We just finished a custom home with 2 - 200Amp panels. I noticed that on occassion the lights throughout the house would flicker very rapidly, almost unnoticable, then after a few minutes they would stop flickering. I began checking loads on everything from individual circuits to total load at the meter. I turned on as much load as possible, including ALL 4 air handlers with as much as 15KW of heat on a single AHU. The lights never dimmed, flickered or anything. Then we turned on the hot water at one of the sinks and the lights began flickering again. The house is equipped with 4 - 60 Amp insta-hot water heaters. I then turned off all other breakers except the insta-hots and the lighting circuits. 2 of the 4 insta-hots would make the lights flicker. With only a total of 48 amps load on the panel the lights would still flicker. One of the water heaters is in a different panel than the lighting circuits but still causes all the lights to flicker rapidly. The 2 panels are fed seperately from a single meter with double barrel lugs. The service is an existing one from when the house was originally built, the house has since been ENTIRELY renovated. The lugs have been checked and seem to be tight. Any suggestions on what the problem may be would be awesome, and I will check the thread regularly to answer any questions, thank you.
Flickering lights????
- Thread starter mgotgame
- Start date
- Status
peter d
Senior Member
- Location
- New England
I think an important distinction needs to be made.
Are the lights truly flickering (erractically getting brighter and dimmer)?
Or are they momentarily dimming? (Getting dim, staying dim, then returning to full brightness)
Momentary dimming is a fairly normal condition when motor loads start. However, flickering is abnormal and suggests some kind of loose connection.
Are the lights truly flickering (erractically getting brighter and dimmer)?
Or are they momentarily dimming? (Getting dim, staying dim, then returning to full brightness)
Momentary dimming is a fairly normal condition when motor loads start. However, flickering is abnormal and suggests some kind of loose connection.
- Location
- Chapel Hill, NC
- Occupation
- Retired Electrical Contractor
If these water heaters have some type of electronics built in then I suspect there is something in there that is causing the load to make and break repeatly until it final catches. I would say similar to a contactor not engaging but I don't it has a contactor but some relay that is doing the same.
- Location
- Chapel Hill, NC
- Occupation
- Retired Electrical Contractor
Sounds like it's time to get the person who installed the units to get a service man out there. It doesn't seem to be your problem unless you furnished the heaters.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
There is a very real possibility that everything is working in a correct but annoying fashion.
I bet that when you turn on all of the various large loads, that the lights get dimmer and the supply voltage drops. But the lights simply get only a little bit dimmer and stay dim, so the eye doesn't notice it.
But the water heaters are large loads that are cycling on and off rapidly. The lights don't get significantly dimmer with these loads rather than the air handlers, but the load is flickering, and the lights produce a noticeable flicker.
I bet that the water heaters have control electronics that regulates the output temperature. This control is intentionally and rapidly turning the heating element on and off to maintain a desired average power level and output temperature. While in theory this control could be the equivalent of a large electronically controlled dimmer, adjusting output each half cycle, there is a very real possibility that this control uses a solid state relay and is cycling the heating element at a much slower, more noticeable rate.
The curious thing is why two of the heaters cause the problem and two do not. It may be the settings; two of the heaters may not need to modulate their output; perhaps with a lower output temperature or lower flow rate they would also cause flicker problems. Or perhaps two of the heaters are broken. If the manufacturer was clever, then they would design a system that modulates the output at high frequency, but has a failure mode that keeps working at lower frequency.
-Jon
I bet that when you turn on all of the various large loads, that the lights get dimmer and the supply voltage drops. But the lights simply get only a little bit dimmer and stay dim, so the eye doesn't notice it.
But the water heaters are large loads that are cycling on and off rapidly. The lights don't get significantly dimmer with these loads rather than the air handlers, but the load is flickering, and the lights produce a noticeable flicker.
I bet that the water heaters have control electronics that regulates the output temperature. This control is intentionally and rapidly turning the heating element on and off to maintain a desired average power level and output temperature. While in theory this control could be the equivalent of a large electronically controlled dimmer, adjusting output each half cycle, there is a very real possibility that this control uses a solid state relay and is cycling the heating element at a much slower, more noticeable rate.
The curious thing is why two of the heaters cause the problem and two do not. It may be the settings; two of the heaters may not need to modulate their output; perhaps with a lower output temperature or lower flow rate they would also cause flicker problems. Or perhaps two of the heaters are broken. If the manufacturer was clever, then they would design a system that modulates the output at high frequency, but has a failure mode that keeps working at lower frequency.
-Jon
dbuckley
Senior Member
- Location
- Canterbury, New Zealand
This sounds awfully like "burst fire" triac control. Light dimmers use phase angle control to control the amount of power flowing to a lamp, and hence it's brilliance. However, phase angle dimmers have the disadvantage of generating large amounts of interference due to the fast current risetime, which generally required chokes to reduce the risetime and thus the interference generated.
Problem is, as current goes up, so does choke size, which is why a typical SCR stage lighting dimmer is flippin' heavy, its all the chokes, which are, at the end of the day, just heavy globs of metal.
So the alternative method of power control is to use burst fire control. This way the triac is switched on the zero cross point of the mains half-cycle, and thus the triac is either conducting or non-conducting for the entire half-cycle. No nasty fast current risetime, so zero interference, so no chokes. Now we can't do this for lights, as they have so little thermal inertia that they flash, but we can and do use it for heating loads.
So thats what happens under the covers, Jon explained the rest of the whys and wherefores above. I suspect that the reason some flicker and others dont is down to volume of water flowing; if the volume is low then the heater has to reduce the output to stop the water from boiling.
I think you only have a small number of options.
If the HO is wedded to instant heat, then you need a lower rated heater unit, such that with the lower volume of water flowing the heater doesn't need to modulate.
Or, swap out the instants for a small storage tank heater.
Or, break out the lighting circuits and supply them through something like one of these. Get it on sale or return though! I know these things, although flogged as frequency converters, they will happily output 60Hz, and should supply stabilised voltage, so the end of the flickers. Or they could have an on-line UPS, which is really the same thing with batteries; that means that if the power fails the lights stay on, whilst the generator starts up
Problem is, as current goes up, so does choke size, which is why a typical SCR stage lighting dimmer is flippin' heavy, its all the chokes, which are, at the end of the day, just heavy globs of metal.
So the alternative method of power control is to use burst fire control. This way the triac is switched on the zero cross point of the mains half-cycle, and thus the triac is either conducting or non-conducting for the entire half-cycle. No nasty fast current risetime, so zero interference, so no chokes. Now we can't do this for lights, as they have so little thermal inertia that they flash, but we can and do use it for heating loads.
So thats what happens under the covers, Jon explained the rest of the whys and wherefores above. I suspect that the reason some flicker and others dont is down to volume of water flowing; if the volume is low then the heater has to reduce the output to stop the water from boiling.
I think you only have a small number of options.
If the HO is wedded to instant heat, then you need a lower rated heater unit, such that with the lower volume of water flowing the heater doesn't need to modulate.
Or, swap out the instants for a small storage tank heater.
Or, break out the lighting circuits and supply them through something like one of these. Get it on sale or return though! I know these things, although flogged as frequency converters, they will happily output 60Hz, and should supply stabilised voltage, so the end of the flickers. Or they could have an on-line UPS, which is really the same thing with batteries; that means that if the power fails the lights stay on, whilst the generator starts up
- Location
- Chapel Hill, NC
- Occupation
- Retired Electrical Contractor
There has to be something wrong with 2 of the tanks. Why do 2 work fine and 2 cause the problem. I still think the issue is in the heater electronic controls in 2 of the units. Call the service guy.
ELA
Senior Member
- Occupation
- Electrical Test Engineer
The explanations of the heaters "duty cycle" controlling Triacs makes sense to me. It also makes sense that a couple of the units may be near temperature setpoint and thus the duty cycle is low and a couple of the others may be far below setpoint and the duty cycle high. With the difference in duty cycles one may cause a more visable light blink than the others.
If you have access to a scope and clamp on current probe ( or line monitor) that would tell the story.
I have always noticed the lights blink rapidly when when my laser jet printer is printing. I have been curious as to how much current they might draw when printing to cause this.
Just for fun I am posting a current waveform since I just learned how to host pictures and wanted to try the link.
This is an example of how the current draw, when intermittent , will enhance your perception of the lights dimming.
The real small peaks are an indication of what normal idle current is vs. the 7.5 amp peaks ( 100mv = 1 amp) when printing, that causes the lights to rapidly oscillate.
If you have access to a scope and clamp on current probe ( or line monitor) that would tell the story.
I have always noticed the lights blink rapidly when when my laser jet printer is printing. I have been curious as to how much current they might draw when printing to cause this.
Just for fun I am posting a current waveform since I just learned how to host pictures and wanted to try the link.
This is an example of how the current draw, when intermittent , will enhance your perception of the lights dimming.
The real small peaks are an indication of what normal idle current is vs. the 7.5 amp peaks ( 100mv = 1 amp) when printing, that causes the lights to rapidly oscillate.
- Status