Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Directional People Sensor

Status
Not open for further replies.

PICMICRO

New Member
Sorry, I couldn't find a more descriptive Title.
I have a home (everybody does, :) ), and an entrance corridor (around 6m wide). Whenever someone walks across it towards the door, I want to ring the bell. Whenever someone leaves the house out through the corridor, I want to sound a different bell / do nothing.
In short, I want a beam break detector with direction sensing. I am pretty optimistic that Direction Sensing Can be easily done with uC interface using two simple beam break detector. Direction can be worked out from the chronology of beam break.
So, actually, I just want suggestion for a reliable, Beam Break Detector, to be used outdoor. (The corridor is outdoor)
I searched around a saw some methods.
1. Using laser diodes and Photo-transistor / LDR. However, I am concerned with the safety, so don't think Laser would be suitable.
2. Using a simple, IR LED/ Photo Diode pair. But, I don't think it can provide 6m distance out door, and I am also concerned with its immunity to ambient light.
3. Using IR LED Transmitting circuit and TSOP reciever ICs. Don't know if it can work.

Any suggestion please.
 
I do not see a simple way to go about what you want using a single beam. Since you want it directional (Enter / Exit) I see a need for two beams. The beams (sensors) can be of your choosing. Using two beams the trick becomes to distinguish enter and exit.

A popular circuit for doing this is a people counter where as people enter a count increments up one and as they exit it decrements down one count. A version is attached. It would need a little hacking but should be able to work. The idea is the logic behind how it works. The sensors would be spaced about 6" apart so one is broke before the other while the first is still broke.

Being at work I don't have access to much but I would likely approach things using a circuit along these lines with two sensors. Maybe someone else has some thoughts?

Matter of fact you could likely use any number of quadrature decoder circuits as to detecting direction of travel.

Ron
 

Attachments

  • People Count.gif
    People Count.gif
    26.1 KB · Views: 491
Last edited:
I think Ron's idea is a winner. I would replace the standard photo diodes with TSOP's and a peak detector to get the range and noise immunity.
 
As Ron suggested, here's a simple (quadrature) decoder circuit that gives one signal for an A-B input sequence and another signal for a B-A sequence. That should do what you want.
 
Thank you people for your co-operation.
But I think, I will be doing the 'quadrature decoding' using a micro-controller. So, I am seeking suggestion here for 6-m long, ambient light immune outdoor Beam Break detector.
ronv suggested to replace photo-diode with TSOP. Thats exactly where I am stuck at. Using TSOP means, I will nead to transmit IR pulses modulated at carrier frequency (often 38 Khz), this I can do. Now, what do I get at the Output of the TSOP, and How I use them?
Has anybody played around TSOP beam break detector?
 
I tried to use TSOP1838. What I did was, I used PIC16F72 and setup its PWM for 38Khz frequency and connected IR Led to it. I then switched the dutycyle from 0 to 50% alternatively using timer0 interrupt, so as to create short brusts of around 1ms duration. The TSOP module responds to it and generates output for first few seconds (2-3 seconds) but after then, it just shuts down. I think its ignoring my signal by interpretting it as noise. Any chance anybody knows more about this and a workaround?
 
Hi, Got all of it sorted out. I used brusts of around 50 cycles and a gap of around 25 ms every 90 ms. (As suggested in the Datasheet). I did use two TSOPs and interfaced with my uC to find out the direction. Its working. Any body wants to ask anything, just do it.
Thanks.
 
Cool, so now it works as advertised? :)

Ron
 
Very good! :) Nice little video.

Ron
 
Status
Not open for further replies.

Latest threads

Back
Top