On 5/31/05, Jinx <joecolquitt@clear.net.nz> wrote:
> > >Ball-bearing + ping pong ball + UGN3503 + magnet
> 
> > Looks good.  How well does it work if the vehicle is tilted when the
> > alarm is armed?
> 
> So far so good. I'm looking at various ball bearing sizes to find what the
> optimum size is. Too small and it will be attracted too much to the magnet
> and not wobble enough with small movements. Big-ish (eg 8mm) seems to
> be better. With any size, there's very little friction between the steel and
> the celluloid, so magnetic attraction is mostly the damping factor. My only
> concern would be the life-time of a celluloid ping pong ball, both due to
> material break-down and the bashing it's going to get from a ball bearing,
> but almost any other small plastic ball with a smooth regular inside surface
> will do. I've added an op amp for AC gain and a little HF noise reduction.
> The UGN3503 output  is very stable so it can be cranked up with gain
> and changes detected with a narrow adjustable 12F675 ADC window

The ball bearing could be damped with a fluid, which would also
prevent damage to the ping-pong ball.  I would think silicone oil,
which is available in different viscosities, might be a good choice;
although, there may be material incompatibilities, the damping will
change with temperature, and it's messy.

I like this design, though.  And with the addition of more sensors you
could know the tilt in multiple directions.  It would be an
interesting project, but at some point an accelerometer seems better
suited cost and size-wise.  It would be interesting to compare the two
methods side-by-side, though.

Bradley

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