GFCI Behavior

[jim6918]

Do they ever get weak? Last night during the show, I noticed that one of my controllers was out. Temperatures were in low teens and about 4 inches of fresh snow from the day time hours. Fortunately, my load center and controllers are all out of the sight of viewers. I found that the GFCI for that controller had tripped. The circuit is a true 20 amp circuit, with only <15 amp load to channels. I reset, it and it stayed on for about 10 minutes and then tripped. I immediately suspected one of my 6 incand. mini trees because the alligator plugs lay directly on the ground, although I have never had an issue in past two years. I unplugged the 6 trees (13.2 amps) and let the controller run for about 15 minutes and it stayed on. I then started plugging the trees back in one at time, with some in between each until I had them all back on. Mind you I was doing this in the dark while the show was running out front. A few minutes later it tripped again. Logic would tell me the fault is with the last tree plugged in. I left that tree off the rest of the night with no problems. Today I plan on unplugging everything except the suspect tree and see what happens. I also plan on clearing off the alligator plugs and drying them out with a hair dryer and zip tying them higher off the ground.

Question: since I have the time and a spare should I replace the GFCI? I am a little suspicious about it since the show ran for about 90 minutes before it tripped, and when I plugged the last tree back in, there wasn't an immediate trip like you should see if there was a direct short. Has anybody else had GFCI issues that could only be traced to the GFCI itself and not something else?

[KStatefan]

Yes GFCI can get weak and trip at a lower than 5ma leakage. There are testers available to test that will tell you at what point they are tripping.

[jim6918]

KStatefan wrote:

Yes GFCI can get weak and trip at a lower than 5ma leakage. There are testers available to test that will tell you at what point they are tripping.

I will have to look into that ASAP. Seems almost as valuable as an amp meter. I couldn't stand it, so I replaced the GFCI this afternoon. Ran show for 3 hours without a trip. I also found an alligator plug to one of the mini lights that the cover hadn't been completely closed. Keeping fingers crossed for tonight. More snow forecast; in fact snowing right now.

[cmoore60]

The current leakage on the trees can be accumaltive to the point of tripping the GFCI. A few years ago my mini trees were tripping the GFCI. I sat one morning and tried many combinations. Decided at that time the current leakage was accumlative.

Chuck

[plasmadrive]

I don't think they are accumulative. :-}

They are pretty instantaneous though. Once you get that imbalance between hot and neutral and it exceeds the trip threshold, it is game over....TRIP

They do get weak however. I have GFCIs that have lasted years and some that last as little as six months. I also only buy the ones rated for outdoor use now. They seem to last longer.

[Dave Pursel]

FYI. I have 12 controllers of which only two are indoors. This is my fourth year and I've just started having a lot of trips where I've never had one before. We have had a lot of rain lately. Not that I haven't in past years. I did about 2 hours of studying GFCI's the night before last. If they are Class A, which most are, they are designed to trip at a 4-6 milliamp inbalance. They don't seem to make an adjustable trip GFCI. I tried to locate a GFCI that had a slightly higher trip range in the 6-8 but to no avail. 9-15 milliamps (depending on male vs female adults) is when your muscles won't let go and the danger factor goes up drastically. Note you will still get a hell of a shock with a GFCI in the circuit, but if it's working it shuts off the current flow to your body before your heart stops. The shock is still just as strong with a GFCI and still hurts I might add, as I've been there, done that. A class B GFCI is out of the question as it's trip range is in the 1000 milliamp range, way too high. Might as well take the GFCI (which I was seriously considering) out of the circuit if you intend to use one of these. If I wasn't worried about someone's child coming into the yard I probably would take them out as I know where to turn off the switch prior to working on them. Anyway, my show was a half hour late starting last night because of the following changes I made. I had four power cords running to two GFCI recepticals with each receptical being on a seperate 15 amp circuit indirectly. I added two additional receptical boxes and bought 4 new $20 Leviton Outdoor/Tamper proof rated GFCI's. Two boxes/GFCI's being on each circuit. Where I had the four power cords sharing the previous two GFCI's I now have just one power cord going to one outlet of each GFCI and leave the other outlet empty. This gives me better control of what each GFCI is powering thus making troubleshooting a little easier when a trip occurs. And if a trip occurs there's less controllers that lose power. On a side note, over the years I have had 2 GFCI's that are down by the lake go bad, but neither would reset at all when they went bad. By the way. I made these changes last night while it was raining and I never had a GFCI trip during the show last night.

[jim6918]

I really hate the unknown. I wish that I had been able to isolate my trips to a particular channel first and then eliminate the GFCI secondarily. I didn't have the luxury of trying a seond night after having pulled the alligator plugs up off the lawn along with replacing the GFCI. All I know is that my show ran last night in the same environmental conditions without any trips. What was throwing me the first night was that the trips were not instantaneous when all the channels were plugged in.

[Jim Saul]

Jim, you can try bypassing the controller with a 6 outlet strip. power up the channels and see which channel is popping the GFI.

[jim6918]

Jim Saul wrote:

Jim, you can try bypassing the controller with a 6 outlet strip. power up the channels and see which channel is popping the GFI.

I am past that point now. I replaced the suspect GFCI yesterday and also pulled the alligator plugs to the suspect mini trees up off the lawn. Show ran fine last night, so I am keeping my fingers crossed. Of course, I still have about 80 alligator plugs laying directly on the lawn on other controllers that could give me fits, but hasn't happened so far.

[Steven]

KStatefan wrote:

There are testers available to test that will tell you at what point they are tripping.

The perfect piece of test equipment here is a differential current meter. Maybe that's not the right name because I couldn't find anything on Google.

This tester would have a plug, a socket, and a digital or analog display. You plug the device to be tested into the socket, and plug the tester into power. The meter would tell you, down to .1mA resolution, the amount of ground leakage.

You would plug it into your LOR controller, and turn each channel on one at a time to see which ones will add up to 5mA when they are all on.

If anyone knows where to get a tester like this, I'm sure everyone here, especially Jim, would really appreciate it.

[-klb-]

I have not found a plug in one like that, but there are expensive beasts like the Fluke 360 clamp meter, designed to do differential current meausurement like that.

[Steven]

-klb- wrote:

but there are expensive beasts like the Fluke 360 clamp meter, designed to do differential current meausurement like that.

That's a great idea! I didn't think about a clamp meter.

There is a cheap one for $9.99 at Harbor Freight, but its lowest range is 20A with 3 1/2 digits. That means its resolution is 10mA, which is not low enough. I'll have to keep looking.

Edit: I found this one from a Taiwanese distributor on alibaba.com. This sounds like the perfect meter for the purpose if the price is reasonable. I have asked for a quote and will update again when I get a reply.

[-klb-]

I have a true RMS Fluke clamp meter, at about half the price of the 360, but below 1A current flux, I have no faith in it. Get up over 2A, and I find that at minimum the precision (repeatability) seems to go way up, and appears good to .1 +/- .1. Not sure about the actual accuracy. But below 1A, it just doesn't seem to have that same repeatability. I'm not saying that it makes sense on any level. Just repeating my experiences.

[Steven]

-klb- wrote:

I have a true RMS Fluke clamp meter, at about half the price of the 360, but below 1A current flux, I have no faith in it.

I understand from what I have read that AC/DC clamp meters use a Hall effect sensor to detect current, which does not usually offer high-resolution. AC-only clamp meters, an the other hand, use a coil, which can offer better resolution at lower currents. That's how that AC leakage meter I posted above works.