I'm doing a project with some tight requirements, in which I need to measure the distance between two points, which might not be in direct view of each other, and on parts that are too small to mount anything on them.
An elegant solution would be to tie some kind of conductive rubber band between them, which would change resistance as it stretches, and then measuring the resistance, but all I've got from Google is stuff about physical resistance of exercise bands.
There ABSOLUTELY are such devices, made of conductive rubber of some sort. For example, HERE, from RobotShop.com
I suspect there is a bunch of hysteresis, and that the resistive properties will change over time and use. Perhaps one of the other suggestions would work better
I'm not aware of any "rubber resistors", but there is a thing called a string potentiometer that implements pretty much the same function.
A spring converts a displacement into a force. You can then use a "load cell" to measure that force. Include words like "miniature", "micro" or "SMD" in your search. There are several manufacturers of devices that fit your space requirements.
You could also use an actual rubber band, but a metallic spring will give more consistent and linear results.
Yes there is. It normally changes its resistance on compression rather than tension, though there may be tension operated versions, and it uses quantum tunnelling.
The first available version was the "QTC pill".
Availability was difficult a couple of years ago but seems to be improving again...
Some background from the Nuffield Foundation
These are manufactured by Peratech who now offer a range of form factors for force sensing, as well as custom products.
While far from a rubber band, you might consider a rotary encoder with a spool of steel wire. Reset at a certain distance, maybe 0 cm, and then count the revolutions of the spool and calculate the distance. Although you would have to hold the device close to one object if too small to mount the device on.
There are three options that I know of for stretchy conductive strings, none very good.
A much better alternative would be something like acoustics, ultrasonics. It can even go other than line-of-sight, if that's a problem to be solved. You don't say what stability is required, either short or long term. Although the speed of sound in the atmosphere is dependent slightly on pressure, temperature and humidity, it is relatively simple to install an additional calibration recevier at a fixed reference distance from the transmitter, and make what is effectively a comparison between the reference and the measured distance.
So, a review of the stretchy strings ...
a) Any metal wire, obviously steel has larger elastic range than copper. The change in resistance will be tiny, and easily swamped by the temperature coefficient. A wire like constantin will have a much lower tempco, and may be more useable.
b) A rubber tube filled with mercury has been used in the past for exactly this purpose, however its toxicity rules out its use in this sort of application these days. There are other less toxic metal liquids available, though they are pricey and difficult to obtain.
An aqueous solution of some suitable salt, measured with AC, say copper sulphate between copper electrodes may be possible, but temperature coefficient would still be a problem.
c) You can get silver-filled rubber cord, for making conductive gaskets. This has quite a marked tension/resistance response, but it is weak, and prone to creep. A broad equivalent is the carbon-filled plastic anti-static bag material. The same problems of weakness and creep apply.
Why not forego the mechanical solution and go with a purely electrical solution? How about using laser distance measurement? It won't have reliability issues and you won't have to worry about tolerances on the piece that measures tolerance. Also with the mechanical solution it's accuracy will most likely decrease over time.
Here's an example of a laser tape measure hack: https://www.sparkfun.com/tutorials/323
