... you're welcome, my comment was not implying anyone endorsed his methods. In my years at eevblog I don't think anyone has made an analysis of leakage current from DMM in high voltage mode. I think he had the right idea in mind but not the right technical implementation, except for his drop test; that meter did hold up well to physical abuse and for that alone, its a plus and I was impressed; thus the effort to discuss it here.
Modemheads comments are also vital, that the videomaker took some safety precautions and yet risked others suggests there was some gaps in his knowledge, that can be serious with kV. Had the meter truly made a light show at high voltage, he was not complete covered and his face and elbows were exposed, also his father's desk was littered with items that could ignite and make matters worse.
Compare to Dave's video of doing the same so long ago, the conditions and methods are proper, plus he made it fun:
The high voltage shock is short and snappy, and the physical result is less the issue than that voltage leaping out of the meter chassis and into the users hands, thus giving them a potentially injurious but short lived kV shock. They can be lethal.
My beef with cheap meters is not so much it doesn't work, aka measurement accuracy while new, but that it may fail unexpectedly as it ages, be it months or years, and when you need it most. Even if the safety is dubious, if you use it only for CAT 1 level and below 30V, most any meter, is safe to use. I once kept such a cheapie in my car for troubleshooting purposes, and found it reading erratically when I actually needed it and then died, and of course, I didn't have a backup because what can happen to an unused meter in a car? All it was subject to was heat, humidity, or severe cold every year and used only once. I subsequently bought a used Fluke 85 for $50 and replaced it and as expected, works every time.
The video says he bought 10 pieces at $3 each, and by probability if I did the same and brought a few I could correlate them when needed to insure at least 2 were reading correctly, and replace whatever broke. Over years a few may die due to assembly defect, I'd still be ahead, but if there is a parts defect from a cheapo source, all can die literally the same year. So now, my $30, I must compensate for the cheapness by doing more brainwork, and carrying more stuff, a simpler solution is get a meter with a solid reputation. That's what we pay big names to do, do truly good QC, and what few rarely discuss: the operational temperature conditions of heat, humidity, vibration and shock that a portable device is subject too.
Conclusion? I don't think the test shown proved the meter is safe to use to meters rated ~1kV nor the results of current overload because the tests for those safety issues were not conducted.
Saturation, thank you for your explanation. When I posted the link, my intention was not to make fun of the poster, but to try and learn what is actually going on with his tests. I'm trying to learn more about input protection so your explanation is appreciated.