Re: [TCML] constructing a metal suit "ideas"

[Jim Lux <jimlux@xxxxxxxxxxxxx>]

On 9/1/11 11:38 AM, James Hutton wrote:
> wow. I have posted A LOT of questions lately... i guess thats what happens when you finish your summer job a week before school and have nothing to do all day...
> I am wanting to enter a tesla coil into a science fair. (basef) After thinking for a while, I think I have come up with a good "problem" (or excuse for using a tesla coil in a science fair :P)"Is it possible to create a lightning-proof suit"... that way, I get the excuse of making an impressive coil and using lightning simulation to test my suit.
> I've been doing a bunch of research on metal suits, (and asked about metal paint on a morph suit... I guess not...).I have seen lots of clunky big looking things and I dont really like the look of those.I came across this video: http://www.youtube.com/watch?v=2aDt7loyEfUThat is exactly what I would want. Something that fits closely to the body and is smooth looking.I'm assuming they are using chain-mail in the video... but I have looked everywhere, and I cannot find full body suits. Not even all the parts to put one together (i.e. gloves, shoes etc..)Most likely those are custom-made and expensive...
> So i guess I'm just looking for the cheapest, but most reliable way to make a full body suit. Some other things that might work is using screen door mesh. but I dont know if that will give me enough protection. (maybe with more layers?)or maybe even simple clothes wrapped nicely with aluminum tape.
> Since this is a science fair, I want the suit to look professional aswell.
> Does anyone have any tips, other ideas, or input?thanks!
>   		 	   		  _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla


First off, excellent for getting started on the science project early, 
assuming you're looking to compete at school and regional level leading 
up to ISEF next May.

What grade level are you?  The expectations and advice are different if 
you're in 7th grade than 12th.

Basic science fair advice:
1) With a tesla coil, you're rowing upstream, a bit, because TC's, per 
se, are overdone.  Using one as a piece of test equipment, as you're 
contemplating, is a good idea.

2) You need a way to do quantitative measurements, and repeatably, in 
the course of your investigation.  If you make your problem "how do I 
make a protective suit", that doesn't lend itself to a quantitative 
answer.  A better kind of question is, "can I make a protective suit 
that weighs half as much as conventional suits"  or "can I make a suit 
that reduces the electric fields inside by a factor of 10" or something 
else.

3) You need some (documented) research upon which you made your 
estimate.  Take the "can the field be reduced by a factor of 10" kind of 
problem:  You need to have research and analysis that says:
	a) what are the fields now
	b) what possible ways could reduce it
	c) what reduction would you get for each of the ways
ANd then, you can go do experiments and see if the actual reduction 
matches what you figured by analysis in (c)

Or, for instance, you say, "I want to make a safe suit that is no more 
than 1 mm thick and weighs less than 1 kg"  (numbers out of thin air, 
for example).. You need:
	a) research on what is "safe" (in terms of things like ANSI limit on fields
	b) what the current "state of the art" for suits is, how thick, how 
much they weigh
	c) What your proposed new approach is
	d) Calculations of what you expect your approach to get
	e) experiments to confirm what you calculated in (d)


This list has lots of people who can point you in the right direction 
for background information, analytical techniques, test methods, etc., 
but ultimately, you've got to be able to stand there and say something 
like:  The state of the art for the protective suits used in Tesla Coil 
demonstrations is 10kg and 5mm thick, as used by A, B, and C.  These 
suits use a traditional approach similar to chain mail or shark suits, 
and reduce the fields to D, E, F when exposed to discharges of G, H, and 
I.  A new and improved approach, similar to that used in RF protective 
gear, as described by J, K in IEEE Transactions on Y, or the conductive 
suits used for "hot workers" on energized HV Transmission lines (see 
IEEE Trans on Power Electronics, v32, no1, 1953), should allow a 
reduction in mass and thickness to ....

You get the idea...
All too many projects show up with "My teacher suggested X, and it 
seemed to work", and while good experimental technique counts for a lot, 
it's the early stage research and calculations and predictions that 
separate the OK projects from the winners.  (i.e. the Hypothesis 
shouldn't just be a guess.. it should be a carefully researched and 
educated guess)


Your measurements MUST be quantitative and repeatable.

No "the sparks looked longer"
The measurements should be in standard units.  If you're reducing 
electric fields, then you'd better be reporting the field in 
Volts/meter.  If you're reducing the "shock current", then you should be 
reporting the current in mA, through a realistic conductive phantom 
(Basically a salt gel or similar with conductive properties similar to a 
human)

You have to do multiple tries, with the same experimental conditions. 
If you are varying something like voltage from 0 to 10 in steps of 1: Do 
the measurement set to 0, then set to 1, then.., and set to 10.  Then do 
it again, starting at 10, and decreasing in steps.  Then do it again, 
using random steps.    that lets you say, with numbers to back it up, 
that one trial didn't depend on the previous one.  And it gives you 
multiple measured values for each experimental test value.


You MUST be able understand the fundamental uncertainty of the 
measurements.  This is more than just "sig figs".  If you make 10 
measurements of the same setup in the same conditions, and you expect 
the uncertainty to be X (2 sigma), then by gosh, the standard deviation 
of your measurements should match (i.e be X/2).  This is a "separate the 
men from the boys" kind of aspect.  Last spring, I saw entirely too many 
projects at ISEF where they cranked their numbers through Excel, put up 
the graph with a straight line fit and said, look, residual error is 
0.1391433.   Not only was the underlying function they were trying to 
fit NOT a linear function, but if their experimental error is on the 
order of 100 times the residual error cranked out of Excel, it's 
meaningless.  If you don't know why you're doing the analysis, then 
don't do it, because doing the wrong thing looks worse than doing nothing.



Go for it... Good luck, and ask questions...

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla