I was thinking of a sensor grid touchscreen type device, but I just can't figure out how to interpret the sensor input into x y coordinated that a computer could read. This sensor grid consists of two perpendicular arms each with 50 photocells and a line laser shining on each arm array. I am trying to find out what grid point is being interrupted by determining which photocells have increased resistance on the x and y arm arrays. I may need to have a separate PCB for each array. Is there any way to determine which sensor is in the center of the group if for example sensors OOO were triggered, how would I get it to only send the grid coordinate of the centre O? could I use an Arduino for this or should I have a custom PCB made with a different microcontroller?
Suppose one were to perform an A-D conversion of the voltage on each photocell. How often must the position be updated? Let's guess 20 times per second. That allows 50 msec to scan 100 photocells and do the analog to digital conversions, giving 500 usec to complete each individual cell scan selection and conversion. That seems to be within the capabilities of an Arduino. At the completion of a full scan, apply some interpolation logic to the values just stored in the array (you will need to program said logic). It may even be possible to achieve higher resolution; for instance, if 2 adjacent cells OO have the same analog value, then the position may be declared as between the 2 cells O*O.
Many years ago the HP-150 Touchscreen computer used an array of LEDs and photodiodes to determine your finger position on the screen. The technique involved looking at which beam or pair of beams were interrupted. Using this they could identify the position of your finger to the resolution of one (of 27) lines vertically and 2 (of 80) characters across it.
The user of the computer needed to very carefully monitor the CPUs steam pressure -- this was a long time ago!
Similar threads
