I want to make pressure dependant resistor.

Glad you haven't given up. What about a solid material like quartz with a signal amp. You know, the pressure on quartz is the basis of lots of our components.....

what type of pressure do you want to use?
i thought of domething like a seringe and with some comtacts attached to it and some springs, or a tube, or maybe something like a tin box with a spring inside and use as a variable resistor a resistor from an old potentiometer , the type wich it like a line, up and down if you know what i mean.
but you can not drive such great loads with these methods.

High power ?

Just been reading the saga and wondered if you could make use of a bog-standard rheostat?
A car petrol-sender uses a float to move the wiper on a rheostat (a few watts I guess).
You could buy or wind your own on an insulating bar, try finding iron, constantin or similar wire to wind it, or get an old wirewound resistor (easily available up to 30W and more if you hunt), old ones had bare windings, the newer (round) ones are glazed (which needs removing).
A wiper could travel its length to give varying resistance.

The clumping you discribed is unfortunate and, I'm afraid typical with powder-y material. I think I would move to materials that have a more natural tendency to return to their uncompressed state. I'm thinking of air, water, oil, and hydrolic fluid.

Consider the possibility of an air cyclinder (like a piston in a cylinder - or - like the hyperdermic discribed earlier). These things are made in an unbelievable assortment of sizes and are designed to extend or retract a rod (similar to a car's piston rod) when air is injected. Now, how about setting the thing up such that when you push the rod down manually, the air inside the cylinder compresses. When you remove the pressure on the rod, the pressure inside the cyclinder returns to normal. The variation in pressure inside the cyclinder could be connected to a sensor (like an oil pressure sensor in a car) because these devices are fitted with standard pipe threads.

This will meet your requirements for a small travel distance, a return to neutral, and a variation of current (of course you will have to build a circuit to deal with your current requirement.

Fortunately, we have more ideas than problems.....

Thank you all for your suggestions.

However, this unit when done, will be in series with
a 12v car battery, it will be operated mechanically,
it will be expected to control about 60 or 70 watts,
maybe more if i make a big one.

The one i have made now will turn a small three watt
twelve volt lamp from completely out up to very near
full brightness from my pressing on it with my
fingers.

But it soon starts to clump up a bit, and then the
lamp wont go right out. Adding a little graphite did
not seem to help at all. I may try putting the
powder in layers between washers or such, but im now
thinking of blotting paper and some sort of
electrolyte in many layers.

But im a bit concerned that liquids and current
almost always produce gas. I dont want those sort of
problems.

This unit will be expected to reproduce the inputted
mechanical pressure fairly faithfully as current
from the 12 volt car battery. It doesnt have to be a
very good reproduction, but it has to be
recogniseable.

I have just been experimenting with wire wool.
Curiously the rusty parts of the wire wool did seem
to control the current a bit.

Maybe all i need is a pile of rust?

Well, i have some more testing to do.

The suggestion by mechie of winding myself a
suitable rheostat is realistic, but this unit will
be providing a constantly varying output, but then
so does the petrol float on a car, so maybe it could
run for a long time without failing. i will have to
give that possibility some more thought.

Just to recap,
this unit should be capable of varying the current
to a car headlamp, from bright to very dim.

Regards, John

I've been reading about your efforts.

You might be able to make coil spring out of HIGH Resistance wire (NiChrome wire ?) that would sort of look like a battery spring.
Coil the wire in a flat spiral first, and then pull the center out a little bit.

The center of the coil is bonded to a contact plate and the other, your second connection, is capped with some foam rubber.

As increasing pressure is applied, more of the spring coils should come into contact with contact plate lowering the resitance.

Another thought, Mabey leaf springs sitting on a flat carbon surface would work in a similar fasion.


P.S.
Personally, being a little lazy, I would use a strain gauge to sense the pressure, feed the signal to a PWM oscillator and drive a power transistor(with heat sink) to handle the load. Such a system could easily handle five times the power demand you require using inexpensive components. Using a fuse would increase the reliability factor.

Hi SPARX,

Many thanks for your suggestion, which i am quite taken by.
This assembly would be expected to operate quite fast,
and also handle currents of around 10 to 12 amperes.

I am most disappointed that i was unable to get graphite
or carbon to do the job, i still feel that it should be
possible to do this with carbon or graphite.

I have heard of something called a 'silver pile' which i
am trying to find out about.

I would be looking for responses at around 4,000 per minute.

Still working on this,
John

I found something called a 'carbon pile'
which i think may be the sort of thing
that i am looking for:

http://avstop.com/AC/apgeneral/924.html

Unfortunately i could not get it to
display on my machine, i wonder if any
one would be kind enough to copy it and
maybe post as a jpg ?

I would like to see what this is.

John

I found this description of a 'carbon pile'
which sounds very much like the unit i am
trying to construct.

**broken link removed**

in paragraph three.

Hi john - I copied the article and saved it as a .doc file because it is too big to post here as an attachment. I intended to send it to you as a PM but I see that medium will not allow an attachment at all. If you would like to PM me with your e-mail address, I would be glad to send it as a normal e-mail attachment. If .doc is not good for you, let me know what you prefer.

Have you tried a bar and spring arrangment with a piece of nichrome wire? You could slide a contact up and down the wire to vary the resistance. Even the wire from an old toaster should handle the current you are looking for with no problem.
Powdered metals and such are going to drive you nuts.