I’ve had an interesting time working with the adhesive copper foil tape that I was so hopeful about recently. The very, very short version is that I don’t think it will work.
After the foil tape arrived in the mail, I eagerly opened the package and examined my new prize. It was just about what I was expecting, although it did have some oxidation stains on the outermost layer (I mentioned this in my previous post). This was pretty easy to remove by cutting off the first 12 or so inches of tape, which didn’t bother me since I have 36 yards of the stuff.
Next, I cut the 34 sensor pieces as outlined in my last post. I think most of my estimated sizes are fine, but the thumb pads ought to be a little bit bigger to accommodate the touch placement of the thumb in various other areas on the hand. All in all, this was a pretty simple process, and I wasn’t worried about making mistakes since the foil tape is so cheap. As a rule, I’m trying to balance the size of each sensor pad with its usability, making it as small as possible while still making it easy to find in a quick finger-tap motion. If the sensors are too small, each tap will require a lot more precision, which I’m trying to avoid. On the other hand, I don’t want to make the sensors so big that they run into each other accidentally. Obviously there will need to be some provision for this in the software, but I’d like to minimize it.
After I got the sensors cut, I cut sensor lead wires from the 26-gauge black solid wire I got a while ago. The finished product shouldn’t need any wire to be more than about 8 inches long, but for testing, I made each one about 16 so I could have a little bit of play. Once the board can be mounted on the glove, most of the wires will be much shorter than 8 inches. I still need to come up with a digram for wire guide placement on the back of the glove. I’m also looking into something like this 34-conductor round cable for conveniently extending the sensor leads. It shouldn’t be necessary on a finished glove, but for a prototype, it might be nice.
Once I had the wires cut, I stripped about 1/4 inch off one end of each of them (as shown in the first picture above) and started the soldering process. I discovered very quickly that the fastest way to do this is to melt a solder bead onto the corner of each sensor first without the wires, and then go back and re-melt it along with the wire inside. This was more efficient than soldering each wire all at once for each pad—mainly, I think, because the pads are so small that they stick to the solder and/or the wire independently during the solder process, and therefore it is very easy to get a crooked wire attachment. After I starting using the 2-step approach, just about everything turned out perfectly.
Next, I used the adhesive backing present on the foil tape to pre-position each sensor on the glove while I was wearing it on my right hand. This proved to be slightly difficult since I am right-handed, and because I had the glove on that hand, I couldn’t really use it for very much of the work. I got all of the sensors on except for some of the 1/4 inch side finger segments, and then decided to try permanently gluing them on (since the tape adhesive is terribly inadequate for a permanent hold on cotton).
This is where the trouble started. First, I have 5-minute epoxy, and I mixed together nearly enough to do the whole glove. This was my own fault, of course, since I wasn’t thinking about the fact that I probably wouldn’t be able to apply 34 sensors very well within five minutes. But that’s beside the point.
I removed the first (little finger tip) sensor from the glove and plopped it firmly in the mixed epoxy puddle to thoroughly coat the back side, and then firmly stuck it back where I thought it should go. That went relatively well, except for a small amount of epoxy escaping out from under the bottom of the sensor. I’d used slightly more than I needed. No big deal, I thought.
Then, I moved on to the second one, the sensor directly below the previous one. I repeated the same glue-sticking process, and then discovered when I tried to place it on the glove that the first sensor was coming off because of the weight of the wire attached to it. Of course, this necessitated my holding both sensors on with two fingers, at least temporarily, while the glue got a little more sticky. This also resulted in more epoxy leakage, some of which got on the glove, and some on my fingers. Ew.
By this point, I was doubting the suitability of my approach. I figured I would try to do the third main sensor for the little finger, then let it dry and see about continuing later. Of course, that’s when I discovered that the entire remainder of my 5-minute epoxy puddle had hardened beyond usability. Oops.
So, took the glove off and thoroughly cleaned my gluing hand while I let the glove dry. A few hours afterward, observing the glove in its two-sensor dried epoxy glory, I concluded that this couldn’t possibly work the way I intended.
Here are the reasons why I don’t think it is feasible:
ul.spaced li { margin-bottom: 1em; }
- Gluing copper foil to the glove with epoxy is messy and imprecise. It is also impossible to undo. If you’ve ever tried getting cured epoxy out of cotton, you’ll know what I’m talking about. You would need to have a tremendous amount of time, or else very precise automated equipment, to use epoxy to glue 34 sensors onto a cotton glove. If you don’t have exactly the right amount of glue applied to the entire surface of the sensor, some will not stick, or some will escape from behind and get on other parts of the glove.
- Cured epoxy is not flexible at all. One of the reasons I liked the idea of copper foil is that foil is flexible. I forgot that the epoxy I’d have to use to attach it would utterly negate that property.
- Small bits of epoxy that escape onto non-sensor areas of the glove make the surface feel very rough. This isn’t the end of the world, but it doesn’t feel like a finished product. It’s really unprofessional. I want this glove to look and feel awesome.
- Foil only looks nice until it gets bent a few times. As I said in the point above, I want this glove to look great. As you can see in the final picture in this post, wrinkly, bent copper foil looks terrible. The effect wouldn’t be quite as noticeable if the foil was all glued to the glove (most of it isn’t in the pictures), but it would still be there.
Some of this is about looks, and some of it is about usability. I believe it is probably possible to use copper foil and epoxy to get the job done if you are extremely precise and good with epoxy (and perhaps use 4-hour epoxy and not the 5-minute variety). However, it’s still going to be a whole lot of work. Therefore, I’ve been looking into alternatives.
In light of what’s happened with the copper foil, I’ve been looking into conductive fabric. Honestly, this is what I should have been doing from the beginning, but I guess I’m too used to thinking of electricity as going with metal, wires, and solder. I found a company called LessEMF which sells all kinds of electrically conductive fabric. I’m particularly interested in what they have labeled Stretch Conductive Fabric, and secondarily the Pure Copper Polyester Taffeta Fabric. They sell a sampler package for $10 that contains 2″x3″ swatches of all of their materials, which I have ordered. Once it shows up, I’ll see which performs the best.
