What Is An Electrical Fire?
This page gives my views on a subject that has not had sufficient research to support or refute my views. I trust you will take them for what they are.
The public has not been encouraged to distinguish fire danger arising from electrically powered things (or their abuse) from fire danger arising out of imperfections in an electrical system (due to design or installation). Fires from this second matter certainly deserve the name “electrical”, though it must still be understood that any fire that is destructive beyond the malfunctioning electrical system equipment itself (destructive, that is, of people or belongings) is actually a structural fire, which depends on the proximity of combustibles to such electrical equipment.
By this definition, for example, fires originating from light fixtures that overheat from bulbs of excessive wattage are not electrical. Nor is a fire from bedding blocking the airflow of an electric heater (unless the heater’s safety device failed to shut it down). Also an arcing short from someone’s having nailed a cord along the surface of a wall would not be an “electrical” fire as I am defining it. Rather, this would be a case of abuse of an electrical appliance, not a case of improper installation of the electrical distribution system.
It could be that over an indefinitely long time, every electrical component installed in every home could deteriorate to the point of presenting a slight fire hazard. The design, approval, and installation of these components does not take into consideration such a long period of time, during which many kinds of conditions can have an effect on their integrity. Nor could there be an economical way of requiring such longevity from the beginning. We assume the need for some maintenance and for eventual replacement — even of the building itself. Whether the maintenance done is preventive or is in response to problems encountered might make some difference in fire statistics, but this has not been shown, that I am aware of. Over-maintenance is possible. Also, since any change made can awaken existing imperfections or contain imperfections of its own, total safety will never be a realistic goal.
For a photographic look at the ugliness, if not scariness, of some overheated wire connections, see my page on Melted wires.
Of all home fires 1994-1998, about 3%* were due to electrical system failures, as I have defined them, compared with 2% from natural causes. I admit my definition is that of an electrician (who is also apologizing for problems with the design and manufacture of the components I install) and I hope that my conscientiousness in installation counteracts the negligence of a few other electricians. I sympathize with some of the distrust the public has developed for electrical things in general, but not all of it, because this has been an area of considerable hype.
*The U.S. Fire Administration categorizes 9.4% as “electrical distribution” fires. Within this category, however, is included lamps, light fixtures, and lamp and extension cords, which are certainly electrical but by their portable nature or product nature should not so easily be lumped with what the electrical trade is responsible for. My alternate figure of 3% comes from breaking down the causes of 149 of these “electrical and lamp” fires studied by the U.S. Consumer Product Safety Commission in the early 1980’s. The subcategories in that report could not always distinguish between mechanical damage and poor installation, and do not reveal enough to mount a campaign against all the true causes. For instance, overheating at a receptacle may be from the looseness of the wires attached by the electrician to it or from looseness of its slots that receive what you plug into it. After a fire it is not easy to tell the difference. Looseness in the slots is common and is a case of mechanical damage, often from plugged-in cords being pulled sideways — as when going too far away with a vacuum cleaner.
There are often special looking electrical receptacles in bathrooms or kitchens that have “Test” and “Reset” buttons — often black and red — on them. These are ground-fault circuit interrupters — GFCIs or GFIs. Their purpose is to protect people from electrocution. They do not prevent shock altogether, only deadly shock.
And they do not prevent overloads on the circuit. That is the job of a circuit breaker at the main panel. (What is the little light on some GFIs?)
Why Can’t I Reset? Is GFCI Bad or Is There a Ground-Fault?
GFI Outlet Diagram — Hooking Up
Is an Unknown GFCI the Cause of an Outage?
Finding a Tripped GFCI Receptacle
Confusing Terms: GFCI, GFI, Load, etc.
Bad GFCI or a Ground-Fault? — Troubleshooting
Is a GFI tripping for a ground-fault? If you are pretty sure you need to troubleshoot a ground-fault itself, you may want to go to Tripped GFI — Why? or Troubleshooting a Ground Fault. Otherwise, keep reading. First, this chart summarizes troubleshooting the health of a GFI receptacle device:
|G F I S Y M P T O M|
|CAUSE:||Pushing test does not pop reset out||Reset won’t stay in when pushed||Reset is out but things plugged in work||Reset is in but things plugged in don’t work||Reset pops out when something is turned on|
|Line and load reversed||—||since 2003||yes||—||—|
|Other miswiring of GFI||—||yes||—||yes||yes|
|120 volts not reaching GFI||yes||since 2003||—||yes||—|
|Button not pushed in well enough||yes||yes||—||—||—|
In over your head? Consider calling a Nearby electrician.
Overprotecting gets confusing. When homes are upgraded, PEOPLE OFTEN THINK GFIs NEED TO BE ADDED in bathrooms, kitchens, and elsewhere. They are under the impression that the outlets there are not already protected. This may not be true. It doesn’t actually hurt to double-protect, but it can get confusing. For instance, if something at the bathroom outlet tripped a GFCI, it could trip the device that had been added right there as an “upgrade,” or else it would trip the original one in the garage — whichever one happened to respond more quickly. If you don’t know the complication, you won’t be looking in the right place to restore power.
Trouble retrofitting. Besides having to hook a GFI up correctly, anytime you introduce GFI protection onto existing wiring and existing loads, you may find unexpected tripping. Although this could be detecting a shock hazard you weren’t aware of before, there are also a few wiring conditions from the past that are not GFCI friendly. One unintentional condition would be where a ground wire is contacting the neutral side of a receptacle. Some intentional conditions would include places where the neutrals of two circuits are unnecessarily (and improperly) combined in a multi-gang box, or where the neutral for lights of a GFCI-protected 3-way switch system is introduced from somewhere ahead of the GFI you put in.
Line and Load.
Tripped or just dead? If a ground-fault interrupter is giving trouble, can you tell the difference between its (re)tripping and its simply failing to reset? The reset buttons of newer GFIs (from 2003 on) will purposely not catch hold if they are not receiving power or are not hooked up right. A tripping button, on the other hand, will generally catch hold for a split-second or at least make a mechanical sound when you try to reset.
Quirks of a common brand. I have found that some of the Leviton GFIs (late 1990’s?) can be tricky to reset fully; if you push that button in more on one side than the other, it may not reset both the hot and neutral contacts; just retrip and try again. As a separate tip, by 2007 Leviton GFIs’ buttons have lost their color, and the reset button is so recessed that it doesn’t “look” tripped when it is. These also may need more force than you are used to, to reset them.
GFI Outlets — Info and Hookup
[GFCI Hookup: The GFCI to the left is hooked up so as to sense and protect other outlets besides itself. The one to the right is connected so as to pass normal power (no GFCI protection) on to other outlets.]
Location. In the U.S., and in some respects in Canada, GFCIs have been required since 1973 to protect more and more locations in new homes. See the Location details below. This does not mean that each location requiring such protection has to have its own GFCI device. One device will usually be protecting several normal-looking outlets downstream from it. So if outlets have gone dead in bathrooms, kitchen, dining room, garage, laundry, or outdoors, there is a very good chance that a GFCI somewhere has tripped off.
Shock prevention. Understand that 99% of the time GFCIs save us by not letting us have their electricity in the first place. Long before you are about to use a protected appliance or light, the GFCI will have sensed most faults and cut power off.
Tripping — inconvenient but not false. Besides preventing shock, it is also common for a GFCI outlet or GFCI breaker to trip for conditions that might not put you in any danger. This is because the device can’t tell the difference between a human body and a wire or a pipe or water. A GFI is very sensitive to small “leaks” of current, even harmless ones, off from the intended circuit. Often the tripped GFI will allow itself to be reset by the (red) reset button, since the cause of its tripping was a passing thing. For a technical treatment of How a GFCI works check this by Sam Goldwasser.
Which button? One of the silly ways a GFCI outlet will trip is by a person accidentally pressing the test button. This may be a visiting child, or it may be that when someone is unplugging something from the GFI, their thumb happens to press on the test button. When you are purposely resetting (to restore power) or testing (to cut off power) a GFCI, you will need to look hard for the little letters etched in the buttons to know which one to push.
Which GFI? It is unrealistic for a homeowner to be expected to know where all the GFCIs are in a home, or to know which other normal outlets will be dead from the tripping of a given GFCI outlet. Still, if you have an outage of the receptacles in one of the rooms mentioned above, it may help to look for GFCI outlets in all of those areas and to reset them all. The reset button will need a forceful push. Avoid the test button.
The search to find all possible GFIs needs to be persistent and thorough. It may be that the bathroom and outdoor outlets are dead from a GFI in the garage that is never used itself, because it is behind a pile of storage boxes. Or a set of kitchen plugs has gone out because of the GFI hiding behind the heavy hutch in the dining room or nook.
Doing the TEST. On most GFIs is a statement to “Test Monthly”. That means pushing the test button, seeing that the reset has popped out, seeing that nothing will run using the outlet at that point, and then pushing the reset. The point of this is to see if the device is still sensing properly, so as to do its job of stopping electrocutions. GFIs DON’T HAVE A HIGH RATE OF FAILURE in regard to sensing ground-faults, and when they do fail in this sense, they will still run things (e.g., a hair dryer). Recent GFCI design requirements seem to be making the test button unnecessary: new GFCIs must fail to deliver power if the GFCI becomes incapable of reacting properly to faults.
Bad GFCI? Of course, they can fail in other ways, as when they won’t run things. But BE CAREFUL NOT TO ASSUME THE GFCI IS AT FAULT. There could easily be an “open” upstream from it. If someone added your now-dead GFCI unnecessarily when upgrading, the more original GFCI may have tripped somewhere ahead of this one. Above all, DON’T THINK THE GFCI IS DEFECTIVE WHEN IT KEEPS TRIPPING or will not let itself be reset. I have come across only two cases of defective tripping. GFCIs trip for reasons, including misconnections and an incompatibility with a few motors, treadmills, UPSs, dimmers, and fluorescent lights. But most commonly it is for various faults downstream in wiring or plugged-in appliances — both dangerous and harmless faults.
I don’t intend this site for giving advice on installation, but many people putting in a combination GFCI-switch device (GFCI outlet plus switch) seem to need help. One of the wires that comes from the switch part attaches to the (black) wire going to the item to be switched. How you hook the other wire up depends on whether the item to be switched is to be on the same circuit as the receptacle. If it is not, keep it and its neutral totally separate from the GFCI wiring. If the switched item is going to be on the same circuit, then how you proceed depends on whether the switched item should be GFCI-protected or not. If protected, its switch’s remaining black would attach somehow to the Load hot terminal. If unprotected, to the Line hot terminal. The switched item’s neutral should attach to the corresponding neutral terminal: Load or Line.
Is an Unknown GFCI the Cause of an Outage?
When a GFI outlet trips off, both the hot and neutral “line” terminals become disconnected inside the receptacle from the “load” terminals. Because this is true, there is A TEST THAT CAN DETERMINE THE LIKELIHOOD THAT A TRIPPED GFI HAS CAUSED AN OUTAGE affecting standard receptacles.
Normally there is continuity between neutrals and grounds at a receptacle, because both are ultimately attached to the neutral bar at the main panel. However, if a GFI receptacle has interrupted power to other “normal” receptacles, one result would be that a continuity test between neutral and ground at such a receptacle would show very high resistance (no continuity). If a test showed such discontinuity, one could not be certain it was due to a tripped GFI, since a poor ground or other condition could show the same result. Nevertheless, it would make a tripped GFCI very likely as the cause. (The condition of the ground could be determined by a different test).
Finding a Tripped GFCI Receptacle
Code History. Hunting for a GFCI of unknown location takes some persistence, but it can often help to know the history of GFCI requirements in the National Electrical Code (NEC). For U.S. homes, the house wiring Code began requiring GFCI protection for receptacles outdoors in 1973 and in bathrooms in 1975; at that time this was more often achieved by a GFCI breaker in the panel rather than a GFI receptacle. In 1978 garage receptacles were added to the list, and the use of receptacle-type GFCIs became more common. The 1987 Code called for kitchen-counter receptacles within six feet of the sink to be protected. 1996 saw all the kitchen counter receptacles come under the requirement. Outlets in crawl spaces, unfinished basements, and near wetbars were added in 1990 and 1993, and those within six feet of laundry or utility sinks in 2005. This pdf-chart expands this history even more.
In Practice. Because GFCI receptacles are an extra expense and take more labor to install, one GFCI device was typically placed where it could feed protection on to the other normal-looking outlets required to have protection. UNTIL 1987, THIS COULD BE ACHIEVED BY JUST ONE GFCI FOR THE WHOLE HOUSE. Then there needed to be more, because kitchen outlets were already required to be on at least two circuits dedicated to the kitchen/dining area. When all the kitchen outlets needed protection (1996), there always had to be at least two GFCI devices for the kitchen, besides at least one for garage/outdoor and one for bathrooms. None of the above may be accurate for Canadian homes; I do not know the Canadian code history enough to give a similar description.
When searching for a hidden GFCI, consider the following. Any circuit, or any part of one, was allowed to be protected. In practice, only the ones required to have protection tended to be wired downstream from a GFCI device, unless an amateur did so unintentionally. But it was sometimes convenient to plant the GFCI device at a location not requiring protection. For example, in the dining room for protecting the kitchen, or indoors for protecting an outdoor outlet. Another common practice has been to avoid giving protection to a refrigerator, since protecting it is not required.
Four commonly overlooked GFCI locations are a rarely-used main-floor guest “powder room,” an entryway closet, a walk-in closet, and the garage. The garage is especially worth combing if the electrical panel is there. A circuit for many of the locations requiring protection would naturally begin by going from the panel to a garage outlet, where the device itself would be placed. Garage outlets become invisible in two ways. Literally, they disappear behind stored things or permanent shelves. Psychologically, some of them become forgotten because they have never been actively used. Garage outlets were often placed about four feet above the concrete floor.
Confusing Terms: GFCI, GFI, Load, etc.
GFCI versus GFI. If you are confused about the terms GFCI or GFI, quite simply those names are interchangeable. I reveal my preference in the Glossary. We have some other confusions also. One is whether either term refers to a receptacle-type or to a circuit-breaker-type interrupter. In fact, people are commonly calling either type a “GFCI breaker” or a “GFCI switch.” Though they both can have the effect of a switch, that is not their purpose and I discourage that way of talking. As for “breaker,” the circuit-breaker-type GFCI does do double-duty as a normal circuit breaker and looks like a breaker and lives in a panel. So it deserves the name “GFCI breaker.” Likewise, a receptacle-type GFCI does double-duty as an outlet. Even though it automatically “breaks” part of a circuit when it trips off, to call it a breaker will confuse anyone who is aware that there is a different device by that name already. So we should call this more common device a “GFCI outlet” or a “GFCI receptacle.”
More. Another confusing phrase I hear is “GFCI circuit.” Where a true GFCI circuit breaker is used, everything it feeds would certainly be a GFCI-protected branch circuit. Where an outlet-type GFCI is used, the branch circuit it is on always includes something — some wire, at least — that is not being protected or shut off by the device. People want to use the word “circuit” to refer to the items that go dead when the GFCI trips. This is understandable, but “circuit” is not the right word. We need to keep using “circuit” to talk about the whole circuit this outlet and its loads are part of. “Load” is the accepted term for the things being protected.
Areas we serve:
Electricians in Charlotte NC
Electricians in Matthews NC
Electricians in Indian Trail NC
Electricians in Weddington NC
Electricians in Pineville NC
Electricians in Waxhaw NC
Electricians in Harrisburg NC
Electricians in Midland NC