27 kV is not 20 kV!
I took 20 kV because it's really the worst human load value I've measured in the 5 years I've had my MONROE Voltmeter. At the time (1985-1990), I had 4 electrostatic discharge simulators:
- Two from KEYTEK, the 3000 model up to 30 kV and the Minizap portable up to 15 kV.
- An old Schaffner 430 single polarity up to 25 kV (from memory) and a more modern 432 dual polarity with relay allowing contact discharges but limited to 16 kV.
After that, I had my hands on just about every make and model of simulator except NoiseKen. Last year I sold a Schaffner 435 and an EMC Partner ESD3000 16 kV, and earlier this year an EMC Partner 30 kV.
Today, I've only kept two devices: an old Schl?der 16 kV that I don't use at all any more, and "THE" Schaffner/Teseq 438, which is by far the best simulator ever made, and which I'll probably take to my grave!
In our case, the man doesn't discharge to earth, but to an ungrounded cart or rack.
So we get a partial discharge, a charge balancing between two capacities, which means that there will never be 20 or 27 kV, but only about ten kV, which is bad enough.
So please stop dreaming up laboratory tests that I can also do perfectly well, and try to think before you transpose what you do in the lab into field tests.
If you want to check that a neon light or a LED can be turned on in real conditions, redo all the tests with only about 10 kV and, above all, don't keep the current standardized value of 150 pF / 330 ohms, but take a more realistic value of 100 pF / 450 ohms, which corresponds to what an average human can represent.
Of course, we need only consider purely passive devices. Active devices do exist, but their price is too high to be discussed here. On the other hand, these passive devices need to work in real-life conditions, not in the laboratory. 27 kV never exists in real-life conditions, otherwise IEC 61000-4-2 would not have been content to define level 4 at 15 kV only.
150 pF / 330 ohms is a bit of a stretch to really allow maximalist testing according IEC 61000-4-2. If you verify in other standards you will see that some other HBM will be used.
F1EKU
