voltage, no current is flowing! Power only flows in the last 1/3rd of the waveform, thus giving the dimmer a horrible power factor!
Another example of this is the computer supply, which IS drawing peak current at the peak voltage. What gives it a low power factor is that it is only drawing power during about 10% of the waveform. In this case, the power draw is in-phase, but disproportionate.

Let's say that a 500 watt load is operating at 33% and is drawing 150 watts.
It is drawing the 150 watts at about 50 volts RMS. That works out to 3 amps. With a pf of 1, 150 watts would be drawing 1.25 amps, so the dimmer is operating at a pf of .42

The reason this is not a factor in dimmer design is that the circuit would be expected to handle 4.12 amps for the 500 watt load. Thus, the 3 amp draw does not require any attention.

Power companies want the voltage and current draws to be in sync. The failure to be in sync is why we end up (on three phase) with triplen harmonics, and the resulting neutral over-current. Again, this is a factor that comes into play due to the off-phase current draw occurring at lower dimmer settings.

Transformers have been developed to reduce triplen harmonics, but with great irony, they can't be used for lighting equipment as their action throws off the ZVD in the dimmers! http://www.etcconnect.com/lightminds/Hidden_Harmonic_Filter.pdf
 
Last edited:
Re: Experience with Philips Ambient LED in the home

OK so are you telling me that if I put a power factor meter on a dimmer then the meter will go from zero to one as I bring up the dimmer, I don't think so, I've never seen any indication of this when using generators but I have a power factor meter somewhere in the factory, I'll dig it out and try it, but the logic of how wave chopping affects power factor,[which is a phase change phenomina] escapes me.
"Let's say that a 500 watt load is operating at 33% and is drawing 150 watts.
It is drawing the 150 watts at about 50 volts RMS"
That assumption is way out because of resistance change in the filament anyway, a subject we covered in depth in the preheating argument.Probably over 250 watts at 33%
Quote
Default Re: Preheating Conventionals?

Please consider, a 1000 watt 240 volt T19 on a dimmer{for 120v thinkers just halve all the figures} It has a cold resistance of 3.8 ohms and a hot resistance of 57.6 ohms. A typical dimmer will feed 1.65v at "0" and drive .35A putting 2.24 W into the lamp of preheat.
@ 13V you can see the barest glow in the filament and are drawing .96A and 12.6W
@30V you can see a real glow and are drawing 1.5A and 45W
at 60V you begin to get some output but at 1/4 voltage you are drawing 2A or nearly half the "full" current
@80V which is 1/3 voltage you are drawing 2.3A which is more than half the full load current and is also the point at which the dimmer starts to interact with the other dimmers on other phases.
This is why running all your dimmers at 1/3 is the worst thing you can do to your neutrals.
 
Last edited:
Re: Experience with Philips Ambient LED in the home

John and David, you are both right, and you are both describing different situations that result in low power factor.

Pulling out the textbook on power factor, we learn that it is the ratio between the active power and the apparent power.

There's this other component called reactive power that is worth understanding, which is a result of inductance or capacitance in the load, resulting in a phase shift between the voltage and the current, and that's the situation that John is describing. It is certainly present in the system to some degree, but is not the cause of the low power factor of switched circuits. If we were dealing with linear loads like motors, heaters, or the like then power factor correction would involve adding capacitors or inductors to remove the reactive component from the equation.

In the case of switched dimming, David's description is more applicable because the low power factor is due to the non-linear waveforms produced, and the harmonics arising as a result. The current required is higher than anticipated for the power available to the load. The current and the voltage are mostly in phase, but the waveform is distorted. Those harmonics use energy but do not contribute to providing useful power to the load. That energy gets dissipated, mostly as heat, in other parts of the system, like the transformer and wiring. So you can think of distortion power factor as a ratio between the fundamental harmonic current, measured as RMS, to the total load current. In North America that would be the 60Hz component versus the current with all the higher order harmonics.
 
Re: Experience with Philips Ambient LED in the home

If (RMS line voltage X RMS amps = watts) then we have a power factor of 1
Otherwise (RMS line voltage X RMS amps X power factor = watts)

With 150 watts being dissipated at a late phase equivalent of 50 volts drawing 3 amps, given rounding, which one of the following equations is true:

A) 120 X 3 = 150
or
B) 120 X 3 X .42 = 150
 
Re: Experience with Philips Ambient LED in the home

power factor is Watts/VA now in the article it states,
"Not only is the current waveform highly non-sinusoidal, but it is also out of phase with the voltage supply. Hence, these loads have a non-unity power factor" but magically at full it reverts to pf 1"
How can the current be out of phase with the voltage, a triac is a switch, it can't change phasing like a capacitor or inductor.
I get that distortion can lose energy and cause a "distortion power factor" but that is not what is being stated.
 
Re: Experience with Philips Ambient LED in the home

power factor is Watts/VA now in the article it states,
"Not only is the current waveform highly non-sinusoidal, but it is also out of phase with the voltage supply. Hence, these loads have a non-unity power factor" but magically at full it reverts to pf 1"
How can the current be out of phase with the voltage, a triac is a switch, it can't change phasing like a capacitor or inductor.
I get that distortion can lose energy and cause a "distortion power factor" but that is not what is being stated.

yes, 120 X 3 X .42 = 150
or
150 /(3 x 120) = .42
or
150/360 = .42 (w/va = pf)
(rounded) Same equation.

Simply put, the "phase" of the outgoing waveform now lags the source AC waveform:
proxy.php

Current draw now lags waveform phase,
 
Last edited:
Re: Experience with Philips Ambient LED in the home

yes, 120 X 3 X .42 = 150
or
150 /(3 x 120) = .42
or
150/360 = .42 (w/va = pf)
(rounded) Same equation.

Simply put, the "phase" of the outgoing waveform now lags the source AC waveform:
proxy.php

Current draw now lags waveform phase,

no it doesn't "lag" it is a resistive load and in phase, the fact that you have a picture with a trailing edge dimmer is not a "lagging" situation, if you use a leading edge dimmer does that give you a "leading" current? clearly not. Using 2d pictures to convey 3d concepts does not work.
Why put "phase" in inverted commas, it's either phase or it isn't and this clearly isn't, any more than ""Not only is the current waveform highly non-sinusoidal, but it is also out of phase with the voltage supply." is totally wrong, I can't figure out why this is so hard to understand.I also can't understand why it keeps getting repeated.
 
Re: Experience with Philips Ambient LED in the home

no it doesn't "lag" it is a resistive load and in phase, the fact that you have a picture with a trailing edge dimmer is not a "lagging" situation, if you use a leading edge dimmer does that give you a "leading" current? clearly not. Using 2d pictures to convey 3d concepts does not work.
Why put "phase" in inverted commas, it's either phase or it isn't and this clearly isn't, any more than ""Not only is the current waveform highly non-sinusoidal, but it is also out of phase with the voltage supply." is totally wrong, I can't figure out why this is so hard to understand.I also can't understand why it keeps getting repeated.

Correct. Voltage and current are completely in sync in time. However, all of the harmonic component while being dimmed is combined into what is called "displacement power factor," as Sk8rsdad mentions. This type of power factor is created with non-linear loads such as power-rectification or power-modification equipment. In fact, that is where we started the thread in regards to LED power supplies but this also applies to fluorescent ballasts, computer power supplies [all switch-mode power supplies], and all wave-chopping power equipment, including phase-chopping dimmers.

I think what may have been confusing here is the description of the chopped waveform. Looking at zero-cross, current and voltage are in syne. Looking at peak energey in current versus voltage, they are not. So yes, the chopped waveform is in zero-cross sync BUT the missing component of the waveform works to help create our harmonic friends.

One note - most dimmable FL ballasts list a power factor of 0.9-0.95 but that does not hold true when dimmed and indeed I have measured them easily going down to 0.5-0.2 in some cases. Switch-mode supplies are such a huge non-real power suck it is amazing that power companies don't do more to partner with electronics designers and the government to help solve the non-usuable energy waste problem. But then, that costs money too.

Back to the topic-ish...

We have to remember that the load on the line feed is the combination of the dimmer and the lamp, and in the case of a phase-controlled dimmer, we now have a non-linear load. A linear load lamp, with a true switch, not a phase-controlled switch, is a linear load.

If this doesn't help in understanding how this type of power factor exists, would screen shots of a PF meter help? Unfortunately I didn't have time to hook this up yesterday but I can certainly do it on Monday. If you can get to your meter, you will indeed see what has been described. I'm really curious what Epimetheus finds on his Watts Up? meter.

David
 
Last edited by a moderator:

Users who are viewing this thread

Back