US near Washington, DC. The giant hole where all US citizens' tax dollars shoot out of like an Eastern version of the Ol' Faithful geyser. (-:
I will probably end up following the CEC's recommendations, anyway, even though the NEC governs here, simply because the Canadian rules happen to make a lot of sense. I'm guessing that frequency of nuisance trips of a new GFCI gets about as low as possible with only one appliance on that circuit. It also seems that accidental trips for any reason are reduced by dedicating a line to the breaker. My own experience backs that up. Two GFCI trips per year with some other heavy-duty equipment on the same line as the fridge suggests the refrigerator does not trip the GFCI itself.
The GFCI unit that trips about every six months is a Slater, not a Leviton, it turns out. I've had it since GFCI's first arrived in the consumer mainstream. It could easily be close to 25 years old which means a more modern version might not even make those 2 nuisance trips a year even with other gear on the line and the problem's solved!
I am going to swap the old unit out and the new 20A Leviton GFCI in its place after I "dedicate" the line from the breaker to the refrigerator. I just feel more comfortable with shock protection on the refrigerator and accept that replacing a fridge worth of food is a possible consequence. It just seems that possibly replacing frozen food is a better option in the long run than finding yourself planning a funeral from a freak accident with a funky fridge. What's that line from "Casino" - "You can have the money AND the hammer?"
Quite simple, really. The air compressor's got a slow leak in the quick-change hose valve (a cheapy Harbor Freight deal). When the unit loses power for more than a few minutes, it's unable to top itself off and the tank loses pressure.
When electricity is restored, the unit automatically comes on to bring the tank pressure back up. It only takes the power being down long enough for the tank's pressure sensor switch to trigger to cause the fridge and the compressor motor to fire simultaneously. That all occurs without anyone being present. When the tank reaches pressure, the compressor motor shuts off. So it could be the dual starting or it could be the abrupt shutoff and reverse EMF spike of the compressor while the GFCI is also under load from the fridge to trip the GFCI.
I hear you, and I am here trying to figure out what the various risks and trade-offs are. Given the choice of taking a lethal shock or a freezer full of spoiled food, I'll take the spoiled food every time. (-:
I can take some serious precautions to prevent nuisance tripping like a dedicated line and a late-model GFCI. I'm also building a home automation remote warning system using HomeVision that calls my cellphone if the fridge goes outside of its temperature range or the alarm system detects a break in. I'm hoping to eventually be able to answer the door intercom remotely by having the home automation controller ring my cell whenever someone presses the doorbell when I'm not home. It will also call if the basement floods, the house goes too hot or cold, etc.
I may even decide to add a second, non-GFCI outlet so that if we're away overnight or longer I can switch the fridge from a GFCI outlet to a standard one. I've had nuisance trips, but they've been with a very old Slater 1st generation GFCI on a non-dedicated circuit. Changing to a new GFCI and a dedicated line should reduce those nuisance trips to near zero. Well, that's the plan. Hopefully I won't have to post a follow-up that says "Mark was right and I've got a freezer full of chum."
When I said that the Nec doesn't require a dedicated circuit for a fridge, I didn't mean to imply that a dedicated circuit wasn't a good idea, or even required by the refrigerator manufacturer. Nec is a minimum requirement. If your fridge is in a kitchen, and the outlet is behind the fridge, gfci protection is not required. If the circuit and outlet are properly grounded, it will be perfectly safe, and not vulnerable to ground fault related anomalies
Sorry to start a cross-border dispute! (-: I should have mentioned North East US.
That was my understanding. Thank you for confirming it.
It's a motivator. Bought two 250' rolls of 12/2 w/G and have been slowly replacing 2 wire outlets and wires from the 1940's. Most importantly, I've moved all the high current devices off the old wiring and left the old two wire outlets to serve things like 2 wire chargers, floor lamps and items that never had grounds to begin with and that don't draw lots of current.
Turns out in an old Cape Cod that doing a rewire from the basement is a lot easier than working with the old wires that went up to the attic and then down throughout the house. Unfortunately, when we pulled down the poorly refinished basement's ceiling and walls, the re-wiring effort was no longer the top priority. For one thing there was a huge hole in the cinderblock wall - it looked like someone had started tunneling out. Next time we buy a house, we'll ask lots more questions if only 1 wall out of 4 in the basement is panelled.
Yes, I knew when I crossed floors I was probably coloring outside the lines. I did it in haste to at least temporarily protect both those areas with GFCI I also tried to make sure that area was serviced by two different breakers so I could still see if the GFCI tripped and took out the lamps on that circuit. It's easy enough to rewire the correct way. Do you know the basis for the rule? Is it that one breaker should not service two floors or that one GFCI can't span floors or both?
'
What worried me is that there may be an underlying small ground current leak - perhaps some insulation is degrading - and it only shows when the GFCI warms up from carrying a larger than normal current and some capacitor or resistor value shifts enough to make the imbalance detection circuitry react differently. A while back Smarthome released new Insteon-brand switches, and IIRC, they only exhibited flashing problems when the load on the unit was beyond a certain limit. People with chandeliers and 300W torchiers ran into serious problems with unwanted flickering and outright flashing that users with small lamps (and probably most of Smarthome's beta testers did not experience. I note that just to point out that devices can behave quite differently under a heavy load than they do under a light one.
Yes, and that's underway as I noted elsehere. The unit in question is perhaps a 25 year old Slater. I will swap it for a 2008 model Leviton.
No problems with any of the testers so far. Everything checks out. I am pretty sure that this issue was caused by having other devices on the same outlet. I'll be switching the fridge over to its own dedicated line later this week and I'll also be switching out the older model Slater GFCI for a newer Leviton 20A model. I'm hoping those two changes will eliminate the nuisance tripping. If the problem occured more than once or twice a year, I'd remove the GFCI entirely, but I'm reluctant to give up the protection it ostensibly affords until it proves itself to be too troublesome to maintain.
I agree. Still, they've taught me a lot. I wonder if they deliberately chose not to focus on electrical work because of all the potential dangers.
Hey thanks for the reminder. 'A GFCI not crossing floors'. However I prefer the idea of a GFCI being in a weather protected environment!
For example: We have an outlet low down outside front door in a rugged weather resistant box, fed with conduit that runs under the front step. It's been there some 35+ years and is convenient and useful for Christmas lights etc.
But would prefer to have the GFCI on it in the basement from which it is fed! Rather than outside in the weather and beyond that old conduit under the concrete front step. There is occasioanlly snow build up in that area.
Welcome any comments/advice. However if necessary (or safer) the outside outlet might/could be eliminated. TIA
I was pointing out the fact that *no one could be using the compressor to blow out the coils* if no one was home.
Sure, the compressor could have been plugged in to the GFCI when no one was home and sure, both the compressor and the fridge could have turned on at the same time when no one was home, but nobody could have been using the compressor to blow out the coils if no one was home.
Don't fret. I miss the point of most jokes made in newsgroups. It's not you.
Depends. My little JRT loves the air compressor. Based on how frantic she gets when I am using it, I'd say she likes it even more than chasing squirrels. Since she's figured out how to unlock her crate, how to shoulder a heavy file cabinet so she could get to her lost nylabone and how to open the refrigerator, I wouldn't be surprised to find out she's learned how to operate the compressor nozzle some day. You try telling her she's no one.
Sorry I missed this and a few others on the first read through. New newserver.
That's a good point about current leakage rates. If the unit still trips with nothing else on the circuit AND a new model GFCI in place, then I am going to begin checking out the refrigerator with an ammeter* to see if there really is a current leak. At that point I'd be willing to believe it's not simply a nuisance trip, but an indication of a problem. Hopefully I'll remember to switch the fridge to a non-GFCI outlet when we're away. If there's a ground fault when no one is home, that's not as bad as a meltdown of all our food.
*(Why that spelling and not ampmeter, I've always wondered?)
Understood. If the Canadians think a dedicated line is a good idea, I assume they've got reasons based on experience. There are lots of good reasons to dedicate a line, like making it easier to supply just that device with a generator or not having to shut it off to service any other devices on that circuit. It's so cheap and easy that it's almost crazy NOT to do it.
Why does having the outlet blocked have anything to do with it? Not sure I follow that reasoning.
Isn't the whole purpose of having something like a GFCI to act as a "last line of defense" in case something very unlikely occurs like a cord getting frayed and passing power to the metal case? Even a little current leakage might be enough to kill someone. From what I recall, it doesn't take much current, if delivered across the heart, to cause death. I would think if there was no real protective value that the NEC would say so, instead of appearing to gradually bring GFCI's into the code in nearly all circumstances.
Now arc-fault interrupters seem to be a more contentious case. The folks I know that have installed them report they are plagued with nuisance tripping. I wonder if it's just another case of it taking time for the manufacturers to fine tune the product?
Good question but how/why do many other units get adopted/used. Is not 'Volt' named after the researcher Volta? Another example is the Bel (or more commonly the Deci-bel =3D one tenth of a Bel). Or more commonly and colloquially as Dbs. ("dee-bees"). Named after Alexander Graham Bell! I think 'watt' and 'ohm' are unshortened? Just a thought.
Kitchen countertop receptacles have to be GFCI protected. A receptacle behind a refrigerator is not a countertop receptacle.
The requirement for commercial kitchens to have refrigeration on GFCI receptacles was because people were getting shocks. (Homes probably have greater care taken of equipment.)
Starting 2008 much more sensitive AFCIs were required. The old ones detected a 75A arc (and would only detect an arc from H-N or H-G). Starting 2009 they have to detect a 5A arc (and can detect a loose arcing connection). I would think detecting a 5A arc without tripping on normal arcs (like turning off a switch) would be a real challenge. At the same time (IMHO a dumb idea) they were required for far more areas of a house.
I haven't heard about major nuisance trips. Have other people?
While the tester is certainly a binary, pass-fail event, nuisance tripping may actually be the GFCI detecting a growing irregularity of some kind. Cords fray slowly and corrosion is no jack rabbit, either. That's my "current" conundrum - deciding whether the nuisance tripping is really just a spurious event or a signal that there's an unhealthy interaction between electrical components that should be run to ground. I guess I've worried myself enough to stop grousing about it and read up about detecting current leaks in appliances.
As a X-10 home automation user, I've learned the the household wiring is truly a network. Plug-in devices can interact very strongly with each other in some very annoying ways. In my case the bad interactions appear as ground loops and degraded X-10 signal strength.
I have investigated a couple of these old refrigerators that trip GFCIs and they were all bad. Most had internal shorts in the compressor. I suspect that is why they use more power as they age and why the freon has a burnt smell when you cut open the line. If you put a scope with a current probe on the grounding conductor you see spikes. I bet there is a mini thunderstorm going on inside that compressor. Eventually this can get bad enough to trip the breaker but if the short is closer to the neutral end of the winding it might run like this forever. The only real danger to the user is if the ground has a high impedance and some voltage gets imposed on the case.
While I'm *definitely* no code expert, I'd agree with you, out of common sense, to try to keep the "controlling" GFCI outlet inside and a much simpler standard outlet outside. That's how I wired my outside outlet. The GFCI controlling it also runs the radial arm saw and it's mounted inside. Not sure if that's code, but the likelihood of my operating the saw and anything else outside at the same time is very remote. The power to the outlet is also controlled by X-10, so I can turn it on and off from anywhere in the house.
The GFCI should have a temperature range rating. It could be very possible that operating it when it's too cold could compromise its protection capabilities. I'd be more worried that you've got it in a bad place with regards to potential immersion. When I searched for guidance in the NEC, all I found was a maximum outlet height limit, which I thought peculiar since they didn't give a minimum one, which I thought would be more important.
If I recall my outdoor box, it was set up for a standard duplex outlet and wouldn't even accommodate a GFCI because they are typically in the Decora style because of the need to access the "Test" and "Reset" button. Look outside - the decision to use an indoor GFCI may already have been made for you! (-:
Then there's the issue of resetting. You may not want to trudge inside and out to keep resetting the unit. There's a lot to be said for local control in these cases. Part of the reason I'm rewiring the fridge to its own GFCI is because I want the reset button right near the protected unit as I have a "blown out" knee (I think that's the proper medical term) and going up and down the stairs isn't much fun at the moment.
How about that poor CPS abbreviation, getting sacked and replaced buy Hz. I can just hear it gloating: "Hertz, donut!"
It sounded funnier rattling around in my brain, I swear! - I was going to say it must have stung like a hit in the family joules . . .
1000 aches = kilohurtz What is it that's black, charred and smouldering and hangs from a light socket? An DIY electrician trying to change a light bulb!
Diode - What everyone hopes they'll do
Somebody, pull the plug on me!
Who's old enough here to remember that TV show with the meter, "Queen Faraday?"
The problem with bad electrical puns is that anyone conduit.
1012 bulls = 1 terabull - what these puns are!
100 buckets of bits on the buss,
100 buckets of bits, Take one down, short it to ground,
99 buckets of bits on the buss.
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.