Human Body Capacitance and Resistance

A repost of a comment I made elsewhere, for discussion here.

The subject is ESD Human Body model values. I'm enamored by a 1989 symposium paper by Richard Fisher, of Sandia Nat'l Labs, where he created a "Severe Human ESD Body Model." His model had worst-case numbers meant for use in electrostatic-discharge circuit-protection analysis, etc.

Fisher's Severe Body Model consists of two parts, the body and an arm with hand reaching out to zap something. The body part has 400pF of capacitance in series with 250 ohms and 0.5uH. Then the arm and hand part bridges the body terminals with 10pF, and finally we have another 110 ohms and 0.1uH in series to complete the model and connect it to the poor real-world victim. The body capacitance is higher than you may see elsewhere first because the body is sitting down, and second because it's a worst-case body. We won't go further into what that means. :-)

You charge the 400pF capacitor to a voltage of your choosing. 20kV is a nice high number. During discharge we get a fast spike of current from the 10pF, with sub-ns risetime to dangerous levels, with up to 5A peak current, and lasting up to 5ns into the "load." This is followed by a slower discharge of the 400pF capacitance, lasting up to 200ns.

This would be followed by, ahem, a postmortem.

As for the effect of high moisture and humidity, as said, these can affect things, e.g., lowering resistances to the low levels we see in Fisher's Severe Model, but it also means the maximum electrostatic voltage developed on the 400pF capacitor is likely to be much lower. I suspect Fisher would prefer to take the dry-air high voltage with the moist-skin low resistances for his Severe case.

--
 Thanks,
    - Win
Reply to
Winfield Hill
Loading thread data ...

With all disclaimers on - not having been in depth into that ever, just having designed numerous HV sources - I think I have a question.

If the 10 pF is modeling the initial, pre-spark/ionization path (i.e the capacitance hand/victim), I can't see how it will be charged in sub-nS time, the hand has to approach the victim in that same amount of time (a really fast hand :-) ).

So the 10 pF must stand for something else I do not understand.

Now how much current limiting resistance is there in the ionized path, or how does it get ionized (perhaps the 10 pF are some part of that process that I do not understand, of course) I have no idea.

Dimiter

------------------------------------------------------ Dimiter Popoff Transgalactic Instruments

formatting link

------------------------------------------------------

formatting link

Reply to
Didi

Dimiter Popoff asked...

I'm sorry for not being clear. The 10pF is already charged up by the 400pF cap's voltage via 250 ohms. Here's an ASCII drawing of Fisher's Severe Human Body Model.

250 0.5uH 110 0.1uH ,---/\/\---UUUU---+---/\/\---UUUU----> > > >

_|_ _|_ ___ 400pF ___ 10pF | | '-----------------+-------------- GROUND

View with a fixed font, notepad, etc.

--
 Thanks,
    - Win
Reply to
Winfield Hill

I get it now - thanks. Not having much to add I guess I can only mention *my* body (not a model of it :-) got somewhat charged by 3 kV once or twice last time I did such a design (older designs have also utilized it but this has been long enough ago to be forgotten :-). BTW, the last time it was a 60W thing, I must have had some luck staying undamaged as a model :-). (

formatting link
)

Dimiter

Reply to
Didi

Those resistances seem to be quite low for the resistance of the skin. If the resistance would be that low, you would get severe burns each time you touched the 230 V mains.

Those resistances in the order of a few hundred ohms would make sense, if we assume that the spark penetrates the skin and the current propagates in fluids under the skin.

RF burns from touching an antenna connected to a 100 W transmitter can be quite painful, put it can be hard to detect, where the RF current penetrated the skin.

With the component values given, it would form a lossy resonant circuit with a resonant frequency in the upper HF range, thus expecting a few cycles at that frequency, until the oscillation dies down.

Reply to
Paul Keinanen

The skin is the big unknown.

Once opened by a wound, etc., the resistance is VERY low.

Open heart defib is only 2mA on the paddles.

Arm-to-arm with open wounds is only 10mA as opposed to 40mA on dry skin to cause a fibrillation.

Area of contact is also a factor in lowering skin resistance (lowering of the interface resistance into the salty blood conductor of our inner body).

As for an electrostatic event causing a fibrillation, it is not very likely until one gets up into the lightning bolt voltages. The time of stroke becomes a factor.

Reply to
Archimedes' Lever

I got 'charged up' by a 50kV X-ray supply once. It was a very good thing I had good shoes on and provided no path as an outlet for the charge feed. Otherwise, instead of charging me up, it would have charged me up to the point where the exit location sparked across the gap to the return point.

I would likely not be around to reflect on the event.

HV bench operations... one hand... the other behind you... remain separated from other bench elements.

Ideally, energize the device inside an HV containment cage. We never did that for anything at 50kV or below, but we did do it for all of our stuff that went up to 400kV.

Reply to
Archimedes' Lever

So0me references, if you don't already have them:

formatting link
formatting link
formatting link

Combining dry ait high voltage with moist skin low resistance sounds like a typical solutiion, when a committee avoids the use of it's individual brains.

RL

Reply to
legg

[snip]

How many times are we able to touch 230V mains in order to find out?

Bob

--
== All google group posts are automatically deleted due to spam ==
Reply to
BobW

I have known of two extreme cases of skin resistance: one person could feel a slight "tingle" of he placed his hands between 120VAC (himself in middle), and the other got killed when he accidentally got himself in series with a SIX VOLT car battery ("good old days") - the current thru the chest was more than sufficient to do the job.

Reply to
Robert Baer

Those resistances may seem to be low but i think are in the correct order of magnitude for a "worst case" based on "average". One person died of excess current thru the chest by "merely" (for him) getting in series with a six volt car battery.

Reply to
Robert Baer

formatting link

Do *NOT* bet on that combo as being unlikely..

Reply to
Robert Baer
[...]

Oddly enough, I could not find any existing LTspice simulation, so I posted the above in abse under "LTspice Human Body Model".

Interesting results. Where does the torso get 500nH?

Here's a couple more ESD references for those who lack an extensive library:

formatting link
formatting link

Mike

Reply to
Mike

Once again Dimbulb is wrong, "always wrong".

Don't you understand Ohms Law, if resistance goes up (dry skin), current goes down, YOU IDIOT!. Also it take the same amount of current through the body (dry skin, moist skin or open wounds) to cause their heart to fibrillate. The closer you are to the heart the less current it takes because more of that current will flow directly though the heart.

Resistance goes down for open wounds(assuming the current goes into the open wound) because there is a direct connection to the internal fluids of the body, hence less voltage is needed to cause fibrillation. With dry skin (high resistance) it takes more voltage to cause dangerous amounts of current to flow.

The current levels that you specified are also wrong!

20 uA of current is considered hazardous to the Heart if it passes directly through the heart.

500 uA of current is the maximum amount of leakage current allowed in medical equipment through ground and if the ground connection was broken, you touch the case of the device and ground, that current will flow though you.

These specs are for patients at high risk, poor health condition.

Reply to
Shaun

[snip]

Correct. 400pF doesn't store much energy at a few kV. At 3kV its 1.8 mJ. Getting stuck to a 3kV DC source is a different matter.

According to

formatting link
it takes hundreds of Joules to defibrillate a heart.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
What color is a chameleon looking in a mirror?
Reply to
Paul Hovnanian P.E.

You are the idiot. Every statement I made was about OPEN skin. You know, intimate access to the blood pathway.

Learn to read, dipshit.

Get a clue while you are at it. Then... grow the f*ck up.

Reply to
Archimedes' Lever

Wrong. The key is the pathway, and how much of the total current actually flows through that part of the pathway that is also comprised of the heart. The heart only needs a couple milliamps through it to fibrillate or defibrillate. As one gets further and further from the heart as far as the current source and exit is concerned, the amount of current needed to get the heart at 2mA increases, because the pathway is millions of parallel resistors of which only a few thousand relate to current flowing through the heart. The current in is one value, and the current out is that same one value, but the current in each of the millions of parallel resistors differs for each and not all contribute to any flow that would relate to fibrillation or pass through the heart. So entry point and egress point are very important. There are pathways that would not cross the heart at all. Mode of entry is also important. Dry skin has a high resistance. Blood does not. If punctures by the voltage or by other means is involved, the current to cause fibrillation lowers because the skin is not longer in the model.

You lose. Again.

Reply to
Archimedes' Lever

That is what I said, dumbfuck.

Reply to
Archimedes' Lever

NOT voltage, idiot. CURRENT causes fibrillation.

You lose, again.

Reply to
Archimedes' Lever

I made HV medical device supplies. I know about what the limits are and what the design constraints are.

Operating room defib paddles run at around 2mA during a cycle.

Grow the f*ck up.

Reply to
Archimedes' Lever

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.