Thought I'd play around and see if I could get any meaningful measurements of soil conductivity/moisture content with a home-built contraption.

Knowing that resistance readings and direct current conduction doesn't seem to work well from my internet info, I decided to use conduction and AC instead.

Started with a couple of new welding electrodes. Not that is probably makes much difference, I grabbed a couple of 1/8" 6010 covered electrodes, and removed the coating from the end where you strike the arc. Only removed ~3/8" from each tip.

The other end is already bare for ~1", where it attaches to a welding electrode holder. The length of the bare end that goes into the soil is probably important.

I started with a 24 volt transformer, which actually put out ~26.5 volts--AC. I spaced the electrodes as far apart as I could in a typical 5 gal container, 8". This distance is very important, and I wanted to keep the far apart to get a better overall conduction measurement than the typical very closely spaced "box store" meters.

I used a fairly high quality AC ammeter that cost ~$50. First test on the container, which had not been watered for 2 days, had such low readings that I knew were way below what the ammeter could accurately measure--~.12 amps.

So, a removed the 24V transformer and wired the electrodes directly to a 125V line. Now I got much better readings, 0.75 amps. Next I added 1/2 gal. water(which quickly disappeared into the black potting soil, and the amps increased to 1.14. That's ~142 watts of power and will heat the soil, so I didn't leave the power on for long.

Next I have to try and figure out how the amps readings correlate to soil moisture.

I'm letting it sit for a while and I'll take another reading, then add another 1/2 gal. water. Will report back, assuming I don't manage to electrocute myself.