florecent fixture on gfi

Status
Not open for further replies.

michael

Member
Location
Florida
Under the cabinet flo fixture on its own switch protected by newly installed GFI.When light is turned on the fixture works properly.As soon as you turn light off it trips GFI.Not every time sometimes,it takes 4 or 5 tries on and off before it trips.GFI is wired correct,switch is correct,fixture is correct and grounded.Fixture has a starter.Any suggestions would be helpful thank you.
 

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
Do as the other say or change the fixture and see what happens. Are you certain there isn't an intermittent problem?
 

cadpoint

Senior Member
Location
Durham, NC
I this a continuation of a circuit without a ground? Granted the GFCI doesn't need one to operator correctly.
But it's been written here many time that flourscents (modern ballast) do need that complete bond.
 

magoo66

Member
Just a stab in the dark here...old flourescent starters have a small capacitor, could there be enough current flow on the neutral when the circuit is opened and the cap. discharges to trip the GFCI? The fact that it does not happen every time could depend on what part of the sine wave is interrupted. Like I said, just a stab in the dark. On a different note, undercab light on a GFCI? You didn't tap off the S.A. ckt in the kitchen did you?
 

TobyD

Senior Member
Not meaning to hi jack the question ,but,recently we've been having a similiar occurence with the bath vent light combo units doing the same thing you described.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110102-2221 EST

The problem is most likely a large transient voltage generated when the inductive circuit (ballast) of the lamp is open at a point in the cycle when maximum current is flowing.

The important identifying point is the tripping occurs on turn off and not on turn on.

e = N df/dt where e is the instantaneous voltage generated , N is a constant at least including the number of turns, and df/dt is the rate of change of flux linking the turns relative to time. Current flowing sets up a magnetic field. No current and the field collapses producing a high rate of change of flux and thus a large voltage spike.

Another way to look at this is: You can not instantaneously change the current in an inductive circuit. If you try, the the inductor will generate whatever voltage is necessary to maintain that current flow. If there is no voltage limiter, then a high enough voltage is developed to cause an electrical discharge somewhere. In some cases this might cause insulation breakdown.

Arcing across relay contacts is a result of an inductive kick.

This principle has been used in automotive ignition systems for over a century.

One solution might be to put a low pass filter between the lamp and the GFCI.

.
 
Status
Not open for further replies.
Top