Saturday, 29 June 2013

Making a digital board game from paper and copper tape

Earlier tests proved successful, but actually making our board game modules is still a bit fiddly (and, let's not pretend, about a quid - ok, 96p - more expensive than it needs to be!). So this time, we tried a design whereby the base of the board would be slightly magnetic, and the playing pieces contained steel/iron pieces. This allowed us to use a much thinner "membrane" between the two layers of copper strips.

We printed our board game layout and put some strips of copper tape along a sheet of paper, following the design, spaced 25mm apart (centre-to-centre). The copper tape crinkles and creases very easily, and even when pulled tight before sticking it down, there are loads of little bumps and creases in the tape. This is no good for us, since the gap between the two copper layers is going to be only a sheet of paper thick, so we need the copper traces as flat as possible.
Luckily an old trick I was taught many, many years ago (while still at school and working weekends in a stained glass craft unit) came in handy: simply rub the tape, quite firmly, with a firm-but-flexible tool. Back in the day, I used to use a wooden peg - it is hard enough to rub the copper tape flat, but not so harsh (like the edge of a steel ruler, for example) as to damage the tape. Our scissors have a "comfort-grip" handle, and they worked perfectly!

(note how the top strip is much flatter and shinier than the strips lower down, which have yet to be rubbed flat)

After this, we loaded our board design into CraftRobo and cut out the circles from a sheet of sticker paper (in the final version, we'll use double-sided sticky sheets, but for now this was all we had to hand).

(ours is one of the very early CraftRobo machines, so the cutting force is pretty poor, compared to later machines. However, it's fine for carving through a layer of sticker paper)

Once the sticker paper was cut, we "weeded out" all of the little circles using a craft knife.

We're still not sure if this was a good idea or not - it made the final sticker quite difficult to handle, but the alternative would be to leave the circles in the sticker (though some may already fall out as the sticker is peeled off the backing sheet) but then they would need weeding off the sheet with our copper tracks, which may be just as cumbersome.

After placing the sticker sheet over the paper with our copper tracks on, we cut and placed little strips of double-sided tape in-between each of the circular cut-outs. In future versions, this should already be in place before cutting out the circles.

The top sheet, with thinner copper tracks at 90 degrees was stuck down, on top of this. It is important while performing this task, that both layers are pulled taught, so that no wrinkles get caught in either of the two layers (which would create a permanent short between two layers, where they buckled together).

Each "square" on the board was then tested using a multimeter set to continuity testing.

First checking that a button press was recorded, and then that it returned to "normally-open" when released:

So far so good. But we did notice one problem - we're effectively working with three sheets of paper, stuck together in parts. During testing we didn't stick the bottom layer down to anything, and if part of the page was allowed to curl (creating a wrinkle somewhere in the paper) then one button, right in the middle of the board, became unreliable. When released, the continuity tested recorded a partial signal:

This is only to be expected, and we can explain why it's happening - but it just goes to show that we need to take care during the construction of each board section; every part needs to be as flat as possible and there's no room for error when sticking the middle membrane down. Even the smallest wrinkle could cause parts of the board to stop functioning properly, so we need to be extra careful.

But, errors aside (we don't mind errors so much if they are repeatable, expected and can be explained) it was a pretty good test for our board game idea. And much quicker (as well as cheaper) than making multiple PCBs, soldering SMT shift registers, messing about with pin headers and ribbon cables and all that! Our last test, before we decide for sure if this is the way forward, will be to create a top layer from photo-quality inkjet paper (so the playing surface can be printed on the top side) and call in a few favours from Nick at CustomStuff to see if we can get it laminated too.

We expect a laminated playing surface to be slightly less flexible than just a sheet of paper, but this in turn might be useful as it will help ensure that the top layer is flat as it is affixed to the rest of the board.