GFCI's

The outlets in the back of the house are on a 20A GFCI. (Note to self: are there any other outlets on that same circuit that I need to chase down?)

Presently using them to light XMAS lights on one of the citrus trees. IIRC, each string is about 250W. With ~700W on the circuit, turning the breaker ON (i.e., using it as a switch -- instead of plugging/unplugging the load) causes it to immediately trip. Repeating the action in short order appears to get it to latch and remain latched (ON), indefinitely.

OK, so the surge when all the lamps are cold can increase the inrush current -- on the short term -- to levels that probably exceed the 20A limit of the breaker.

"In theory"

OTOH, in years past, I've run the circuit at close to its capacity (~10 strings) without this problem.

I've changed breakers (swapped with one feeding another circuit) and the problem persists.

[There are no leakage paths in the wiring OUTSIDE]

This suggests something in the wiring/fixtures. I'll start isolating outlets, tomorrow (dark here, now). Perhaps some insect (leaf cutter?) has opted to nest in one of the receptacles.

Anything else I should explore?

Reply to
Don Y
Loading thread data ...

For those who didn't connect the dots, this is a GFCI *breaker*, not a GFCI *outlet* daisy-chained to feed the others.

Reply to
Don Y

Does it trip with nothing connected? If not, it's not a dauber or anything on the main circuit. It has to be on the "load". Outdoor lights on a GFCI can be a problem - lots of possibilities for "leakage"

Reply to
clare

Since you are swapping stuff around, try it on a non-GFI circuit. That will tell you if it is a ground fault. Usually this tracks back to water/bugs in a box.

I have one nightmare GFI circuit here that is longer than the design spec for GFIs but it works when everything is dry. When it fails, I end up splitting the circuit up to isolate the failure. Bear in mind, it can be a ground fault on the neutral. On outside boxes, make sure all of the wirenuts are pointed up, near the top of the box and that they are not too close to the box. It mitigates the water they all collect eventually.

Reply to
gfretwell

If there is absolutely nothing connected, a neutral fault will not trip the GFCI. The slightest load will trip it tho. (it splits the neutral current) The other thing to remember is the faults are additive. If you are leaking a couple ma in a couple places, it will work until the total gets up around 5ma or so.

Reply to
gfretwell

Appears not to. OTOH, it won't trip with smaller loads.

The problem is, I can't tell if it is tripping from a momentary "over current" condition *or* a "leak". The fact that the circuit *holds* on the second attempt suggests it is related to the turn on transient (all those ice cold filaments now having had a chance to warm up -- even only momentarily -- to higher operating resistances).

Note that the other GFCI breakers have operated flawlessly for exactly the same length of time -- with more frequent usage (the outdoor circuit is rarely used) located in the same (outdoor) electrical panel. That, coupled with swapping breakers suggests the problem isn't in the breaker (or panel).

I'll try moving the load to a different receptacle and isolating this portion of the branch circuit. That will require a "longer" extension cord (actually, the current cord is long enough; I'll just have to unwind it from its "storage spool") so I will first verify the "bad" outlet fails with that lengthened cord before trying the lengthened cord in an upstream outlet.

Reply to
Don Y

My 1979 house near Baltimore was built with one GFCI circuit that powers the front outdoor receptacle, the recepts in each bathroom, and iirc the one near the sink in the kitchen.

I bought the house when it was 4 years old, and within a couple years, the breaker kept tripping, but the replacement has only tripped twice in 30 years.

For the last 12 years I've had a heavy orange extension cord plugged into the front outdoor outlet, or into an extension cord plugged into it. The female end of both has lain in the grass for the last 12 years, 365 days a year/24, through rain and snow and sleet and gloom of night. And the breaker has only tripped twice. Many different locations, wherever I feel like throwing it at the end of the day.

I do make a point to pick it up a foot from the end when it's wet.

I use a B&D elelctric lawnmower with no problem, but i don't know its rating or other things about your question.

Reply to
Micky

Try just unwinding it from the spool and trying it again. You may just be coupling current into the grounding conductor.

Because of that, they do have specs for different wiring methods and the total length of the circuit.

Reply to
gfretwell

You're ignoring the fact that the breaker latches and HOLDS on the "second attempt". This suggests the problem is related to the change in load characteristics from the previous "brief lighting event" (fraction of a second when power surged into the lamps just before the breaker trips on the first attempt).

I'm at a loss to explain how anything else could appear and then disappear so readily!

Reply to
Don Y

Again, nothing changes on the spool between the first breaker closure (that trips) and the second (that holds).

Same extension cord used in previous years. Tree is in the same location that it was -- as is the electrical outlet (i.e., so, no more or less cord on the reel than in past years). Same strings of lights (though far FEWER of them). Same ambient temperatures. etc.

I'll have to play with it some more to get a better characterization of how it behaves (different size loads, different TYPES of loads, how long the circuit needs to "rest" before the problem will reproduce, etc.)

Reply to
Don Y

Moving the strins of lights to another circuit would take the existing wiring *and* GFCI out of the calculus.

OTOH, I could see if there's a breaker nearby that I can "borrow" onto which to move the existing wiring. I.e., I've tried changing JUST the GFCI breaker with another GFCI breaker (keeping everything else constant); this would allow me to try nonGFCI vs. GFCI (again, keeping everything else constant).

Trick will be to see how far I can reach in the electric panel to pick up a different breaker.

Boxes are recessed into block/cement walls. Faces sealed with foam gaskets. Outdoor "in use" covers keep out direct rain/water.

Bugs, OTOH, can always crawl in through the hole in the cover intended for the cords to exit.

But, bug would lead to suspected GFCI issue. And, would persist (not "clear" itself after the first breaker trip/reset)

Again, see above. IMO, it has something to do with the initial transient. I need to find a load that is more PURELY resistive to see what it's like when the load is X from the moment the breaker is FIRST closed.

Reply to
Don Y

What are yiou using for lights? It sounds like they are incandescent C9 bulbs in strings of 25 bulbs. (which is actually 225Watts). [25 times

9watts].

700W is not even close to 20A. A 20A breaker can handle roughly 2400 watts. (Of course you dont want to run it at it's maximum load). Either way, 700W is only around 30% of the capacity of the breaker. Yea, starting current is always higher, but even if it was double (1400) watts, you're still way below the allowable load limit for a 20A circuit.

Check, check twice, check 3 times and check again for any minor breaks in the light strings. One very tiny spot of bare wire, a bulb base slightly exposed and touching a tree branch, or the soil, and so on.... Christmas light strings are notorious for electrical leakage and if they're old, even more so.

Also, years ago, when I used to decorate outside of my parents house, using those C9 (and C7) strings, more than once one of the bulbs seemed to develop an internal short. Apparently the filament broke, but welded itself back together but it was shorter and drew excess current. That bulb would eventually burn out, but sometimes it would blow a fuse when it died.

So, you might check every bulb for any sign of defect, which is easier said than done. Or replace all the bulbs in one string at a time.

Better yet, just buy some C9 sized LED strings, and if you have them on every night until the end of December, they will probably pay for themselves in electrical savings. Just a guess, but a string of 25 LEDs probably use about 20 to 30 watts.

Three strings of incandescent will use 675W (Roughly 700W) every hour. So whatever you pay per KWH, is spent approximately every 1-1/2 hours.

As an example, if you pay 15 cents per KWH, Six hours per day will cost you 60 cents per night. That's a cost of $18 for one month.

Three LED strings will use around 80W per hour. 480W in 6 hours, which amounts to about 7 cents per day or $2.10 for an entire month.

You save $15.90 in that month.

Reply to
Paintedcow

Unless the first application of power heated and "dried" whatever was causing the leak. Not a very likely scenario, for sure - but possibly just within the realm of possibility??

Reply to
clare

Make up a "soft start box" getr an electric heater. take a short extension cord (or make one uo) and cut the live wire, inserting a second plug in series with the existing one. Plug the heater into the second plug, the lights into the first. This will reduce the current on startup. Wire a switch across the second plug so you can short the heater out of the circuit after the lights come on. This will tell you for sure if it is a cold surge problem. If you get a 750/1500 heater you can select how hard or soft the start is.

Reply to
clare

But the current spike on LEDs can be pretty healthy too- depending what kind of driver circuit. Just a series string, not so much - but LED drivers? OUCH!!

Reply to
clare

I'm at a loss. If there was a "partial short" (leak) that somehow cleared itself, then I wouldn't expect it to return hours later -- and the same scenario play out again. And, again.

I need to get out a watch and see how long I can wait *after* the first breaker close before the *next* closing will NOT hold. I.e., how long it takes the "problem" to reappear.

It it's on the order of a few minutes, it's unlikely that there's a "partial short" (dead bug, etc.).

Note that this is a dry time of year for us -- it's not like there are any external (or internal!) sources of water/moisture (other than body fluids of bugs :> )

Reply to
Don Y

"plug" --> "outlet"

Given that the load is so light, I can use an electric hair dryer and fiddle with the temp setting (cold/warm/hot/blistering) to gradually increase the effective load. A switch shunting the hair dryer's connection to remove it completely.

It's worth a try. I'm going to gather more data, first, so I know exactly how I can reproduce the problem (e.g., if I have to wait

2 hours for the "short" to redevelop, then any tests more frequent than that would be foolish.
Reply to
Don Y

So there are 3 strings of lights?

Try switching the breaker on with 2 out of 3 strings plugged in, i.e one string unplugged..

THen change to another 2 out of 3 etc i.e another one unplugged.

If it still trips with 2/3 , it's leakage. If it doesn't, it's surge current.

Reply to
makolber

install a regular switch or a timer after the breaker.

not only more convenient, but long term using a breaker as a switch can lead to breaker failure.

Reply to
bob haller

There are three strings "in series" (this is a misnomer; they are obviously WIRED in parallel, but, each string plugs into the string "upstream" from it. I.e., the first two strings act as an "extension cord" for the third; the first string acts as an extension for the second and third; etc.

Note my comment that, in years past, we would have up to 10 (effectively) strings on the same *single* outlet. E.g., extension cord would run to the base of the tree. There, a "cube tap" made 3 receptacles available from the ONE that is present on the extension cord. Two strings were plugged into that cube tap. Two *more* strings were plugged into the "ends" of each of these first two strings. And, two more plugged into the ends of each of those.

So, "two strings of three". Each string has 25 (?) 9W (nominal) lamps so ~225W/string or 675W per string-of-3; 1350W for the pair of these.

The third outlet on the cube tap would either feed a (65W?) flood light sited beneath the tree (again, the goal is to cause convective air currents from the *heat* generated by the lamps -- incandescent lamps being notoriously inefficient at generating VISIBLE light).

Another extension cord would run from the other half of the same duplex receptacle off to another tree -- with a set of 3 strings, there.

Or, variations on that theme.

I.e., we would typically be pushing the 20A circuit as far as we could with a "static" load.

As there's no "switch" to turn these on and off, it's easiest to just flip the breaker for that circuit (there's nothing else on it). So, we would turn the breaker on each (late) evening and off each (early) morning -- in much the same way you'd turn lights on and off in a house.

*We* have "changed" nothing. Yet, something *has* changed. We're not seeing extraordinarily cold weather (hovering around 30F, recent nights) whereas we've seen high teens in years past.

It's not raining -- whereas we've had the lights ON in rainy weather, previously (hence the use of the GFCI -- lest we dance the "electric bugaloo" when we step outside!)

The house is masonry construction so doesn't "move" to affect the wiring in any way. Outlets recessed into the block walls so not directly exposed to the elements.

It would be silly to add more strings just to prove it can't handle an even BIGGER load...

Reply to
Don Y

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.