I’m investigating field failures of lamp cord where the wires are breaking, right behind the strain relief overmold. The obvious problem is that I have a relatively small diameter, flexible cord entering a relatively large diameter less flexible overmold. I can change the strain relief shape and make it softer but I’m not sure that I’ve completely solved my problem. Here’s what I’ve got:

22 AWG lamp cord
Each conductor has 16 strands of 0.016 mm bare copper wire
UL style 2468
Insulation thickness 0.5 mm

I am observing that the copper wire alone limits the elasticity of the cable. The insulation, by itself, is very elastic. When I pull on the cable I realize that all of that force is transferred directly to the wires. Is this normal?

I’ve noticed that some manufacturers include plastic filaments with the copper wire. I assume that these filaments take some of the stress off the copper wire.

I would appreciate your advice on:

1)How much of a solution is addressed by just changing the over mold?
2)How should I specify plastic filaments that would be added to the cord?
3)Any reference information about this out there?

Very best regards

- Bill

Hi Bill,

This could a very serious problem because there could be fires and a threat to life and limb if the strands were breaking randomly and leading to hot spots. I would think that a full recall is required and UL should be notified immediately.

Perhaps I am old fashioned but just so you know 22 AWG is a telephone wire size (19/22/24 and 26AWG) and 18 AWG is bell wire size. Does UL really allow the use of 22 AWG as a circuit size? Perhaps this is a LED lamp.

Certainly the standard thermoplastic resin you describe is quite elastic and presumably the bunched copper strands are acting together as a composite material under axial tensile load unless the insulation can very easily slide over the two conductors. In that case, the conductors would rapidly take all the tensile load.

Copper is a very soft metal and repeated bending will create work hardening and ultimately a failure. I am however unclear as to the way the conductors are failing, (Repeated bending as described or a tension failure generally because people are pulling or yanking on the cord instead of the plug.) Please keep in mind that you sure don't have very much copper in your cable.

Anyway please let us know what part of the process your company is involved with. (Wire manufacture, lamp cord manufacturer, lamp manufacturer, lamp wholesaler or lamp retailer.)

It is the lamp cord manufacturer that is the key here and this company should be discussing the problem with the cable manufacturer.

Kindest regards,
Peter J. Stewart-Hay
Principal
Stewart-Hay Associates
www.Stewart-Hay.com

Hello Peter -

Thank you very much for the response. Fortunately my situation isn’t as dire as I made it sound by my choice of the word "lamp cord". It just describes the cable itself and not my application. I do get power out of the cable but it is from an energy limited SELV source.

You asked how the wires broke. I am trying to make that determination. I’m not a wire engineer so I had to look on the Internet to find articles with cable failure analysis information. I didn’t find much but I did find some good information at Test & Measurement World from May 2007. The microscopes that I have here are great for soldering but don’t have the magnification to get a real good look at the tips of the broken wires. I do see signs of flex failures and tension failures. I believe some of the flex failures are older. You can see where the wire tip has rounded like it has been hitting it’s old connection point. Can you point me to more information on wire forensics?

Here’s a clue to my problem. I use the same cable with different plugs. I have seen a small number of failures across the board. The cable is failing in unacceptable numbers has a plug that requires the user to pinch and release a locking feature before the plug will un-mate from product. The product is a consumer electronic device. I assume that repeated yanking of the cable (and the cable not unplugging from the product) causes the greater failure rate.

I plan to address the problem by changing the overmold design, making the overmold material softer and adding the plastic filaments. Making these changes show a marked improvement in a flexures to failure test. However, I’m not sure that a weighted flex test is a good indicator of the failure mode. Can you recommend to different test?

Lastly I have a number of completed assemblies where I am considering re-working the product instead of scrapping it. The problem is, I can’t replace the cable itself (adding the plastic filaments). I can only cut off the old plug and apply a new plug / overmold with the new overmold design and material. Again, I’m encouraged by improved flex test performance but not sure how good of an indicator that test is.

I definitely appreciate your comments.

Best regards

- Bill

Hello again Bill,

Thanks for the information. Now I understand what you are doing with this cable. It reminds me somewhat of my pinch and release special USB cable which connects my SLR camera to my computer.

Now, the conductor failures:

The College of Engineering and Applied Science, Department of Mechanical Engineering at the University of Colorado, Denver; and the University of Denver, School of Engineering and Computer Science, Department of Mechanical and Materials Engineering don't seem to have a metallurgical engineering degree program but I'll bet they have metallurgical engineers on staff. If you take in a couple of damaged cable samples that have not already been mechanically compromised by previous investigations, they should be easily able to tell you how the strands parted. All it takes is a good metallurgical microscope and somebody trained in what to look for. Metals all fail in ways that are well known. Moreover they will probably do this for free as a service to the surrounding community.

Second, the fix:

Once the failure is known, the cure is probably straightforward. If these conductor are tension breaks, you need more strength in the cable jacket and that could mean longitudinal strands (under tension) of perhaps Kevlar or some other high-strength, preferably non-metallic material. The wire and cable product design engineers should be able to give you free new design cable samples to experiment with. My concern is that unless people really pay attention as to how to release the plug, the stronger cable may just move the problem along to the next weakest component. Perhaps a two color plug highlighting the pinch feature is needed.

If the conductors are flexing too much for some reason and then work hardening and fatiguing to failure, I believe you are already on the right track by overmoulding with a with a new and a much longer tapered stress relief shape at the plug and by the use of a softer plastic.

Best regards,
Peter J. Stewart-Hay
Principal
Stewart-Hay Associates
www.Stewart-Hay.com

Very good advice Peter, thank you.

Peter, Bill in Denver

Is this a UL labeled product? If it is then they can probably do some testing for you. There are also a couple of labs out there that can evaluate your problem etc. I don't remember the name but one was located in the St. Louis area.

rb