Gus Calabrese wrote:
>
> Subject:
>  Sensor questions (maybe [OT])
>     Date:
>           Wed, 8 Apr 1998 12:47:47 +1000
>    From:
>           Clyde Smith-Stubbs <clyde@HTSOFT.COM>
>
> I need some suggestions on a couple of sensors;
>
> 1) I need to measure water level (and thus contents) in a concrete tank.
>
> The tank is about 2m deep, so a float is probably not practical. My
> thoughts are either an ultrasonic setup to detect the water surface,
> or a pressure sensor.


What resolution is needet ???

Do I am missing somwhat ? Here is my idea: sonds a little bit dumb but
very simple and cheap.

Why dont detect the presnce of whater between 2 probes ?
To prevent any chemical processes in the liquid or on the probes
only AC should be used. There are special chips for this as mentioned
befor.

I have done this with simple CMOS logic gates ( invertors with
schmidt-trigger
inputs) of the 74HCT14. One Gate is used as an multivibritor and is
coupled to one probe trough an Cap and the rest can detect the signal.
(no whater between the probes, no signal).

You can increase the resolution by increasing the number of probes.
For more then 4-5 probes you even can use a PIC at the sensor !

Use some wires with stripped ends as probes. The longest wire ends
at the bottom of the tank and is used as input. Make the striped
ends of the other wires ends in a fixed distans, say every 10 or 20 cm.
So you have 20 or 10 steps for the 2 m tank. Hook this wires trough
caps to PIC pins. For every elektrode generate some pulses on the
pin and look if there is a signal on the sensing elektrode. The PIC
can serial transmit the level value to a controller/display station.


I suppose this method should be immune to dirt in the water.
For clean water maybe silver pleated or carbon electrodes should
be used.


just my 2 cents on the long thread.

St.

wi