Originally Posted by dknguyen

1. Did you think of your physical circuit implementation to detect something on a resistive sheet? The only way I can come up with is to line the edges of the sheet with contacts and then measure the resistance between every single opposing pair of contacts and then choose opposing X-pair with the lowest resistance and the Y-pair with the lowest resistance and call that the position of the playing piece. A few problems arise with this method. First of all, it is IMPOSSIBLE to detect the position of more than one piece because there is just too much coupling between the outputs and the positions of the pieces. This pretty much puts the nail in the coffin for a gaming board.

I may be able to time the signals to distinguish them using this model. However, I’m not sure if I really understand this model. Let’s say I introduced electricity into a sheet at a location corresponding to the location of the game piece. Now, if the edges of the sheet are lined with contacts – as this model suggests, would the contacts nearer the game piece – specifically the contacts closer to a perpendicular bisector intersecting an axis and a game piece, receive a larger measure electricity than those contacts further away along the same edge? What electrical property of the sheet would have to be changed in order to change the difference in measure of electricity between contacts on the same axis? What would be the reason for this? In order for this difference in measure of electricity between contacts to be practically measurable, would the sheet have to have certain electrical properties? Using this model, would I have to time the circuit connected to the contacts on the x axis on a different interval than the circuit connected to the contacts on the y axis – so that small changes in the amount of electricity received from contacts on one axis is not equally compensated by small changes in the amount of electricity received from contacts on the other axis at different locations where electricity enters the sheet?

An experiment was conducted to determine if a household conducting sheet could be used - instead of a sheet prefabricated to have special electrical properties, to measure distances in terms of resistance. Measurements of resistances associated with one of two probes that contacted a sheet when each of these two probes were located in different locations in relation to each other as well as a third location where electricity traveled to or from the sheet were taken. This was done using a sheet of foil commonly used for cooking, probably made mostly of aluminum or tin. The sheet of thin foil had a width between one and two feet and a length that made the sheet approximately square. A wire connected to one part of an electrical source was wrapped around approximately the middle of one side of the approximately square sheet of foil. Two other open circuits were also involved in the experiment – one consisting only of a wire and the other consisting only of a resistance measuring device. Ends of each of these two other open circuits were connected to another part of the electrical source. The other ends of each of these two other open circuits were used as probes on the sheet of foil. The combined result of two trials was not accurate because the meter being used had an ohm adjust dial that was set to a same arbitrary setting that was different for each of two trials conducted. The experiment was also not accurate because the person conducting the experiment did not know how to read the meter scale. The scale on the meter was between 0 and 1K ohms. The select switch that identified this scale was labeled as X1K. XIK was interpreted as meaning a maximum of 1K ohms or that the scale was measured in units equal to 1/1000 ohms. In other words, X could either mean ‘max’ or ‘times’. During each trial, the types of measurements associated with the largest and smallest distances were taken repeatedly until the person conducting the experiment felt comfortable that the precision of the measurements at larger and smaller distances were about the same. For the types of measurements associated with the largest and smallest distances, the precision of each of - if I remember and calculate correctly, four measurements in the first trial was approximately .5 x 10^-1 units, and the precision of each of two measurements in the second trial was approximately 1 x 10^-1 units. The first trial, if I remember correctly, had an average measurement of about 1.4 units. I’m pretty sure the second trial had an average measurement of about 2.9 units. Though exact values relating distances, the precisions of the measurements at each combination of locations, the differences of measurements and precisions between each combination of locations were not recorded - the average differences in measurements between just about all of the larger and smaller distances, for both trials, were about the same as the precision measured at each location. For these reasons, the experiment was inconclusive, possibly either because the voltage was not to scale, the area of the cooking foil was not to scale, and/or the meter was not precise enough, or the sheet offered no resistance. For both trials, the locations of the two probes were switched – resulting, as suggested, in no estimated change in measurements. For this reason, it may be possible to conclude that this experiment applies to measuring resistances between two probes instead of three. Specifically, in this experiment, the two probes were placed once - per each part of the trials, in all possible combinations of locations 1) near the wire wrapped around approximately the middle of one side of the approximately square sheet of foil, 2) the corner on one side of the sheet of foil nearest this wrapped wire, 3) the middle of the side described in location 2, and 4) the other corner of the side described in location 2 and 3. From this experiment it can only be concluded that resistance in some cases can not measurably be detected using some household sheets of foil with probes – or game pieces.

Correction to the 3rd January 2009 post. 2*4^2 measurements were taken for each run. There were four runs for the first trial, and two runs for the second.

Originally Posted by blueroomelectronics

Did you ever find a solution?

The best thing that I thought of so far involves wiring a keyboard so that switches that used to correspond to keys correspond to areas on a board that can be identified when contacts on the pieces are made with conductors on the board. Unfortunately, I don't have the skills to make the switches sensitive to the weight of the game pieces and therefore momentary. I'm still interested in finding a sheet with the right resistance.