Sounds interesting - cut into 18mm discs you should get an almost indestructible variable brightness switch. The paper seems to think that heating isn't an issue.
It's really cheap http://www.mutr.co.uk/product_info.php?cPath=418_464&products_id=1144, and the science principle underneath are delectable.
however its application for lights is somewhat limited by the nonlinear resistance to force (though this kind of works for lighting), the magnitude of forces needed, residual resistance, and the fact it's still a resistor. A light basically has to be fairly well made I think to make it work.
The best really cheap way to do variable lights is already available either with the mode controller/pwm or through the sense pin on regulators, but it's a problem of motivation rather than any tech limitation that they're not common.
Ok, I've thought about it, and it probably wouldn't be that bad even for the cheap lights we have. Basically you just glue it to the center + terminal of the driver boards and it should more or less work pretty well.
Anyone in the UK interested in a group buy? It's only out to a few bucks for each person given it's like 80 or pills so for 10pounds. I'd buy it myself but their international shipping looks a bit nebulous. Or I suppose if enough people in the US were interested I can ship out and split the costs from the UK since these basically fit in any 1st class envelope.
Very interesting in did, so in theory you can make a direct driver variable just twisting the body if the rubber pill is located where agentex say above.
The quid is how 'sensible' this stuff is since the battery need some force to make contact anyway.
The very old rheostat's way to control power is not that bad, the Led Lenser headlamp have one and work pretty fine.
I have a CA little engine from 1900 with a rheostat wired in a rectangular piece of stone working good despite his 110 years old.
Installation is tricky. Obviously you can't use a non-conducting adhesive and these little things seem to leap for freedom every chance they get. I'm going to try a sandwich between two coins of less than 14mm diameter to see if that works better. Or look at some light with a spring post rather than a spring at the tail end.
Maybe glued over the edge on a equally thick rubber with a hole made 'ad hoc' could work.
Would be interesting know the resistance of the stuff half pressed and full pressed, the range must be between 35 ω or less to 0 ω if not has this range I doubt could be used to control a power led.
Not that easy to measure with a multimeter, but I got a reading of 2.5megohms with the probes just touching it. Which may be the conductivity of the wooden desktop.
Sandwiching it between two coins to prevent the very sharp meter probes from destroying the rubbery material I got a resistance of more than my meter could measure >50megohms down to <0.1ohm with a couple of kilograms of pressure.
That's a good idea, I think I have some old inner tubes somewhere which might suit for attaching the stuff