Question on Christmas light strings and GFCI's continuously tripping???

[jango]

Can I get some feedback/input on why GFCI receptacles continously trip on a project of a friend of mine where they have Christmas light strings installed, via timers, throughout their landscaping (in trees, etc.). This seems to be a problem every year and I really want to get to the bottom of it. Any help or suggestions would be greatly appreciated.

 

[John120/240]

Maybe the problem isn't with the GFCI's. They are doing the job they were installed for.

More than likely there is a ground fault in the lights.

 

[davedottcom]

Can I get some feedback/input on why GFCI receptacles continously trip on a project of a friend of mine where they have Christmas light strings installed, via timers, throughout their landscaping (in trees, etc.). This seems to be a problem every year and I really want to get to the bottom of it. Any help or suggestions would be greatly appreciated.

Unplug everything, reset the GFCI, start plugging the lights in again, one thing at a time... when the GFCI trips you found the set of lights with the problem.

 

[kwired]

These "holiday" light sets are full of potential places for a ground fault to happen, and likely a big contributor to why GFCI protection is required for outdoor receptacles.

The GFCI is likely doing exactly what it is supposed to do and there is a fault in the plugged in equipment.

 

[ELA]

I am guessing it is intermittent in nature? Does it happen more frequency when the cords are wet?
Of course there can be a string or timer with a very real fault.

Beyond that GFCI's can only drive a certain amount of load capacitance. The more cords and timers that are added the more leakage.
It is possible that he has so many cords that the capacitive leakage is very close to making the device trip all the time. Then when you add some additional leakage due to moisture/noise ... ??? Then they intermittently trip.

The solution I recommend and have implemented previously with success is to install multiple GFCI outlet devices to feed the Christmas display. Then divide the load up amongst several devices. That way the total capacitive leakage is divided up between devices. No one device experiences the total capacitive ( or moisture related) leakage.

If you want to get real serious purchase a meter that measures leakage current. They are spendy but they can tell you exactly what type of leakage you are dealing with.

 

[G._S._Ohm]

. . .GFCI's can only drive a certain amount of load capacitance.

. . .purchase a meter that measures leakage current.

About 900' of Romex w/gnd at 100 pF/ft. You can check this by putting 120vac into the hot lead and have a milliammeter from the gnd lead to gnd. Use a very small fuse (1/100 A) in series with your meter.

Or open up a bad GFCI and use it to measure the current imbalance.
Probably the toroid and IC inside still work but you have to find a pin on the IC that has a voltage that corresponds to mA of leakage current. Nowadays, if you have the numbers off the IC you can pull up a block diagram of the gizzards.
You can test it by using various resistor values to gnd.

 

[ELA]

About 900' of Romex w/gnd at 100 pF/ft. You can check this by putting 120vac into the hot lead and have a milliammeter from the gnd lead to gnd. Use a very small fuse (1/100 A) in series with your meter.

100pf/ft sounds a little high. Can you please cite where that comes from? I may have to adjust some of my previous estimations based on a lower number.
Extension cords typically have a higher capacitance than romex.

In a real world scenarios such as the Christmas light scenario there is much more at work than a simple theoretical romex estimation might predict.

Or open up a bad GFCI and use it to measure the current imbalance.
Probably the toroid and IC inside still work but you have to find a pin on the IC that has a voltage that corresponds to mA of leakage current. Nowadays, if you have the numbers off the IC you can pull up a block diagram of the gizzards.
You can test it by using various resistor values to gnd.

Been there, tried that:

A few months back I took a Leviton GFCI apart to better trace out the circuit in hopes of taping into it so that I could use the pin 5 (op-Amp) output as an indication of the differential current indication. My hopes were that I could provide some instructions as to how an electrician might modify a GFCI outlet to use as a cheap tester for when they want to quantify leakage currents.
They could bring the pin 5 out to a DV meter and use a lookup table to measure differential currents.

For my own purpose I had hoped to interface it to a Microcontroller with an LCD to directly indicate the current values.

Purchasing a leakage current meter is out of the question for my occasional desire to use one and I am guessing that is probably true for most electricians as well.

Unfortunately I found that they do not actually use the LM1851 as is shown in the LM1851 datasheet . Pin 5 actually drives the GND/NEUTRAL COIL and they utilize a totally different technique to sense a downstream neu-gnd connection.

Pin 5 goes to C9 in series with C3, both sides of C9 connect to the Inject coil (200:1). I have traced this out pretty thoroughly.

Here is the link that explains a little more about why this is connected this way: ( text in between schematics)

http://www.idea2ic.com/GFI/GFICs.html

 

[G._S._Ohm]

100pf/ft sounds a little high. Can you please cite where that comes from?


A few months back. . .

I measured the reactive current of 10' of NMB .

I guess then I'd search for some voltage on the IC pins that corresponds in a linear or non-linear way to leakage current.
After the holidays I hope I get time to put my actions where my mouth is. By then I will have two GFCIs to mess with.