At 04:21 PM 5/18/2005, Debbie wrote:
>Thanks for those ideas Mike. That NVE website looks pretty interesting too.The
>app will be - **if** it goes ahead? - to photograph cars on a country road.
>It's a big station abutting a nat park and they want to monitor who uses roads
>at back of their property by photographing vehicle rego plates. It's semi arid
>Outback country so I think dust will be the killer - how do you photo thru a
>dust cloud thrown up as the vehicle moves away? If go for a forward 
>shot, you'd
>have probs with the vehicle's headlights at night. Plus flash/illumination
>probs. Still thinkin' about that

Solid state cameras are pretty amazing that way. You can get the 
pixels inbetween the overexposed ones, IF you disable any sort of 
auto-exposure, and use your own ambient light sensor, and illumination.

>Don't want to use inductive loops or anything that needs cables 
>spread out - it
>needs to be smallish & easy to deploy so solid state sensors look attractive.
>The range of the sensor would be an issue - it needs to detect a vehicle from
>the side of the road. That's not far as we're talking dirt tracks & large 4x4
>vehicles mostly. don't need auto rego-number reading. Hmmm ... a laser bounced
>off a corner reflector is a possibility, maybe.

Bugs?

Anything wider than about half the beamwidth will trigger it.

Critters and rain can be kept off the optics by the same sort of 
thing they use for windshield wipers on ships.
Spinning plexiglas disc throws anything that lands off radially.

>Would you say solid state MR sensors could detect a vehicle in a 
>situation like that?

It would be best, if the vehicle comes between you and the south pole.
You could also add a local bias magnet.

Be prepared to see this set off by geomagnetic storms.

You can also use a large-ish loop oriented like a sign, not buried.
With two of these, say 10' apart, you can go differential, and geomag 
problems should null out.


Then there's the TX-RX method, send 1kHz into one coil, and use 
another to pick it up, using synchronus boxcar detector integrator if 
need be, to eliminate noise.  Level shifts tell you that the amount 
of magnetic coupling has changed.

A tracking window comparator (I can give you details offlist) will 
eliminate hits as the coupling varies slowly over time.

Third, you can set up the coils so that the field at the receiver is 
nulled out, and adding metal unbalances it.
A bit trickier to set up, but very easy to produce in numbers.



-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist