Spatial Chess

Posted on Nov 17, 2013 in Physical Computing
Spatial Chess


information to be added locally

for the moment, refer back to Saki’s original post:


information to be added locally

for the moment, refer back to the larger system sketch (rough, rough!):

And for a more extensive photo array of tests see:


(pardon, will rotate in a few)

So above is a sketch describing the chess board ‘base’ connections and the ‘top’ piece copper contacts. We’ve moved from autonomous pieces to a spatial registration. As the resistors above suggest, each corner of the board has a different resistance to its contact. So, when the upper piece completes the circuit, we can just grab a single analog reading (resistor contacts in parallel) and be able to tell how the architectural block is oriented. (Testing success!)


The non-successful contacts shown above– the rings– seem to be a little small and hard to align. The idea was to have the circuit feed into one ring, cross (in the top piece) to another ring, and thus complete the circuit for: ring a and c- the unique resistance that identifies all the separate architectural chess pieces, ring b and d- the LED light (pwm) that would dimly lit for any connection and be bright/pulsing when you replicate a saved sequence (all via code which Saki has brought together). We did get a resistance reading, but it began obvious (w/ 500, then 900 as the Serial, print() reading, that we we getting just the resistance circuit (900) sometimes, but when the alignment was off, we were accidentally closing the LED circuit (with it’s built in 220 ohm and thus getting the 500 reading).


As you can see from above (things that will be built into the chess pieces), thus we have three indicators/circuits per piece:

  1. the corner orientation connection (blue wire) which is simple circuit completion w/ 4 R options: heads into A0
  2. the unique resistance identifier…. new resistance value for each chess piece: heads into A1
  3. the internal glowing light…. visual confirmation, highly programmed: pwm in from D11 or equivalent

The larger chess board would have on/off, perhaps an indicator light to that effect and a momentary light to trigger the saving of block positions on the board (w/ internal LED).

We’re still toying with whether that will be enough feedback, but I think there’s something nice/awesome about a) Saki’s tactile use of wood, b) the way we can cut into the base and create much better copper contact surfaces, c) and the idea of getting veneer so that the pieces glow internally when on the board. (I’ve spent some time w/ it… so there are veneer glue issue to be worked thru…)

But think, this would make a lovely glowing architectural game, slightly less of a warm glow….


This weeks tests and prototyping was really the first round of tinkering. (Like, have we mentioned that people aren’t allowed to drink or sweat over the game? Might short-circuit those copper contacts. đŸ˜‰


[yep, snowflake contract templates!]


[super simple wiring, two analog outs, one pwm in and a bunch of pull-down ground contacts]


We did some play with the magnetic switches…. they’re been stored with their magnets and are thus themselves slightly magnetized. There’s a code way around that, but even if we create separate battery/LED/switch circuits in the top, we wouldn’t really be able to fully program them, since we’ve moved toward the chess-board registration strategy.

There were some successes:

As Saki’s video, below indicates, just using 100, 330, 1k and 1.5k resistors we got good corner readings, and good copper contacts.

As you can see from the second value… we seem to jump between the 500 and 900 on the ID, so we’re going to move the light contacts (which weren’t working) and double the unique ID contact to see if it’s a spatial issue or something else.

I’ll not make you watch all the magnetic tests, but here’s the interesting voice over-version:



Granted we need to get a complete working prototype this week… so along w/ widening the ID contacts, we’re going to try placing the light power contacts on a vertical lip… w/ ground contacts on the base, between the corners (as shown in the really messy sketch below). This way the edges that  keep each piece centered will also enable the light to work. 2 birds, 1 stone.



So after some sketching and discussing immediate prototype work…. we came up with a rough (rushed) schedule and a simple bill of materials.

Saki’s google calendar, below:


And our initial project estimate…. for a 2×2 (i.e. 4 square) grid, with three unique architectural chess pieces:


(Meg & Saki)