Advances in alternative energy technology over the last four decades have made it both easier and cheaper to live off-grid. This series provides a personal perspective on the pluses and minuses of living with technology in the forests of Northern California.
I stood precariously in the medium-sized Sterilite plastic storage container, carefully aiming the shower head down to where the maximum amount of water would fill the tub. As I carefully bent down to get the shampoo out of my hair, I wondered how long it would be before the spring would run again.
It was a hot summer, although it had, unusually, started indecisively, like an elderly car spluttering into life. May and June had been uncharacteristically damp, with a couple of weeks of sunshine in July, and then a couple of days more rain. In late July, and through August and September, the sun blazed down. Eventually, our spring trickled to a halt.
Fortunately, our neighbor lets us take the overflow from his high lifter pump fed from his own spring that runs all year, when this happens, so our tanks could be kept topped up.
But this year, even his spring seemed to have quit. The tanks were down to the last 3,000 gallons. That may sound a lot, but when you consider that the average American gets through 176 gallons of water per day, it’s not that much. And there were five of us, including the girl and her sister who rented another cabin on the property, dependent on that last three thousand gallons. If we all behaved like typical Americans, the water in the tanks would barely last half a week.
Of course, there were no flush toilets. We had started doing our laundry in town at the laundromat as soon as the tanks started dropping, but even so, it could be two months until the rains came and the spring started flowing again. So, we rationed and conserved what water we did have. We used biodegradeable paper plates and cutlery, to cut down the washing up. As previously mentioned, we showered while standing in Sterilite containers, sparingly, and infrequently. The gardens went unwatered: corn withered, potatoes turned brown, squash shriveled up.
It turned out that the neighbor’s spring was flowing steadily, but a connector taking the high lifter overflow down to a catchment basin that fed into our water system had come off. The connected had possibly been knocked off by a thirsty raccoon or skunk hearing precious water flowing. After reconnecting the connector, the tanks soon filled up again. But for two weeks, we could recreate in our own homes the glorious inconvenience of being stingy with water, which is usually only associated with Burning Man.
It would have been a perfect time to have already built my remote water-level measuring system. However, after a few promising breakthroughs, I’d rather hit a brick wall with my head. The 101 mundane chores that living off-grid entails (turning the compost heap, watering the garden, taking the garbage to the tip, fixing recalcitrant electrical systems, etc.) had stolen time from my invention itinerary, so it wasn’t until October that I managed to get back to the workbench. Ironically, this happened when the rains had started to come.
There are many ways of measuring water levels, but whatever sensors I chose for the shortlist had to share some common characteristics: low cost, low power, low complexity (minimal modifications to the storage tanks and as few moving parts to go wrong as possible), and high robustness.
Ultrasonic or pressure sensors are either too complicated, too pricey, or take too much power. I had an idea to submerge a vertical pipe, open at both ends, with a float inside with magnets mounted on it that would trigger magnetic reed switches on the outside as the level changed. This method had the advantage of being relatively electronically simple (on/off), low power, cheap, and easy to configure. Unfortunately, it turned out to be mechanically more complicated than expected, with the location of the magnets being crucial, low resolution, and electrically fairly complicated (a lot of soldering and wiring required for the reed switches). I built a test version, which you can see didn’t work that well (apologies for the grainy phonecam footage):