Cheap paper accelerometers? Put us down for a dozen to start. They’re not quite ready for mass production yet but it looks like they’re on the way.
[George Whitesides] led a team to develop the new technology that uses simple manufacturing methods to produce the sensor seen above. Graphite and silver inks were screen printed onto heavy paper. The single limb sticking out from the body of the sensor is a separate piece of paper that bends the carbon area when force is applied. This changes the carbon’s resistance which is measured using a Wheatstone bridge constructed by gluing resistors to the device.
It sounds unsophisticated compared to most of the accelerometer modules we’re used to, but if you need a sensor that detects sudden motion this sounds like the perfect part. Now who wants to be the first person to replicate this in their basement?
20 thoughts on “Paper Accelerometers For Pennies In The Works”
I hope there isn’t a lot of trouble replicating these in a more domestic setting and that it isn’t too tricky.
Because everyone knows ohm is where the art is.
I’ll be here all weekend folks.
Try the fish!
oy. That was horrible, strider! horrible and awesome.
This seems like a flex gauge with one end cantilevered. While that’s a good idea, flex gauges are not pennies per. I wonder how they plan on getting this one cheaper? Simpler design, I guess.
I’ve already tryed replicating that, but I need a multimeter with better precision for the test. I’ll have to wait until monday evening !
Sign me up for some… Thats awesome… Now of we could get some cheap paper GPS and cheap paper compasses to go along with it Id put an xbee on my cat, car, wife and everything else mobile in the house.. Then map out the place with one of those XBOX 3d mapper power drills covered in an earlier article and I’d be set up to sit at my computer and play The Sims in real life.
it only detects acceleration on one axis and no way to tell direction and over time of use it would degrade
i see it used in military applications like in smart missiles or other one time use devices but i would not put it in any of my projects where long term use is the key XD
This looks like the beginning of any number of bad practical jokes (if not terrible puns). The real problem isn’t the circuit but the beam’s substrate, paper is notoriously finicky stuff – anisotropic, non-linear and hygroscopic. Calibrating one of these, much less a whole production run, would be a nightmare.
> i see it used in military applications
you certainly don’t know how seriously militaries take “quality insurance”, specially on things they designed to use only once.
Scott : you can imagine that paper is only a prototype, but imagine the same thing with a plastic sheet. Many application don’t need precise calibration, automatic calibration can be possible (e.g. to detect the fall of somthing/someone, to count heartbeats or people using a stairway…)
This could be very useful in the shipping industry. Imagine having a small sticker that you could scan and it would tell you what forces the package was subjected to. You could even track a fragile package on the ups site and it would update with where the package is and what forces. While as interesting as it wold be to see the data it still wouldn’t protect your package any more.
are there more details on it anywhere?
We used to have the phrase “Good Enough” plus several variants.. as a lead- in for qualifiers of both positive and negative aspect. IIRC the first few generations of some audio crystal devices used Rochelle Salts as a Piezo medium. Despite it being an incredibly hygroscopic material that makes paper seem inert by comparison, it was more than “good enough” for cheap phonograph cartridge transducers and microphones in toys.
The elegance of some things like a Toroid CT often belies how many steps it took to get from prototype number zero to a component used in many cheap PSU designs for that elegance. And from that- how we got from mechanical Gyro systems needing to be “caged” as a routine event, to MEMS sensors- and now these paper bases accel sensors.
Look at Blood Sugar testers having gone from human eye optical reading to the newest *TWO* sensors on each strip designs.
Paper accel sensors could work well with -anywhere a moving,cheap, disposable device needs to sense accel within the DOP these can produce.
Which, is still a damned huge range of open ended questions as to ALL the variables of Cost, Accuracy, Drift, Lifetime Usability overall etc. Those are just to name a few non-trivial data items WrT these sensors that we’ve not even approached determining- yet.
I was dropping questions here and there about sensors in Fab@Home type gear and I had been focused on sensors in the Fabber itself. Not on sensors in stuff we’d be Fabricating! These paper accel things may be a new level of “Disruptive Tech” at a component level.
Part of my thoughts are to Open Source by post as Copyleft/Patent Poisoning every wild hare we get raised by articles like the original design. Though if any of us becomes a Gigazilionare etc from something born of Hackerdom?- hopefully they will repay the Karma somehow TO all of Hackerdom.
Why would you need an accelerometer for a penny?
I work for UPS and trust me you dont want to know what kind of forces and stresses your packages are put under, and I dont think we would make that kind of information available even if we could. Do yourself a favor and package everything like you were going to kick it down the stairs, cause thats a good aproximation of what happens.
The sensor might cost pennies but the electronics to read the millivolt level output signal won’t.
you apparently never worked with a military engineer
“cheap and cheerful” is a phrase that they use quite a bit … if it works and its cheap there all over it … at least US military
remember the JDAM missile GPS head recall?
There are indicators available for packages that will change color if a certain force is applied. I’d imagine they are fairly cheap.
The various mechanical element or membrane/tube “frangible sensing/spill=position sensing” devices are priced accordingly to commodity volume levels for industrial duty. Simply put- a $10 Shock or Tilt stick on of the tube turns non-resettable red,costs $10 due to how few are made. Compared to a whole dollar store pedometer or calculator- or thermometer?
Ok- there’s a Hacking challenge:> Hack those into a shock tattletale! A shock indicator for what “could be” ballpark $1 might be “less hassle” to design on budget with a single pennies Accel sensor like these paper ones.
But,wait- there’s MORE, as shill pitchmen are fond of screaming. There are several “reasons” for electronic and mechanical accel/tilt sensors to both have distinctly different application niches.
EMP is “not yet” a non-military concern outside of stuff that could see energy levels comparable to EMP specs. Like Food Irradiation or Medical Sterility by Radiation to list 2 off the top of my head. Paper sensor/electronics surviving those energy levels? questionable but? it shows that Hacks Va Industrial Design can be different universes indeed.
After the CIA Dragonfly became known, several people in the picolight aircraft groups had the moment of “everything we thought we knew got disproved” almost like what these paper sensors will do when they become an item in catalogs. Or? printable by Fab gear..
midi breath controller?
To replicate this “at home”, try this….
First cut a piece of paper as shown in the photo. Next take a pencil and draw a line from the left side onto the stick and back down onto the right side. Next blacken as much as you can of the left side with the pencil without coming near the right side. Same for the right side. The important area for the measurement is the place where the line goes up onto the “stick” and back.
Next is the problem of attaching wires… How about punching a small hole in the contact area sticking through a small multicore wire, and putting some scotch tape on the other side?
Next, measure the resistance. See if you can see it change if you bend the paper. Hopefully a normal multimeter will suffice. If so the cange is more than 1/1000, and say an arduino can easily measure the results!
The trick to do this accurately is as follows. The device is connected between two pins. Put a capacitor on one of the pins. Take the value to be such that the RC time becomes about 1ms.
Make both pins outputs, make both pins zero (now wait a little while. Say about 10 * 25ohms * your-capacitor-value), make the pin with the capacitor an input and enable its pin change interrupt, make the other pin high and clear a timer. Now about 1ms later, the pin change interrupt routine can register the value of the timer and you have a measure of the acceleration!!!!
To make this three axis is easy too. Make two sticks on the long side of a rectangle. Make one on the short side. Next fold over the short side. Not where the stick starts but more on the rectangle. About 5mm on the rectangle should be enough. If you want, drop a blob of hotglue in the corner to keep the bend at 90 degrees. Next make a 5mm cut into the rectangle between the two sticks on the long side of the rectangle Next bend one of them at 90 degrees as well. Again drop some hotglue in the corner to fix it at 90 degrees. With some careful planning, you need only 4 connections to your arduino. The common pin can be the one with the capacitor, allowing you three axis accellerometry with only 4 pins.
I made a drawing of the template… http://prive.bitwizard.nl/paper_accelerometer.png
You can use a blob of bison-electro to glue the wires to the paper.
are you saying that any strip of conductive material on a flexible substrate will change its resistivity when subjected to stress? Or is there a mechanism between a conductive and non-conductive layers?
I wonder if it is possible to do the same thing on a glass slide with some conductive inc. I will try this also with a metal (aluminum, gold or tantalum) sputtered or evaporated coating on a regular glass slide.
If you would get in contact with me, perhaps we can work on this together.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)