Pulsing lights
- Thread starter cjf
- Start date
- Status
- Location
- Bremerton, Washington
- Occupation
- Master Electrician
When you ssy its a new house you did, do you mean you wired? We don't allow DIY posts, I will lock this, send me a message if I have misjudged this
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
- Location
- Bremerton, Washington
- Occupation
- Master Electrician
A reminder to fill out your profile.
To the OP, have you checked voltage, with all loads off, and then with power on. Then check voltage with washer running
To the OP, have you checked voltage, with all loads off, and then with power on. Then check voltage with washer running
I only had about 30 min there yesterday so all I got to was checking the connections and they’re good. I’ll be stopping back to do a more thorough check next week. Laundry is a dedicated circuit and the lights are on a 15 amp lighting circuit. But it’s only two lights doing it. I’ll take any suggestions to not be there all day since this is on us seeing we warranty all work for a year. Plus I don’t like doing things wrong and want to make sure I do everything I can on my end. Thanks
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
What kind of lights are they?
Ragin Cajun
Senior Member
- Location
- Upstate S.C.
It they are LED, swap them out/around.
- Location
- Bremerton, Washington
- Occupation
- Master Electrician
If you have two of the same lamps, swao one that flickers with one that does not. See if the flicker follows or stays.
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
See if one incandescent bulb has any effect on the flicker.
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
In addition to what Tom mentioned, I suggest unplugging the washing machine and plugging in a portable heater instead. Then check the voltages at the washing machine's outlet and at the lighting circuit, with and without the heater turned on. A few volts drop to the heater would be acceptable, but any change on the lighting circuit should be quite small.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
210209-2003 EST
cfg:
I think you have not had much training in electrical theory and troubleshooting. Sometimes problems can be simple to solve and other times very hard. When it seems hard, then it is good to follow a procedure that will hopefully lead you to a problem location.
I would start at the main panel. Here your voltages will change the smallest amount for load changes. I like to use a 1500 W spacer heater as a test load. At about 120 V this is a load change of about 12.5 A between ON and OFF.
At the main panel the voltage difference between the Neutral bus bar (common) (which is connected to the center tap of a 240 V center tapped single phase transformer, or neutral wire from a three phase transformer), and the EGC bus should be down in the millivolt range.
I would use the neutral bus at the main panel as my voltage reference point. If EGC to neutral voltage is small, there are no appreciable currents on EGC, and EGC is good to all locations, then I can use an EGC anywhere as a tested lead back to the main panel neutral. If this is not the case, then use an extension cord or a single piece of wire from neutral at the main panel as a long test lead to where you want to make a voltage measurement relative to the common of the main panel.
Suppose I get a voltage change at the main panel of 1 V between a hot phase conductor and the common bus for a 12.5 A load change on that phase, then I expect less change on the other phase. Possibly 1/2 as much. You won't see much of any light flicker in an incandescent with this 1 V voltage change at a nominal 120 V. I suspect that you won't find, at this amount of source impedance at the main panel, that there is much flicker at the main panel from the washer starting and stopping.
With background on the voltage change, at the main panel on the phase the washer is on, from starting and stopping of the washer, then look at the voltage change at the washer. This is possibly a number of volts.
Where your light is located see what the voltage change is from the washer cycling.
Next you look for what part of the light and washer circuits are common. Could be a neutral only or both neutral and hot. Use voltage measurements to narrow down where the commonality is. Difference in change in voltage between the washer and the light may give you an idea of where commonality exists.
Using EGC or a separate test lead wire to get a reference back to the main panel can help a lot in searching for commonality.
See where some measurements like these take you.
.
cfg:
I think you have not had much training in electrical theory and troubleshooting. Sometimes problems can be simple to solve and other times very hard. When it seems hard, then it is good to follow a procedure that will hopefully lead you to a problem location.
I would start at the main panel. Here your voltages will change the smallest amount for load changes. I like to use a 1500 W spacer heater as a test load. At about 120 V this is a load change of about 12.5 A between ON and OFF.
At the main panel the voltage difference between the Neutral bus bar (common) (which is connected to the center tap of a 240 V center tapped single phase transformer, or neutral wire from a three phase transformer), and the EGC bus should be down in the millivolt range.
I would use the neutral bus at the main panel as my voltage reference point. If EGC to neutral voltage is small, there are no appreciable currents on EGC, and EGC is good to all locations, then I can use an EGC anywhere as a tested lead back to the main panel neutral. If this is not the case, then use an extension cord or a single piece of wire from neutral at the main panel as a long test lead to where you want to make a voltage measurement relative to the common of the main panel.
Suppose I get a voltage change at the main panel of 1 V between a hot phase conductor and the common bus for a 12.5 A load change on that phase, then I expect less change on the other phase. Possibly 1/2 as much. You won't see much of any light flicker in an incandescent with this 1 V voltage change at a nominal 120 V. I suspect that you won't find, at this amount of source impedance at the main panel, that there is much flicker at the main panel from the washer starting and stopping.
With background on the voltage change, at the main panel on the phase the washer is on, from starting and stopping of the washer, then look at the voltage change at the washer. This is possibly a number of volts.
Where your light is located see what the voltage change is from the washer cycling.
Next you look for what part of the light and washer circuits are common. Could be a neutral only or both neutral and hot. Use voltage measurements to narrow down where the commonality is. Difference in change in voltage between the washer and the light may give you an idea of where commonality exists.
Using EGC or a separate test lead wire to get a reference back to the main panel can help a lot in searching for commonality.
See where some measurements like these take you.
.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
210217-2247 EST
It appears the original poster has disappeared. This is not unusual, but does not provide the readers of the thread with useful information that might benefit the reader.
.
It appears the original poster has disappeared. This is not unusual, but does not provide the readers of the thread with useful information that might benefit the reader.
.
ptonsparky
Tom
- Occupation
- EC - retired
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
210221-1645 EST
ptonsparky:
Following are some results of experiments at my home:
My pole transformer is a 50 kVA 240 V center tapped single phase fed from a 3 phase delta. About 90 ft of service drop wire. 50 to 60 ft is wimpy, and the remainder 0000 copper. My main panel is a 200 A SqD with copper bus bars. My electric meter is a bolt on vs plugin.
My measuring instruments are a Fluke 27 with MIN-MAX, a Rigol scope, and a Fluke 200 A Hall current probe. Test loads are 1500 W 120 V space heater, a Montgomery Ward 1/3 HP 120 V AC induction motor with no external mechanical load, and a 15 W 120 V incandescent bulb.
Space heater at main panel ---
Heater is switched on phase 2. Voltage measured at the main lugs at the panel input.
Phase 1 voltage changes from 123.6 to 123.9, a +0.3 V change, with the about 12.5 A load change.
Phase 2 voltage changes from 123.6 to 122.9, a -0.7 V change, with the about 12.5 A load change.
These changes make sense. Think about the actual circuit from the pole.
Space heater is now at the end of a 50 ft run, from the main panel, of #14 copper Romex ---
Heater is still on phase 2 and no need to measure change of phase 1.
Phase 2 voltage change at the end of the 50 ft run is from 123.0 to 118.0, a 5.0 V change, with about 12 A load change.
Very little flicker in the 15 W bulb at the end of the 50 ft Romex.
Note: that unloaded initial voltage changes somewhat because the power company primary supply has voltage changes from other users load changes, or even other changes within your home. What we are interested in is the change in voltage that results from our load change. Thus. one has to wait sometimes to see that input voltage is moderately stable, and hope it is stable during your experiment.
Next we look at results from motor starting.
Fluke measurement.
The 1/3 HP motor is switched at the end of the 50 ft Romex ---
Fluke measurement in MIN-MAX mode is 122.9 V dropping to 119.9V, a 3.0 V change. Load change is 50 A. Doesn't seem to correlate. Reason is the Fluke does not respond fast enough.
Scope measurement.
The 1/3 HP motors is switched at the end of the 50 ft Romex ---
Peak voltages and currents were measured. Calculated RMS values are shown. Obviously reading of scope produces some errors.
The starting current duration was 5 cycles of 16.7 mS each, and a 1 cycle transition to turn off of the starting coil. So we will call starting duration as 6 * 1/60 = 0.1 second.
Voltage before start is 180 * 0.707 = 127.3 V. During start this drops to 160 * 0.707 = 113.1 V. Thus, a voltage change of 14.2 V. There is very apparent flicker of the 15 W incandescent.
This should be enough for now. If we knew more from the original poster it should be possible to help him find his problem.
.
WITH NO EXTERNAL MECHANICAL LOAD, 4444
ptonsparky:
Following are some results of experiments at my home:
My pole transformer is a 50 kVA 240 V center tapped single phase fed from a 3 phase delta. About 90 ft of service drop wire. 50 to 60 ft is wimpy, and the remainder 0000 copper. My main panel is a 200 A SqD with copper bus bars. My electric meter is a bolt on vs plugin.
My measuring instruments are a Fluke 27 with MIN-MAX, a Rigol scope, and a Fluke 200 A Hall current probe. Test loads are 1500 W 120 V space heater, a Montgomery Ward 1/3 HP 120 V AC induction motor with no external mechanical load, and a 15 W 120 V incandescent bulb.
Space heater at main panel ---
Heater is switched on phase 2. Voltage measured at the main lugs at the panel input.
Phase 1 voltage changes from 123.6 to 123.9, a +0.3 V change, with the about 12.5 A load change.
Phase 2 voltage changes from 123.6 to 122.9, a -0.7 V change, with the about 12.5 A load change.
These changes make sense. Think about the actual circuit from the pole.
Space heater is now at the end of a 50 ft run, from the main panel, of #14 copper Romex ---
Heater is still on phase 2 and no need to measure change of phase 1.
Phase 2 voltage change at the end of the 50 ft run is from 123.0 to 118.0, a 5.0 V change, with about 12 A load change.
Very little flicker in the 15 W bulb at the end of the 50 ft Romex.
Note: that unloaded initial voltage changes somewhat because the power company primary supply has voltage changes from other users load changes, or even other changes within your home. What we are interested in is the change in voltage that results from our load change. Thus. one has to wait sometimes to see that input voltage is moderately stable, and hope it is stable during your experiment.
Next we look at results from motor starting.
Fluke measurement.
The 1/3 HP motor is switched at the end of the 50 ft Romex ---
Fluke measurement in MIN-MAX mode is 122.9 V dropping to 119.9V, a 3.0 V change. Load change is 50 A. Doesn't seem to correlate. Reason is the Fluke does not respond fast enough.
Scope measurement.
The 1/3 HP motors is switched at the end of the 50 ft Romex ---
Peak voltages and currents were measured. Calculated RMS values are shown. Obviously reading of scope produces some errors.
The starting current duration was 5 cycles of 16.7 mS each, and a 1 cycle transition to turn off of the starting coil. So we will call starting duration as 6 * 1/60 = 0.1 second.
Voltage before start is 180 * 0.707 = 127.3 V. During start this drops to 160 * 0.707 = 113.1 V. Thus, a voltage change of 14.2 V. There is very apparent flicker of the 15 W incandescent.
This should be enough for now. If we knew more from the original poster it should be possible to help him find his problem.
.
WITH NO EXTERNAL MECHANICAL LOAD, 4444
Thanks for the input. I just went back today and did some testing on the light that was pulsing when the washer was agitating and there wasn’t any voltage change. I decided to put incandescent bulbs in opposed to led and there was no more pulsing. Any idea why led does it and incandescent doesn’t. Thanks
Flicker Index
Senior Member
- Location
- Pac NW
- Occupation
- Lights
Every ballast design behaves differently in response to spikes, notches, other dv/dt events and voltage variation. Even some 120/277 ballasts that can handle a wide input range can flicker when subject to abrupt voltage change.
The integrated ballast in LED lamps have to be cheap for the lamp to be affordably priced and there are many, many different LED ballast designs in use. Put in different LED lamps into a desk lamp into the same outlet as a vacuum cleaner. You will find that not all lamps experience the same level of flicker as the motor starts.
There's generally no consistency within the same LED lamp brand, but often there's similarities between the same product lineup. If it only involves a few lamps, try different lamps from your bin-o-bulbs.
Generally speaking, the ballast in integrated ballast LED lamps have inferior power regulation than an external T8 electronic ballast and experience dramatic power consumption difference at 110v vs 130v
The integrated ballast in LED lamps have to be cheap for the lamp to be affordably priced and there are many, many different LED ballast designs in use. Put in different LED lamps into a desk lamp into the same outlet as a vacuum cleaner. You will find that not all lamps experience the same level of flicker as the motor starts.
There's generally no consistency within the same LED lamp brand, but often there's similarities between the same product lineup. If it only involves a few lamps, try different lamps from your bin-o-bulbs.
Generally speaking, the ballast in integrated ballast LED lamps have inferior power regulation than an external T8 electronic ballast and experience dramatic power consumption difference at 110v vs 130v
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
210227-2140 EST
cjf:
I am running some experiments relative to your question. I am working with a 15 W tungsten filament incandescent, and a 9 W CREE LED. I have ability to adjust pulse width and amplitude change to the bulbs, and a reasonably capable Rigol oscilloscope. I expect you have no oscilloscope, and probably no access to one. However, an adequately capable scope might assist you in troubleshooting this problem.
The reason I am using CREE is that they seem to have few peculiarities. An incandescent bulb is a very simple device to characterize, an LED bulb is not. However, a single LED chip is easy to define. The drivers for LEDs is where one gets into trouble.
Some suggest that LED bulbs are more sensitive to flicker because they can produce rapidly changing light output. May not be true.
.
cjf:
I am running some experiments relative to your question. I am working with a 15 W tungsten filament incandescent, and a 9 W CREE LED. I have ability to adjust pulse width and amplitude change to the bulbs, and a reasonably capable Rigol oscilloscope. I expect you have no oscilloscope, and probably no access to one. However, an adequately capable scope might assist you in troubleshooting this problem.
The reason I am using CREE is that they seem to have few peculiarities. An incandescent bulb is a very simple device to characterize, an LED bulb is not. However, a single LED chip is easy to define. The drivers for LEDs is where one gets into trouble.
Some suggest that LED bulbs are more sensitive to flicker because they can produce rapidly changing light output. May not be true.
.
kwired
Electron manager
- Location
- NE Nebraska
This is a problem with LED's in general, especially lesser expensive ones and/or ones that are conventional incandescent replacement lamps.
Someone on this site once had a great explanation that went something like "these loads have 'low inertia'" meaning they are easily disturbed by fluctuations caused by higher intensity loads (even pretty minor fluctuations) if not designed to handle those situations.
You have a power conversion device that is drawing primarily from the peaks of the input wave, then when you introduce something that lessens the peak you are basically starving that conversion device.
Someone on this site once had a great explanation that went something like "these loads have 'low inertia'" meaning they are easily disturbed by fluctuations caused by higher intensity loads (even pretty minor fluctuations) if not designed to handle those situations.
You have a power conversion device that is drawing primarily from the peaks of the input wave, then when you introduce something that lessens the peak you are basically starving that conversion device.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
LED junctions require a constant current DC power source to operate properly, but the power wiring in a home is (approximately) constant voltage AC. So all LED lamps include some sort of driver circuit to function.
As others have mentioned, cheap or poorly manufactured driver circuits can be especially sensitive to electrical noise.
On top of this there is a difficult design problem: the simplest LED lamp driver circuits naturally supply _constant_ current to the LED junction, making their light output steady even as the supply voltage changes. This is _great_ if you want a steady light, but a real problem if you want to be able to dim the lights. So many LED lamps have some sort of circuit that attempts to detect if the supply is being dimmed, and then adjusts the output based on detecting this dimming signal.
To the OP:
You have already checked for voltage drop on the problematic lighting circuit when the washer is running, and have not detected a voltage drop problem.
You also installed incandescent lamps and saw no problem.
This suggests to me that electrical noise is triggering the dimming capabilities of the LED lamps being used. Since customers like the efficiency and life of LED lamps, you may wish to try a _non-dimmable_ lamp to see if it doesn't flicker or pulse.
If you have access to a line voltage oscilloscope, you might be able to see noise spikes on the circuit in question. You might even be able to demonstrate that the source of this electrical noise is the washer rather than a wiring error.
Also, thanks for coming back with the current status of your question! Many people ask a question and then vanish.
-Jon
As others have mentioned, cheap or poorly manufactured driver circuits can be especially sensitive to electrical noise.
On top of this there is a difficult design problem: the simplest LED lamp driver circuits naturally supply _constant_ current to the LED junction, making their light output steady even as the supply voltage changes. This is _great_ if you want a steady light, but a real problem if you want to be able to dim the lights. So many LED lamps have some sort of circuit that attempts to detect if the supply is being dimmed, and then adjusts the output based on detecting this dimming signal.
To the OP:
You have already checked for voltage drop on the problematic lighting circuit when the washer is running, and have not detected a voltage drop problem.
You also installed incandescent lamps and saw no problem.
This suggests to me that electrical noise is triggering the dimming capabilities of the LED lamps being used. Since customers like the efficiency and life of LED lamps, you may wish to try a _non-dimmable_ lamp to see if it doesn't flicker or pulse.
If you have access to a line voltage oscilloscope, you might be able to see noise spikes on the circuit in question. You might even be able to demonstrate that the source of this electrical noise is the washer rather than a wiring error.
Also, thanks for coming back with the current status of your question! Many people ask a question and then vanish.
-Jon
- Status