How to Make a Ridiculously Cheap Analog Pressure Sensor

Tired of paying exorbitant amounts for a simple analog pressure sensor?  Well here is an easy smeasy way to make an incredibly cheap analog pressure sensor.  This pressure sensor won’t be terribly accurate in terms of measuring precise weight or things of this nature, though it can be calibrated somewhat and if you choose to coat it in something like Plasti Dip some of the variables such as humidity and the like can be minimized.  However, what this analog pressure sensor is best for is for creating things like bumper sensors that can read variable levels of pressure and various other touch / pressure sensor applications.   Find more interesting things like this in the How-to section of my website TodayIFoundOut.com

Materials:

• Any static dissipative foam (If you’ve ever ordered any IC chips, you probably have some lying around.  IC’s are often set in this foam for shipping.) or if you don’t have any, you can pick it up from a variety of places, such as this
• Wire
• (optional) Plasti Dip Rubber Coating

 

Step 1: Step 1

Step 1:

Cut the foam to the size you like.  You can cut it quite small and still get a good range of resistance levels.  The foam in this picture is cut to less than half an inch square and about 1/4 inch thick;  once completed these two each produce a range of around 2.6K Ohms down to 400 Ohms when squished completely.

Comments

[amason9]

Hi

These would be easy to hook up to an arduino, right? What would I need to do in the system to pick up the resistance change on the circuit?

Cheers

[roastprof]

You could build these into a potential divider to get a change in voltage as an analogue input. Or you could build a Wheatstone bridge to find the unknown resistance.

[harristotle]

This is a great idea.
In my hands, when I built a version that had the conductive foam sandwiched between sheets of aluminium (ie very good surface area), the resistance jumped around a lot, even when no force was applied or the same force was applied.

I am wondering whether there is much difference between different types of conductive foam?
Aren't they just graphite in polyurethane foam?

Any ideas to make the sensor more stable?

cheers

[hiskeyd]

So did you have the wire soldered to the aluminum?  I've only done the "poke the wires into the foam" way and I've never seen the value jump around; it's generally pretty consistent in my experience.  If you have the wires poked into the foam and the plates are just somehow causing the resistance to vary, you might try putting an insulated later between the aluminum and the foam and see if that fixes it.

hmmm. ya i can't think of anything else here why this might be happening.  it seems like even if you have the leads hooked up to the plates, it should give pretty steady values assuming good contact with the foam on both sides.  If the contact between the foam and plates isn't very good that might cause a little jumping when it's just sitting there, but I expect it would steady out if pressed and be proportional to pressure from then on.  Sorry I couldn't be more helpful.

[harristotle]

Hi, thanks for your reply.
I had good contact between croc clips, multimeter and aluminium.
I had very good contact between aluminium and foam.
I also had moderate pressure in the foam at all times due to my cable ties (see attached picture).

I just wonder if there are different grades of conductive foam. It is very strange, don't you think!

[Mark Rehorst]

You'd probably be better off using a metal that doesn't form a nonconductive oxide layer upon exposure to air. That means almost any metal BUT aluminum.

[bikeboy]

In Belgium, anti-static foam is apparently harder to find. I had some that had a silver shine to it and came wrapped around an usb pci card, but sadly it didn't conduct any electricity (how strange is that?).
Then found some active carbon foam on ebay (used in kitchen hoods), which comes in large black sheets. I soldered electrical wires to 2 pieces of 5 eurocents, put some carbon foam between them and sandwiched that between 2 squares of duct tape. Really sturdy and worked fine on my Arduino. Needs quite a bit of pressure (fingerwise), but you can always press on the edges, they're more sensitive there. Use it to modulate music parameters in Pure Data, Fruity Loops, etc... via Arduino (patch available in Pure Data). And when no pressure is applied, it doesn't conduct (although that might not be the case with large (A4 size) surfaces, I've recently discovered).
I'm thinking of making a Really Big Pessure Sensor (2 x 2 meters), hook it up to the Arduino and jump around on it, possibly combined with a homemade graphite paint to make it position sensitive, but that another story. Feel free to steal the idea and help me over the obstacles.

[brendanmccloskey]

I am hoping to build my own large FSR, approximately laptop-sized, to detect both X/Y position and pressure; a fellow PhD student is also hoping to do something similar, but on a much larger scale. Have you had any success with this?

[bikeboy]

I made and an A4 sized XY prototype pad a year ago. But I it didn't give the signal I hoped for. The pad was a sandwich of (from top to bottom) 1) an A4 sheet of aluminum foil connected to +5V 2) A4 sheet of activated carbon foam 3) A4 sheet of paper painted with homemade graphite paint ( graphite or charcoal powder with acrylic paint medium or white wood glue ) and with a strip of aluminum foil on each side (applied when paint still wet). The strips didn't touch each other in corners and were each connected to a grounded resistor.
The reasoning was that this would form 4 variable voltage dividers: when pressure was applied to the aluminum foil and foam underneeth, a current would flow from the +5V aluminum foil through a variable lenght of the graphite paint, through the fixed resistor and to ground. The voltage messured over the fixed resistors would vary in function of the position of the pressure point. But messured values didn't make much sense, maybe because the paint was not homogeneous enough? Maybe the values of the fixed resistors were too low and drained the hell out of the voltage source (thus dropping it's voltage).
Anyway, I did something similar with only one axis on a strip of paper that was darkened to saturation with a pencil (took me half an hour of scribbling) and it worked well as a ribbon controller. I'd suggest you start scribbling on a small square of paper and go from there. Keep me updated on [e-mail].

[brendanmccloskey]

Thanks for this detailed and comprehensive reply; I will keep you informed of my progress (this research is leading to an important element of my PhD in creative interaction design)

[bikeboy]

Off course it didn't make sense (and I had discovered this before but forgot it all together): Let's say that we'd move our “pressure point” in a perpendicular line towards one of the aluminum side strips, expecting to see an increase in the voltage over the resistor connected to that strip (because the length of "resistant paint" in front of the strip would gradually decrease and so would it’s resistance). This would only be true if it weren’t for the two strips parallel to our line of motion (remember the four sides have strips). They conduct almost all of the electrons “emitted” by our pressure point towards our measured resistor, with a their negligible resistance which is not really changing either in function of our movement (1 cm length of alu strip will conduct pretty much the same as 10 cm) , so our length of resistor paint will be of almost no influence on the total resistance, as resistance will always be very low due to these strips. So the voltage measured over the fixed resistor that is connected to the perpendicular strip will stay almost constant ( = almost source voltage).
What could work is the same setup with four electrodes in the four corners of our resistant paint sheet, attached to their respective resistor. Shame on me...
Some has got time to test this?

[8bit]

Thanks for the tip! 

[docboggle]

By the way, if you want to order a small amount of Conductive antistatic foam material for very low cost, go here

http://www.elexp.com/ant_2asf.htm

at $0.90 per 4" x 5" piece, how can you go wrong?

[techiebot]

Seems like you could make drum pads.  But I think using piezo elements is a better way.  Look up todbot and his tutorial on "Spooky Arduino".  There are also other sources using a microprocessor and piezo element for a drum-like instrument.

[Sam Grove]

I suggest placing the foam between two squares of thin pcb to which the wires are soldered.

[harristotle]

BTW, it is moderately linear - see attached


As you suggest, the errors decline with force.

[Foxtrot70]

Excellent work.  I like the fact your not stuck in overly technical terms,  the use of the everyday word "squished" helps to explain on a broad level.  Keep up the good work!

[ProfessorJWN]

Yeah, unless you have to talk to engineers, torsional fatigue, compression,  stress concentrators, etc. all are a pain to non engineering "folk".

Bend, crunch, snap, stretch, much better.

"Hulk Smash!!"  not "Hulk Longitudinally compress!"

 "Hulk will now attempt to exceed durability of the structural envelope.!"

HAH

Great instructable !!!

[piddle01]

Haha I want to major in mechanical engineering and maybe minor in electrical engineering.

[Fred82664]

cool they could be used in a lot of robotic crickets  

[twojima]

so with these, I could make my own drum pads for something like rockband or guitar hero, or am I way off?

[lampajoo]

check out the next instructable in the series: DIY landmines

[shalow]

Do your homework next time, they are already on instructables ;)

[Thav]

Similar instructable here
http://www.instructables.com/id/DIY-Force-Sensitive-Resistor-FSR/

Neat to see you don't need the copper board though. Good idea sealing it.

[frollard]

Wonderful build with professional looking results!  Excellent!

[troseph]

I'm curious just how this works.

[frollard]

if I recall the weatherstrip is very slightly conductive - and pressing harder makes a) better contact between the copper wire and the foam, and b) better contact between individual cells of the foam, lowering its internal resistance.

[thecheatscalc]

Pretty good idea!
I'd bet it'd be even easier if you put two plates on either side to squash it. (ie foam sandwiched by two copper plates, with wires soldered to it. )

[pirateaboard]

Great info- I had to do a quick demo because I didn't realize it'd be so easy!

[lukaj2003]

Nice. I made a similar thing when making a Nandhopper (another 'ible) :P 

[shoppe]

Me too.  Very nice.

[omnibot]

Nice, I like.