Andy Shaw wrote:
>
> Hi Well I've tried to build one of these things and...
> a) What you fill it with depends on how you expect it to work. From what
> I've seen you basically have two choices. One is to use some kind of
> substance that does allow a current to flow through it and then make sure
> you use an equal AC current to ensure that you don't get all sorts of odd
> things happen to the stuff! Second is to use some sort of liquid that does
> not conduct and use a purely capacitive effect. My understanding is that
> most of the commercial sensors use the first of these options (this is the
> way I've tried doing things).
> b) I've seen articles that use water (possibly de-ionised not really sure),
> with a small amount of alcohol added to help prevent the surface tension
> problems.
> c) The big problems I had was getting the damn thing sealed up (you have to
> have a really, really good seal otherwise they eventually dry out. Most of
> the things you would like to use as the liquid have a nasty habit of
> screwing up most of  the things you would like to use to seal them!
> e) Sounds like you are going to have a very small space. I would take great
> care with checking that the surface tension/capillary action problems do not
> screw you up.
> f) Check the resolution you need most of the commercial units I've seen get
> better resolution by going from a wider gap between the electrodes. Sounds
> like you won't have that option!
>
> Good Luck I spent ages trying to do this and then gave up and got a
> commercial part (but then the facilities (and skill) I had available may not
> be anything like yours!). Let us know how you get on.
>
> Andy


Hi guys, I am having difficulty understanding why this is the best
way to meaure tilt, etc? Using ac, variable capacitance, etc etc?

Why not get a little container with a translucent liquid, and
two simple cheap opto trans/rec pairs. As the tilt changes the
liquid thickness between each pair would change, changing the
light amount getting through. The transmitter is just the leds
and a power source. The receiver can tie the optodiodes
straight into the analogue pins on a PIC with a pull up resistor
and for almost zero parts count you have a good reliable analogue
tilt sensor.

But if you start using ac and liquids you get problems like
corrosion, dissimilar metals reacting, salts from electrolysis,
etc etc. What is the benefit of the ac driven system?
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads