Hey guys, I've been reading this forum for a long time as in-the-closet wannabe electronic builder. I have a ton of experience with prgramming and computer, but this is my first actual step inside.

I was wondering if there is a way to detect a broken laser beam, about 2' from the point it was launched, WITHOUT the mirror reflecting beam back for detection, or detector on the "receiver" end.
Is there a way to "see" if the beam was broken ahead, from the "sender" point of view ?

Thanx a lot !
J.

 

Think of covering your eyes and ears, and shooting a bullet. Without something, sight or sound or touch (shot your foot?) coming back and being detected, there is no way for the shooter to know that the bullet has met up with something out beyond him/her.

Ken

__________________
"To invent, you need a good imagination and a pile of junk."
Thomas A. Edison (1847 - 1931)

 

Until you enter... The Twilight Zone

__________________
Pay it forward.

 

Only if whatever is breaking the beam reflects enough of the laser light back so that it can be detected at the sender location.

__________________
Carl

 

Of course you are right (unless you had a pretty intense LSD period in your life ... ), and as I was reading my post I saw I didn't explain myself properly.

If you look at the image, I'm trying to 'know' where my hand it along an approximately 5' linear wide area. With laser I'd need setting and resetting and aligning and aiming, and it would kill the idea.



Now I'm thinking instead of shooting a "laser curtain" and somehow hoping to know where it was broken, to use 2 proximity sensors 'looking' at each other, and by using both side distance, i can come up with an accurate llocation.

Any ideas which proximity sensors to use ?

Thanx for the reply
J.

 

It's making a little more sense now. You only need to know the horizontal distance for one(?) hand in the 5 foot field. How accurate does the "location" need to be? How quickly does the detection have to occur (I'm thinking scanning)?

Ken

__________________
"To invent, you need a good imagination and a pile of junk."
Thomas A. Edison (1847 - 1931)

 

You got an interesting point. If I detect 'from above' and each 'line' indicates broken, I can 'know' multiple occurrences along red vector. However, if I detect along that vector, 'from the side' I have only one line of sight.
I would like to know two hands or even more which is impossible this way, so i'll give up this feature to get the project going.
Scanning would be a smart idea, say every 200ms (5 a second) and accuracy is always important, say to 2-3 inches.
Still, I'm trying to think about the laser option. If I 'shoot' diagonally a series of 20-30 lasers to the red surface. What are the chances I will get a reflection back somehow to know it was broken ? I
J.

 

You could sequentially excite each laser (which would be similar to a scan). That way the time occurence of reflections back from a hand (which you should be able to detect) will tell you which beam(s) were broken,and thus the location of the hand.

__________________
Carl

 

Won't natural hand angle and motion possibly break and reflect beam anywhere else but the beam origin ?
As complex as it may be, I have no problem writing code to figure out which receiver got which beam with all the time differences and delay nuances.
But, I need to know the beam will travel back

I came across this very super-cool and simple detector:
http://www.windmeadow.com/node/5
Do you feel it would feet my needs ?

 

So, crutschow, referring to the sketch I put up in a previous post:
Say I create this 'curtain' by sequencing a linear array of lasers, where each output has a sensor right next to it.
Do you still feel a human hand inside the 'kill zone' will reflect a beam back to the sensor ?
I'll have no problem writing the algorithm to decide where the hand is, I just need to know that the hand generates a solid indication on the laser array side.

Also, do you think this easy sensor will do ?
http://www.windmeadow.com/node/5

Thanx a lot.
J.

 

My idea:

Instead of using a bunch of lasers I suggest using just two of them. Mounted on a swivel? (rotating within certain limits forth and back) with accurate angle measurement and determination the beams will hit the "target" resulting in a crossover of bearings. Triangulation will give the position accurate to a mm.

Boncuk

 

Hey, thank you for the idea, i will surely give it a thought.

 

Always welcome.

Saves lots of lasers and also on the electric bill.

 

Boncuk, I have a followup ....

Ditching the 'curtain idea', just pointing one laser along the same vector toward the wall; if I 'shoot' it and place the sensor right next to it, do you still feel a human hand inside the laser line will reflect some of it back to the sensor ?

With the angle I'm shooting the laser, and having human hand breaking it, I can't convince myself it'll even work.

Do you think this sensor will do ?
http://www.windmeadow.com/node/5

J.

 

For a beam coming from the top, the beam deflection from your hand will be diffuse (that is it radiates fairly equally about a 180 degree hemisphere) thus it should reflect back to the detector. However, because of this diffuse nature, the amount of returned light is significantly less than the original beam, so you probably need a sensitive detector. This may require a phototransistor type rather than a photodiode. (Phototransistors are more sensitive but slower than a photodiode). You also may need a light filter to filter out ambient light and only let the laser light through to maximize sensitivity.

Since the reflection is diffuse you should need only a few detectors to cover the full width, not one for every laser.

The power required for this scheme would be less than a scanned laser since only one laser is on at a time and no mechanical scanner power is required.

Also eliminating any mechanical scan would be simpler and more reliable. And accurately determining the position of a scanned beam is not trivial.

__________________
Carl