It is proposed, by the code committe, to add a sentence to these sections requiring GFCI's to be located in 'readily accessible' locations. I have a problem with this,
The committee refers to the manufacturers' instructions to test the GFCI monthly. I'm sure that all of us have alos spent hours running down a 'power problem' that was the result of a very well-hid GFCI tripping. I sympathize.
Yet, there are situations where you want access to the GFCI to be restricted. One is the situation illustrated in the most recent IAEI issue: the GFCI (now required) supplying power to outdoor vending machines. Why would you want these receptacles to be difficult to reach? Simply to protect the power cords from damage, and the machines from being maliciously unplugged.Hence the practice of setting the machine right in fromt of the receptacle.
Another situation is what I encountered ina prison remodel; you want to GFCI protect the prisoners, yet place the GFCI where it will not be subject to abuse.
Then there are receptacles in hard-to-access locations: up near the gutters for ice melting equipment (or Christmas lights), and on rooftops for HVAC servicing. For the first situation, you would have to run the circuit out of it's usual path in order to place a GFCI where you can easily reach it. So, you say, just use a GFCI breaker in the panel? Well, apart from the cost issue ... do you really want to make the HVAC guy run to the panel every time he has to reset the thing? The roof - where he is working - is, by definition, not readily accessible.
The roof is an interesting example, since the guy using the receptacle is already up there. It would be more disruptive to have the reset button anywhere else. I agree these probably will not ever get tested. In prisons I only saw breakers, in locked panels. I would not want a GFCI up on the soffit, no matter what. A good place would be inside on the wall opposite the outside outlet at regular receptacle height. In my house I have a dedicated GFCI circuit for our holiday lighting on a timer. It is next to the panel.
I thought all GFCI's since 2003 were equipped with continuous internal monotoring, self-test circuitry that locks out and won't pass voltage if a malfunction is detected, which would seem to make the monthly test a moot point anyway. I believe this is why a standard 15A GFCI receptacle cost $12.00 instead of $6.00 like they did back in 2002. The TR and WR are even more expensive now. Seems this new compliance concern from the CMP is about seven years to late to make any real difference as far as safety goes. Now it will most likely just be an added inconvenience for installers to comply with.
GFCI devices can fail "ON" and be incapable of tripping. The cord type start off and have to be reset to work but you wouldn't want that in a device that is installed in the wiring. You would have to reset them all every the power glitched
To my way of thinking, when a circuit loses power because a GFCI tripped, the cause of the problem should be both obvious, and easy to access. For example, when it trips in the bathroom, you ought not have to go up three levels and canvass every bath in the house to fine the ONE GFCI that protects all the bathrooms.
When GFCI's came out, we often had only one for all the 'protected' receptacles - and that was usually found buried in the garage. From lurking in the gareage to lurking in one of multiple bathrooms; I'm not sure I see the progress made here! Nor do I see how moving the GFCI to the breaker panel on the outside of the house improves matters.
Yet, either of these scerarios meets the 'readilly accessible' standard. Having the GFCI IN the crawl space, next to the furnace you're servicing, does not.
There's no substitute for good design. Yet, good design and code compliance are different topics. I think the code committee is over-reaching.
You can always say "but some fool will ..." True enough; but the NEC, in Article 90, already makes clear that it's not a design manual, and that 'code compliance' may not result in the best design.
I am thinking of a "McMansion" I worked in, where there was a pair of mysterious blank-face GFCI's in the bedroom closet, mounted about light-switch level. It turned out that they protected the pumps, etc., for the whilrpool tub two rooms over in the bathroom; no, they were NOT on a common wall. Readilly accessible - but good luck if you ever need to find them!
Greg, I thought there was a UL listing requirement change in 2003 that required all new GFCI's to utilize the Smart Lock design, which would not pass voltage if an internal malfunction was detected or if the device was reverse wired during initial installation? I can remember an approximate 50% price hike for GFCI receptacles right at the time this change was introduced.
Although... I've still seen a couple of blown Smart Lock GFCI's and blank face devices that passed voltage after they were completely fried, so I'm sure this design isn't 100% effective, but it does generally seem to work in most cases that I've come across.
The only thing that bothers me about requiring GFCI receptacles to be "readily accessible" is that I can already see one inspector's interpretation that the entire receptacle must be able to be accessed for service and another one that will say you only need to be able to visually inspect and reach the test/reset buttons. I'm wondering about what happens with factory installed GFCI's located below hot tubs in WP control panel enclosures or a GFCI receptacle located under a whirlpool tub with the removable covers held on with the Velcro tape? Same thing with a GFCI receptacle located behind a washer or gas dryer in a basement or bathroom laundry closet. We've all had those trouble calls where the outside receptacles don't work and we end up tracing the circuit and finding the tripped GFCI receptacle located in the inside hallway, behind the three legged table with the Tiffany lamp on it, or behind the 7-foot tall entertainment center in the living room, but I'd hate to see a few cases of installer stupidity be the main reason for a rule change like this.
I think the question "readily accessible by who?" is an issue.
I'm sure everyone here knows that the average person has no clue they even have GFI's in their house, much less have a diligent monthly testing schedule in place. With this in mind common sense would suggest that those who need to know about GFI protection should dictate what "accessible" means. Maybe they have future plans to send a "testing crew" door to door to enforce such procedures. Can't be too safe you know.
I just finished installing two 20A GFI's in a 2gang right at the access panel of a jacuzzi tub.(pre tub install) The access panel being under the adjoining vanity. No fun for anyone who needs to service the pump or heater for sure. No fun for me if one of those GFI's goes out anytime soon either. But they are accessible, just not readily so. I could have put them in the closet (no adjoining wall) and have before, but the last time we built this house the tub had the pump and motor coming into jbox on the tub and I had to use a 2 pole GFI breaker and a double pole switch under the tub deck. (not in my bid BTW) This time we have a different plumber and a different tub. One of the realities is how do you set up for these unknowns?
If a GFI tripped odds are the attached equipment is in need of a looksy. In 99% of those cases the nearer to the equipment in question the better for those doing the service.
In the case of the spa tub, when it stopped working it doesn't matter if they called the plumber or the electrician first, they are both going straight for the access panel as the first step in troubleshooting. We would, of course, pull the plugs and see if the GFI's reset. The plumber would try to reset the GFI and if it doesn't, then tell the owner to call us. Then we do as described above. If they reset we tell them they have a ground fault in their equipment and it needs to be repaired/replaced. Bring back the plumber. As bad as all of that already is do we need to complicate it further by hiding the protection from the serviceman?
If they want to add that kind of language then they should go the full distance and define different types of usage and what "readily accessible" means to each type. Your average kitchen, bath and exterior GFI protection is usually, by nature of efficient design, readily accessible anyway. Those are the only ones the above average owner is likely to test. You will always have to do a bit looking to find the GFI even in these cases unless you want to get ridiculous and install a device in every box in these areas. Maybe that's what they will push next.
Or they could just leave it the heck alone. Right, not going to happen.
For what it's worth, Hubbell introduced both completely self-testing 15A and 20A GFCI receptacles and 20A blank-face devices this year with their Home Select line. These devices continuously self-test once every 60-seconds and also have the standard reverse wiring and Smart Lock type protection. Just as with anything else, I'm sure most other manufactures will soon follow with their own offerings. I would probably make more sense all around for the NEC to just require all future GFCI receptacle and blank-face devices to be self-testing than to follow through with this misguided "readily accessible" change, at least in IMO, anyway.
I have taken apart GFCIs that were "tested" with a bolted L/G fault and the contacts were welded shut. The circuit was opened by the breaker. No test in the world would open that circuit. Maybe insuring that the listing standard that tests the ability to interrupt AFC on a 20a circuit would be in order.