motion detection / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

motion detection

1 2 3 next

Move Thread

LAN_403

ragnar

Wed Aug 15 2007, 12:23PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Hi all,
I'm curious as to how one might be able to build a small 'analogue motion detector' without a microcontroller.

For example:
you wave your hand past (for example) a photodiode (?), and a motor spins up and down to reflect the stimulus,
but if you keep your hand in front of the photodiode, the motor spins down anyway,
but if you then move your hand away from in front of the photodiode, the motor spins up and down again, reflecting the 'change of light' getting into the photodiode.

I guess I'm looking for an analogue output which is proportional to a rate-of-change of light intensity.

Can anybody analogise this better than me, or suggest an avenue of research, or the particular 40XX chips I should be playing around with? Photodiodes? Light dependent resistors? Phototransistors? What should I really be using? :P

Carbon_Rod

Wed Aug 15 2007, 12:57PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

IR motion sensor from security yard lights are about $15 from the hardware store.

Optics and all... =)

ragnar

Wed Aug 15 2007, 01:01PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

I'm looking to build several (e.g. tens) of these into a tiny space -- the load may as well be an LED, if that gives a better idea of scale.

There's gotta be someone out there who has done this with a chip and a cap... or something. :)

Carbon_Rod

Wed Aug 15 2007, 01:07PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Ok, maybe reflective phototransistor + IR LED .... fast and crudely analog....
( 555 and a 567 PLL could tune the trigger of each unit as long as there is no harmonic selections. A PIC12/PIC10 could also do this with no external parts in RC mode.)

Avalanche

Wed Aug 15 2007, 01:31PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

I've just been having a think about this, and I scribbled the following on an envelope:

Obviously not a working circuit, just a possible approach more than anything. In theory, whenever the light level changes, a voltage will be briefly developed across R3... the polarity depending on whether or not it was getting darker or getting lighter, and the peak voltage depending on how quickly the light level changes... so you could calculate R3/C1 for the required sensitivity / speed. Across R3 you'd put something to detect a voltage differential, and use the output of this to trigger a monostable 555 set for 1 second or however long the motor runs for.

Replace TR1 & 2 and the pullups with an arrangement which would actually work

Marko

Wed Aug 15 2007, 01:50PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hey BP

Yes, for a sensor definitely go for a reflective phototransistor+led.

Problem is, that you will not get enough sensitivity without modulation unless you are running it in total dark. You'll need some kind of oscillator and demodulating bandpass filter.

Even with this approach I never achieved ranges better than few centimeters.

The motor thing can be done simple, just use an amplifier (totem pole?) and a DC-blocking capacitor in series with motor.

Now, if you can give us more insight in your project it may tell more about level of crudeness you need...

Steve Conner

Wed Aug 15 2007, 02:01PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Yeah, what do you *really* want to do? A photodiode followed by a transimpedance amplifier and differentiator (both made from op-amps) would be my first line of attack, but without knowing about the project, it's hard to say.

ragnar

Wed Aug 15 2007, 02:04PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

I just want to have a play around with interactive arrays of 'stuff'. I guess I was initially inspired by examples such as this:
http://multi-touchscreen.com/, but I'm not looking for any degree of complexity or depth, just to fiddle.

Sulaiman

Wed Aug 15 2007, 04:43PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

That's amazing!

For a dense array you could (possibly?) have a jFET such as BF245C
with a led in the drain to +9V, source to 0V, gate to "aerial" and 33 MOhm to 0V
The "aerial" would be a bit of wire/foil etc.
The aerial would pick up mains hum from your body and modulate the light
unfortunately it would be all ON with darkness following you.

I've not 'played' with small mosfets, but if it works that would be light following you.

Doing similar with an op-amp would, although a little more complex,
would follow you with light - just a mains hum pickup amplifier feeding each led.

Wouldn't be as cool as the video though.
I guess position sensing with a computer/display would be easier/cheaper/more versatile
i.e. 1024 x 768 display = 786,432 x (led + "amplifier" + board-space + effort) = $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$

Simon

Thu Aug 16 2007, 12:39AM

Registered Member #32 Joined: Sat Feb 04 2006, 08:58AM
Location: Australia
Posts: 549

BlackPlasma wrote ...

I'm looking to build several (e.g. tens) of these into a tiny space -- the load may as well be an LED, if that gives a better idea of scale.

There's gotta be someone out there who has done this with a chip and a cap... or something. :)

Don't drop the IR sensor idea so quickly. They are based on pyroelectric crystals, which can be found in small packages - TO39 kind of thing. Pyroelectric crystals have an inherently high-pass response, hence the use for motion detectors.

A photodiode circuit might be easier to get parts for, though.

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob