Mythbusters - (k)not lightning protection

[Poe]

My high-school electronics teacher (vocational school) told us that a knot tied in the electrical cord of an appliance would act as an effective choke to lightning strikes.  He showed us a power strip from "a relative's home" as evidence.  It was burnt beyond recognition from plug to knot.  Beyond that it was like new, albeit slightly yellowed.

I want to test the concept.  Any ideas as to a realistic method?  Not only theoretically, but practically... if even possible?

Might make a good video?

[jwm_]

Do you have a scope?

Far from a test but you might start by seeing if a knot has any effect on a signal at all. Start with a circuit  that gives a nice sharp pulse like the one i use here http://notanumber.net/archives/110/experimenting-with-a-homebrew-time-domain-reflectometer
And send pulses down a knoted and unknotted wire. If you can't see any change with one knot, wind it up several times and see what happens.

[Psi]

Sounds like the lightning jumped from the cord to earth at the point of the knot.

I suspect this was more to do with physical position than anything electrical.

ie, The plug is in the wall and the wire is hanging down, the knot is the lowest point so is touching the floor before anything else.

[Whales]

Physics teacher, I do not understand.  You have told me that it happens, but now how or why.  This is unscientific -- please help me.

[HackedFridgeMagnet]

Maybe it was just an 'almost' short (cause by the knot) in the cable that produced heating effects upstream of the knot.

A lightning surge would have probably wrecked other stuff in the house not just part of one cable on one sub circuit.

Busted.

[lewis]

Could be insulation breakdown as a result of the pressure of the knot on the insulation around the internal cores coupled with several kV surge. Insulation breaks down causing a short and when the mains voltage returns to normal a large current is drawn melting the cable from the knot to the plug.

Just read HackedFridge's post - basically this.

[electr_peter]

If cable is overheating, it will burn down and short at the point of knot (because of heat energy concentration). I doubt knot would have significant effect on lighting strike, proper experiment is needed.

[Rerouter]

another way to look at it from how you describe it could be a mis-accredited image, if you where to tie a knot in a cheap chinese power board cable and try and draw the rated current, the insulation would likely melt at the knot and shot active to neutral,

if said relative had also replaced the fuse with a piece of coat hanger or grossly overrated wire for the fure (very commonly done in houses with blade fuse holders) then it would have melted and likely started a fire on the plug end to the knot leaving the rest unaffected,

[Psi]

burnt beyond recognition from plug to knot.  Beyond that it was like new, albeit slightly yellowed

This does sound like a lightning hit to me,  either that or a 11kV/15kV arc over to the 110/220v side which also creates a charred mess wherever the current tracks.

[alho]

Arcing happens at sharp points and the knot is "sharper" than rest of the cable.

[HackedFridgeMagnet]

burnt beyond recognition from plug to knot.  Beyond that it was like new, albeit slightly yellowed

This does sound like a lightning hit to me,  either that or a 11kV/15kV arc over to the 110/220v side which also creates a charred mess wherever the current tracks.

But then why would the surge stop at the knot.  The impulse would still travel further along the cable.
Sound like a more gradual heating thing to me. Especially with a power board involved.

I want to test the concept.  Any ideas as to a realistic method?  Not only theoretically, but practically... if even possible?

High voltage capacitor and something to switch a short circuit. Thyristor maybe.
maybe 50 kamps and an 8u sec rise time IIRC is a standard class II surge.
I seen/used this type of equipment but not something for your home lab.

[Seekonk]

I think a knot could easily be seen on a TDR.  It is a loop and therefore an inductor.  Others have wrapped cords on pipes.  A communications guy I know swears by it.  Nothing really provides protection from a real strike. 

[SeanB]

A knot would be a single turn common mode choke with an inductance of around 1nH, give or take depending on the cable. A lightning strike might generate a transient across this of possibly 10V, but you still will have a good couple of kV of spike going to the rest of the circuit. Still will kill the stuff on the end.

Only way it will not is if the length of wire after the knot happens to form a quarter wave stub on the cable at the equivalent frequency of the rise time of the pulse, and the knot was then a node with high voltage, while the appliance end was effectively short circuit and thus had no voltage across it. Not likely, I read of a church that had been converted to a business, where lightning struck the roof, at the lightning rod. the current arced through the multifoot thick stone wall at the point where an internal mains wire crossed the direct lead down of the lightning earth. the current flow was enough to vapourise sections of wire except at nodes in the ring mains, and blew all the equipment inside aside from one tape recorder ( it was a recording studio)  that was faulty, and which was sitting in the middle of the floor with it's mains lead wrapped around it awaiting collection for repair.

The voltage drop across the lightning earth was enough to cause the voltage to arc through to the mains wiring. The maximum allowed resistance of a lightning earth is here 8 ohms, while a mains earth is allowed a maximum of 1 ohm.  The lightning earth mat where I stay was built and specced to be under 1 ohm, and it is actually lower than the mains earth that is also bonded to it. It is likely the equipment attached to it will survive a direct strike, though you will have a big burn on one antenna that will need repair.

[LabSpokane]

I am not buying that the house took a direct hit. Otherwise the electrical panel would have likely blown out as well. The best ground in the system is at the meter - or ought to be. :-/

The likely explanation - as already stated-  is that the insulation failed at the knot and grounded out the circuit. That is why it was only melted between the knot and the plug. If the building had actually shunted a direct hit through the outlet,  all of the wiring between the panel and outlet would have fried as well and likely caught the house on fire.

Tying knots in power cords is just a silly piece of advice for lightning protection. First off NO in line device protects against a direct hit.  Lightning arrestors and surge protectors only protect against induced currents from nearby strikes.  And the only good protection against direct hits is a lightning rod system or its equivalent coupled to a large conductors going to a redundant grounding system.

Grounding and lightning protection is its own little science and well worth studying if you're going into commercial power systems.

[LabSpokane]

Oh and by the way. The best way to test this "for real" would be in conjunction with your local electrical utility. Some.of them have facilities to do this type of testing.  The level of energy involved in this type of test is more than you really want to play with at home.