Thor
Newbie level 4
- Joined
- Apr 3, 2005
- Messages
- 7
- Helped
- 0
- Reputation
- 0
- Reaction score
- 0
- Trophy points
- 1,281
- Activity points
- 1,471
Hey,
I'm planning to use a grid of IR diodes to measure the position of one IR emitter. My idea is to wire up the diodes to the A/D converter of a PIC microcontroller, and scan over the diodes in the grid in rapid succession, measuing the exposure of each. I'm not sure if I need to amplify the signals. Maybe feeding a low reference voltage to the A/D is sufficient? The emitter will be quite close to the diodes, around one inch or less. I imagine a 4x3 grid with the diodes one inch apart or less, mounted on a board together with the PIC. I figure that I'll be able to measure great enough of a difference in voltage between the diodes to track the X/Y position of the emitter fairly accurately (I have already scoured data sheets and found diodes + emitters with wide and even emission characteristics). The idea is to use an algorithm to calculate the emitter's 2D position down to a millimeter or less, using four diodes at a time.
I figure that if I can calibrate the electronics properly, do some low pass filtering to get rid of noise, and work out the maths, this is entirely possible.
Will this work?
I'm planning to use a grid of IR diodes to measure the position of one IR emitter. My idea is to wire up the diodes to the A/D converter of a PIC microcontroller, and scan over the diodes in the grid in rapid succession, measuing the exposure of each. I'm not sure if I need to amplify the signals. Maybe feeding a low reference voltage to the A/D is sufficient? The emitter will be quite close to the diodes, around one inch or less. I imagine a 4x3 grid with the diodes one inch apart or less, mounted on a board together with the PIC. I figure that I'll be able to measure great enough of a difference in voltage between the diodes to track the X/Y position of the emitter fairly accurately (I have already scoured data sheets and found diodes + emitters with wide and even emission characteristics). The idea is to use an algorithm to calculate the emitter's 2D position down to a millimeter or less, using four diodes at a time.
I figure that if I can calibrate the electronics properly, do some low pass filtering to get rid of noise, and work out the maths, this is entirely possible.
Will this work?