[Elecraft] ESD mat measurements

[Alan Bloom]

A few months ago I reported here on some measurements I made on the
Radio Shack portable ESD mat (see below).  Basically I found that it
doesn't bleed off static charges as it is supposed to.  (i.e. It fails
the ESD Association resistance specification for ESD mats.)

So I recently bought another ESD mat from Jameco.  It's their
19.5x23.5-inch anti-static mat, P/N 10584, $16.45.  Tonight I measured
it using the same technique that I used with the Radio Shack mat.  It's
even worse!  After a half hour, the capacitor was still charged to 75%
or so versus 50% with the RS mat.

So what's going on here?  I can think of three explanations.

(1)  There's something wrong with my measurement technique.
(2)  Perhaps you're supposed to treat the mat with some kind of
conductive material before use.
(3)  Cheap anti-static mats are a fraud and are worthless for their
intended purpose.

I can't figure out how it could be (1).  As a sanity check I confirmed
that the sauce pans I was using as probes are conductive and the
capacitor is indeed 0.1 uF.  I doubt it is (2) - I can't believe that
the mats are supposed to be untreated as they come from the factory.

I suspect (3).  Years when when I was a components engineer at Hewlett
Packard, I was measuring some of those rubber heat sink insulators and
found they didn't even come close to meeting their thermal resistance
spec.  So I came up with Bloom's Law:  "The harder a specification is to
measure, the more likely it is to be a lie."  If you buy a 1k, 10%
resistor it will almost certainly be within spec since anyone with a DVM
can easily measure it.  Measuring surface resistivity by the ESD
Association method requires an uncommon, expensive piece of test
equipment that people who buy $16 ESD pads are unlikely to have.  So
it's easy for the manufacturer to cut corners without getting caught.

Al N1AL


On Wed, 2008-02-27 at 09:51, Alan Bloom wrote:
> Well, I've convinced myself that the Radio Shack portable ESD mat, P/N
> 276-2370 doesn't work properly.
> 
> The "ESD Association" http://www.esda.org has promulgated an
> industry-standard test for ESD mats, ESD S4.1.  It is the standard
> specified by most commercial mats.  I decided not to spring for the $70
> to buy a copy of the standard, but other information I found on the web
> describes the test in general terms.  It uses two circular electrodes,
> each weighted with 5 pounds, spaced 10 inches apart on the mat.  The
> "Point to Point Resistance" is specified to be:
> 
> At 40-60% RH: 10^6 - 10^7 ohms
> At 20-40% RH: 10^7 - 10^8 ohms
> At 10-20% RH: 10^8 - 10^9 ohms
> 
> I don't know what the RH here in Santa Rosa was yesterday when I did the
> test, but I don't think it was very low since it has been raining
> recently and the ground is still damp.  For sure the resistance
> shouldn't be below 10^9 ohms (1 gigohm) and probably more like 10^8 or
> 10^7 (100 or 10 megohms).
> 
> I measured 2.5 x 10^10 ohms (25 gigohms), which puts the Radio Shack mat
> way out of spec.
> 
> Test procedure:
> 
> I didn't find a specification on the electrode size, but in the photo of
> a popular tester they look to be maybe 3 or 4 inches in diameter.  For
> my test, the electrodes were two saucepans, each about 7 inches in
> diameter and weighted with 5 pounds.  They were spaced 10 inches apart
> on the mat (3 inches edge-to-edge).  I connected a 0.1 uF film capacitor
> between the two pans and charged it to 15V with a power supply.  
> 
> I set my ancient Simpson analog volt-ohm meter to 60 uA full scale.  If
> I touch the leads across the capacitor immediately after charging, the
> needle momentarily jumps to about 6 uA (1/10 full scale) as the
> capacitor discharges through the meter.  If I wait half an hour (1800
> seconds) for the capacitor to partially discharge through the mat
> resistance, the needle jumps to about 3 uA.
> 
> An R-C network discharges to 3/6 of original voltage in about 0.7 time
> constant.  So the time constant must be 1800/0.7 = 2571 seconds.  That
> implies the mat resistance is 2571 sec / 0.1 uF ~= 2.5 x 10^10 ohms.  
> 
> Al N1AL
> 
> 
> _______________________________________________
> Elecraft mailing list
> Post to: Elecraft at mailman.qth.net
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft    
> 
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com