Wikipedia:Reference desk/Archives/Miscellaneous/2020 February 10

Welcome to the Wikipedia Miscellaneous Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

February 10

space heater or lamp ground fault?

I posted last week about a GFCI circuit breaker randomly tripping a lot. It started tripping more and more often, so I unplugged stuff one group at a time, except for a hardwired ceiling light fixture which I left connected. The last thing I unplugged was a lamp and a space heater, both of which were plugged in but turned off. Both of them use 2-wire cords. The breaker hasn't tripped lately so these two devices are suspect unless it trips again (they are both unplugged now).

Questions: 1) is it even possible that a 2-wire device has a ground fault that trips a GFCI? I don't have a complete understanding of how GFCI works but I thought that unless there was a ground path through a wet floor or something, leakage in a device was likely to the ground pin of a 3-wire cord. The floor seems dry as far as I can tell, though I will check it for stains. 2) Is there a way I can test these two devices with a multimeter, to check for such faults? Thanks. 73.93.155.38 (talk) 00:37, 10 February 2020 (UTC)

• 1) is it even possible

Yes.

This is the main purpose of GFCI (RCD). The fault diverts current to ground (literal earth), probably through you.

In the earlier pure three-wire (with ground) non-GFCI system, the fault current was expected to divert to the earthed case and return through the ground wire). This is a low impedance path, a high enough current to then blow the protection fuse. This might work for table lamps with low 1A fuse ratings (if correctly fused, on the British system), but it worked very badly for either heaters (with large fuses) or non-British installations reliant on fused radials, fused at 15A and upwards.

You can test these faults with a multimeter. You might even find them. But you can't test the device completely without a machine such as a PAT tester (for most appliance types). See Wikipedia:Reference_desk/Archives/Science/2018_March_6#earthing Andy Dingley (talk) 01:17, 10 February 2020 (UTC)

Thanks, but the heater and lamp are both just sitting on the floor: where could the ground path be? That's what I'm having trouble understanding. The PAT article is helpful and I'll check the plugs and stuff, and will also open up the cover plate and check the outlet itself. Your post from March 2018 is great too. The next suspect after the lamp and heater may be the inside-walls wiring, which I think means call an electrician. I think the breaker has not tripped since I unplugged that stuff, but I was out of the house for a while, so it's possible that it tripped and someone reset it. I'll ask them when I can. 73.93.155.38 (talk) 01:56, 10 February 2020 (UTC)

Hmm, I believe the heater has an indicator lamp that glows when the heater is plugged in even if it's turned off. So if the outlet is miswired, maybe the indicator lamp pulls enough current to trigger a fault: does that make any sense? There is an outlet testing gizmo with three led's (or neon lamps or something) that can detect a miswired outlet.[1] I guess I could get one of those if it seems worth trying. 73.93.155.38 (talk) 02:05, 10 February 2020 (UTC)

• It's possible that it could be anything - I just can't see from here. But GFCIs do have a failure mode with age where they become sensitive and prone to nuisance tripping. It would also depend what its current rating is - a modern sensitive one (maybe 30mA) can easily be triggered by damp.

I find the idea of a space heater (presumably quite powerful) on a 2 wire cord terrifying - but then I have a horror of US electrics.

If you have a multimeter, then test both appliances, as best you can (you can't prove they're safe with limited tools, but you might find the fault). I don't much like those simple outlet testers because they don't tell you numbers, so I prefer one of these: [2] The same thing, only £50 (!) but it tells you roughly the earth loop impedance too. Every live band needs one!

Testing the appliances depends on whether they have a metal case or not. If it does (or a touchable part to the lamp holder) then test that with a multimeter - there must be no measurable conductivity to either of the conections. Even a few thousand ohms can trigger a GFCI. Proving that they're safe would require a Megger (high voltage insulation tester). If they're plastic, then focus on a visual test to look for damage. Heaters are especially prone to their insulators or heating elements getting covered in dust, dirt or conductive grime. I'd try to look inside it, if it were mine.

You can test the socket outlet, but again, that's non-trivial testing. Sometimes visual inspection with the cover off will show a fault though. The cheap neon testers are always better than not testing, but don't assume that their "no faults found" is proof they aren't there despite! Andy Dingley (talk) 02:48, 10 February 2020 (UTC)

Thanks, I'll try to find out what is in those fancy testers and whether I can fake it with a multimeter. I'll also examine the lamp and heater and outlet. Meanwhile, any idea why there is a GFCI breaker at all? It doesn't seem to be running anything in any bathrooms. The bathrooms have GFCI outlets of their own, but they are not on the same circuit as this breaker. GFCI breakers are pretty expensive so if it is the breaker itself being flaky, I wonder if it's bad juju to replace it with a regular breaker. 73.93.155.38 (talk) 06:25, 10 February 2020 (UTC)

• I would always (except when a GFCI is inappropriate) replace a GFCI with another GFCI, not just remove it. They are a significant improvement to electrical safety. Here in the UK our '18th edition' of the wiring regs has done what the last five have done over 40 years and required more use of them, in more places. We're now getting to a point where each circuit is expected to have its own, rather than protecting the whole panel with one or two of them. The cost looks a lot when comparing component prices, but it's not much in an overall system, or if you factor in labour.

Testing your circuits now is likely to need a Megger, more than a simple multimeter, because you're looking for a leakage current which might just not show up at low applied voltages. This testing also needs to be carried out by 'a competent person' (not necessarily qualified and certified, but that's the implication). Another useful UK publication is something called an "On-site Guide" which is the toolbox guide book on a condensed version of the wiring regs. it costs £15 rather than £80 and is 50 pages rather than 500. It lists the test sequence for carrying out this test. It's also good practice to measure the earth loop impedance (although that doesn't sound like your fault today) and that often requires a specialised test meter (annual certification on mine to use it for trade costs as much as a decent new multimeter), or at least a multimeter which can accurately measure fractional ohms, and a rather awkward testing process. Andy Dingley (talk) 11:46, 10 February 2020 (UTC)

GFCI is required by code is many places that other than bathrooms. In the US, each successive edition of the National Electrical Code has increased the types of locations where they must be used. Local jurisdictions may have other specific requirements (or at least declare which NEC edition is in force, and how much can be grandfathered). DMacks (talk) 18:56, 10 February 2020 (UTC)

Thanks both of you. I'm more and more suspicious of the space heater (and to some extent the lamp), since the breaker hasn't tripped since I unplugged them yesterday. Before I unplugged them the tripping had gotten up to around 1x per hour. I'll keep trying to figure out what is wrong with them. 73.93.153.166 (talk) 19:28, 10 February 2020 (UTC)

Try plugging them into a different GFCI circuit and plugging in (and using) other devices on this original circuit. That could help decipher "device" vs "circuit/GFCI" as the problem. A wiring fault somewhere between the wallplate and the GFCI breaker could allow current leakage from neutral to a path to ground. From the GFCI's perspective, this is an equivalent circuit to a ground-fault in a device that only occurs when a device on the circuit sends current to neutral. DMacks (talk) 05:05, 11 February 2020 (UTC)

2-wire appliances in the U.S. must be double insulated, or otherwise not be at risk of exposing a user to live from a single fault. U.S. appliances with switches/controls use polarized plugs to ensure a fault in the switch leaves the appliance dead rather than live. --47.146.63.87 (talk) 21:05, 10 February 2020 (UTC)

• That's generally true for appliances worldwide. However it's still widely not the case for older appliances, many of which are still in use.

In many cases these can't be brought to modern standards, or made anything like adequately safe, for 2-wire connection, but can be safely converted to a 3-wire connection. Andy Dingley (talk) 00:10, 11 February 2020 (UTC)

Sure, I was just elaborating: only applies to things "up to code" of course. I (in the U.S.) had one electric heater go kaput some years back, and if I recall correctly what happened was it would instantly trip the GFCI breaker when plugged into a GFCI outlet and switched on. I'm not 100% sure; this was years ago. I suspect the poster has the same issue, but in my case it was a lot easier to diagnose! --47.146.63.87 (talk) 03:48, 15 February 2020 (UTC)

extent of copying

Lets say I created an article on a poem, but its copyrighted. Would I still be violating copyright guidelines if I only copied 5% of it? For example the entire poem has 100 lines and only 5 lines of the entire poem are inserted into the article. 79.67.74.81 (talk) 17:03, 10 February 2020 (UTC)

There are no numerical definitions per se, and we can't give legal advice here, but the relevant doctrine (in the US) is fair use which is a 4-prong subjective balancing test: 1) what is the purpose of the copying (article about the poem), 2) how much of the copied work are you using (5% in your example), 3) type of work being copied (a published poem I'm presuming), and 4) effect on the work's value (depends on the specifics I guess). See the "U.S. fair use factors" section of the fair use article. If you're asking about writing a Wikipedia article about a well known poem, it's reasonable to just look at comparable articles and follow them as examples. E.g. the article Howl quotes a dozen or so lines from the Ginsberg poem with commentary and seems like the type of thing the fair use doctrine is intended to allow. 73.93.153.166 (talk) 18:50, 10 February 2020 (UTC)

What article? If you are referring to a wikipedia article, you should worry about more than copyright. See WP:NFCC for a start. For textual work, any content which isn't your own work should be clearly indicated as such. For content not available under a CC compatible licence, the content should only be included if it significantly adds to the article, and needs to be clearly marked e.g. as a quotation. For an article on the specific poem, there may be merit. For an article on love often there won't be any merit for such an inclusion. Nil Einne (talk) 11:43, 11 February 2020 (UTC)

The only part of the WP:NFCC policy that is relevant here is this one sentence:

Articles and other Wikipedia pages may, in accordance with the guideline, use brief verbatim textual excerpts from copyrighted media, properly attributed or cited to its original source or author (as described by the citation guideline), and specifically indicated as direct quotations via quotation marks, <blockquote>, {{Quote}}, or a similar method.

The rest is about other media, such as images and audio clips. "The guideline" referred to, WP:FAIRUSE, says this about the acceptable use of text:

Brief quotations of copyrighted text may be used to illustrate a point, establish context, or attribute a point of view or idea. In all cases, an inline citation following the quote or the sentence where it is used is required. Copyrighted text that is used verbatim must be attributed with quotation marks or other standard notation, such as block quotes. Any alterations must be clearly marked, i.e., [brackets] for added text, an ellipsis (e.g.(...)) for removed text, and emphasis noted after the quotation as "(emphasis added)" or "(emphasis in the original)". Extensive quotation of copyrighted text is prohibited. Please see both WP:QUOTE for use and formatting issues in using quotations, and WP:MOSQUOTE for style guidelines related to quoting.

No quantitative criterion can supplant a common-sense application of the fair-use criteria, which depend on the why and how of use. In some cases 5% will be too much, in other cases 10% will be fine.  --Lambiam 22:32, 11 February 2020 (UTC)

I disagree that the only part that is relevant is that. Note also that I said, start with. The lead and rationale section of NFCC apply to everything. They make it clear that our restrictions on the usage of copyrighted works intentionally goes beyond the requirements of fair use under US law. While this issue mostly arises in relation to images, it's important that any contributor to wikipedia understands that even when it comes to textual work. It helps editors understand why we are very strict on such inclusions, and why it's a serious mistake for them to think, even if for example they are a lawyer experienced in such matters, 'my use of the copyrighted text in this article is okay under US fair use provisions, so I"m fine'. I do agree there will never be any clear numerical threshold. Nil Einne (talk) 03:51, 12 February 2020 (UTC)

BTW, one point I was also trying to make without going into much detail is that we should also distinguish between 'copyrighted' and WP:NFCC material. The latter requires quotations etc. But copyrighted material can also be added to article if it's under a CC BY-SA 3.0 compatible licence in certain circumstances. In that case as we are following the licence for reuse, it does not need to be as a quotation, but you do need to do some things to ensure licence compliance. All my contributions including these comments, are copyrighted where eligible for copyright. I've agreed to licence them under the CC BY-SA 3.0 and GFDL my writing them here, but I retain the copyright. Nil Einne (talk) 03:59, 12 February 2020 (UTC)