Negative Pf residential Home

[Rock86]

Co-worker has been complaining that the lights in his newly built house keep flickering. He has had the utility company out there, 3 or 4 electricians, and has asked us EEs in the office what could cause it. Utility company connected him to a new transformer, New ground rod and grounding electrode conductor because they were connected to rebar in concrete. Our boss (retired) went out there and connected his log meter to the line side conductors and came back with this on Leg 1. He has a negative power factor.

Any thoughts as to what could cause this? We have been going around with different theories for him, but I not had a chance to look at the system, but everyone else has been out there, so I would assume they know a lot so far, but know one has a clue.

 

[synchro]

It looks like the PF is wildly flipping polarities. Are you sure the connections are good and it's not just measuring noise. How much RMS current and voltage are indicated on Leg 1?

 

[hbiss]

New ground rod and grounding electrode conductor because they were connected to rebar in concrete.

-Hal

 

[zbang]

Throw that graph away and get another one done properly. (Is there any solar involved? Loose neutral?)

 

[winnie]

Negative power factor seems to mean power flowing in the reverse direction.

Unless there is something which can supply power from the house to the grid any time of day, I'd suggest instrumentation error.

Jon

 

[Rock86]

Throw that graph away and get another one done properly. (Is there any solar involved? Loose neutral?)

Will do. The utility company is coming out next with their equipment. They said it would need to be on their for 2-weeks. I was curious about a loose neutral from the beginning, but according to my coworker's response that was taken care of by the utility company and electrician already. Not do bad mouth either, but I have been in that boat where the utility company said they fixed it, and the next pole down the neutral was still broken.

 

[Rock86]

It looks like the PF is wildly flipping polarities. Are you sure the connections are good and it's not just measuring noise. How much RMS current and voltage are indicated on Leg 1?

 

[ptonsparky]

-Hal

NS. I had to go back and reread a couple times.

 

[winnie]

I've been trying to come up with something that could be found in a home which would actually be regenerative.

Was the monitor on both legs if the service, and can it report power consumption vs time? Was the power actually negative when PF showed as negative?

Obviously something like solar...but if the homeowner had solar the would know about it and would probably ask if this could cause a problem.

Is there a well pump or a pressure booster pump? I'm thinking a bad check valve could mean a pump running in reverse and acting as a generator.

Is there a large fan that sometimes acts as a windmill?

Are the flickering lights LED?

Jon

 

[AdrianWint]

While agreeing that this is probably down to bad or incorrect connection - Don't forget the convention ...... +ve pf is often taken to mean 'inductive load' and -ve pf to mean 'capacitive load'.

I often see -ve pf at home when the load is predominately switch mode power supplies (think PCs, TVs, many consumer electronics, LED lights etc etc) and not much good old resistive loads like heaters etc

 

[ptonsparky]

Does instrumentation error equal operator error?

 

[winnie]

While agreeing that this is probably down to bad or incorrect connection - Don't forget the convention ...... +ve pf is often taken to mean 'inductive load' and -ve pf to mean 'capacitive load'.

Whoops. I spaced that.

The power factor is probably simply close to unity, as expected in a residence. But changing from slightly leading to slightly lagging.

Jon

 

[gar]

230303-0839 EST

I would take my two channel digital scope, power the scope from an isolation transformer. My scope has probe common directly connected to the scope EGC. Also before starting I would make sure that both neutral, and EGC to my isolation transformer are less than several volts.

Using a divider probe, 10x, connect scope channel 1 to neutral and one 120 V phase. Sync the scope to channel 1. On my scope I would shift t = 0 to about 1 or 2 major divisions from the left side. To start set time base to possibly 2 mS/div. This makes 8 major divisions one full 60 Hz cycle. Connect a current transformer to channel 2. Set scaling as needed.

How does current relate to voltage over a sufficiently long time to encompass your time frame that relates to where you see the phase reversal?

Report back.

.

 

[gadfly56]

Whoops. I spaced that.

The power factor is probably simply close to unity, as expected in a residence. But changing from slightly leading to slightly lagging.

Jon

I think you're right. In the graph as provided, it's going to swing violently because of how the axis is set up, but you're only slightly moving the wave forms. It wouldn't take more than a few degrees back and forth, yes?

 

[busman]

As far as I know, there is NO SUCH THING by the definition as "negative" power factor. Since kW and kVA are not negative terms, it cannot be negative. If the system produced energy, it could have a PF greater than 1.0

From Fluke:

Power factor is an expression of energy efficiency. It is usually expressed as a percentage—and the lower the percentage, the less efficient power usage is.

Power factor (PF) is the ratio of working power, measured in kilowatts (kW), to apparent power, measured in kilovolt amperes (kVA).

 

[winnie]

LED lights require a driver to provide the constant current that the LED junctions require. These drivers do a fantastic job of isolating the light output from voltage variations...with the downside that this makes the LED not dimmable.

Most modern LED lights have an additional circuit to detect dimming and adjust the LED driver current.

This makes different brands of LED sensitive to different electrical noises, anything that triggers the dimming circuitry.

I'm wondering if there is some load that is causing current spikes right near the AC zero cross, confusing the power factor measurement and messing with the LED dimming.

Jon

 

[zbang]

As far as I know, there is NO SUCH THING by the definition as "negative" power factor. Since kW and kVA are not negative terms, it cannot be negative. If the system produced energy, it could have a PF greater than 1.0

PF a cosine ratio so it varies between 0 and 1.0; if the system is producing energy the PF can swing negative but I at least hardly ever hear people referring to a negative PF.

From Fluke:
Power factor is an expression of energy efficiency. It is usually expressed as a percentage—and the lower the percentage, the less efficient power usage is.
Power factor (PF) is the ratio of working power, measured in kilowatts (kW), to apparent power, measured in kilovolt amperes (kVA).

IMHO the first sentence is wrong (it's not a percentage), second is right.

 

[synchro]

It looks like the PF is wildly flipping polarities. Are you sure the connections are good and it's not just measuring noise. How much RMS current and voltage are indicated on Leg 1?

View attachment 2564299

Do you have a log of the current for L1 during this time? Just speculating, but perhaps the CT for L1 was actually placed on the neutral conductor instead of L1 . The neutral current will change polarities relative to the L1-N voltage, depending on which phase has the higher L-N current. Therefore, if the CT was on the neutral conductor it could show alternating PF polarities over time.

 

[GeorgeB]

Negative power factor seems to mean power flowing in the reverse direction.

Jon

I've only been at this for about 50 years so many know more than I, but I'm quite familiar with leading (capacitative loads) and lagging (inductive loads) power factor. If we go back to PF as the cos(angle of lead or lag), and limit that angle to 90 degrees, it will always be positive.

I'm with others who seem to suspect noise rather than data collected. Golly, I've never (cough cough) had that problem before.

 

[gar]

230303-1324 EST

Ran a simple induction motor test. A 120 V induction motor run with no load, worst case. Viewed current. The current lagged voltage by about 2.7 mS, or 0.65 * 90 = 59 degrees.

If I were to put the motor on the opposite phase ( 180 degree shift ), but not change the phase my voltage was on, then the current would for motors on this other phase shift to leading.

Otherwise it would take a lot of capacitance in parallel with the motor to shift to a leading phase angle.

The original poster needs to get a scope and current transducer on his circuit and see what the current waveform is doing in comparison with a fixed location voltage reference.

Get a scope and learn how to use it.

.