Go Down

Topic: speed of photodiode (Read 5641 times)previous topic - next topic

calvingloster

Nov 12, 2013, 11:38 pm
I'm building a lap timer for our motocross track. The whole circuit will be signaled by a laser switch which is a laser diode shining onto a photodiode. This will give the signal for my arduino to start and stop timing etc. So I have a few concerns, firstly will the photodiode change its resistance when the laser light is taken away for a few split seconds. U can imagine if a bike comes flying past its going to block the light for a very short time.

I set up a basic circuit with a battery, an LED and a photodiode. I shine the laser on the photodiode and the LED glows brightly. When I take the laser away the LED dims but slowly. So I'm concerned when the bike rides past the signal of the trip circuit will not change at all? Any suggestions? Or am I missing something?

RthuR

#1
Nov 12, 2013, 11:53 pmLast Edit: Nov 12, 2013, 11:55 pm by RthuR Reason: 1
This may not be entirely relevant, but I built a ski timer which followed the same principles. I ended up using an infrared gate, the datasheet from the manufacturer specified the response time though. I ended up using  this. I believe the response time was 10ms, so if you use speed = distance / time you should be able to find out the maximum speed that it will detect (distance being the width of a rider / length of a bike).
If I recall correctly, assuming a skier had a leg width of 15cm, the max speed was 15 m/s or about 50 km/h)

DVDdoug

#2
Nov 13, 2013, 12:00 am
Speed shouldn't be a problem.   Check the specs, and I'll bet the photodiode works at MHz rates.   In fact you might have to slow it down, or take several readings in a row to make sure it doesn't get falsly triggered by a bird, or flying piece of dirt, etc.

I think we need to see a schematic...  You've got a laser and a photodiode, so what's the LED got to do with it?

It probably will take some experimentation.   You have to make sure the sensor is not turned-on  by sunlight (or other light), but that it's always turned-on by the laser, and turned-off when the laser is blocked.  At outdoor events, of course the lighting will change through the day.

And in a tough environment like this, you might want to have a 2nd redundant system running in parallel, or at least one or two people with stopwatches.

MarkT

#3
Nov 13, 2013, 01:12 am
Quote

I set up a basic circuit with a battery, an LED and a photodiode. I shine the laser on the photodiode and the LED glows brightly. When I take the laser away the LED dims but slowly.

Then there's something wrong with that circuit, or you've overheated the photodiode,
expect switching times measured in microseconds.  With the correct circuit using
reverse-biased photodiode you can go even faster.

Or are you using an LDR and calling it a photodiode?  Photodiodes are usually
made of silicon (although an LED can be used as a photodiode)
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

polymorph

#4
Nov 13, 2013, 01:26 am
An LED and a photodiode? What part number is that photodiode?
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8

calvingloster

#5
Nov 13, 2013, 07:28 am

Speed shouldn't be a problem.   Check the specs, and I'll bet the photodiode works at MHz rates.   In fact you might have to slow it down, or take several readings in a row to make sure it doesn't get falsly triggered by a bird, or flying piece of dirt, etc.

I think we need to see a schematic...  You've got a laser and a photodiode, so what's the LED got to do with it?

It probably will take some experimentation.   You have to make sure the sensor is not turned-on  by sunlight (or other light), but that it's always turned-on by the laser, and turned-off when the laser is blocked.  At outdoor events, of course the lighting will change through the day.

And in a tough environment like this, you might want to have a 2nd redundant system running in parallel, or at least one or two people with stopwatches.

I just put the components in a basic circuit to see how it works and I realized that the reaction time was so slow. The part I'm using is a LDR 10k. So what's the difference between a photodiode and a LDR?

As for the circuit, the laser diode will be powered separately and will not be part of the circuit. The LDR will be connected to an analog pin and a 5v pin so when the laser light is blocked the 5v signal will drop and the analog pin will read the drop.

calvingloster

#6
Nov 13, 2013, 07:47 am
I've been reeding up on photodiodes, light dependent registers and phototransistors, and what I'm thinking is I should use a phototransistor?

UnoDueTre

#7
Nov 13, 2013, 09:06 amLast Edit: Nov 13, 2013, 09:08 am by UnoDueTre Reason: 1
Yes, LDRs are slow compared to photo diodes and transistors.

As DVDdoug pointed out, how do you intend overcoming the possibility of false triggers or missing triggers due to  natural light which contains both visible and infra-red light?

You will need to place the photo detector in a tube painted black (pvc pipe is ok) so as to limit the detectors field of view.
Experiment with the diameter and length of the tube.
Simply using a laser (or other light source in the visible or infra-red spectrum) is not sufficient and you will have to modulate the laser with a carrier (similar to what is done with remote controls) then demodulate that carrier and use it's absence as your trigger.

calvingloster

#8
Nov 13, 2013, 09:25 am

Yes, LDRs are slow compared to photo diodes and transistors.

As DVDdoug pointed out, how do you intend overcoming the possibility of false triggers or missing triggers due to  natural light which contains both visible and infra-red light?

You will need to place the photo detector in a tube painted black (pvc pipe is ok) so as to limit the detectors field of view.
Experiment with the diameter and length of the tube.
Simply using a laser (or other light source in the visible or infra-red spectrum) is not sufficient and you will have to modulate the laser with a carrier (similar to what is done with remote controls) then demodulate that carrier and use it's absence as your trigger.

Thanx for your help, can you maybe just explain "in simple terms" why the laser light on its own won't work? I don't understand what u mean when you say modulate with a carrier? Sorry I'm new to this so please bare with me.

calvingloster

#9
Nov 13, 2013, 09:30 am
As for false triggers I'm not really concerned about that, what I would do is design the program to basically discard any trigger signals that are not close to average lap times. So if a lap takes 1minute and 20 seconds I would design the program to discard anything less than 1.30 and 1.10 so if a rider rides past and his arms and body triggers the system the program will take the first trigger and discard the second one

UnoDueTre

#10
Nov 13, 2013, 09:42 am
What will discarding triggers based on a timing window going to solve?
Optical interference from natural light is not limited to time, it has nothing to do with the "beam" being interrupted by a riders hand, foot, leg or any other part.

UnoDueTre

#11
Nov 13, 2013, 09:52 amLast Edit: Nov 13, 2013, 10:01 am by UnoDueTre Reason: 1

.....can you maybe just explain "in simple terms" why the laser light on its own won't work? I don't understand what u mean when you say modulate with a carrier? Sorry I'm new to this so please bare with me.

OK, assuming visible light, your sensor is not going to know the difference between natural light (which contains a broad spectrum of colours) and the light from your laser.
To overcome that, you modulate a carrier onto the laser and in this way when you demodulate the carrier, you can tell if the light is indeed from your laser.
Think of it this way, you are giving your light source a unique "signature" to be able to identify it from all the other light.

As for the darkened tube, it gives your sensor a narrower field of vision so as to minimize unwanted natural light falling on it and desensitizing it.
Think of it this way, you are trying to look at an aircraft flying overhead but the sun is in the same position in the sky, your eyes will be completely swamped by the sun's light that there is no chance of spotting the aircraft.
An extreme example but I'm sure you get the point.
Since your laser is more like a point source, the narrower the field of vision of your sensor the better (within reason).

The reason the tube should be black is to minimize reflections within it.

crazypj

#12
Nov 13, 2013, 09:59 amLast Edit: Nov 13, 2013, 10:02 am by crazypj Reason: 1
Are you building a lap timer to work out speed or a lap counter to count 'the field'?
In my experience, some days you can have a lot of people 'fall down' and miss a lap or two
I think you will need to do some math and work out 'average'  bike length?
Wouldn't you get multiple interupts from leading edge of front tyre, trailing edge, frame (including rider, etc), rear wheel; before you get next 'clean' signal when bike has past sensor?
Setting upon the slowest corner would also give any sensor more time to react, (rather than being fet up on the start finish straight)
Quickest and cheapest way to find out how fast MX bike is going is get a bicycle speedometer  fitted to it

UnoDueTre

#13
Nov 13, 2013, 10:06 am

In my experience, some days you can have a lot of people 'fall down' and miss a lap or two

Absolutely.

Quote

I think you will need to do some math and work out 'average'  bike length?
Wouldn't you get multiple interupts from leading edge of front tyre, trailing edge, frame (including rider, etc), rear wheel; before you get next 'clean' signal when bike has past sensor?

Yes the OP would need to take into account the length of time the beam is broken which is related to the bikes length and speed.

calvingloster

#14
Nov 13, 2013, 12:10 pm

What will discarding triggers based on a timing window going to solve?
Optical interference from natural light is not limited to time, it has nothing to do with the "beam" being interrupted by a riders hand, foot, leg or any other part.

Well there is no issue in my understanding in the first place. False triggers can happen and that's kewl with me, if they happen out of the time range I specify they will do nothing to the timing process of the program.

Go Up