Testing Extension Cords/Outlets

[MrAl]

Hello there,

I have been testing extension cords to see how they fair with medium currents like 5 to 8 amps. What i found was the the receptacle end gets warm but the plug end doesnt.
These are the general purpose extension cords, the kind with the three outlets on one end and a plug at the other end, length 6 feet to 12 feet, but mostly the 6 foot length.
The wire on most of these is 16 AWG gauge, but one has 14 AWG gauge wire.

Anyone else notice this or are these cords all faulty? It's hard to believe that they would all do this. Note however that they do not get hot hot, they just get warm, maybe very warm but not really hot, so say around 50 degrees C. The plug end stays cool or nearly so.
Even the higher priced model with 14 gauge wire did this, which was very surprising.
This doesnt mean they dont work, but it is strange that the receptacle end gets warm while the plug end does not get warm at all or hardly noticeable.
Also keep in mind that 5 amps isnt that much and neither is 8 amps for a 15 amp rated cord.

Run time has to be about 2 hours or longer for the thing to reach it's max temperature. The longer the better however, like 5 to 10 hours.

Does it sound like a faulty receptacle? Well, when i replaced one end with a regular single outlet receptacle it never got even a little warm after that. Even after 10 hours run time it stayed as cool as the plug end. But it is hard to believe that the contacts themselves are the problem, as they seem very clean (almost new extension cord) and have a tight feel when you plug something in.
The only thing left then is the connection between the wire cord and the receptacle. They are probably crimped on, but then again the plug end probably is too, so why the receptacle end and not the plug end that gets very warm? Could it be that they are made so different that they crimp poorly on the receptacle end and ok on the plug end?

I know i can replace the receptacle end and the problem is fixed forever, but i would like to igure out what causes this anyway. Also, for a three outlet extension cord they dont seem to make replacement three outlet sockets, only single outlet sockets, so you loose some functionality if you go to a single outlet for the fix.

Anyone else notice this?
Any ideas what causes this in the first place?

 

[Ian Rogers]

Are these extension leads coiled? It they are stored in a coil even 3 amps will generate heat... The specs say that you MUST uncoil the whole extension lead to use it!!

There is also a thermal fuse in some.... This will get hot as it is there to trip when an overload takes place..

 

[MrAl]

Hi there Ian,

Very good point...this is exactly the kind of thing i want to hear
which is the pro's and con's of using extension cords too, as well as the heating problem.
There is too much baby testing of the cords on the web which is misleading as to their go no-go status.

The cords i was testing were all straight type, uncoiled, non fused, your standard type cord, except for that one which was heavy duty 14 gauge.

I also have a 10 gauge triple tap type but that seems ok, but i havent tested it that well yet. I am mostly trying to find out what happens to those run of the mill type because i find this end heating thing with all of them, even when new.
The test current has to be at least 5 amps though, and not more than 8 to make this an effective test. More is ok too i guess, but i would expect heating at 15 amps which is the rating, which BTW i would not use it for.

Thanks, and thanks to anyone else who has any ideas about these things too.

 

[Tony Stewart]

For every extension discovered as fraudulent ratings, there are probably many more degraded by poor plating thickness, surge currents from SMPS chargers inserted pitting the plating and poor contact design for reliable contacts.
**broken link removed**
**broken link removed**
**broken link removed** Known to Be Sold at: Ollie’s Bargain Outlet, Allentown, PA

https://ulstandards.ul.com/standard/?id=817 Devices produced integrally with flexible cord or cable, covered by the Standard for Cord Sets and Power-Supply Cords, UL 817;

AC receptacles use thinner flexible contacts and thus higher thermal resistance, meaning higher temperature rise than the plug.

I see this as a marginal quality standard with substandard cost-reduced components. Better designs are used in other countries with more contact area, but I see this as a compatibility cost dilemma for North America to fix.

 

[MrAl]

Hello Tony,

That's interesting. So somehow these fake UL tested cords get through.
Now i noticed that this "heavy duty" cord i got from Home Depot doesnt even have a UL rating on it anywhere to be found. Three labels hanging off the wire, stating how to use an extension cord safely, but no UL rating. This one has a faulty socket that gets too warm. Other cheaper ones only get a tiny bit warm, so there is a noticeable difference, and this is at less than 1/2 it's current rating.

LATER:
I cut the socket apart without damaging the metal or wire, and noticed that the metal seems good, crimp seems good, contacts tight, everything looks good. This means it must be the wire where it exits the socket body. I'll have to strip the wire to find out though, which i'll do next after another test with it apart.

 

[KeepItSimpleStupid]

Mr Al:

Take a look at the blade thickness of what your plugging in.

With a Hoover Vacuum cleaner, I have the opposite problem. The cord near the plug gets warm easily. I'll bet the wrong wire gauge.

You might TRY to measure the resistance of each conductor including the contact resistance. What about which outlet is used of the three? Is the center one better?

I have a two foot 3 wire adapter thingy, but instead of using it on the end of the cord, I use it on the beginning to give be easier access. I've never checked the temperature. Likely uses are a string trimmer, electric chain saw and demolition hammer.

 

[MrAl]

Hi,

Thanks for the ideas.

I checked the blade thickness, looks pretty good. It's a tight fit in the socket. There is a slight difference in tightness from socket to socket, but i used the tightest one for the tests.

With the vacuum cleaners i think they expect short time usage so they use thinner than normal wire, but if it gets hot at only one end that means there is a problem with the end somewhere. My guess is that the wire bends back and forth near the area where the wire exits the plug body and that stresses the strands. They can break one by one inside the wire, or even stretch, and a stretched wire ends up with a smaller wire diameter than when new.
I suspect that is the problem with this one cord i am testing, but i dont see how that could have happened because it was almost never used, so it may be the wire is just defective inside near the end...maybe it was pulled on or squashed during production or something.

I will test the resistance next before anything else.

LATER:
Ok tested. It came in at 56mOhms per wire over the entire length. The calculated value should have been 37mOhms, so there is a problem, in the wire, both wires actually, the black and the white ones.
They do look like 14 gauge wires, so there must be a problem with the wire such as broken strands or stretching.
Also of importance though is the wire coating is much thicker than what we normally see i think. The normal 16 gauge extensions have two wires where you can see both wires, while this one has a round cable with three wires inside, one is ground. The round cable covers all three wires but it somewhat thick, so it probably causes some heat buildup because the wire metal to ambient thermal resistance is going to be much higher than the more normal extension cord. The normal extension has some insulation on both wires, but this round cable has some insulation on both wires PLUS the outside cable coating which is at least as thick as the two wires individual coatings. That means double the thermal resistance, so that cant be good.
This doesnt answer the question of why it only gets hot near the one end though. That's asymmetrical, which means there could also be a problem neat that end. Maybe that end of the wire is causing the higher resistance, but it doesnt look like the crimp is doing it as that looks really good.

 

[Tony Stewart]

Measure the contact resistance. As you probably know Hotspot temperature rise is a function of series Electrical Resistance, Rs times effective series Thermal Resistance of junction to ambient, Rja.

or ΔT= I²Rs* Rja in units [W] * [°K/W]

where for increments of ΔT, units °K and °C are interchangable

e.g.
Use different plugs with a short circuit in different socket positions.
Driving another socket with a plug wired in series with any heavy current. (AC or DC, 5A~15A)
Measure series Current on 10A scale.
Measure voltage across an unused socket for load plug pair contact resistance only.
and then at source plug screws or pins for both plug pair contact resistance. (i.e 4x R_contact_avg)
Reverse the plugs and repeat for verification. Note how R increases with Temp and I^2R.

With a ~10A measurement and 1mV resolution on voltage drop, you have 100 uΩ resolution easily, perhaps 10 uΩ with a 4 digit DMM (0.1mV).

I would expect new contacts 50 mΩ to be acceptable for a 30 deg C rise. ( not sure on stds but all should be based on temp rise at rated load)
A better method of specifying cordsets might be hotspot rise after accelerated environmental aging.

.. and I would expect contact life cycle might only be 500 cycles subject to some standard load.

If marketting says "heavy duty" it might have the same connotation as with carbon pile batteries.. ie. much weaker than alkaline.

In the end, consumer cordsets are only supposed to be safe from fire or shock, not lossless and lossy cordsets are safer in event of accidental short circuit due to inherent R current limit from a 10kA ampacity panel, so arguments to improve the standard may be futile.

FWIW
Life is a simple application of Ω's Law.

Tony Stewart

 

[KeepItSimpleStupid]

A, sort of, related story that I investigated was a large number of electrical failures in our building. It really came to light when the ground potentials were about 120 V/2 . This is a magic number and it's usually generated by the symmetrical leakage resistances to ground in an RF filter in a 3 prong appliance. Hence when the ground lets go, a very low current capable voltage source is developed. If something is using ground as a reference, that stuff can be destroyed.

Iv'e seen it happen in outlet strips and also electrical outlets. The brand that was in the building where I worked had 420 of them. The design was such that the ground was made by friction between the plated piece that runs down the center of the front of the outlet and a brass piece that made the ground contact. If the tolerances were not right, AND two plugs were inserted, it was possible that one of the grounds would "let go". I did identify a means of testing.

The powers that be made a decision to "replace most". Nn valuable areas (Labs/Hallways) all were replaced and to evaluate when computers or valuable equipment were added to other areas. It was past Warranty time.

 

[MrAl]

Hello again,

Now that i have the socket cut apart i can test it in more detail and find out exactly where the heat is originating from. I might be able to do more testing tomorrow sometime, hopefully.

I am temped to just cut the end off and replace it with a single socket made for fixing extensions, but i am trying to hold off until i can test it more and pinpoint the fault.

 

[audioguru]

If the wire was crimped to a contact that is not clean (or a little corroded) then there is a higher contact resistance and more heating. Maybe the wire was also corroded because the extension cord was made during a monsoon.

 

[MrAl]

Hi,

Thanks ag.

I did check the crimp and wire at the crimp site, and it looked exceptionally clean. That was surprising to me because the cord was hanging unused for at least 2 years now.

What it is starting to look like is something we didnt think of yet. I had it plugged in for about an hour now to test it again, and no signs of heating. Not even lukewarm. But now i have the socket split apart and open, so there is plenty of free air flow that easily reaches the metal that makes up the receptacle contacts.
But what else i did inadvertently was plugged the test plug into the socket that is CLOSEST to the wire, which was a big mistake because i am suspecting that the brass that makes up the contacts is the culprit. Brass has resistance about 4 times that of copper, and because it was totally enclosed it had almost no free air flow to cool it, and the plastic they used is very thick stuff (strong though).
That's my hypothesis, and so i moved the plug to the LAST socket at the end of the row of three sockets. Now the entire brass strip that makes up the contacts is involved in the conduction of the entire current, unlike before when the plug was right near the wire. So next i will test the temperature rise of the brass and see what happens. I just did this a few minutes ago so it's going to take some time before i see anything worth mentioning. But seeing the cord end cool for the very first time is a nice change.
If the hypothesis turns out to be true, then i guess i will have no choice but to change the ends when this happens. For this cord, i could have epoxied the socket back together, but even if that works i think i would be more comfortable with replacing the end entirely now, with one single socket.
Also if this turns out to be true it means that other cords made like this probably have the same problem, and the ones with the three sockets closely spaced will have the same problem only to a lesser degree, unless totally defective of course.
Does a little heating in the end mean the cord is no good? Probably not, if it is not too bad, but it could lead to faster tarnishing which might make it worse over time. This would lead to a bad failure after some time like 6 months, a year, maybe two years, etc. The way i see it is if we can find ONE cord that DOESNT heat up at all, then they all should be made to be at least as good as that one. That's the safest possible route i believe.

 

[KeepItSimpleStupid]

Mr. Al:

I have this one **broken link removed** from Home Depot. I used it today for about an hour with a chain saw and no warmth.

 

[MrAl]

Hi,

Thanks for the link. I checked their site and they dont have this one listed anymore. It would be listed as "15 ft., 14/3, red and black", and they dont have that same one. They do have one that looks like it but the cord is black, so that's a different model. If it's made the same i dont know but i wont be buying another one to test.

A longer test reveals the cord is now completely cool from end to end. If any temperature rise it has to be 1 degree but no more. The only thing that can make this much of a change is the cooling of the guts of the socket, the metal contacts. There's nothing else i can think of that would cause this because the only difference now is the socket is wide open, with the top section folded back when before it was molded closed which completely encloses the contact metal except for the very small slots that make up the holes where the plug prongs stick through.
Alternately, playing with the wire so much maybe some of the strands in the cord are touching better now, if they were broken to begin with. This is a little harder to believe though.
There could be a difference in the depth of travel of the prong that goes into the contact 'fold' that is made to contact the prong. As the prong goes into the fold, the plastic socket enclosure will prevent it from going deeper than a certain depth. With that part of the enclosure removed, the prongs can probably go deeper into the folds of the contact metal, but it's hard to measure this because the socket enclosure does not go back together exactly the same as when new.
This is just a guess of course as it may not make a difference either.

 

[MrAl]

Hello again,

I decided to check the plug prong depth after it is plugged into the outlet side. I had to push the plastic halves back together tight to do this. What i found was that the plastic is so thick that it only allows about a quarter inch prong depth into the socket metal. The way to tell is to gently plug the plug of another cord with nice heavy prongs into the socket, and slowly and gently push it into the socket while feeling for any resistive force. That is, the plug will push in freely until the very tip meets the first socket prong (the two could be slightly different). That is where the measurement starts, and from a cross side view we see some of the prong still exposed, the part that did not yet enter the socket plastic. For this extension, when this occurs the exposed prong length is about 1/4 inch. That means that when the plug is then forced all the way in, only 1/4 inch of the prong makes contact with the internal socket contact metal. For another socket tested this is different: about 3/8 of an inch remains exposed, meaning once it is pushed all the way in it will make 3/8 of an inch contact, or at least farther up on the prong past the usual small hole at the end of the prong. By contacting only the last 1/4 inch, it is subject to missing the contact area that makes up the hole. There is no metal there, so it can only contact the edges. This limited insertion distance could contribute to heating.

So for this cord, the plastic is thick which makes it very strong, not subject to crushing under foot or the like, but it also limits the distance the prong can enter into the socket which means less electrical contact area.

I have since then replaced the end with a somewhat high quality single outlet socket end. What surprised me about this high quality outlet end though was that it does not have a mechanism for cord relief. It depends on the strength of the three wires. They are pretty strong however, being 14 AWG, but that's not how it should be done really as there should be some method of cord relief. No room for an "electricians knot" either.

 

[KeepItSimpleStupid]

When I think "high quality", I think Hubell. https://www.stayonline.com/detail.aspx?id=6806

For BOTH ends, I want something that you can grab onto. I remember when i was a kid and used to go with dad to the junk yards where you could actually walk around in and buy stuff. I found a 2x14 AWG x 4 x18 rubber cable that i thinnk I paid $8.00 for say 40 years ago (scrap value of copper at may be %0.33/lb back then. The ends cost me some big bucks. The cord is a bit heavy though and I still have it.

I use a flat 100' yellow cord with a smaller ground combined with an orange cheapy 25' cord. When i use te hedge trimmer, I have a coiled cord that clips onto the belt.

Then these things https://www.amazon.com/Cord-Connect...437986931&sr=8-9&keywords=extension+cord+lock (the orange things that lock to cords together) don't work quite right. I tried to repair one with real hinges and it, sort of, works better.
It's a really nice product, only if it were bigger and had real hinges.

 

[MrAl]

Hi,

Yes the Hubbell ends are the best probably, but also the highest price. I see others made in a similar way but less expensive, maybe 8 bucks, or less made by Leviton.
The one i am using was 5 bucks but they might be more expensive now. It's pretty nice because it is a little smaller than the Hubbell and similar, but still has grips on the sides for when plugging and unplugging.
https://www.amazon.com/Pass-Seymour...438005149&sr=8-6&keywords=extension+cord+ends

Another way we used to do it was to connect a regular steel wall box to the end of the wire complete with two to four outlet sockets. That was pretty tough, but looked kinda unfinished. Wonder if they make special covers for boxes with outlets that dont actually mount on the wall. Using a regular face plate hangs over the edges and doesnt look right anyway, and isnt that strong.

The way i did one cord was to cut off the socket end, then disassemble a not-too-expensive power strip, then replace the wire from that power strip with the better wire from the extension. That gave me a long wire for the strip, several outlets, a built in circuit breaker, and a couple added LEDs to show power present and power 'on', as well as the power switch. It was an unlighted switch so i used LEDs instead.
This works out pretty nice, and the new 'end' is soldered onto the end of the wire so it's a pretty good connection. For temporary use only.

That cord end 'container' looks interesting. What we used to do was tie the ends together and then plug the two in. That would relieve all the stress off of the connectors. We'd loose about a foot of wire doing that because of the knot.

 

[KeepItSimpleStupid]

There are cover plates, I think like this **broken link removed** one that you would use on a outlet box that might be surface mounted in the cellar. The boxes have rounded corners. You might get away with an outdoor receptacle box too and a cord grip.

Cord grip: https://www.mscdirect.com/product/d...&pxno=33614944&refnum=33614944&rItem=05773668

You can use https://www.mscdirect.com/product/details/54119805 on a weatherproof box.

The "knot" trick doesn't work well on fat cables or less flexible cords like 14 AWG.

 

[MrAl]

Hi,

Thanks for the links.

The thing i didnt like about the face plates was they they hang over the edge of the box. That is because the edge of the plate is made bigger than the box so that it hides the space between the box and the sheetrock in a normal installation.
This looks really crappy for an extension cord. If we could find a plate that would fit the box perfectly, that would be great. I know i have seen a four outlet one a long time ago but dont know if they still make them or not. They mount on top of a square box (four outlet size) and they look like they belong there.
Maybe a custom made cover? Didnt want to have to do that, but if it's not too hard to do then it's not too bad.

 

[KeepItSimpleStupid]

These **broken link removed** basically, should work.

The "boxes" on the end of cords look better with "weatherproof boxes".