Electronic – High voltage DC power line for batteries

This is a common problem in audio amplifier connections especially where there is digital noise and switching regulators. There is no ideal way around this problem without using differential inputs.

Why do differential inputs work - the signal return back to the box on the left won't carry any power supply currents because it feeds a high input impedance differential amplifier (or a transformer). Obviously you have to wire it correctly.

Without a differential input (silicon or transformer) it's difficult to find a signal return path that won't pass digital currents.

This problem has nothing to do with dodgy earths in your home or needing to put a decoupling capacitor somewhere.

To begin with the potential of the upstream system needs to be known and the distance and transformers (impedance) that exists.

Once that is known and the closest inline fuse or circuit breaker device is known - some modeling and calculations can be performed.

One metric which is modeled is ISC or Instantaneous Fault Current. This is also referred to as a bolted fault - as in if the phases were bolted together creating a line to line short - what threshold would the current reach and for how long before the upstream overcurrent device could clear the fault. ISC usually has the quickest reaction time IF the fault current gets that high.

Another Metric is LSI or Long term Fault current. I think of this as the long time overload trip on a motor. You don't want this to trip during a short duration starting current but want it to trip if this value is sustained for a period of 5 or 10 seconds.

Depending on the system (which I doubt the utility overhead lines have) it may have LSIG or Line to ground fault monitoring.

The video your refer to was most likely a Line to ground fault which results in a lower current threshold for LSI and ISC detection.

You can see from the video there are arcs that are sustained for a few seconds and then stop for a few seconds.

To answer your question IF the fault currents and interrupt times for the utility were modeled - the event probably had a lower current threshold to pickup on ISC and a shorter time duration than can be detected through LSI. IF the system had LSIG the fault may have been cleared, but again I doubt if the utility has that fault detection.

Fault clearing devices usually have little to do with power dissipation and more to do with sensing current over a time duration.