Fabrics with electrical functionality have been around for several years, but are very rarely used in mainstream clothing. The fabrics are very expensive and the supply can be unreliable. Frustrated by this, [Counter Chemists] developed PolySense, simple open-source technology to make any fibrous material into a conductive material that can be used to sense pressure, stretch, capacitive touch, humidity, or temperature.
PolySense uses a process called in-situ polymerization, effectively dying a fabric to become piezoelectric. This is done by first soaking the fabric in a mixture of water and the organic compound pyrrole, and then adding iron chloride to trigger a reaction. The polymerization process that takes place wraps the individual fibers of the fabric in conductive polymer chains.
Instead of just uniformly coating a fabric, various masking techniques can be used to dye patterns onto the fabric for various use cases. The video after the break shows a range of these applications, including using polymerized gloves and leggings for motion capture, a zipper that acts like a linear potentiometer, and touch-sensitive fabric. The project page lists sources for the required chemicals in both Europe and the US, and we look forward to seeing what other applications the community can come up with.
The project is very well documented, with a number of scientific papers covering all the details. [Counter Chemists] will also be presenting PolySense at the 2020 Virtual Maker Faire.
This technology can also be used to make a fabric piano with a lot less effort. On the more mechanical side of things, you can also 3D print on pre-stretched fabric to make it pop into 3D shapes.
21 thoughts on “Dyeing Fabric To Create Sensors”
VR body glove incoming ?
It would be great to see it indeed!
Certainly looks plausible and somewhat practical with this method. Got to wonder how enduring the coatings really are as there’s alot of mechanical and chemical threats clothes have to deal with. Not read the papers yet, but certainly is interesting and even if the effective duration isn’t that high or you need to do alot of on the fly calibration as it ages its got merit.
Very good point. We conducted a few tests* and all the problems can be avoided or stabilized:
– This polymerization works well on most natural fibers but not as well on synthetic (in both cases it needs to be well washed).
– UV light degrades the performances but just for a couple of hours (then it has no more impact)
In summary, once the functionalized material is well washed and UV treated, it should be stable.
*see Limitations section page 10: https://counterchemists.github.io/files/PolySense.pdf)
Sounds good, I assume the UV will still be degrading performance just vastly into diminishing returns with all the most exposed areas getting degraded fast. But nothing lasts forever if it doesn’t fail in the lifetime of the underlying fabric that is a soild win.
I don’t see any mention of how it deals with sweat particularly though humidity is mentioned (still not read all the documents properly). Has this be studied yet? I would think for garment use that would be the biggest problem.
Haha, we noticed the humidity effect during a presentation at CCC camp: it was vert hot and sweat introduced a drift in resistance measures.
…but we should characterize it, that’s a good point.
How could it be objective and easy to replicate though?
Should we fold a sample in a glass vertically, measure its resistance, and add water progressively? (a bit like a soil moisture sensor)
It seems that there should be a better way…
As sweat is not just water I think you would have to use actual sweat to verify the water experiment is at least a good indicator.
Could perhaps just squash at a known force some of the doped material between glass? sheets and dip an end in water – capillary action should distribute the water you add that way.
Atmoised water in a seal box from one of those funny ultrasound gizmos might do work well – never played with one but they seem to the job.
Or adding steam – a hot plate with x mm of water dropped on it and resulting steam fed into the known volume container (or if you have a big enough container a hot plate inside with a burette to feed known water quantity) should give a good repeatable moisture content i would think? Would need some good temperature controls to be able to get exactly the same result in any room (colder the room more the water will condense on the side walls) but done right it should be a good characterization I’d think.
Many things to try!
I guess we would have to look in the literature to see what humidity sensor researchers use for their characterizations…
Thanks for your ideas, ping us if you want to keep chatting further:
Yeah, I’ll just wait to rent one…
*Cue, hosing out the bodysuit scene on “Upload” series on prime video*
Funny one indeed!
We need a bit more work to reach their level of accuracy but we tried heating the polymerized materials and it’s pretty good:
They did it!
They reverse engineered eeonyx!
This is mad!
But how expensive is it?
If I remember well, the cost of the chemicals would be about $10-$20 per square meter (it depends on the material that you’re using).
The smallest quantities of the pyrrole and the iron chloride you can order in Europe will cost you ~ 24 €. With this you can polymerize 3+ square meter of cotton.
That’s very promising! Thanks!
PS: look for the “Chemical products sourcing” section:
…and feel free to drop a comment if you have any better solution!
Example: some already look for DIY pyrrole synthesis:
How does it’s conductivity compare to,,,say,,,copper? I’m thinking about a stretch-tune-able rf antenna,,,
Ah, copper being the best conductive material (for its cost), it’s difficult to compare, but you could approach the problem from the other side:
There are conductive textiles that are pretty good and you can etc them as you would for a PCB. Our great Hannah made an article about this:
Typo fix: you can *etch* eTextiles like PCBs
That’s pretty cute. But I was thinking more along the lines of a half-wave dipole that I could stretch from…say…10 meters to 40 meters…no antenna tuner…constant feed-point impedance…
If use a zig zag approach it could work on a stretchable fabric too:
Thanks for covering this :-)
Quick note: as cool as it would be, the resulting fabric is not piezoelectric. It is piezoresistive though :-)
(Possible that auto-correct swapped those words, first line of second paragraph).
Please be kind and respectful to help make the comments section excellent. (Comment Policy)