PSA: don’t use small and/or shoddy power cables!

Thought I smelled something funny and before I knew it, there was a lot of smoke and sparks. It was like a kids sparkler except about 10x stronger.

Thankfully, the only thing damaged was the power cable itself and my nerves.

Looks like it failed right in the strain relief.

Moral of the story… use a properly sized cable and also one that’s got proper UL/CE pedigree.

4 Likes

Wow… scary!

2 Likes

I’m having flashbacks to finding C13 cables jammed in where a high temp C15 cables where needed in lan closets. Despite there being a notch to prevent that.

I have the cable that came with the PSU labeled as such.

Yeah. I know better than this. I was lazy and just grabbed one from the bin when I was rewiring, apparently. Lesson learnt.

1 Like

Using good cables and not an el-cheapo-delux brand from a dollar store bargain bin is always sage advice, but, any cable can do this with enough flexing / abuse.

Don’t make sharp bends, don’t let things sit on the cable, dont pull on the cable to disconnect only pull on the “shell” of the ends, etc.

Put something (eg block of wood) behind your PC so if its pushed back against a wall it isn’t the power cord that stops it (inducing a sharp bend) which is what this looks like may have happened.

And remember - this almost fire would have still happened even if the hardware was OFF. :skull::fire:

2 Likes

It actually was not bent at all, at least when it died. But I have no clue what its history was. I was running some benchmarks on the PC so I was hitting it really hard and it probably got hot and snowballed from there.

The cable itself is of unknown origin. And the likelihood of the regulatory markings being legitimate is pretty low.

I dodged a bullet. I need to go buy a lottery ticket.

1 Like

Yikes! Never seen a cable suffer that fate. Makes you think twice about your electrical PC setup. Surprised its the only thing that suffered (excluding your nerves, they are likely asking for a vacation), the power supply likely used its safety features to protect the rest of your PC.

Glad you’re safe and that it happened while you were in the room, could have gone very bad

1 Like

Here’s the root of the issue. If you’re not confident about whether a cable has been abused in the past, cut the wall outlet end off and toss it.

So much for “sim flying isn’t dangerous” :fire:

2 Likes

I was going to use it to wake my son up in the morning. The bare end to the toes oughta get him up. :wink:

Of course I’m kidding. I am going to keep the parts as an example for future reference and lessons to others.

1 Like

Put behind glass in a shadow box labeled “8 to go”

2 Likes

Shouldn’t the circuit breaker have tripped as soon as the short circuit occurred?

This is why I keep the OEM power cords with their respective PSU’s and don’t reuse them if I upgrade. A cord for a basic 500 watt PSU may plug into an 800 watt PSU, but that doesn’t mean it’s capable of running 800 watt loads. Weaker PSU’s can have smaller wire sizes without the user realizing it.

3 Likes

It doesn’t have to be a short in order to spark. It just needs to be too few connected wire strands for the current draw. In essence, the few remaining strands became the equivalent of the filament in an incandescent light bulb.

3 Likes

That makes sense!

This thread makes me wonder whether I reused the power cord from the 850W P/S I replaced with a 1000W unit.

Even though it’s drawing the same amperage (that’s determined by the computer) it still makes sense to use the cord that came with the P/S.

I was a physics major in college almost 30 years ago and am a design engineer now. I know better than to use such a cable. It’s kinda like a dentist getting a cavity because he doesn’t brush his teeth.

I’ve actually had to specify materials with very particular flammability ratings (i.e. UL 94-V0). I had to blow out the little flames on the cable, which shouldn’t be necessary. It should be self-extinguishing. This one probably was never even tested/certified, even though it had markings that suggest it was certified. I can print out a certificate that says I’m certified as a brain surgeon, but that doesn’t mean I actually am certified. Low cost imported stuff often has fake markings.

Back to the sparks… To be a real “arc” the energy is actually jumping an air gap. It is electrical potential finding a path to a lower potential and the “spark “ you see there is the electrons using the air as a wire which becomes ionized (a.k.a. plasma). The key thing here is that an Electrical arc needs a destination too. And with the US household 120VAC potential, an arc can only jump a very small air gap. Way smaller than the distance between the two “prongs” on a wall plug.

The more I’m thinking about this in hindsight, it was actually combustion of the cable materials that was spewing sparks and little bits of burning material. So it really was like a kid’s sparkler. So what happened was that a few of the strands in the cable were broken already and I was hitting it with a big stability test load so lots of current. ANY current heats up the wires, but usually there’s enough surface area on the multiple strands to handle it and keep the heating within limits. That’s also why electrical transmission lines are high voltage so that the current is reduced, thereby reducing the energy wasted on heating up the wires.

As an interesting aside… the current actually only moves along the surface of a conductor, rather than through the bulk copper (or rarely aluminum). So stranded wire can handle more current than the same diameter solid core wire.

So the few strands that were still attached heated up and became the effective light bulb filament. They overheated and went from filament to fuse one by one, thereby exacerbating the problem by making the remaining filaments carry even more current. That heat caused the cable material to combust and apparently the way it combusts is by throwing out sparks.

The fact that my UPS and circuit breaker didn’t trip most likely means that it wasn’t actually an electrical short from hot to neutral or ground. Technically, there might have been little arcs between the broken ends of the strands as it was failing but that’s not really a “short to ground”.

Shame on me for using a cable that wasn’t up to par.

Well, you know, there’s a way to test if it was the combustion of the material…

Not that I’m recommending it. But there’s a way.

1 Like

I was fascinated when I saw this video. I always had the ‘water pipe’ model of current flow. Electrons in creates pressure (current) that pushes electrons out the other end. Totally wrong.

1 Like

For the sake of science, I might just do that. No electricity required.

1 Like

That is a great video!! Back in the day, I became very familiar with Maxwell’s equations and the Poynting Vector. So yeah, even my somewhat complicated explanation above is an oversimplified version that isn’t even technically correct. But having said that, it’s a good approximation of how it actually behaves. A mental “crutch”, if you will.

E&M, as we called it was really hard. Basically 4 dimensional vector calculus to describe the time dependent fields. And as Uncle Max always said, div B = 0. No magnetic monopoles exist! :wink:

And then there’s quantum mechanics. The powers that be still don’t agree on what’s going on there. The math is even uglier. Science is cool. Especially physics, but I’m biased.

1 Like