power factor questions

Status
Not open for further replies.
ohmhead

ohmhead

Senior Member
Location
ORLANDO FLA
Well question i see formulas and post on power factor lots of different views on this most give examples of there own with math use to clearly show there understanding of it heres one question i have to ask .

Please explain this imagine a 3 phase service pick a voltage loaded on all three phases they are all resistively loaded but one has a slightly leading PF and another has a slightly lagging PF in this case by calculation or by wording which formula would be the one to correct PF to unity or closely ?

Keep in mind that maybe the VAR of opposite polarity may cancel out the other but will it in the real world if the loads switch on /off at different times is it now going to be a issue what would be the end result can a average be found meaning PF correction what is correct in this condition .

And if switched caps are used can this add to the PF problem due to charge and discharge times of these cap in a bank .
 
ohmhead said:
Please explain this imagine a 3 phase service pick a voltage loaded on all three phases they are all resistively loaded but one has a slightly leading PF and another has a slightly lagging PF in this case by calculation or by wording which formula would be the one to correct PF to unity or closely ?
Click to expand...
If they are really resistively loaded, the power factor would be unity. There would be nothing to correct.
 
If you have unbalanced ?-? connected resistive loads, you will have leading power factor in one phase and lagging in another. Overall power factor will be unity. To correct the power factor, balance the loads.

As a simplified example, consider a 10 ohm resistor connected from ?A to ?B, with VA = 120V @ 0? and VB = 120V @ -120?.
VAB = 208V @ 30?
IAB = 20.8A @ 30?
IA = IAB = 20.8A @ 30?
IB = -IAB = 20.8A @ -150?
pfA = cos(30? - 0?) = 0.866 leading
pfB = cos[-150? - (-120?)] = cos(-30?) = 0.866 lagging
 
100215-1019 EST

With sine wave excitation to a resistive load if you measure the power to the load, the voltage across the load, the current to that load, and the phase angle you will find exactly a power factor of 1.000000, no other possibility.

.
 
gar said:
100215-1019 EST

With sine wave excitation to a resistive load if you measure the power to the load, the voltage across the load, the current to that load, and the phase angle you will find exactly a power factor of 1.000000, no other possibility.

.
Click to expand...

If you don't do any switching, this is correct. If you switch on/off, even a resistive load can have low pf. SCR's are turned on at some point and turn off at zero cross. This creates displacement power factor.
LowPF.jpg
 
PowerQualityDoctor said:
... SCR's are turned on at some point and turn off at zero cross. This creates displacement power factor.
Click to expand...

My understanding is:
1. pf is defined as (Real Power)/(Apparent Power),
2. Apparent power is defined as V x I (I'll deal with I* later)

and

pf is not unity because the switching transients introduce harmonics which increases the current which increases the apparent power, which decreases the pf.

Is this true (or close enough)?

If it is, is the pf caused by SCR switching leading or lagging?

cf
 
gar said:
100215-1019 EST

With sine wave excitation to a resistive load if you measure the power to the load, the voltage across the load, the current to that load, and the phase angle you will find exactly a power factor of 1.000000, no other possibility.

.
Click to expand...

If you were going to measure the power factor of Phase A, you would measure the angle of the Phase A current with respect to the Phase A voltage, right? What if the load is between phases? What is the phase angle between the phase current and the phase to ground voltage?
 
100215-1301 EST

PQD:

I explicitly said sine wave excitation.


Cold Fusion:

If I have a circuit with a bridge rectifier connected to a large capacitor input filter with a nominal resistive load such that the RC time constant is long compared to a half cycle of the input AC, then the AC input current waveform is very peaked and nearly centered on the peak of the input AC voltage waveform.

Thus, the current is very symmetrical with respect the voltage waveform.

I think it is inappropriate to assign any angle to this waveform to try to associate it with power factor. My oscilloscope has such an input AC current waveform and a power factor of 0.72 .

An SCR turned on later than a zero crossing has a delayed waveform and is not symmetrical with the voltage waveform and again there is no meaningful way to associate an angle with this that corresponds to the power factor.

.
 
100215-1316 EST

jghrist:

If you take line to line as the excitation to the resistive load, then you must measure the current thru the resistor, and the voltage line to line. Otherwise you are mixing apples with oranges.

There may be an approximate relationship of line to neutral voltage phase angle to line to line voltage, but it has nothing to do with the resistance line to line, and the voltage and current relationships in the resistive load.

Look at the definition of power factor. It is the relationship of the VA to a load and the real power to that load. Obviously it means the volts and amps to that load, and the real power to that load. Not some arbitrary volts off in the wilderness. Further, although I have never seen it in writing, it implies the RMS value of voltage and current.

.
 
In the example that I gave the excitation is sinusoidal, but I added an SCR to show that even resistive load can have PF other than 1.00

As a matter of fact there are two types of PF:
DPF = displacement power factor, the cosine of the angle between the voltage and current of fundamental frequency, which is equal to the ration between kW and kVA of fundamental frequency.
TPF = true power factor, which is the ratio between kW and kVA with all the harmonics.
They are related according to the formula (assuming THDv is low):
TPF = DPF / sqrt(1+THDi^2)

In the previous example, the phase shift causes DPF to reduce and the harmonics causes the TPF to be even lower.

About the question of measuring PF and powers between phases - I am sorry but there is only one company that does it. I explained it in previous post with links totheir explanations and deserved deletion stating it is advertising. I am not going to repeat it. Ask the moderators to do so.
 
gar said:
100215-1316 EST

jghrist:

If you take line to line as the excitation to the resistive load, then you must measure the current thru the resistor, and the voltage line to line. Otherwise you are mixing apples with oranges.

There may be an approximate relationship of line to neutral voltage phase angle to line to line voltage, but it has nothing to do with the resistance line to line, and the voltage and current relationships in the resistive load.

Look at the definition of power factor. It is the relationship of the VA to a load and the real power to that load. Obviously it means the volts and amps to that load, and the real power to that load. Not some arbitrary volts off in the wilderness. Further, although I have never seen it in writing, it implies the RMS value of voltage and current.

.
Click to expand...

It is quite complicated to explain without linking to other places that may be considered as commercial, nor possibility to upload pictures, so I will try to be descriptive.

Assume a single phase, line to neutral network. You have source, SCR and resistor. The current through all the points is the same, as it is in series. The voltage between the source and the SCR is sinusoidal and chopped between the SCR and the resistor. The PF of the resistor itself is unity while the PF on the SCR (and to the network) is less than unity.

See my previous post about line to line and line to neutral measurements.
 
100215-1337 EST

PQD:

It is not the resistor that has a power factor other than 1 with sine wave excitation, but rather a new device or circuit which is the combination of a resistor and SCR. You could just as well use a diode and resistor as this new device.

.
 
Correct, except that a diode is not commonly used and I used a common device for explaining that even network with resistors only (as loads) can have lagging PF.
 
All of the discussion of SCRs and distortion power factor is very interesting but I do not believe that it addresses the OP
Please explain this imagine a 3 phase service pick a voltage loaded on all three phases they are all resistively loaded but one has a slightly leading PF and another has a slightly lagging PF in this case by calculation or by wording which formula would be the one to correct PF to unity or closely ?
Click to expand...
This seems very likely to be completely and simply explained by unbalanced ?-? connected loads, and corrected by balancing the loads. The OP discusses "one phase" and "another", clearly meaning the power factor of a single phase, not the power factor between phases. If you have a combination of loads, some connected ?-N and some ?-?, does it become impossible to measure the power factor? This may not be a theoretically pure measurement of current through a load and voltage across a load, but in the real world, many loads are a combination of individual loads and can indicate non-unity power factor without distortion or SCRs.
 
I provided comprehensive reply, discussing both Star and Delta networks. Unfortunately, I mentioned the only company in the world that their meters can do it, so it was deleted (see the post between #2 and #3). I also discussed this in post #10.

It was just too long to write again and "win" another deletion...
 
I believe that jghrist is on the right track as far as the original poster is concerned.

A pure resistive load with no switching involved has a unity power factor relative to the voltage applied to its terminals.

However the terminal voltages at the loads are not the only voltages to be considered, and even a perfect resistive load may present a non-unity power factor to the system. If you have a resistive load connected to two terminals of a 'wye' source, then the supply legs of that wye source will see a .866 power factor. This is not simply a mathematical oddity; this is a real difference between the VA and the W delivered by the transformer coils.

The load placed on coil A is the _series_ circuit consisting of coil B and the resistive load, so coil A is not seeing a simple resistor. Similarly coil B sees the series circuit consisting of coil A and the resistive load. The net result is that while the load has unity power factor, it is not presenting a unity power factor to the transformer.

If the OP has simple resistive loads connected line to line, and they are not balanced, then metering of the source will indicate a non-unity power factor. The solution is to balance the loads.

-Jon
 
jghrist said:
If you have unbalanced ?-? connected resistive loads, you will have leading power factor in one phase and lagging in another. Overall power factor will be unity. To correct the power factor, balance the loads.

As a simplified example, consider a 10 ohm resistor connected from ?A to ?B, with VA = 120V @ 0? and VB = 120V @ -120?.
VAB = 208V @ 30?
IAB = 20.8A @ 30?
IA = IAB = 20.8A @ 30?
IB = -IAB = 20.8A @ -150?
pfA = cos(30? - 0?) = 0.866 leading
pfB = cos[-150? - (-120?)] = cos(-30?) = 0.866 lagging
Click to expand...

Power factor is the relationship of the current in A-phase and the voltage in A-phase, similar in phases B and C, or using your Phase-to-Phase values, Vab to Iab and therefore is unity as they are both in phase. You cannot relate phase-to-neutral and phase-to-phase values.
 
richxtlc said:
You cannot relate phase-to-neutral and phase-to-phase values.
Click to expand...

Why not?

Consider a system consisting of a wye transformer and a line-line load. This is a common everyday installation.

If I want to know about the power factor of the load, then I will look at Iab and Vab. If I want to know about the VA requirements of the transformer, then we need to look at Ian and Van.

These two can clearly be related, as the equations above show in a very simplified case.

-Jon
 
winnie said:
Why not?

Consider a system consisting of a wye transformer and a line-line load. This is a common everyday installation.

If I want to know about the power factor of the load, then I will look at Iab and Vab. If I want to know about the VA requirements of the transformer, then we need to look at Ian and Van.

These two can clearly be related, as the equations above show in a very simplified case.

-Jon
Click to expand...

I meant you cannot relate a phase-to-phase value to a phase-to-neutral value, you are correct in what you said above, I mistyped my answer.
 
Well not a normal example or thought just a load consisting of a perfect one ohm resistor in series with a perfect infinite value capacitor if we drive this load with 1 volt RMS sine wave then the amps are 1 the watts are 1 watt the var is zero ? YES /NO


But if we drive the load with 1 volt DC at the same time what happens ?

RMS volts are now 1.414 volts the current is the same the capacitor blocks the dc and the watts are the same so VA is now 1.414 but the watts are 1 and VAR must be 1 adding dc created a reactive power ? YES /NO

So can we say a VAR is a real thing ?

No formulas needed just look at this and tell me if its wrong in thinking this way .
 
Status
Not open for further replies.
Top