I have a plan

... and a prototype. This is a brave, new prototype for my dancepad sensors, using an off-the-shelf weight/force sensor from a cheap digital scale in a Wheatstone bridge, driving up the voltage differential with a op-amp and then using the other half of the dual op-amp to make it more digital with a Schmitt-trigger.

The protoboard probably has a few extra components that aren't really necessary, the opamp should have enough power to drive two transistors, but in this version the op-amp drives an LM311 comparator, that has an open-collector open-emitter output. Since the differential amp + Schmitt-trigger are connected inverted, this arrangement suits perfectly, as the comparator collector can be inverted using a pull-up resistor.

Now "all" that's left to do is finalize the calculations for gain-resistors in the diff-amp, draw a pcb, order a few parts and duplicate this circut 18 (or 16) times = 2 pads, 9 or 8 contacts each. I haven't really decided on the center button yet. It would be nice, but it's not really that necessary in the near future. The good thing is, if I make the pad-structure properly I could add the 9th sensor pretty easily.

Oh, one thing I forgot in the above list: build the pads. That means lots of plywood, several smaller chunks of wood, polycarbonate and a wide variety of other strange pieces of hardware.

I also have a cunning plan for the player menu controls for the pad, as I don't like the foot-operated switches so many DIY dancepad projects use. And since I'm using force-sensors here, you can be assured that the other parts are going to utilize something weird too.