Well still dealing with some leg issues and was stranded in bed most of the weekend. But I did make a little progress on the watering system.


First thing I mixed up a batch of potting soil. I used a 2.5:1 topsoil to vermiculite. Decided to go down this path when reading how most sensors are sensitive to mineral content. Most of my plants are planted in a similar mix so this should be a good representation of the final application. From there I created 3 containers of dirt and weighed them out to ~38.3 oz each. One container is left as is and will be my moist sample. The next container I added 6oz of water and will be the wet sample (I may have added way to much water top is a little muddy). The last container I dumped out in a tray and baked at 350deg F for several hours mixing every so often. This will be the dry sample.


While the third sample was backing I did some playing around with the nominal sample. To start with I created a pair of probes out of some insulated 12 gage copper wire. Using my Sencore Z meter I checked the capacitance of the probes in free air and got around 47pF. When inserted in the dirt I got weird results. About every other time I would get around 189pF and the next the meter would error. Not exactly sure why the meter was acting up. Double checked the probes were fully insulated (coated the cutoff end in enamel and eventually just folded the probe in half so the cut end was not contacting the soil at all.) Basically all I was able to prove at this time was the soil does increase the capacitance between the probes and the value looks to be in the ball park of 200pF for slightly moist soil.


I also tried sticking 2 2" square patches of metal foil tape on the outside of the container. My hope was the increased surface area of the foil tape would result in a larger capacitance value. But I got the opposite result and got a slightly negative value. More than likely the calibration had drifted on my meter slightly and that resulted in the -1.8pF when the foil tape was hooked up. Going to revisit this setup once I get the 555 timer circuit built and tested.


I started looking into using a 555 timer to build a astable oscillator. Going with 200pF as the soil capacitance and 1M ohm resistors for R1 and R2 would result in a 2.5kHz output frequency. Normally I don't see 555 timers used with such small caps or large resistors. I have some concerns that the leakage current from the timer it's self may be a significant issue at this point and may not function properly. Either way I though I would give it a go and see what happens. Unfortunately the leg started to act back up and called it a day.

Next up I want to finish up the 555 circuit and see what sort of results I can get out of simple round probes (wire). If that does not work out will have to come up with a better probe solution. One option is the foil tape on the outside of the container. The other is building a set of flat probes with more surface area that can be installed in the soil. The biggest issue with the flat probes will be making them porous enough that water can disperse into and out of the soil trapped between the probes and waterproof at the same time.