On Sat, 1 Feb 2003, Anand Dhuru wrote:

*>Peter, I notice the empahasis on the 'insulate' part; should they be or
*>should they not?

They should be.

*>I had fabricated a sensor using a pair of aluminium strips the height of the
*>tank, separated by plastic spacers ever few inches. The strips themselves
*>were NOT insulated from water. Worked as expected, but the problem came from

Aluminium is 'self insulating' in water. Water has a high epsilon-r but if
it has a lot of impurities and is very conductive the capacitance will be
mostly from the insulation on the electrodes (and much higher than
expected). Electrolytic capacitors are built that way. There are ways to
compensate for this (f.ex. second small permanently immersed insulated
capacitor probe, used in the other branch of the metering bridge - or as
reference in a digital system).

*>elsewhere; the sensor was too sensitive to noise, even when connected using
*>a shielded, or co-ax. cable; the readings would go haywire if one touched
*>even the plastic insulation on the connecting wires.

The trick is to make a coaxial sensor. The inner electrode is shielded by
the outer one.

*>Despite the issues involved, I eventually settled for a conducting type of a
*>sensor; a ladder of 10 100K resistors in series; as the water level rises,
*>it progressively 'shorts' out the resistors underneath. This sensor was then
*>used as the lower half of a potential divider and fed to the ADC on the PIC.
*>Cant complain with the results, its been working for over a year now!

Your water must be very clean ... I would not try this here. With ~2V
across the immersed parts I would not be able to find the resistors after
a few weeks (washed down the drain).

Peter

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