I've been testing cables and electrodes now for a year, to weed out leaky safety gaps.

It's brought me praise from the people who I test for even when I have  to give them bad news.   It's brought me the threat of a lawsuit from a  seller of violet wands and accessories.  At different events, I have  seen anywhere from 10% failure to 50% failure!

I never expected the fail rate would be so high.  Fail defined by a  device which is obviously intended to be gapped but reads leaky when  tested by a 1000V megger. This year at Thunder, I was talking to a couple who were looking  at some electrodes.  The lady was asking about sensations, and she  described a scene where they were using a body contact cable, and  something or someone touched her metal bracelet and she got a bad shock  that made her hand go numb. I thought about it, and I asked if he had that particular body cable  with them at the event, and he said that he did.  I asked him to please  bring it down for testing. When he brought the cable down later, we tested it and it was leaky. A leaky cable itself isn't enough to cause a bad shock.  In order for  line current to flow through the leaky safety gap, there has to be a  path from the hot side of the outlet to ground or neutral. The cable looked fine externally, but it was definitely failing, and it hurt his lady!

Needless to say, he's getting a new one.   Pictures and more information are available on my Fetlife profile.

I wasn't there when the incident happened so I can't say for sure, but I can propose a scenario: Reversed hot and neutral in the outlet or extension cord, or in the  wand itself.  This would bring the hot lead through the resonator coil,  out to the collet.  The failing body contact would occasionally allow  line current to jump the gap.  This would become more frequent as the  scene progresses due to heating of the failing safety gap.  If the  person then touches ground or neutral somehow, then they could receive a  bad shock. This is exactly what the safety gaps are supposed to prevent, and in  this case it did not.  

Fortunately there was no permanent damage. Her  bracelet may have been sitting just above one of the nerves in the  wrist. This kind of a bad shock event could be a scene ender, could turn  someone off permanently from violet wands when they would otherwise  enjoy them, or could result in significant to serious injury.  In an  extreme case death is possible. A shock from hot to neutral by the way, won't be picked up by a GFCI  since they are designed to trip only on a situation where the current on  the hot lead is not the same as the current on neutral, typically a hot  to ground connection. 

Electrical play IS edge play, and the responsibility for safety rests with YOU!