This instructable documents my first attempt to 3D print conductive snaps on to fabric. I wanted to 3D print a female snap which would connect to a regular metal male snap.
The file was modelled in Fusion360 and printed on a Makerbot Rep2 and a Dremel using Black Magic 3D conductive graphene PLA.
The snaps are YKK 'Snapet' open prong snaps (size 12L) and are 7.5mm across. They are often used by eTextiles practicioners as they are the smallest available. You can buy them in different sizes from different suppliers, but they seem to be a standardised design. Buy size 12 here.
My aim is to explore ways to make a garment which is conductive and stretchy and preferably uses no hard metal parts. Making connectors which are compatible with existing purchasable snaps will make it easier to test and iterate.
This test worked surprisingly well and the file is worth printing, but it definitely needs a lot more tweaking. For now it can be printed and tested as-is, but is definitely a proof of concept rather than a fully functioning snap which can be printed reliably e.g. the PLA tends to shrink, and the snaps have a limited lifespan.
If you print this file please leave a comment and tell me your findings!!
These explorations are part of a larger project called Second Skin, a prototyping suit for eTextiles. I will upload all the files, patterns and documentation as they are completed. You can follow the project on here, or via my website: http://www.rachelfreire.com/second-skin-login
You should also check out Rewear by Lara Grant. She has been working on a modular system for wearables based around a breadboard of snaps 3D printed on to fabric. She also has a focus on the sustainability of these techniques which is something I also see as integral to their future development. We will be integrating our experiments into a dedicated etextile connectors website soon, so make sure you check out Lara's website and Instructables too!
Step 1: Fusion360 Files
The file was modeled quite quickly using Fusion360.
I took as many measurements as possible from an existing snap and made a rough design. Because the snap is so small, some of the inner proportions were made using guesswork and therefore will need more playing around.
Link to download current version here: https://myhub.autodesk360.com
The file attached to this Instructable was my first attempt. It worked quite well. The linked file (above) was tweaked, making the base of the snap more solid. the idea being it would help it stick to the fabric better. While this did help somewhat, both files are still worth testing if you want to print a version of this. I had success and failures with both.
If you want to use larger snaps (such as 15mm which are more common) I would expect this file can be resized and printed at the correct dimensions and will also snap to larger versions of this snap design. I haven't tried this yet as I am trying to make everything as small as possible.
These metal snapets are amazing, but often the die are hard to find. I use Prym vario pliers for applying snaps by hand and its hard to find the snap die to fit. So I made a printable fusion file for the 12L Snapets ;) Again, its not perfect as 3D prints tend to shrink and warp and eventually break. But I've just been printing new ones when this happens! Files are attached for the inside (connector part) and outside (ring attachment) die. One is a fraction larger than the other. If you use them the wrong way round, the snap will stick in the die.
Step 2: Print Test and Setup
This first snap was printed by Lara Grant. She is working on a similar project making a snap fabric and has a great Instructable about 3D printing on to fabric. You should also check out her wearables classes
It is Black Magic 3D graphene filament and was printed on a Makerbot Rep 2 with the print temp and extruder set to 220°
We have both been testing a technique whereby you print a base layer of filament, pause the machine to insert fabric then continue printing. This means the filament will melt around the fabric and create a seal. You can see this in the second image; there is filament on the underneath of the fabric. This layer was printed on to the bed first, then the printer was paused and the fabric was inserted. The printer was then unpaused and the print continued.
It worked amazingly! First attempt using the file I had made 10 mins earlier.. And it even snapped really snugly!
This snap you see here was printed on to powernet. It is a material I use a lot and am using for the related project Second Skin which uses stretch circuits. It is 4 way stretch and used for lingerie and dancewear. It works well because it is a fine synthetic mesh. It is usually made from polyamide so the filament melts the surface and adheres well to it. The filament can also melt into and around the surface of the micro-fine mesh itself.
Powernet has good tensile strength and if it is tensioned with tape when you lay it on the bed it doesn't get snagged by the extruder.
Step 3: 3D Printing on to a Conductive Fabric Trace
This brilliant fabric is a stretch jersey knitted with conductive traces. I believe it is the sorcery of Hannah Perner-Wilson and Mika Satomi of Kobakant and was custom made. I was given some at eTextiles summercamp and we decided this would be a great thing for testing the electrical connection between fabric and print.
It is jersey fabric, and it looks like the fibres were coated before they were woven, rather than the conductive coating being printed on after manufacture. It is too thick to print through the surface (as in the earlier test) as the filament would not connect in the same way as it does through the holes in the powernet.
We set the Makerbot to print straight on top of the fabric. what you see above is the first test print on this material.
Lots different people have been testing 3D print on to fabric, and it seems to differ according to the nature of the print, the materials and machines used. Most successes seem to involve meshes as the weave is loose and the filament can sink through the fabric to create a bond.
Some people lower the nozzle of the printer. This smashes the extruder into the fabric and forces filament into the fibres but can drag the material. Another option is to raise the extruder starting point up to start the print, meaning you slice the print in such a way that it starts just above the thickness of the fabric. I imagine this would work well if your fabric was thick. As ours are quite thin and flat, we printed straight onto the material with the default settings, just turning off the rafts and any support.
It worked beautifully! This may have been due to various factors:
- the surface of this particular fabric was ideal for the print to adhere
- the extruder just happened to be at the perfect temperature at this point in time (this filament can be very inconsistent)
- the gods of 3D printing were in a good mood and we got super lucky
Clearly, this needs more testing.
Step 4: Testing Conductivity
This test used a stretch eTextile connector made with Karl Grimm conductive thread. Inside the black connector there is a zigzag of conductive thread isolated by fabric layers of either side. Each end has a male snapet. All of these materials have quite low resistance.
The resistance across the 30cm connector, through the snap and across about 8cm of conductive fabric appears to be about 10 ohms. This was surprising and seemed to stay pretty stable even when stretched. I'm not sure this is an accurate and repeatable reading!
Step 5: Printing on Different Surfaces
Next I decided to try printing on a Dremel. This is mainly because the Makerbot was having a hissy fit, but variety is always good. Again, the print temp and extruder were set to 220°
I have been working on bonded, insulated, stretchy conductive traces for eTextiles. These fabrics use bonded textile layers with Bemis SewFree fusing, a super fine heat bonding film. This means the fabric samples were thicker than the previous tests. The conductive traces are isolated inside powernet fabric and have only the ends exposed as round pads.
When I first printed the file with default settings it smashed into the surface of the fabric and distorted the print. You can see the result in the first image. The snap didn't work this time.
Jonathon showed me how to slice the file in Cura and raise the start position of the extruder by 0.4mm.
For the next test I also added a layer of SewFree bonding to the surface I was going to print on. This was to see if this made any difference to how the print adhered.
It initially worked really well, as you can see in the last image. Unfortunately after a couple of snaps, the snap loosened from the fabric and fell off.
Step 6: Multiple Print Test
I next tried printing multiple snaps to see how current flowed through two snaps at either end of a conductive trace. As I had only one working snap on the previous test, I couldn't check. Maybe the print Lara had made previously was a fluke.. I made a quick panel to try multiple prints.
As this was a test, I decided I would print each snap individually, rather than trying to print multiple snaps on one piece of fabric.
1. I didn't want to invest time in making a layout file as the fabric circuit I was printing on to was made imprecisely
2. The prints often fail
3. I didn't want rogue filament dragging across the fabric
I lined each snap up to a centralised point and printed them one by one. Each one came out perfectly.
I added SewFree fusing to some of the conductive pads. You can see this in the images as white circles and semi circles. This is the paper backing which gets peeled away. I left it on so it is easier to see in the images. I thought it would be good to see how the fusing affected the adherence across the same print. They all turned out pretty similar. Most stuck, and a few fell off. Not sure why, but I assume it is due to minute differences in the thickness of the fabric layering. They were all printed in quick succession on the same printer with the same settings.
the resistance across a 15cm conductive trace through two resistive snaps was around 50 ohms. This was done immediately after printing and seemed super conductive, so we needed more tests..
Step 7: Reading Resistance
The readings I took from the snaps seemed to differ a lot. This also changed over time.
Step 8: Re-attaching Snaps With Conductive Epoxy
Some of the snaps fell off after a little use. They didn't adhere so well to the tight bonded material as with the earlier tests.
At this point, it's worth investigating another option: can the snaps be printed and then stuck to the fabric afterwards.
It may be true that snaps can be printed on to certain fabrics but need to be glued to others. This could still be a workable option.
I used conductive epoxy and glued two of the snaps back in place to see if the glue can make a bond and conduct reliably.
Unfortunately this didn't adhere well to the fabric at all. The epoxy is quite chalky and does not like the dense synthetic material. Though the glue allowed a small amount of current to flow, the snaps fell off after one snap.
Step 9: Conclusion and Next Steps
This snap design worked really well for a first test. It snaps securely, can conduct a small amount of current and is a good proof of concept.
Unfortunately they didn't give consistent conductivity. Some were ok and others didn't work at all. It seems using tightly woven fabric is an issue, so this doesn't work so well for my bonded fabrics. Using more open weave like the jersey, and especially the powernet seems the best option. The issue with that is that the less dense a fabric, the worse to conductivity is for etextiles.
There are quite a few practical issues with the PLA. It tends to deform and shrink. Some of the snaps worked immediately, some needed a few initial forced closures before they will comply, seemingly to stretch the print a little. Some seemed too small to snap at all.. It was all a little inconsistent.
I have also been reading that the conductivity of these materials can change over time. In this case I would say that the pressure of the snapping itself may affect this. Also running current through the snap may permanently increase the resistance. This will definitely involve more testing.
there's a pretty good overview of Black Magic 3D filaments here
I want to use this snap idea in a glove design. I want to find a way to make detachable connectors for stretch sensors. The idea would be that this snap file can be integrated directly into a 3D printed sensor to connect it to a circuit.
In review I found this process interesting and informative. It is not stable enough to produce consistent measureable results and I would like to explore further in more controlled experiments.
If you try any of these prints, please leave a comment!