Photodiode for an IR LASER

Well I'm trying to time how long an object is in the air, using the breakbeam to start and stop the timer. The objects could easily be traveling around 40mph so the time resoution has to be very fast. Especially if the object does not travel very high, because it will break the beam on the way up and "turn around" in less than a second to re-break the beam and stop the timer on the way down.

I would like to have a resolution of 250hz or better.

I know with the optex units you can have mutliple beams, you just have to move a switch to like 1 of 4 positions. So, it seems like they use modulation, but someone was mentioning to me that you have to run a wire between the sender and receiver if you do modulation, and this is not the case with the optex units.

John,

Could I just have a column of IR LEDs each modulated at a different rate (no optics or filters). And then have one IR photodiode 650 feet away (or a column of photodiodes) focused so that it can only see the column of IR LEDs. And then use software to pick out the individual LEDs based on their modulation? So, it wouldn't be like a laser, but if something was blocking the view of LED #2 I would know that the beam (#2) had been broken? Would something like this work?

Its very important that I can pin point the exact height of the object breaking the imiginary line stretching from the center of the LED straight back to the center of the photodiode.

With this setup, if an object was very close to the photodiode it would block the entire column of LEDs, which would be a false positive. If the object was very close to the LEDs it would only block the view of the LEDs that it was actually in front of, so in this case it would work. How can I make it work in all cases?

You can modulate the beams at different frequencies and demodulate the detectors with tuned circuits with no other signal between source and detector, just the way you can tune 3 radios to 3 different stations even though they are sitting side by side. You don't get the full lock in performance, but they can still work. The problem with any modulation scheme is that demodulating and filtering the result slows the response time.

Not with just a digital output from each detector. The light seen by each detector will be a combination of energies from many sources, so you would have to process the signal as an analog processing task, to sort out the contributions from each source and decide which source signal you want to respond to.

Beats me.

## ## = object in motion

^ = Direction of Movement

Possible Setup (650feet distance between LED and Photodiodes):

Freq1 LED PD PD LED Freq2 Freq3 LED ^ PD PD ^ LED Freq4 Freq5 LED ## PD PD ## LED Freq6 Freq7 LED ^ PD PD ^ LED Freq8

What do you think about this setup? Would it be possible using some kind of lens to make the IR photodiode so that it can only see an area the size of the LED it is paired with?

What do you mean by "you don't get the full lock in performance?"

result slows

How slow are we talking? In this case demodulation and filtering is the same thing right? If you filter for a specific frequency you are demodulating, correct?

Maybe I'll start a new thread with a more focused question.

Thanks John.

If you focus an image on the detector, it will see the part of the image that falls on the detector. But, unless you use something like a telephoto lens, it will be difficult to have one LED's image fall on the detector, but not the image of the one next to it.

Been through that, already. A bandpass filter will let energy through that falls within the band. Measuring that energy takes some time.

A lock in amplifier rejects everything that is not at the right frequency and at nearly the right phase. But that rejection also involves averaging over some time so it also doesn't give you fast response.

result slows

Then you have to rectify the signal and pass that through a low pass filter to clean up the noise. Both the bandpass and low pass filters cause signal delay. The wider the band of the pass filter, the less the delay. The higher the cut off frequency the lower the delay. But in both cases, the lower delay filter also passes more stuff that you want to ignore.