#1 I have not be hired to do this work so buying that kind of gear for one time use is out, and #2 it is not my house. Our office is not furnished with that kind of equipment. Our data logger is as fancy as we go. From what my coworker said, the utility company will be going back with their equipment.
Negative Pf residential Home
- Thread starter Rock86
- Start date
ptonsparky
Tom
- Occupation
- EC - retired
Most of us don't have a scope let alone the training to use one that would give more than questionable results.
AdrianWint
Senior Member
- Location
- Midlands, UK
From a commercial power meter logging the incoming feed to my residence....
Notice the -ve (ie. capacitive) Reactive power & the -ve power angle.
Since a capactive reactive power is below the origin, how would others express a power factor which has a reactive power line low the origin?
How would others refer to powers which are not in the upper right quadrant of the the real/reactive power diagram?
Notice the -ve (ie. capacitive) Reactive power & the -ve power angle.
Since a capactive reactive power is below the origin, how would others express a power factor which has a reactive power line low the origin?
How would others refer to powers which are not in the upper right quadrant of the the real/reactive power diagram?
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
Some background review to put this in perspective, hopefully not just for engineers:
The complex power S is the sum of the real part P (real power component) and the imaginary part jQ (where Q is the reactive power component). And so S = P + jQ. Typically, P is in kW, Q is in kVAR, and |S| = √(P2+Q2) (the apparent power) is in kVA.
Having an imaginary part on a complex plane (instead of a straight line that represents only real numbers) is a math technique used to represent the portion of the power resulting from current components that are at + or - 90° from the applied voltage due to reactive circuit elements such as inductors and capacitors. We can thank Heaviside, Steinmetz, etc. for this.
The power factor is the real power divided by the magnitude of the of the complex power S (apparent power), or PF = P/√(P2+Q2) = cos(θ), where θ is the angle between the apparent power S (e.g., in kVA) and the real power P (e.g, in kW).
A positive reactive power component Q will cause the the apparent power to lag the real power (i.e, an inductive load), and therefore the complex vector S will be in the upper right quadrant of the complex plane.
A negative reactive power component Q will cause the the apparent power to lead the real power (i.e, a capacitive load), and therefore the complex vector S will be in the upper lower quadrant of the complex plane.
Because PF = P/√(P2+Q2), when real power P is positive the power factor will also be positive in both the upper right quadrant (Q positive, lagging PF) and lower right quadrant (Q negative, leading PF) of the complex plane because of the term Q2. However, if the real power P is negative (i.e., when the complex power S is in the left two quadrants), then the power factor will be negative.
By convention, a positive reactive component Q (inductive) will consume reactive power and a negative reactive component Q (capacitive) will supply reactive power.
An example of a "load" with a negative power factor would be a grid-tied PV inverter when it's operating. Instead of consuming real power from the grid, the inverter is supplying real power to the grid. The real power component P above is then negative, and therefore PF = P/√(P2+Q2) is also negative.
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
Typo. That should be "A negative reactive power component Q will cause the the apparent power to lead the real power (i.e, a capacitive load), and therefore the complex vector S will be in the lower right quadrant of the complex plane.
namesdontmatter
Member
- Location
- Hawaii
- Occupation
- EE
I agree with the posts that this is switching back and forth between Capacitive and Inductive. I'll bet if you could take the raw data and plot all (-) values as (+) values the graph is continuous. Alternatively, run a handful of box fans for a few hours and see if it stops switching signs and stays positive.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
230304-0827 EST
Whoever designed that graph is incompetent in being able to convey information to a viewer on the problem.
When one studies that graph for while it appears to be trying to convey information on power factor as one changes from a small laging to a small leading value. It does a poor job of this.
In a normal home or most small businesses switching from an inductive to capacitive load is not likely to occur? From an instrumentation view point --- if one measured voltage from neutral to one phase leg of a split phase system, and measured current on the neutral, then for only resistive and inductive loads cycling on and off on the system you would see what appears in this graph.
We need a better description of what loads are in this home, and how the instrumentation is setup.
.
Whoever designed that graph is incompetent in being able to convey information to a viewer on the problem.
When one studies that graph for while it appears to be trying to convey information on power factor as one changes from a small laging to a small leading value. It does a poor job of this.
In a normal home or most small businesses switching from an inductive to capacitive load is not likely to occur? From an instrumentation view point --- if one measured voltage from neutral to one phase leg of a split phase system, and measured current on the neutral, then for only resistive and inductive loads cycling on and off on the system you would see what appears in this graph.
We need a better description of what loads are in this home, and how the instrumentation is setup.
.
Jpflex
Electrician big leagues
- Location
- Victorville
- Occupation
- Electrician commercial and residential
I’m not sure if it’s proper to refer to this as a negative power factor. Ac current normally switches polarity or positive/ negative per alternating cycles.
You would likely either have a lagging (inductive) low power factor (undesirable) or leading power factor also (undesirable) capacitive loads or synchronous motor over excited.
Goal is 0.95 % power factor typically inductive current lagging voltage, in order to avoid utility charge penalties.
In industrial applications a capacitor is placed between phase legs to counter inductive reactance to boost power factor closer to 1 unity. However going to 1 has a risk of having a leading powe factor which is just as bad as lagging.
The flickering lights could be due to too low power factor causing low line voltage.
You would likely either have a lagging (inductive) low power factor (undesirable) or leading power factor also (undesirable) capacitive loads or synchronous motor over excited.
Goal is 0.95 % power factor typically inductive current lagging voltage, in order to avoid utility charge penalties.
In industrial applications a capacitor is placed between phase legs to counter inductive reactance to boost power factor closer to 1 unity. However going to 1 has a risk of having a leading powe factor which is just as bad as lagging.
The flickering lights could be due to too low power factor causing low line voltage.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
230304-1102 EST
We need more accurate information on how this circuit was instrumented, and what are the typical known loads.
If one assumes that the system has only incandescent lights, typical induction motors, and some other various types of loads, then it is unlikely to see net capacitive loading on the system at any time.
I took a look at several loads:
1. Unloaded single phase induction motor --- one was about 3 mS lag, and the other about 3.2 mS. These will have less lag when fully loaded, but it won't be near zero lag.
2. An off brand LED, 10 W, about 1 mS lead. A CREE LED was about 0 mS.
3. A Stancor constant voltage transformer, 0 mS.
We need a much better description of the original poster's actual circuit, and this also means a better way to describe the measurements.
.
We need more accurate information on how this circuit was instrumented, and what are the typical known loads.
If one assumes that the system has only incandescent lights, typical induction motors, and some other various types of loads, then it is unlikely to see net capacitive loading on the system at any time.
I took a look at several loads:
1. Unloaded single phase induction motor --- one was about 3 mS lag, and the other about 3.2 mS. These will have less lag when fully loaded, but it won't be near zero lag.
2. An off brand LED, 10 W, about 1 mS lead. A CREE LED was about 0 mS.
3. A Stancor constant voltage transformer, 0 mS.
We need a much better description of the original poster's actual circuit, and this also means a better way to describe the measurements.
.
ptonsparky
Tom
- Occupation
- EC - retired
Low PF causing low voltage? For a residence? I don't think so.
Then again that stuff is out there for sale on the Internet. Buy an extended warranty for you auto while you're at it.
Then again that stuff is out there for sale on the Internet. Buy an extended warranty for you auto while you're at it.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
230304-1740 EST
As I ponder this more my thought is we need plots on the same graph vs time of:
1. real power on a scale of 0 to some maximum. That is one plot.
2. A plot of power factor relative to some base line of 1.00 where lagging PF is positive incrementally from 1.00, and leading PF is negative relative to that same base line. A positive, lagging, PF of of 0.98 would plot as 1.02, and a leading PF of 0.95 would plot as 0.95 . This is a second plot.
.
As I ponder this more my thought is we need plots on the same graph vs time of:
1. real power on a scale of 0 to some maximum. That is one plot.
2. A plot of power factor relative to some base line of 1.00 where lagging PF is positive incrementally from 1.00, and leading PF is negative relative to that same base line. A positive, lagging, PF of of 0.98 would plot as 1.02, and a leading PF of 0.95 would plot as 0.95 . This is a second plot.
.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
The problem is that we have a 4 quadrant measurement (or a complex valued measurement) but are trying to portray it with a single value vs time graph.
True negative power factor requires reverse direction power flow. But as synchro very clearly described in post 24 you can have - or + reactive power flow on top of - or + real power flow, and as also noted earlier in this thread sometimes + power with - reactive power (capacitive load) will be represented as 'negative power factor'.
Getting back to the OP's question:
The time scale on your charts covers several days.
The period of the power factor jumps is irregular but on the order of an hour. (Maybe 20 minutes, maybe 2 hours, but not crazy rapid.)
Are the lights flickering associated with the power factor jumps?
When I look at the voltage chart, if I squint just right I can see little spikes that seem to have roughly the same periodicity as the power factor jumps.
My guess: You some true capacitive load that is switching on and off. Maybe an HVAC system with oversized capacitors or the like. Perhaps a well pump. The leading vs lagging power factor is causing large regulation changes in the supply transformer, which make the lights flicker.
-Jon
True negative power factor requires reverse direction power flow. But as synchro very clearly described in post 24 you can have - or + reactive power flow on top of - or + real power flow, and as also noted earlier in this thread sometimes + power with - reactive power (capacitive load) will be represented as 'negative power factor'.
Getting back to the OP's question:
The time scale on your charts covers several days.
The period of the power factor jumps is irregular but on the order of an hour. (Maybe 20 minutes, maybe 2 hours, but not crazy rapid.)
Are the lights flickering associated with the power factor jumps?
When I look at the voltage chart, if I squint just right I can see little spikes that seem to have roughly the same periodicity as the power factor jumps.
My guess: You some true capacitive load that is switching on and off. Maybe an HVAC system with oversized capacitors or the like. Perhaps a well pump. The leading vs lagging power factor is causing large regulation changes in the supply transformer, which make the lights flicker.
-Jon
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
If you zoom out the picture in post #1 of the PF from the data logger so that it fills up your screen, there appears to be some symmetry in the + and - swings of the PF around the zero baseline when it's flipping more often. It reminds me of the symmetry in the envelope of amplitude modulation like what's pictured below:
I'm thinking that if the sign of PF indicated whether the current was leading or lagging, then it would be unlikely to be flipping back and forth with the same magnitude but opposite sign. For example, between a 0.85 lagging and a 0.85 leading power factor. This would require the inductive reactance to be tracking the capacitive rectance.
However, say if there were inductive L1-N and L2-N loads, and which one was largest changed back and forth over time, on the neutral conductor you might see an alternating PF of a similar magnitude but with opposite signs. That's because the neutral current would be the difference between the L1 and L2 currents, and it would be changing polarities relative to the line voltages.
But it would be helpful to see current waveforms associated with the results already shown. When the line current is relatively small, it wouldn't take much to shift the PF around. And so I wouldn't put much weight on the PF that is shown during such times.
I'm thinking that if the sign of PF indicated whether the current was leading or lagging, then it would be unlikely to be flipping back and forth with the same magnitude but opposite sign. For example, between a 0.85 lagging and a 0.85 leading power factor. This would require the inductive reactance to be tracking the capacitive rectance.
However, say if there were inductive L1-N and L2-N loads, and which one was largest changed back and forth over time, on the neutral conductor you might see an alternating PF of a similar magnitude but with opposite signs. That's because the neutral current would be the difference between the L1 and L2 currents, and it would be changing polarities relative to the line voltages.
But it would be helpful to see current waveforms associated with the results already shown. When the line current is relatively small, it wouldn't take much to shift the PF around. And so I wouldn't put much weight on the PF that is shown during such times.
**UPDATE**
Last night, another electrician, engineer and my coworker hooked up a data logger and watched it in real time. With the main switched off, they showed +1 for both legs.
With the main switched on and all breakers off, +1 both legs.
Main on, certain individual circuits turned on, drop to -1 on the circuit leg.
Main on, multiple circuits on, motor load like a freezer kicks on, leg would go from -1 up to +(something) and the down to -1 when the load stopped.
From what i was told, they checked the meter multiple times to make sure it was on the right way.
Last night, another electrician, engineer and my coworker hooked up a data logger and watched it in real time. With the main switched off, they showed +1 for both legs.
With the main switched on and all breakers off, +1 both legs.
Main on, certain individual circuits turned on, drop to -1 on the circuit leg.
Main on, multiple circuits on, motor load like a freezer kicks on, leg would go from -1 up to +(something) and the down to -1 when the load stopped.
From what i was told, they checked the meter multiple times to make sure it was on the right way.
The lights are flickering. This is what started the investigation.
Well pump, aerator pump, and larger pumps were off during the most recent test when an issue occurred. Said pumps showed no issues according to the meter when they were turned on.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
+1 power factor with main switched off is basically meaningless. Power factor is relevant when current is flowing. Main off should mean no current.
Same for main on, breakers off. You might have a bit of current flowing for things like GFCI/AFCI breakers.
From what I see above, the power factor jumping from + to - is almost certainly a very small change in reality, going from leading to lagging in response to small loads.
-Jon
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
While everyone was there observing things, did they correlate the light flicker to the power factor swings? The power factor swings happened when various loads switched. Did the lights flicker at the same time?
-Jon
Thats what I told them. It didn't sound right that the power factor would exist without current.
From what I was told, no. Now being told that only when the washer runs, does the lights flicker. Freezer and Refrigerator become faulty shutting off for longer periods of time and warming higher than the programmed settings.
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
That's apparently fairly common, especially with washers that use VFD's and when there's LED lighting. Have incandescent lamps been placed on the lighting circuits to see whether they also have noticeable flicker with the washer running?
Regarding the changing power factor, perhaps there is some type of LC resonance in the system that's causing issues. Perhaps you could try placing relatively large resistive loads like portable heaters on some of the circuits so that the real power they draw would dominate any reactive power. This might also provide some damping if there are any resonance effects happening. See if this reduces the flickering to any extent. Then again, the root cause of the flickering may have more to do with LED internal dimming circuits falsely responding to noise as winnie mentioned in post #16 of this thread, rather than from power factor issues.