how to detect laser reflection with photodiodes

[degs]

Hi to all,

am starting a project where I want to detect the reflection from a laser by means of photodiodes or phototransistors. The idea is to point a laser pen dot at an object and have photodiodes or transistors pick up the reflected light from the object --- call it the "target".
This would be like a toy version of the military laser target designation and laser target seeking systems.

The specs for photodiodes and transistors are in terms of luminance required, which is fine, but I have no clue what diatances may correspond to a given luminance -- or even if the reflection from a laser dot at point blank range has sufficient luminance to be detected at all.

So, would like to know if what I have in mind is feasable at all -- perehaps someone on the forum has done a similar project? Any advice or comment would be very much appreciated.

Thanks, Sean

 

[BobK]

I am not sure I understand what you are talking about. Where are the photoreceptors? Are you talking about making a device that can follow the dot, pointing itself at the target?

Bob

 

[davenn]

I am not sure I understand what you are talking about. Where are the photoreceptors? Are you talking about making a device that can follow the dot, pointing itself at the target?

Bob

hey Bob,

he's talking about a laser guided device. As in the military where they use laser guided missiles etc, a guy on the ground paints a target with a laser, the missile ( or whatever ordenance) detects the reflected laser off the target and homes in on it.

Degs

the detection is definately going to be the difficult part
what sort of distance are you looking at being able to detect over ?
what is the detector going to be in --- I hope NOT a missile ?

Dave

 

[(*steve*)]

The secret is partially in what they paint the target with...

 

[degs]

Thanks Bob, Dave and Steve for you replies.

Dave is exactly correct as to what I have in mind --- but my budget does not run to a missile! the seeker will be a gutted toy vehicle which should go to where the cheap laser pen dot is reflecting off.
For example point the laser at a point on the room wall and the vehicle should go towards the reflections from the laser dot.

The distances involved need be no more than a couple of yards / metres, basically to operate within a room of a normal home.

 

[BobK]

I don't think you could accomplish that with simple photodectectors. A video cam with a microprocessor scanning the image to look for the dot would be a possibility.

Bob

 

[sirch]

I guess the problem you are going to have with a visible light laser in a normally lit room is that the laser light frequency will be in the ambient light spectrum. But it may be possible with some optics – i.e. a lens to focus the dot and a filter that filters out everything but your laser colour. An array of photodiodes could then be used to allow your car to seek until the brightest point is in the centre of the array. This of course assumes that your laser dot is the brightest thing at that colour.

However as BobK says a video processing solution is probably easiest - consider using a second-hand Android phone, you get a camera, processor, etc. for not much cash.

 

[degs]

Thanks Bob and sirch for your recent replies.

I do take your point that may well be more practical to go for vision ( camera ) approach, but I wouldlike to take a shot at the laser reflection detection method first anyway.

So I think I will go ahead and get a selection of phototransistors, a red laser pointer 1mW, some reflective tape to put on "target", make a clocking circuit to drive the pointer at , say 10Khz?, and a matching filter at the output of phototransistor, and see what I can detect

I will start test at point-blank range, and if get that working at least, then push out the range to whatever the limit turns out to be.

Will update the forum when I get something working.

Thanks to all for your replies/advice, I really appreciate it.

 

[BobK]

The detection should be possible, but I wonder how you would determine the direction to the target?

Bob

 

[degs]

The detection should be possible, but I wonder how you would determine the direction to the target?

Bob

Have got two methods in mind --- one have sensor on rotating turret with shaft graduated in degrees. The other have several sensors aligned with vehicle N,NE,E,SE,S,SW,W,NW and N for rough cut direction. Then when vehicle oriented in the right general direction have few more sensors front facing to get more accuracy.

 

[sirch]

Could you not just have a front facing array covering say a 90 degree zone and then have the vehicle drive round in a circle uintil the laser light is "seen" by the array?

 

[gorgon]

You could do this with two sensor spaced a little apart, looking in the forward direction. You get the 'scanning' by turning the vehicle. The main problem is the type of light you use. If it is visible light you may have a problem with light 'noise' in the room. You may fix that with some form of modulation or pulsing the light.
The sensor then only need to tune to maximum level on both sensors to go in the right direction. To make a directional sensor just put it in a tube that limit the visiual angle of the sensor. The sensor can be a normal optotransistor. To help with sensing you should make the sensor gain adjustable from the controller of the vehicle.

 

[degs]

Thanks sirch and gorgon for your advice.

I did not think of having the vehicle act as its own turret in effect! a great idea.

As regards detecting the reflected light and determining from that the direction to move in.With 2 sensors yes I can see how it would work. Signals feed into a comparator and its output determines GO LEFT or RIGHT.

But what if there were 6 sensors --- can't use comparator to select the strongest. Without some analog to digital conversion followed by storing the values and comparing them, is there a way to select the strongest? that is really sticking with analog circuits, no memory, determine which of the 6 directions to go for ?

 

[sirch]

I am probably missing something but assuming you want 6 sensors to give you a wider detection angle, can you not just group them, 3 left of center and 3 right of center, it doesn't matter how far off center the laser spot is, only that you need to keep turning until there is an equal amount of light in left and right sensor clusters.

 

[degs]

Thanks for that insight sirch.

I was looking at it in a different way - for example with 6 sensors have 6 possible angles or directions from which signal may be detected --- so can get rough cut at direction to target.
So would like a way to determine which of the 6 sensors has the strongest recieved signal. How to do that without need to sample each one, digitise each one, and then save the values and then compare them to find the highest value.
Is there a way to achieve the same result by analog circuits alone?
If not I will go for the LEFT or RIGHT choice using several sensors to get a wide angle view.