2020-10-01

Blinky gloves

I’m experimenting with wearable computing as a way to get back into programming. It’s fun because it involves more than just pixels on a screen; you get to mess with real-world stuff like LEDs and circuits and sensors and whatnot. And it’s low-risk because it’s hard to shock yourself badly or make stuff blow up when you’re working with 3 to 5 volts.

Eventually I’m planning to make a turn-signal jacket for biking, much like this one, but I wanted to get a sense of how to work with the various necessary bits — sewable processors, LEDs, conductive thread and so on — before adding programming to the mix. I thought I’d start by just adding some uncomplicated LEDs to the back of some gloves so that when I stick out my arms to signal a turn there’s a little bit of extra light there to make it yet more obvious what I am doing.

I was going to just sew some LEDs to my gloves, using the conductive thread as a resistor and connecting them directly to a battery holder for power (so the LEDs would be on whenever the battery was in). This is a good enough plan, but then I found tiny boards pre-programmed to light up 4 LEDs in random blink/fade patterns and I liked the idea of that better. So I bought two Lilypad twinkle boards and a bunch of sewable LEDs (normal LEDs are also sewable of course, but not so much washable which for gloves is definitely a factor) and some stainless-steel thread. I wanted LEDs in a colour other than red or white (which have legal meanings for bike lights) but only red and white were available. I used red, then, so they look like taillights and not headlights, which I thought might be confusing for people behind me.

Here’s a picture of the circuit concept. There are 4 LEDs, each with their positive end connected to a petal on the board (the white and green clips) and their negative end connected to a wire connected to the negative port on the board (the yellow and black clips). The wire then connects to the negative end of a battery. The red clip connects the positive port on the board to the positive end of a battery (not shown). When the circuit is closed, the LEDs blink and fade in random patterns.
Circuit proof of concept

Since the proof-of-concept worked I started sewing, connecting the LEDs each to one of the four ports on the board by sewing a very short seam with stainless-steel thread. It’s terrible stuff to work with; very high friction, very sharp, very prone to tossing off tiny thread fragments that short out your circuits, very hard to seam-rip. After my first experiments with it I started to wrap several of my fingers in first-aid tape to protect them. You can use stainless-steel thread in a sewing machine as a bobbin thread, which is probably a less personally damaging idea but not one that would really have worked with gloves. I tested each seam after I finished it to be sure it worked.

Testing each sewn connection

After all four LEDs were attached to the glove and the board, I ran a long seam from the negative port on the board to each of the negative ends of the LEDs, testing each one as I went.

Testing the thread connections on negative board point

Once that was all working it was time to attach the battery and sew the seam from the positive and negative ports to the battery case. I put the batteries on the inside of the gloves to shield them from the weather somewhat.

Showing the battery case on the inside of the glove

They work! The patterns are mildly mesmerizing so I don’t look at them while I’m riding.

Finished gloves

Once I was satisfied that they were working I covered the exposed stitches with black fabric paint to help protect the thread and keep it from throwing off the aforementioned circuit-shorting fragments.

It’s a good and successful first experiment but there are a couple of obvious improvements I could make:
– Move the batteries farther away from my wrist. I put them close to the wrist elastic on the gloves to keep them protected and to keep them from getting joggled around too much but I think it would be easier to pull the gloves on and off if they were closer to the far edge of the glove.
– Add a switch so I don’t have to pop the battery in and out all the time. This is just a matter of sourcing a suitable switch; they do exist, but not easily through Canadian suppliers. The joys of Canada Customs await.

Fortunately I have helpful cats.

Carson feels a need to help with the electronics