Maker Pro
Maker Pro

Wiring mishap?

T

Travis Evans

Jan 1, 1970
0
I'd like to think that I'm reasonably cautious and safety-conscious
around this stuff, but despite my best efforts it sometimes just seems
to hate me. Today I turned on my 70-watt M98 metal halide fixture
(after having used it several times within the last few days without
incident) and was greeted with a really loud POP, like a firecracker.
The lamp still appeared to start up, but I quickly shut it off because
it scared me.

First I visually inspected the lamp, whose arc tube seemed to possibly
be significantly more blackened now than a few days ago, though I can't
be positive since this is relying on eye and my (not too terribly
reliable) memory. The lamp in question isn't too terribly old, though
its starts-to-burning-hours ratio may be fairly high.

Electrically, something was definitely not right. Without a lamp in
place and with power applied, my meter showed power consumption of
around 70-100W, possibly intermittently. I was also seeing continuity
between neutral and ground at the plug, which didn't look right given
that the ground and neutral feed wires inside the fixture are not
connected together at any point.

As I disassembled the fixture, I really should have checked for any
obvious signs of loose/dangling wire ends, etc., that would have
explained what had happened, but I didn't think of it. Instead I just
took everything apart, checked continuity of individual components for
anything odd (didn't find anything), and then disconnected and
reconnected all of the wiring from scratch, again checking that the
connections were all tight (all of which I had thought I had done the
last time I was inside).

After reassembly, everything seems normal again for now--no path between
neutral/ground and power consumption without a lamp of only a few watts.
Finally, I carefully powered up for a couple of seconds with the MH lamp
back in place. It appeared to start normally.

I suspect that something in the circuit had somehow gotten shorted
(maybe a neutral somehow came in contact with the grounded chassis?).
It's just not totally clear whether this was from a wire falling out of
a wirenut or what have you. My main concern now is with how safe
resuming operation would be--i.e., is there any increased chance of
components like the ballast, ignitor, and especially the lamp, failing
catastrophically later after this event. The lamp is unprotected, and
while the fixture is enclosed and designed for this type of lamp, I
don't really need any more "excitement" for a good long while. :)

So I thought I'd ask for advice before doing anything else.
 
A

Andrew Gabriel

Jan 1, 1970
0
If the capacitor is not shorted, which seems reasonable because you
say the fixture now works, the loose wire theory sound plausable. What
is the brand and model number of the ballast? Is this a multi-voltage
ballast that has "extra" leads for alternate voltage mains that need
to be capped off?

Capacitor could have shorted and self-fused.

Something that can happen with mercury-based discharge lamps is
that they can operate as a mercury arc rectifier for short periods
during run-up. When this happens, the ballast will pass much more
current, although this normally just results in much louder hum
for a few seconds. If the unit wasn't fused with this in mind, the
fuse might pop.
 
T

Travis Evans

Jan 1, 1970
0
If the capacitor is not shorted, which seems reasonable because you
say the fixture now works, the loose wire theory sound plausable. What
is the brand and model number of the ballast? Is this a multi-voltage
ballast that has "extra" leads for alternate voltage mains that need
to be capped off?

It's a 120v/60Hz residential-type fixture. No extra unused leads on the
ballast. No separate capacitor, just an ignitor and ballast. I'll have
to get the brand/model labeling on the ballast once I have a chance to
get back inside.
 
T

Travis Evans

Jan 1, 1970
0
It's a 120v/60Hz residential-type fixture. No extra unused leads on the
ballast. No separate capacitor, just an ignitor and ballast. I'll have
to get the brand/model labeling on the ballast once I have a chance to
get back inside.

Ballast labeling: Pollux Cat. No. MHA70
(70W M98 Metal Halide)
 
T

Travis Evans

Jan 1, 1970
0
I'd like to think that I'm reasonably cautious and safety-conscious
around this stuff, but despite my best efforts it sometimes just seems
to hate me. Today I turned on my 70-watt M98 metal halide fixture
(after having used it several times within the last few days without
incident) and was greeted with a really loud POP, like a firecracker.
The lamp still appeared to start up, but I quickly shut it off because
it scared me.
[...]

I think I may have figured out what was going on. During reassembly
again earlier, I checked between neutral and ground on the electrical
plug with the continuity tester from time to time. Once it was fully
assembled, I again noticed a neutral-ground short.

Started disassembly again, and the short disappeared as I started
loosening one of the chassis screws. I also noticed that the nearby
feed wirenuts, especially the neutral one, have some of the conductor
visible from the bottom which I hadn't noticed before. It definitely
looks like the end of that screw sticking into the chassis (holding two
metal covers together) was able to touch the exposed neutral strands at
times, which would short it to ground.

I'm definitely going to be fixing this before attempting to use the
fixture again. Maybe I should get a GFCI, too? I'm glad that
things didn't go as badly as they could have given the right
circumstances.

Thanks to everyone for the responses to this and all my other posts
here. This is an excellent educational experience. :)
 
T

Travis Evans

Jan 1, 1970
0
I see that you have found that the ignotor is not the issue.

The fixture is supposed to be grounded. In that case, a GFCI does not
provide any additional protection against shock, but it will signal
future wiring problems.

The other issue, of course, is why there is condustor exposed outside
the wire nuts. Is this an original connection, or one you made inside
the ballast housing when you installed the fixture? You say "feed
wirenut" which makes me believe this a connection you made. If not,
then the manufacturer has a serious problem.

It is a connection I made inside. Some of the wires had too much
insulation stripped away for the length of the wirenut. I've trimmed
them to a more appropriate length and now know to be more careful in
this regard in the future.

As far as I can tell, none of the other components appear to have
suffered any ill effects from the event. I'm not sure what caused the
loud pop unless an arc jumping from e.g., the exposed neutral conductor
to ground through the mounting screw could produce such a sound.

The fixture in question does have a UL label.
 
Travis Evans said:
Started disassembly again, and the short disappeared as I started
loosening one of the chassis screws. I also noticed that the nearby
feed wirenuts, especially the neutral one, have some of the conductor
visible from the bottom which I hadn't noticed before.

I am not an electrician and this is not electrical code advice.

Some fixture manufacturers supply wire nuts that are IMHO kind of small
for the job of connecting the (usually) 18 gauge fixture wire to the 14
or 12 gauge wire in the house. They seem to particularly like the
orange ones that are about 7/8" tall; the ones I have have "P3" molded
into them. I can't find specs online for exactly this model, but the
spec for what I think is a similar model from another manufacturer
(Ideal/Buchanan WireTwist WT3) show it has UL listed combinations of
1 #14 with 1 #16, and 1 #14 with 1 or 2 #18, at 600 V. It doesn't have
any listed combinations with 12 gauge wire.

For hooking up fixtures, I usually use the yellow wire nuts, which are
the next size up; they are usually about an inch tall. It seems like I
have to do less work to get all the wires to fit into that size. Also,
they tend to have slightly longer skirts, which makes the strip length
of the wires a little less critical; it's still wrong if any exposed
conductor is outside the wire nut, but it's a less likely to happen.
One model (Ideal/Buchanan B-cap B1) is UL listed for 1 #14 with 1 to 3
#18, 1 #14 with 1 to 3 #16, 1 #12 with 1 or 2 #18, 1 #12 with 1 or 2
#16, all at 600 V.

The red wire nuts, which are the next size up from yellow, are almost
too big for this job. They have the right UL listings, but sometimes
it's hard for me to get them to start threading onto the wires.

Some manfacturers now have "red-yellow" wire nuts, which are supposed to
cover all of the wire combinations of the yellow and red wire nuts. I
have never used these but I have doubts about how well they work.

Some wire nuts call for this and some don't, but I find that if one of
the wires is stranded, it helps if the end of the stranded wire sticks
out a little (like, less than 1/8" or 3 mm) past the ends of the solid
wires, before screwing on the wire nut.

Most of the time the house wire will be 14 gauge. Some cities, even
for residential, mandate 12 gauge wire (and 20 amp breakers or fuses)
minimum, so sometimes the house wire is 12 gauge.

Standard disclaimers apply; I don't get any money or other consideration
from any companies mentioned. This is based on experience in the US.
I am not an electrician and this is not electrical code advice.

Matt Roberds
 
A

Andrew Gabriel

Jan 1, 1970
0
I am not an electrician either. I have used wire nuts correctly many
times and incorrectly once that I know of!
Wire nuts work but there is no way to inspect them. You can't tell
what happened to the wire because you can't see it. Did it stay
twisted? Is it making contact, or hanging by a thread?
When I lived in Europe they were unknown. Everybody used the
"chocolate block" which made a far more positive connection, and one
where you could see if you'd done it right or not. I can't see any
advantage to the wire nut over the chocolate block.

They were never common in the UK, but had been phased out by the 1930's
in mains wiring because of poor quality connections. They only remained
for things like low voltage bell wiring.

They were replaced with something which looked similar, but is actually
a one-sided chocolate block connector with a grub screw to clamp the
conductors inside a brass insert. From that, the current range of more
familiar chocolate block connectors developed, two-sided and multi-pole.
 
T

Travis Evans

Jan 1, 1970
0
In the US, the amateur is most likely to rely on the wire nut, when
he's exactly the person most in need of a more dependable method.
[...]

I definitely qualify (at least for now) as an amateur. I had
been using a few good practices like tugging on the wires after
application to make sure they're attached firmly, but as I've learned,
wirenuts are still definitely not foolproof. It's nice that they
require no tools to use, but as was mentioned, it's often not visibly
obvious whether the connection is sound or safe.

Also, it seems that attaching more than about three leads to a single
wirenut can be be a real pain, especially when some of the wires weren't
made quite as long as they should have been (the neutral/common
connections in my HID light fixtures often seem to be guilty of this).
Unless possibly you happen to have at least three or four arms. :)
 
T

Tomsic

Jan 1, 1970
0
Travis Evans said:
In the US, the amateur is most likely to rely on the wire nut, when
he's exactly the person most in need of a more dependable method.
[...]

I definitely qualify (at least for now) as an amateur. I had
been using a few good practices like tugging on the wires after
application to make sure they're attached firmly, but as I've learned,
wirenuts are still definitely not foolproof. It's nice that they
require no tools to use, but as was mentioned, it's often not visibly
obvious whether the connection is sound or safe.

Also, it seems that attaching more than about three leads to a single
wirenut can be be a real pain, especially when some of the wires weren't
made quite as long as they should have been (the neutral/common
connections in my HID light fixtures often seem to be guilty of this).
Unless possibly you happen to have at least three or four arms. :)

Where I want to be sure of a good connections (such as in wiring installed
outdoors), I use pliers to twist the wires together, then solder that
connection, and then screw a wire nut over the top for physical protection.
I've never had one of those connections go bad in 20-30 years. Sometimes, I
add an overwrap of vinyl electrical tape to keep moisture out. There's also
a liquid polymer available for making a waterproof connection. You dip the
connection (including the wire nut) into the liquid, wait until it stops
bubbling and then let it dry. It's then as weatherproof as insulated wire
itself.

I also like another connecting device that's almost as good as soldering.
It consists of a short, small-diameter copper tube and a rubber cap. Strip
a bit of insulation off the swires, put the wires together, slip the tube
over the wires and crimp it with a crimping tool. That makes a solid
connection that you can easily test for strength. Then place the rubber cap
over the connection and lock it. The cap has a rubber tab with a loop in it
so it encloses the connection -- even the bottom -- and the rubber tab goes
over the cap to lock it in place. I use that type of connection for indoor
work. It's easy to work with such connections to make wiring modifications.
Just remove the rubber cap, snip off the wire at the end of the copper tube
and make a new connection -- very little wire is lost.

Soldering, however, is my "gold standard". If I want the best, most
reliable connection possible, I solder it.

Terry McGowan
 
G

Glen Walpert

Jan 1, 1970
0
Travis Evans said:
On Nov 16, 11:29 am, [email protected] (Andrew Gabriel)
wrote:

They were replaced with something which looked similar, but is
actually a one-sided chocolate block connector with a grub screw to
clamp the conductors inside a brass insert. From that, the current
range of more familiar chocolate block connectors developed,
two-sided and multi-pole.

--
Andrew Gabriel [email address is not usable -- followup in the
newsgroup]

In the US, the amateur is most likely to rely on the wire nut, when
he's exactly the person most in need of a more dependable method.
[...]

I definitely qualify (at least for now) as an amateur. I had been
using a few good practices like tugging on the wires after application
to make sure they're attached firmly, but as I've learned, wirenuts are
still definitely not foolproof. It's nice that they require no tools
to use, but as was mentioned, it's often not visibly obvious whether
the connection is sound or safe.

Also, it seems that attaching more than about three leads to a single
wirenut can be be a real pain, especially when some of the wires
weren't made quite as long as they should have been (the neutral/common
connections in my HID light fixtures often seem to be guilty of this).
Unless possibly you happen to have at least three or four arms. :)

Where I want to be sure of a good connections (such as in wiring
installed outdoors), I use pliers to twist the wires together, then
solder that connection, and then screw a wire nut over the top for
physical protection. I've never had one of those connections go bad in
20-30 years. Sometimes, I add an overwrap of vinyl electrical tape to
keep moisture out. There's also a liquid polymer available for making a
waterproof connection. You dip the connection (including the wire nut)
into the liquid, wait until it stops bubbling and then let it dry. It's
then as weatherproof as insulated wire itself.

I also like another connecting device that's almost as good as
soldering. It consists of a short, small-diameter copper tube and a
rubber cap. Strip a bit of insulation off the swires, put the wires
together, slip the tube over the wires and crimp it with a crimping
tool. That makes a solid connection that you can easily test for
strength. Then place the rubber cap over the connection and lock it.
The cap has a rubber tab with a loop in it so it encloses the connection
-- even the bottom -- and the rubber tab goes over the cap to lock it in
place. I use that type of connection for indoor work. It's easy to
work with such connections to make wiring modifications.
Just remove the rubber cap, snip off the wire at the end of the copper
tube and make a new connection -- very little wire is lost.

Soldering, however, is my "gold standard". If I want the best, most
reliable connection possible, I solder it.

Terry McGowan

Electricians on the other hand almost never solder any connections, and
most of them get excellent reliability also, The most common reason for
wirenut failure (according to one of the wirenut manufacturers IIRC,
exact source forgotten) is pre-twisting the wires with pliers before
installing the wirenut. It is very important to let the wirenut twist
the wires. Most electricians and mechanics can do this by hand, most
others need to tighten wirenuts with pliers.

General method of installing wirenuts which is used by most of the
electricians I have worked with:

Buy good quality wirenuts with tightening wings. The round ones are hard
to tighten adequately and the cheap ones sometimes strip out before
adequate tightness. For example:

http://www.idealindustries.com/products/wire_termination/twist-on/

Read the directions - never use any wirenut for a wire combination not
specifically listed on the manufacturer's instructions, get the right
size. Strip the wires accurately to the strip length in the instructions.

Be sure you have enough free wire to make a tight, straight bundle long
enough to hold on to easily; 1-1/2 to 2 inches.

Make sure the ends of all wires are exactly even and parallel. While
holding the bundle tightly so that no wire can slip (it can help to tape
the bundle if slippage is a problem), start the wirenut and continue
tightening until the insulated wire between your grip on the wire bundle
and the wirenut is well twisted. 14 gage will twist a lot, 10 gage or
larger not as much. This will probably require pliers unless you have
stronger than average hands or the wire is 14 gage or smaller.

I suggest practicing with some of the wire and wirenuts you will be using
and tightening one to failure (stripout, fracture or wire breakage), so
you know how tight too tight is. Tight enough is around half of that.

Stranded wire is an exception. I believe the only proper use of wirenuts
with stranded wire is to join a single solid wire to one or two stranded
wires. If two or more solid wires must be joined in the junction box of
a fixture where stranded wire must be connected, the solid wires and a
short solid wire pigtail can be connected with one wirenut, and the
pigtail connects to the stranded wire. Due to the ease with which
stranded wire bends, a good connection requires that the strip length of
the stranded wire be enough longer than the solid wire strip length that
the stranded wire can have it's normal strand twist tightened up by hand
then, with the insulation end lined up with the solid wire insulation
end, the stranded wire should be pre-twisted about one full turn around
the solid wire, with about 1/8" left over beyond the end of the solid
wire. The wirenut must squeeze the end of the stranded wire over the end
of the solid wire to prevent the stranded wire from pushing back. Not a
bad idea to tighten one of these to failure to get the feel of it - they
strip out at lower torque than solid wire connections.

Wire layout is important for reliability also. The wires need to push
back into the junction box easily without jamming into both sides of the
box, to avoid excessive stress on the wire insulation and connections.
Depending on the dimensions of the box a U or Z bend, started in all
wires before making the connections, should both allow good working
length and allow a low-stress fold-up into the junction box.

Properly installed in a junction box suitable for it's location, the
wirenut will be protected, and if strip length was correct there will be
no exposed copper at the base and no need for any electrical tape. Tape
may improve vibration resistance, but a crimped connection is much more
appropriate for high vibration connections such as in motor junction
boxes. (Crimped connections have better vibration resistance than
soldered because of better wire strain relief. The wire tends to break
at the edge of a soldered connection well before it will break at a
properly crimped connection under the same vibration levels.) Tape won't
help in corrosive environments because of the impossibility of sealing
between the wires; a sealed connector is needed for reliability in
corrosive environments.

Might sound complicated and may take a bit of practice, but once mastered
it is a very fast and easy way to make code compliant and reliable
electrical connections in non-corrosive, low vibration environments, used
successfully by thousands of electricians every day.

(It is much easier and faster to show someone how to do this right that
to explain with text only - I could have made at least 20 connections in
the time it took to type this!)

Regards,
Glen
 
TimR said:
Can you even get chocolate blocks over here?

If you're talking about something like this
http://en.wikipedia.org/wiki/File:Leuchterklemme_75dpi.jpg
you can get them at most of the "big box" home improvement stores in the
US; some hardware stores also have them. Usually they have nylon
insulation and come as a 12-circuit strip; you can cut this apart to
make whatever size you need. The 12-circuit strip costs maybe $3 to $5.
I've never seen them sold in big buckets of 2- or 3-circuit blocks in the
US, which I would guess is a common item at the electrical supply store
in the UK and maybe continental Europe as well. (The one or two junction
boxes in Germany that I have seen the insides of also seem to use this
kind of connector.)

What do you do with this kind of terminal when you need to connect more
than two wires? Is it "legal" to stuff two bare wires under one screw,
or do you crimp multiple into a pin terminal first and stuff that under
the screw, or do you use a stamped metal jumper to connect several
circuits together, or...?

Matt Roberds
 
A

Andrew Gabriel

Jan 1, 1970
0
If you're talking about something like this
http://en.wikipedia.org/wiki/File:Leuchterklemme_75dpi.jpg
you can get them at most of the "big box" home improvement stores in the
US; some hardware stores also have them. Usually they have nylon
insulation and come as a 12-circuit strip; you can cut this apart to
make whatever size you need. The 12-circuit strip costs maybe $3 to $5.
I've never seen them sold in big buckets of 2- or 3-circuit blocks in the
US, which I would guess is a common item at the electrical supply store
in the UK and maybe continental Europe as well. (The one or two junction
boxes in Germany that I have seen the insides of also seem to use this
kind of connector.)

They come in strips of 12, which you cut down, and those come in
bulk packs of 20 or more strips.

You can also get ceramic (high temperature) ones in blocks of 1,
2, or 3, but they can't be cut down and are significantly more
expensive.
What do you do with this kind of terminal when you need to connect more
than two wires? Is it "legal" to stuff two bare wires under one screw,

Yes. The connectors are rated in cross-sectional conductor area,
and you can put in as many conductors as you like, totaling up to
that cross-sectional conductor area. In practice, you can put in
whatever fits. They work best when all the conductor strands are
similar size, so if you have wildy different sized conductor strands,
put them in opposite ends. Stranded flex should have a bootlace
ferule applied unless the chocolate block connector is specifically
designed for flex (in which case either the grub screw end is matched
to the curvature of the bottom of the channel, or it has a metal leaf
protection spring under the grub screw).
or do you crimp multiple into a pin terminal first and stuff that under
the screw, or do you use a stamped metal jumper to connect several
circuits together, or...?

If you're going to use a crimp, you might as well crimp all the
conductors. That's also perfectly permissible.

Crimped (or soldered or brazed) connections are allowed where they
will be inaccessible after installation. Screwed connections must
remain accessible. Wirenuts (or Screwits as we called them, which
was one of the manufacturers' brand names) are not allowed anywhere
for the last 70 or so years. The general principle of connections
is that the contact pressure should be high enough to be gas-tight
(which prevents oxidation). That can't be achieved with a wirenut;
it needs a screw clamp or a crimp.
 
G

Glen Walpert

Jan 1, 1970
0
I think you're correct with solid wire. The failures I've seen have
been related to solid wire. For stranded i doubt that it matters.

It is very important to let the wirenut twist

Yup. The failures I've seen were related to that. But those are only
the ones that burned off - who knows how many other flakey connections
are hidden away?


It sounds and is complicated. That's why I'm suspicious it isn't the
best method for the DIYer, though as you note the pros get it right.
Mostly.

I have worked with power distribution for several decades, including a
decade supervising a facilities design group with 6 power distribution
designers, and have been on site for initial checkout and power-up of
many facilities with thousands of wirenuts in low power lighting
circuits, and I have never seen a wirenut connection fail. I have seen
several screw connections fail, due to over or under tightening, and a
few cases of insulation failure due to being jammed hard against a box or
another conductor, but never a wirenut failure. I have worked with
dozens of electricians, but never once heard any complaints about wirenut
reliability. (I will start asking about this.) So I have to wonder, is
your experience a fluke, or mine?

Perhaps the difference is exposure to DIY wiring vs professional wiring.
I have to wonder how many of the approximately 65,000 fires attributed to
electrical causes annually in the US are the result of DIY wiring.
<http://www.nfpa.org/assets/files//PDF/Proceedings/
Ahrens_presentation.pdf>

It seems that all of the electricians I know who do residential work get
a good percentages of calls for a quote on a job and never hear back.
Once the homeowner gets a detailed quote with a list of exactly what is
needed to hook up whatever needs hoking up, and the price quote with the
$50 for the permit and the $100 for the inspection, the homeowner does it
himself or calls uncle Joe over to do the work, no permit, no inspection,
and it seems nothing to be done about it - although one electrician I
know has stopped putting any details about the work to be done on his
quotes.

Mind you I am not at all opposed to DIY work, as long as the homeowner
learns how to do the work correctly and has it inspected. Pity that
happens so infrequently.

Regards,
Glen
 
G

Glen Walpert

Jan 1, 1970
0
On Mon, 21 Nov 2011 21:10:15 +0000, Andrew Gabriel wrote:

... Wirenuts (or Screwits as we called them, which was one of
the manufacturers' brand names) are not allowed anywhere for the last 70
or so years. The general principle of connections is that the contact
pressure should be high enough to be gas-tight (which prevents
oxidation). That can't be achieved with a wirenut; it needs a screw
clamp or a crimp.

70 years ago there was a type of wirenut which was ceramic and had a
molded-in conical thread, which did not work nearly as well as modern
wirenuts with a metal conical threaded insert. I am not sure where you
are that they are illegal, but in North America almost every building has
wirenuts in lighting wiring in multiple places, and all but perhaps a few
badly installed wirenuts have redundant, gas-tight connections both
between the wires and the conical threaded insert, and directly from wire
to wire, with each contact area being larger than the wire cross
section. Take one apart after 40 years or more and you will still see
bright copper at both contact areas, solid proof of a gas tight
connection.

In discussions of the detailed causes of electrical fires, I have never
heard any reference to wirenuts being a significant problem, nor did a
web search turn up anything, so as of now I see no factual basis for the
claim that there is any problem with wirenuts other than the possibility
of faulty installation by untrained amateurs - and that happens with all
wiring methods!

Regards,
Glen
 
?

.

Jan 1, 1970
0
Glen Walpert said:
On Mon, 21 Nov 2011 21:10:15 +0000, Andrew Gabriel wrote:



70 years ago there was a type of wirenut which was ceramic and had a
molded-in conical thread, which did not work nearly as well as modern
wirenuts with a metal conical threaded insert. I am not sure where you
are that they are illegal, but in North America almost every building has
wirenuts in lighting wiring in multiple places, and all but perhaps a few
badly installed wirenuts have redundant, gas-tight connections both
between the wires and the conical threaded insert, and directly from wire
to wire, with each contact area being larger than the wire cross
section. Take one apart after 40 years or more and you will still see
bright copper at both contact areas, solid proof of a gas tight
connection.

In discussions of the detailed causes of electrical fires, I have never
heard any reference to wirenuts being a significant problem, nor did a
web search turn up anything, so as of now I see no factual basis for the
claim that there is any problem with wirenuts other than the possibility
of faulty installation by untrained amateurs - and that happens with all
wiring methods!

Regards,
Glen

The reason that the National Electrical Code (US) now requires "arc fault"
circuit breakers that feed outlets in certain areas of homes -- such as
bedrooms and living rooms -- is because of problems over the years with poor
electrical connections either on the load device or in the outlet wiring
that eventually arc, heat the surrounding materials and start fires. I
don't know that the faulty electrical connections which prompted the code
requirement involved wire nuts, wiring device connections or combinations of
those; but clearly there was concern on the code-writing panels about arcing
connections.

Terry McGowan
 
G

Glen Walpert

Jan 1, 1970
0
The reason that the National Electrical Code (US) now requires "arc
fault" circuit breakers that feed outlets in certain areas of homes --
such as bedrooms and living rooms -- is because of problems over the
years with poor electrical connections either on the load device or in
the outlet wiring that eventually arc, heat the surrounding materials
and start fires. I don't know that the faulty electrical connections
which prompted the code requirement involved wire nuts, wiring device
connections or combinations of those; but clearly there was concern on
the code-writing panels about arcing connections.

Terry McGowan

True enough. Such an expensive code change only went through because a
lot of deaths were attributed to fires caused by arcing faults. The arc
fault breakers do not protect against high resistance connections which
can get hot enough to cause a fire without arcing, but they are expected
to save a lot of lives by interrupting power to those that do arc.

At least in wiring installed by electricians you will rarely see wirenuts
in receptacle circuits, as standard duplex receptacles will connect 2 or
4 wires to both hot and neutral terminals, allowing all reasonable power
splits without wirenuts, so I do not think they were a significant factor
in the decision to require arc fault protection. A neutral wirenut at
the switch for switched receptacles is the only normal use of wirenuts in
receptacle circuits I can think of. Wirenuts are very commonly used in
lighting circuits, where they are used to connect neutrals in almost
every switch junction box and again in the fixture on hot and neutral,
but ground fault protection is not required for lighting circuits.

The two dozen or so overheated electrical components I have seen, none of
which actually caused a fire although many could have if near flammable
material or if not detected in time, in order of decreasing frequency:
- bad receptacle-plug connection with high load, caused by corrosion and/
or poor receptacle condition with worn contacts or weak springs. Poor
quality receptacles are often involved.
- loose connection at a screw terminal of receptacle or plug, usually
caused by bad initial assembly sometimes by abuse.
- pinched insulation shorts from bad assembly or abuse.

While I have heard of wirenut failures and read about their causes (and
how to avoid them) I have never actually seen a failed wirenut
connection. If the manufacturer's instructions are followed completely
they are easy to install and will work adequately over a fairly wide
range of tightness, unlike screw connections where in my experience
beginners rarely tighten correctly unless they use a torque screwdriver
which few of them own. For the past 35 years I have done once a year
volunteer electrical work at a large event put on by a non-profit
organization, where I often train new volunteers how to do safe
electrical work. Without supervision they will under tighten about 70%
of the time and over-tighten to the point of severe thread damage about
20% of the time, so I always make them use a torque screwdriver a few
times to get the feel of correct torque. If everyone did that a lot of
failures would be prevented. (I almost always use a torque screwdriver
or wrench myself, even though my feel is pretty good.)

I always give beginners pretty much the complete wirenut spiel as my
previous post, with demo, and they never have the slightest bit of
difficulty with them. But if anyone keeps statistics on specific
components involved in starting fires which would provide other than
anecdotal evidence that wirenuts are or are not a significant problem, I
haven't been able to find it.

Regards,
Glen
 
?

.

Jan 1, 1970
0
Glen Walpert said:
True enough. Such an expensive code change only went through because a
lot of deaths were attributed to fires caused by arcing faults. The arc
fault breakers do not protect against high resistance connections which
can get hot enough to cause a fire without arcing, but they are expected
to save a lot of lives by interrupting power to those that do arc.

At least in wiring installed by electricians you will rarely see wirenuts
in receptacle circuits, as standard duplex receptacles will connect 2 or
4 wires to both hot and neutral terminals, allowing all reasonable power
splits without wirenuts, so I do not think they were a significant factor
in the decision to require arc fault protection. A neutral wirenut at
the switch for switched receptacles is the only normal use of wirenuts in
receptacle circuits I can think of. Wirenuts are very commonly used in
lighting circuits, where they are used to connect neutrals in almost
every switch junction box and again in the fixture on hot and neutral,
but ground fault protection is not required for lighting circuits.

The two dozen or so overheated electrical components I have seen, none of
which actually caused a fire although many could have if near flammable
material or if not detected in time, in order of decreasing frequency:
- bad receptacle-plug connection with high load, caused by corrosion and/
or poor receptacle condition with worn contacts or weak springs. Poor
quality receptacles are often involved.
- loose connection at a screw terminal of receptacle or plug, usually
caused by bad initial assembly sometimes by abuse.
- pinched insulation shorts from bad assembly or abuse.

While I have heard of wirenut failures and read about their causes (and
how to avoid them) I have never actually seen a failed wirenut
connection. If the manufacturer's instructions are followed completely
they are easy to install and will work adequately over a fairly wide
range of tightness, unlike screw connections where in my experience
beginners rarely tighten correctly unless they use a torque screwdriver
which few of them own. For the past 35 years I have done once a year
volunteer electrical work at a large event put on by a non-profit
organization, where I often train new volunteers how to do safe
electrical work. Without supervision they will under tighten about 70%
of the time and over-tighten to the point of severe thread damage about
20% of the time, so I always make them use a torque screwdriver a few
times to get the feel of correct torque. If everyone did that a lot of
failures would be prevented. (I almost always use a torque screwdriver
or wrench myself, even though my feel is pretty good.)

I always give beginners pretty much the complete wirenut spiel as my
previous post, with demo, and they never have the slightest bit of
difficulty with them. But if anyone keeps statistics on specific
components involved in starting fires which would provide other than
anecdotal evidence that wirenuts are or are not a significant problem, I
haven't been able to find it.

Regards,
Glen

The National Fire Protection Association organized a study about 5 years ago
which looked at the aging of electrical components in homes. What the NFPA
did was to pull the electrical parts out of condemmed homes of various ages
and then analyze those parts from outlets to wiring materials, connections,
sockets, etc. for safety. I've not seen the study results, but it could be
a source of some wirenut info.

Anyone have access to it?

Terry McGowan
 
?

.

Jan 1, 1970
0
. said:
The National Fire Protection Association organized a study about 5 years
ago
which looked at the aging of electrical components in homes. What the
NFPA
did was to pull the electrical parts out of condemmed homes of various
ages
and then analyze those parts from outlets to wiring materials,
connections,
sockets, etc. for safety. I've not seen the study results, but it could
be
a source of some wirenut info.

Anyone have access to it?

Terry McGowan

Turns out the final report from the study can be downloaded at no charge
from http://www.nfpa.org/assets/files/PDF/Research/RESAReport.pdf I've just
looked through it quickly and didn't see any particular reference to wirenut
problems, but plenty of references to splices and connection failures. Lots
of scary pictures too!

Terry McGowan
 
G

Glen Walpert

Jan 1, 1970
0
On Wed, 23 Nov 2011 09:38:47 -0500, . wrote:
Turns out the final report from the study can be downloaded at no charge
from http://www.nfpa.org/assets/files/PDF/Research/RESAReport.pdf I've
just looked through it quickly and didn't see any particular reference
to wirenut problems, but plenty of references to splices and connection
failures. Lots of scary pictures too!

Terry McGowan

Nice find, thanks. I read the whole thing, and there is no mention of
any wirenut (twist-on wire connector) related problems being found, and I
think they were pretty complete in their failure reporting. The splice
problems were related to inadequate insulation and twisted, unsoldered
connections. (Wirenut insulation is more durable and damage resistant
than electrical tape). Compare to "Of the 254 receptacles tested, 55
receptacles passed the initial temperature test with a temperature rise
of 20° C or less." These were all receptacles with inadequate retention
force, as measured with an inexpensive receptacle tension tester. Test
your receptacles lately?

Looks like code violations and failure to replace receptacles with
inadequate retention force are the biggest problems found with installed
home wiring in this report, and comparing their installed wiring fire
statistics (6400 annual) to the NFPA statistics for all electrical cause
fires (65,000 annual), most electrical fires are started by portable
wiring and appliances.

Regards,
Glen Walpert
 
Top