Less than unity power factor?

Status
Not open for further replies.
Absolutely, I agree. IMHO that is an error in teaching. 'Displacement power factor' is one thing, 'distortion power factor' is something else, both related by the fact that the RMS current carried by the wires (which causes heating) is greater than the minimum required to deliver the power consumed. Displacement power factor might be the first type of PF that someone encounters; they should not be taught that it is the only kind.

(responding to my duty cycle example)


Can I be honest? Up until recently I never heard of distortion power factor- in fact up until this thread I didn't even know what it fully meant. This is one those times when I am truly learning something new.





Agreed, in most common resistor 'on-off' regulated loads. This is in some sense 'distortion power factor', but only over multiple AC cycles; during a single AC cycle the PF is 1 or no current is flowing. Maybe it deserves a different name.

But consider a dimmer which functions by turning the load off for part of the AC cycle. This can introduce both distortion and phase displacement, changing the power factor.
-Jon

I think the hardest part is grasping how it does it- or rather why it quantifies as such...

Bear with me folks, this is the first time I am making this connection :blink: :dunce:
 
More like a several hemispheres and brain stem firing. This is so new and alien to me its like I'm dreaming. Also forgive the questions that are about to follow. :lol:
But it is not a new phenomenon. I have a technical paper on the topic dating from the 1930s. It concerned harmonics caused by Mercury Arc rectifiers.
 
The mercury vapor arc itself has inductive reactance. Generally with a magnetic ballast this gets buried in the inductance of the ballast itself. In a self ballasted lamp, the “ballast” is done with tungsten resistive elements that themselves have no inductance, but that then exposes the raw inductive reactance of the mercury vapor arc.
 
The mercury vapor arc itself has inductive reactance. Generally with a magnetic ballast this gets buried in the inductance of the ballast itself. In a self ballasted lamp, the “ballast” is done with tungsten resistive elements that themselves have no inductance, but that then exposes the raw inductive reactance of the mercury vapor arc.

I do not think I see the mechanism for an inductance in the arc that is any greater than the inductance of the same diameter, current, and length wire. There will be a negative resistance component, just as there is for a fluorescent bulb (which is arguably just a mercury arc plus phosphors.)
 
An 'inductive' power factor could simply mean that the current waveform lags the voltage waveform, without any of the energy storage of a true inductor.

I tried to find the current waveform for a self ballasted mercury lamp, but no luck.

-Jon
 
181021-1352 EDT

mbrooke:

I have no idea where over time, what classes, or self learning I came to some of my knowledge about power factor. One part of the process is clear. I grew up with Weston and Simpson meters. Before I new about electrodynamometer meters (Weston) I worked with Weston and Simpson DC meters with a rectifier for AC voltage. These rectifier meters were calibrated to read the RMS value of an AC sine wave. Whereas an electrodynamometer meter reads true RMS directly within its frequency capability, and works on both AC and DC with some trouble in the near zero frequency to possibly 20 Hz because of the mechanical averaging time constant. A hot wire meter also reads RMS directly and up into the MHz range.

Before about 1950 there were no electronic true RMS meters. True RMS requires multiplication.

RMS measurements relate to power in a resistive load. In the early 1950s we were doing signal detectability work, and this meant measuring power of a random gaussian signal (noise). Using an average reading meter (one using a simple full wave bridge, Simpson 260, or HP wider bandwidth) required a correction factor.

In the study of what RMS was and how to calculate its value from a waveform was another step. One does not need calculus to do an approximation, but it does require an understanding of what calculus is all about.

RMS is not mentioned much in DC circuit analysis, but it does apply. RMS starts to appear in AC circuit theory, physics, math, and calculus courses.

The concept of power factor appeared in the late 1800s with the advent of AC machines.

On an instantaneous basis p = v*i, and average power is the result of adding many small time increments of power over time, and then taking the average value. Calculus is nothing more than a means of taking infinity small increments and adding them together when you have an equation that defines the function. In general power at one instant of time (steady state) does not affect power at some other time (a scalar quantity), and therefore each separate increment of power can be added to others with simple addition.

When your load is not a pure resistance, then v = i*R may not hold true, and then v and i might not be well correlated. When this happens Vrms times Irms many not equal average power.

I believe power factor was originally defined in the 1800s as I defined it in an earlier post. In some ways power factor is a strange animal in the it may relate two uncorrelated values to some other value. In some circuits knowing the waveform of voltage does not allow you to know the waveform of current without detailed knowledge of the circuit. And detailed knowledge of the circuit may allow you to easily know the current and voltage relationship. A simple RL load is easy, but a bridge rectifier with capacitor input filter and a non-linear load can be difficult.

.
 
181021-1352 EDT

Timeout problems.

Before about 1950 there were no electronic true RMS meters. True RMS requires multiplication. Possibly Ballantine had a logarithmic based multiplication method in the 1940s.

.
 
Question, is it possible for a load without an inductor or capacitor to have a PF less than 1.00? Here is why I ask:

Yes. You don't need an inductor or capacitor...you only need inductance and capacitance. I know I'm splitting hairs here. But when I see inductor or capacitor, I think of a device designed to produce a given amount of inductance or capacitance. When I see inductance and capacitance, I think the of the inductance or capacitance that actually exists in reality. Some (hopefully most) of this is from the inductor or capacitor designed into the system. It's the stuff not designed into the system, but that creates inductance of capacitance, that can create these situations.

In short, the light may not be designed with an inductor or a capacitor, but I'll bet a pretty penny inductance and/or capacitance exists when it's energized. It is this inductance/capacitance which is manifest in the listed PF.
 
181021-1614 EDT

I do not know what is in one of these bulbs. However, it is an arc discharge.

If an arc discharge device is driven from a voltage source, then there must be some sort of current limiter between the voltage source and the arc, or very large current will be drawn limited by the source impedance of the voltage source.

An arc is somewhat of a constant voltage load. Thus, in an AC circuit current does not flow continuously and proportional to source voltage, and power factor will be less than 1.

.
 
Early in engineering school we study resistors, inductors and capacitors as discrete elements. Later in the process we learn that no real world component is purely any single one of those things.
 
181021-1614 EDT

I do not know what is in one of these bulbs. However, it is an arc discharge.

If an arc discharge device is driven from a voltage source, then there must be some sort of current limiter between the voltage source and the arc, or very large current will be drawn limited by the source impedance of the voltage source.

An arc is somewhat of a constant voltage load. Thus, in an AC circuit current does not flow continuously and proportional to source voltage, and power factor will be less than 1.

.



Just a mercury arc tube and filament resistor in series acting as the ballast. The 120 volt versions have "cathodes" (like florescent lamps) that are removed by a bimetal switch once the arc tube warms up. The 240 volt versions are much simpler, literally a classic mercury vapor lamp arc tube and halogen bulb in series. Here is one to clear up any confusion:


http://www.lamptech.co.uk/Spec Sheets/D MB XingJi HMB500.htm


XingJi HMB500.jpg
 
But it is not a new phenomenon. I have a technical paper on the topic dating from the 1930s. It concerned harmonics caused by Mercury Arc rectifiers.

If you have those papers in pdf or HTML, I'd certainly be interested in having a look.

The mercury vapor arc itself has inductive reactance. Generally with a magnetic ballast this gets buried in the inductance of the ballast itself. In a self ballasted lamp, the “ballast” is done with tungsten resistive elements that themselves have no inductance, but that then exposes the raw inductive reactance of the mercury vapor arc.


I guess the part that makes me scratch my head is how an arc tube can be inductive without an iron core or at least a magnetic field (like an air choke)...
 
181021-1904 EDT

Once you deviate from a sine wave, then you have harmonic content.

A steady state waveform can be viewed or studied from either its harmonic content or from its waveform.

If I want to build a phase locked oscillator synced to another waveform, then I want to view that waveform from a harmonic content perspective. If I want to determine RMS current, then I may want to work from the waveform.

See https://en.wikipedia.org/wiki/Fourier_series

.
 
And now you have peaked my interest. I can understand harmonics, but how do they produce vars? Ive always understood that you need something that "stores" energy like plates or a magnetic core to produce phase displacement of current in relation to voltage. But I admit my understanding is thin.
As has been stated, because it is not producing real power but still has a V×I component.


S*S = P*P + Q*Q + D*D (could not find carat symbol on this phone)

where

S = apparent
P = real
Q = reactive
D = distortion

It is the age-old problem of why capacitors can't always fix a non-unity power factor. displacement vs distortion.
 
mbrooke:

Your post #33. That is a very innovative design. An arc lamp requires a current limiter when driven from a voltage source. One can use a resistor, inductor, or capacitor for this limiter. If a resistor is used, then there is a great waste of power.

By using a halogen bulb as the resistor and including it in the bulb use is made of what would otherwise be wasted power, further visible light output is added that is substantially at the red end , and spread thru the spectrum to add to the line spectra from the mercury vapor. Also heat is added to the mercury part of the bulb.

.
 
As has been stated, because it is not producing real power but still has a V×I component.

I think I need an oscilliograph to visualize this one.


S*S = P*P + Q*Q + D*D (could not find carat symbol on this phone)

where

S = apparent
P = real
Q = reactive
D = distortion

It is the age-old problem of why capacitors can't always fix a non-unity power factor. displacement vs distortion.


Is this the same as what harmonic filters try and do? Or am I way off.
 
mbrooke:

Your post #33. That is a very innovative design. An arc lamp requires a current limiter when driven from a voltage source. One can use a resistor, inductor, or capacitor for this limiter. If a resistor is used, then there is a great waste of power.

By using a halogen bulb as the resistor and including it in the bulb use is made of what would otherwise be wasted power, further visible light output is added that is substantially at the red end , and spread thru the spectrum to add to the line spectra from the mercury vapor. Also heat is added to the mercury part of the bulb.

.

I agree. But not only that, consider that mercury arc tubes like to be in a low pressure fill to achieve maximum thermal efficiency (minimal heat loss) while filaments like a high pressure to limit evaporation. In this case you get a win-win, the outer envelope has a fill ideal for the mercury arc tube, while the filament uses halogen which greatly prolongs life expectancy. Very simple and elegant design common in China, Asia and a bit in South America.
 
Status
Not open for further replies.
Top