My tank is for drinking (as well as irrigation) water, and it’s quite high in minerals, especially iron, so it’s plenty conductive enough; I’ve tested it with a simple voltage divider on one of my Photons. I was originally thinking of using gold wire wrap pins, since I thought I had a bunch of those around from the “old days”; but alas, I can’t find them. Another approach I was considering was to use a donut shaped float containing a magnet that would ride up and down on a pvc pipe containing a bunch of hall effect devices; this would only expose the water to the plastic parts, and the electronic components would all be in the dry space. Either of these approaches will require the use of multiplexing since I’ll probably want to have 1" spacing (1" = ~30 gallons) for the top 2 or 3 feet, with wider spacing below that (I do have several CD4067 16 channel multiplexers lying around).
I’m still considering the heating to control condensation on my HC-SR04’s. Wondering if I could paint the sensors black, and use several IR LED’s to provide the energy for heating? I looked at a some more expensive ultrasonic sensors that are supposed to be self-cleaning; self-cleaning it turns out is a heated sensor – those used ~70 ma continuous to heat the sensors. One other problem I’m worried about is whether I will get mineral deposits on the sensor which could compromise its long time reliability. My tank has an ozone bubbler in it (for sanitation and oxidizing iron) which does cause some aerosolization in the head space of the tank. A lot of variables to think about for what would seem to be a fairly simple project.