Trip sensor -- line of sight

The problem with optics in that sort of application is keeping it clean and dry. If you want to aim a laser beam round the circle, it also has to stay aligned.

I'd try one or two nice bright LEDs on each post, driven at a few kilohertz, and a small solar cell plus transimpedance amp for detection. Put a tubular sunshade on the detector, sloping downward slightly like a mailbox, and use a DC restore loop to get rid of the offset from daylight.

Run that into a simple lock-in made from a CD4066, and test the output with a comparator or a spare op amp.

A lens in front of the LED will help, without being too directional.

Alternatively, you can put the detector and LED in the same box, and put a bit of high-brighness retroreflecting tape on the other post. It returns light directly back, with a cone angle of about 0.5 degrees, so you wouldn't lose too much light, and you wouldn't have to run the clock signal between boxes.

The limitation will be the shot noise of the sunlight photocurrent.

Cheers

Phil Hobbs

Or use near infrared LEDs and a cheap IR long pass filter. At a pinch unexposed slide film tails will do.

The lens itself also needing shading.

The rear reflectors intended for bicycles aren't bad either. I'd favour this solution if the range requirement isn't too excessive.

Which is why I'd be inclined to filter out as much of it as possible and black flock line the tube leading to the sensor.

I suspect these days an off the shelf area coverage motion sensor webcam would be more reliable and very much easier to get working.

On a sunny day (Thu, 3 Apr 2014 07:23:12 -0700 (PDT)) it happened mkr5000 wrote in :

Modulated laser pen, a few mirrors, sync detector?

Any guess at how much light you would loose in the retro- reflecting tape scheme? (I've never used reflecting tape) (And what's your fav reflecting tape... something from 3-M?)

Re: not running clock signal to each box. You could make three of the boxes repeaters.. lock on the the signal and blink an LED at the same frequency. Then everything comes around to one master box that has the LO in it.

George H.

If the signal is big enough to lock to, sure. The PLL would have to be designed well, with an analogue phase detector.

Good quality retro tape is about 3000 times brighter than Lambertian. I've used Scotchlite 3000X and Scotchlite 7610. Reflexite also makes stuff that sends light back with a fixed offset angle, which can be pretty handy.

Cheers

Phil Hobbs

The garage-door safety things seem to be pretty good and rugged. I've seen web sites that analyze their interfaces (mostly to fake them out!)

I'd recommend four links, rather than mirrors.

How big is the space?

This is what I was afraid of, that I was falsely dreaming that in 2014 ther e would be a solution "off the shelf" -- I can say that for starters anyway , I'd like to keep the garden small like a 12' by 20' rectangle minimum. Th is may open up a new adventure -- have never experimented with anything las er before.

Near infrared either. (Less prone to sunlight problems?)

Can I modify a motion detector to act better during daylight I wonder?

john -- you're something else -- those garage door things may be it !

seems to me they don't have a problem with sunlight either -- I have a couple in my garage, will have to test the distance -- may be an easy solution

also just thought of sort of a long rotating wand (or 2) in the center of the garden, free to rotate but if something stopped them from turning, presto.