Low Smoke & Fume cable versus P.V.C. cable overloading test. Shed Experiment.

A good demo but not perhaps totally representative to heat only one core, usually the flow and return will be in the same cable and once they touch, the ADS should kill the power. Had he applied the welder power at one end and a short at the other, he would have probably found he did not need more than 100A to burn the insulation off the 2.5mm2

The PVC problem is in part that the fumes are horribly poisenous, rather than just that there is a lot of smoke- not all smokes are equal ;-)

Mike.

The cables did though take many seconds to actually become damaged. That surprised me. Admittedly the wires were in a cooling air flow designed to remove the fumes away from the test bench.

Z.

High currents through the cable are perhaps not very representative of what happens in a conventional building fire either - in this experiment as soon as the plastic reached melting point it fell away from the source of the heat (the conductor) - in a real fire with the whole space engulfed in fire, the insulation would continue to be heated & burn until its final destruction whether it fell away not.

    - Andy.

Yes these demos are neither pure adiabatic, being far too slow - the heat clearly has time to get out of the copper into the plastic. (for be solidly adiabatic  kA and milliseconds are the right units) nor are they small overload of long  duration, which would be more like 40 A on a 2.5mm2 cable for an hour. In my limited experience of this sort of thing the 'cheese-wire" effect means that some form of ADS dead short detection operates before the cable actually flames.

However, as it is intended to illustrate the differnet performance of  low smoke cable it is still a nice demo.

Mike

P.S.C.C.s and earth fault currents around here will I believe be limited to a few hundred Amps according to my testing. Normally less than 1kA. But I suppose that nearer to substations the figure might be larger. And in towns and cities the figure nay approach a few kA.

Z.