Agitation cycle of washer causing lights to pulse

[Asiel]

I have a 3 yr old direct drive washer that is causing lights on the phase to pulse when agitating. The Washer is on it’s own dedicated 20a circuit that’s 25 ft away from the main service. I have placed the washer on a gfci circuit in the kitchen that’s on the other phase and it has the same effect on the lights on different circuits, but same phase.

I swapped out leds for incadesents and the pulse was still observed.
At the service panel I observed zero current on any of the circuits EGCs. The currents on the neutrals match the currents on there respective hots. Voltage from neutral to ground is zero or a few millivolts depending on what meter I use.


The amperage of the washer while on agigtation varies depending on what meter I use.
Extech showed spikes of 9amps

Fluke on min max showed as high as 13.7a

Fluke 43b showed 19a. Using sags and swells.
Washer outlet min voltage- 123.5
Washer outlet max voltage - 124.7
Service panel voltage same phase as washer-123.8 min 124.7 max

Min max at washer outlet 124.7 max 123.1 min 124.3 avg using fluke1587

Light circuit voltage matches panel voltage.

Logic tells me I should be seeing a bigger voltage drop but I’m not.


I measured my dryer since it’s 240v load with neutral.

Non load voltage 125.6 max 125.2 min


Non washer phase on start up showed max 125.5 min 121.7 with current up to 38a settling at 25a 124.4 max and 124.0 min

washer phase on start up 126.9 v max 1241. Min amperage 4.32a settling at 23a with 124.6 max 123.8 min


I have plugged in a 10.5a vacuum on washer outlet and observed a 1.5v change at the outlet. I observed a volt change on the same phase at the panel. The other phase increased by. 3v.

I don’t see anything nefarious. Can I assume the washer is causing the issue?

 

[Little Bill]

I had a similar situation a while back with the washer causing lights to flicker. Turns out it was the neutral at the utility pole/transformer that was the problem. I would suggest you get the POCO out to check things on their end. Then have an electrician to check your end if the POCO's end is ok.

 

[Dennis Alwon]

If the only time you have the issue is with the washer then it is a good chance the washer is the problem.

 

[gar]

211219-1945 EST

Asiel:

The washer is the cause of the flicker. Your washer motor current ramps up and down. I have never seen one and have no idea on the period and magnitude of variations in current consumption. However, I have some potentially useful comments.

Some experiments I have run in the past, and just now partially repeated.

I like to use a 15 W 120V incandescent bulb as a flicker detector. It is a fast responder, and not excessively bright. A 2 V RMS change is detectable, but relatively insignificant in noticeability. Signal known exactly and I can detect the flicker. Without signal known exactly I might not notice the flicker. Here signal known exactly means timing. The 15 W bulb is a good detector of both rises and falls of potential.

I have a 1/3 Hp induction motor that at 120 V has a starting current that lasts about 6 cycles, and a peak current value about 50 A, or 35 A RMS.

At my basement workbench I have about 50 to 70 feet of wire from the main panel. 50 ft about is #6 copper, the remainder is #12 copper, and several breakers ( 3 ) in series. With my Fluke 87 I can not read a short dip, but I can read a short rise. I have a split phase system meaning two different phases that are 180 degrees relative to each other. My neutral wire from the main panel is common to both phases. Thus, a load on one phase will produce a voltage drop on that phase, and a voltage rise on the other phase.

I realize that a single phase service will use three wires from the utility transformer to the main panel. Therefore, a single wire is the common for both phases. Assuming the service drop is made up of three wires with each being the same material and size, then when one phase is loaded there is an equal drop on neutral to the drop on the phase wire. With no change in load on the opposite phase it means the voltage on the unloaded phase will appear to increase by the magnitude of the drop on the neutral. This assume the power transformer can be considered to of zero impedance.

#10 copper wire is about 1 ohm per 1000 ft. If you had a loop path of 100ft ( circuit distance 50 ft ), then 50 A at 0.1 ohm is a drop of 5 V.

You need to know the min and max current to the washer as it cycles.

Then you need to do a voltage change measurement at the many panel with a 1500 W space heater as a test load. This is about 12 A. Measure the voltage directly on the supply wires to the main panel. This gives you an idea of the source impedance to the main panel. Now you make judgements as to what is going on.

.

 

[LarryFine]

I have plugged in a 10.5a vacuum on washer outlet and observed a 1.5v change at the outlet. I observed a volt change on the same phase at the panel. The other phase increased by. 3v.

The only thing that can cause a voltage increase on the other phase is voltage drop on the neutral.

Your issue must be along the neutral pathway, either a bad connection or a compromised conductor.

 

[synchro]

I have plugged in a 10.5a vacuum on washer outlet and observed a 1.5v change at the outlet. I observed a volt change on the same phase at the panel. The other phase increased by. 3v.

The only thing that can cause a voltage increase on the other phase is voltage drop on the neutral.

Your issue must be along the neutral pathway, either a bad connection or a compromised conductor.

I'm thinking that since the 10.5A L1-N load current of the vacuum cleaner raises the L2-N voltage at the panel by 0.3V, then there is an 0.3V drop on the neutral conductor. The L1-N voltage drop is 1 volt, of which 0.3V is on the neutral. And so the voltage drop on the L1 conductor would be 1V - 0.3V = 0.7V, which would indicate that it has 0.7/0.3 = 2.3 times the impedance as the neutral conductor does.

As gar has indicated above, the resistive load of a space heater would be better for testing. A vacuum cleaner current may have significant reactance, harmonics, and noise that could make interpreting the results more difficult.

 

[Asiel]

I'm thinking that since the 10.5A L1-N load current of the vacuum cleaner raises the L2-N voltage at the panel by 0.3V, then there is an 0.3V drop on the neutral conductor. The L1-N voltage drop is 1 volt, of which 0.3V is on the neutral. And so the voltage drop on the L1 conductor would be 1V - 0.3V = 0.7V, which would indicate that it has 0.7/0.3 = 2.3 times the impedance as the neutral conductor does.

As gar has indicated above, the resistive load of a space heater would be better for testing. A vacuum cleaner current may have significant reactance, harmonics, and noise that could make interpreting the results more difficult.

I don't have a space heater. Would a hair dryer work?

Here is snap shot of the my Fluke when I was measuring the amperage and voltage at the main panel on the line side of the breaker while my washer was running.

PQM line side main while washer running

 

[GoldDigger]

I don't have a space heater. Would a hair dryer work?

Here is snap shot of the my Fluke when I was measuring the amperage and voltage at the main panel on the line side of the breaker while my washer was running.

PQM line side main while washer running

A hair dryer would work, but is not likely to be available in as high a power. A high powered heat gun would be more likely to be useful.

 

[Asiel]

With a 1600w blow dryer 13.36A

unloaded phase went from 126.2v to 126.6v
loaded phase went from125.9v to 124.9v

I then switched it around so the loaded phase would become the unloaded phase.

unloaded phase went from 126.0v to 126.4v
loaded phase went from 126.0v to 125.1v

 

[synchro]

I have a 3 yr old direct drive washer that is causing lights on the phase to pulse when agitating. The Washer is on it’s own dedicated 20a circuit that’s 25 ft away from the main service. I have placed the washer on a gfci circuit in the kitchen that’s on the other phase and it has the same effect on the lights on different circuits, but same phase.

With a 1600w blow dryer 13.36A

unloaded phase went from 126.2v to 126.6v
loaded phase went from125.9v to 124.9v

I then switched it around so the loaded phase would become the unloaded phase.

unloaded phase went from 126.0v to 126.4v
loaded phase went from 126.0v to 125.1v

So does this mean that the voltage rise on the unloaded phase is not producing nearly as much flicker as the larger voltage drop on the loaded phase? If this is true, could most (if not all) of the lights with the problem be shifted to the phase which is not feeding the washer? And then other 120V circuits which are less sensitive to voltage sags might be moved to the phase that's feeding the washer. That is, unless the lights are on both phases of multiwire branch circuits, which would prevent this from being done.

 

[Asiel]

I don’t have any multibranch circuits. So I could do some rearranging in my panel to get all the lighting/ room circuits on one phase. I don’t know how unbalance my panel would be after that, but I could play around with it.

I’m just trying to cover my bases and make sure I’m not overlooking a safety hazard which maybe causing my lights to pulsate. I only notice the pulsation of the lights when the washer is agitating.

I’m curious if the amp draw that I’m capturing with the fluke 43b which is as high as 19 amps is normal for these direct drive washers. My fluke amp clamp meter can capture 13.7 amp when set to min max. I tried calling LG and they sent me right to chat where they proceeded to tell me they didn’t have that information. Label states 9A.
At work I would just go over to another device and measure and compare, but I don’t have that luxury at my house.

 

[gar]

211220-0739 EST

Asiel:

The voltage changes you are now showing for the two phases make sense. Whereas previous values you presented did not. Your changes are quite typical of what I might expect for a 15 A load change.

Why do you care if your load is balanced? In my home I have a 200 A service, and I seldom go over about 2.5 kW load. 200 A at 240 V is 48 kW. Lighting load is probably not a big part of your load unless you are in my range of power consumption.

A repetitive pulse variation at the voltage change levels you are seeing may be more noticeable than single events.I will have tun some experiments.

A possible solution to your problem might be to have a 15 A load that is switched on when the washing machine 15 A load is off.

.

 

[retirede]

Since no one has asked this yet, you’ve had the washing machine 3 years. Did the problem just start now, or has it always done this?

 

[Asiel]

Since no one has asked this yet, you’ve had the washing machine 3 years. Did the problem just start now, or has it always done this?

I have had the washer for 3 years. I just noticed the pulsation in the beginning of November.
I did change out my cfl lights to LEDs at the beginning of October, but there has been LED lights in my bathroom for the last four years that are pulsating now.

Gar:

Thank you. I did a poor job of presenting my information.

Before I move my lights over to a different phase I’m going to try to find out if these amperage’s are normal for the washer or not.

 

[synchro]

If the pulsing has gotten more noticable recently, perhaps there are electrolytic capacitors after the rectifier of the VFD in the direct drive washer that have degraded. That might allow higher peaks of current to be drawn from the AC power line, which could then accentuate light flickering. ... Just a thought.

 

[Asiel]

If the pulsing has gotten more noticable recently, perhaps there are electrolytic capacitors after the rectifier of the VFD in the direct drive washer that have degraded. That might allow higher peaks of current to be drawn from the AC power line, which could then accentuate light flickering. ... Just a thought.

That’s a good theory. I’ll pull the washer apart and do some testing when I have time later this week.

 

[kwired]

A hair dryer would work, but is not likely to be available in as high a power. A high powered heat gun would be more likely to be useful.

heat guns or portable heaters are typically only 1500 watts max (if they have a 15 amp 120 volt cord cap). 1600, 1800 or even 2000 watt hair dryers are pretty common anymore.

 

[gar]

Asiel:

Some information. At my bench in the basement I have about 50 ft of #8 copper from the main panel to the bench area, then 15 to 20 ft of #12 copper. In this path is also 3 breakers in series and the main panel main fuses.

At my bench on the unloaded phase I get a rise in peak voltage from 171.6 to 175.5 from a 50A peak current pulse of 6 cycles on the opposite phase. This is a 3.9 V peak change, increase, or 2.8 V RMS change on the unloaded phase. The 15 W incandescent on this unloaded phase showed negligible flicker.

In tests I have previously performed I felt the amount of flicker from a CREE 9.5 W LED was about the same as a 15 W incandescent for same pulse signal change. Both bulbs were in parallel. I prefer the 15 W incandescent because it has no electronic garbage associated with it.

I see no reason for you to dig into your dryer at this point. You need to learn more about some basic characteristics of various components and circuits.

.

 

[Asiel]

Asiel:

Some information. At my bench in the basement I have about 50 ft of #8 copper from the main panel to the bench area, then 15 to 20 ft of #12 copper. In this path is also 3 breakers in series and the main panel main fuses.

At my bench on the unloaded phase I get a rise in peak voltage from 171.6 to 175.5 from a 50A peak current pulse of 6 cycles on the opposite phase. This is a 3.9 V peak change, increase, or 2.8 V RMS change on the unloaded phase. The 15 W incandescent on this unloaded phase showed negligible flicker.

In tests I have previously performed I felt the amount of flicker from a CREE 9.5 W LED was about the same as a 15 W incandescent for same pulse signal change. Both bulbs were in parallel. I prefer the 15 W incandescent because it has no electronic garbage associated with it.

I see no reason for you to dig into your dryer at this point. You need to learn more about some basic characteristics of various components and circuits.

.

How quickly are you cycling it on and off? All of my light fixtures have at least two bulbs so maybe the amount of light makes it more noticeable.

Here is a screen shot at my main panel on the loaded phase while my washer was on. PQM screenshot

Are you suggesting the issue is not in my dryer and is in my house wiring or are you making the assumption that I cant read a line diagram?

I popped the top-off the washer's filter assembly today and verified resistance was 0hms from incoming hot across the fuse and the isolation transformer and verified O ohms for the neutral across the isolation transformer. I verified that 120v at the main PCB board from the filter. I'm unable to actually test the Suppression capacitor in the circuit. These boards are not made for me to remove the cap from the circuit to test for leakage. These boards are suspended in this clear gel? in a plastic tray. Not sure if its worth the effort. I did notice a sound coming from the filter assembly that sounded similar to chatter associated with a relay when power was returned. It wasn't like a hum you usually hear with a transformer. I didn't see any evidence of heat or disfiguration of any components.

I verified that the motor windings were good. W to U, W to V , U to V were all aprox . 10ohms. No significant differences which is in range according to the manufacturer.

 

[gar]

211220-2343 EST

Asiel:

What I am suggesting is that you play with some items where you pretty well know how they actually work. A this point you dryer is somewhat of an unknown device.

I don't think two or more bulbs on the same circuit is much different than one bulb. The reason I like the15 W incandescent is that it is a simple device with very consistent known characteristics, and not extremely bright. In fact I would like to reduce its brightness with a optical filter.

Your waveform of motor excitation has what duration? It may appear to be a large number of 60 Hz cycles. Possibly 15 seconds or much longer.

When you see lights dim is it just momentary, or is of the same duration as the time the motor is powered? I can expect a motor to have a short duration of starting current, like 6 to 20 60 Hz cycles. If dimming occurs for a very short time compared to the motor run time, then the current shown in your plot may not indicate how large starting current is.

If you used a Fluke 87 in pulse mode ( 1 millisec ) a large starting current, greater than 15 A, might be detected.

An easy test is to put a 15 W test light on the unloaded phase and watch what kind of light pulse occurs.

.