Yeah, my main problem is that the sensors have to be submersible
(there are no holes in the bottom of the cistern, for obvious
reasons), and have to be compensated for atmospheric pressure, as an
inch of mercury is about a foot of water depth...
Looks like it's not as trivial as it seems.
If you can't submerge the sensor, the easiest way is to use a 'dip tube'
type of setup. A small plastic tube with a weight on one end to hold it
down in the bottom of your tank. On the other end, a 'tee' with a gage
pressure sensor on one leg, and a source of very low pressure air on the
other. Trouble with this is you must adjust the air flow to *very* low
flow, so just a bubble every second or two percolates up from the bottom end
of the dip tube.
If you want continuous measurements, you need continuous air supply. If you
just want to check the level once/twice a day, a tiny hand pump can be used.
Pump the air until you're sure the line is cleared, then stop pumping.
After you stop, the pressure in the line will stabilize and you can take a
reading. For intermittent use, you can get just a direct-reading pressure
gage of the right range. To help stabilize the readings, you can attach a
sort of 'diving bell' to the lower end (think of an inverted drinking cup).
This will provide a larger volume of air that has to leak off before
affecting the reading very much. Of course, it takes more weight to hold
such an air volume under water. Maybe permanently attach it on the inside
wall while the cistern is empty for maintenance.
Using a 'gage' pressure sensor means it will sense the difference between
the sensing tap and the atmosphere around it, thus compensating for changes
in the weather. If you can only find 'absolute' instruments, then you must
get two and use one to measure the atmospheric pressure, and one to measure
the pressure in the sensing 'dip tube'. Then the pressure difference is
caused by the depth and density of the liquid being measured.
Along with the various differential pressure schemes, there is also the use
of a float pulling up/down on a string/wire and looping the wire around a
form of revolution counter (these can be direct reading, or electronic).
Also the float with a magnet riding up/down a tube with a series of
reed-switches. Or, an insulated wire inside a free-flooding tube to measure
the different capacitance constant between air and water (the insulated wire
forms one plate of a capacitor, the tube the other plate, with air/water as
the dielectric). Although a bit high-tech, I've also seen the ultra-sonic
range finders such as some cameras used to have to measure the distance from
the top to the surface of the water.
Using two uninsulated wires and measuring the relative conductivity can be
pretty impractical if you don't know the variability of the water's purity
(although a separate 'reference cell' in the same tank could be used in some
form of bridge). Best to use a very low voltage AC in such a setup to
minimize glavonic-like corrosion of one side.
The capacitance and resistivity probes can have calibration issues if there
is any growth of 'organics'. And obviously, floats sometime sink, and
tubing sometimes can clog. They all have various advantages/disadvantages.
As you can see, there are *lots* of ways to measure the depth of the water
besides differential pressure.
daestrom