Imagine getting home from work, its winter, its cold. You have a shower to warm up and half way through your shower… your water stops. You wrap a towel around yourself, remove the shampoo from your eyes and go out into the cold to swap the tanks as the water tank you were using had run dry. This was a common occurrence in my house… but not anymore. Now I can keep visually see the level of water in my water tanks on my iPad and receive an email when the tank level gets low.
Awesome project! Just wondering whether it would be possible to use the Raspberry Pi Zero W - provided there is wifi access to the tank, and how reliable “reconnecting” etc. would be. Had any look at that anytime? It would shave a few dollars off the project cost… though reckon wired ethernet would be probably more reliable… ??
I’m sure Denver will be able to provide better insight, but using a WiFi connection shouldn’t be an issue at all. Provided it’s in range, it has built in handling to ensure that data is sent and received correctly.
It can absolutely be run from a RPi zero W. I tested it on a zero as well. I only used the RPi 3 and the Ethernet as I didn’t have power at the site where it was to be installed, so I used POE to get power there. I am about to build another one that runs on an onion from battery and solar. That one will be wifi only.
@Andrew54467, I am lately getting fluctuations in readings. The existing program uses a single reading to get the level. i am going to modify the program in the coming weeks to take an average or mean of 5 readings to get better accuracy. I will let you know what i come up with.
Was inspiration for mine, making up a part of an automated irrigation system.
Although I am using Python code to take the distance measurement to the water level in the tank (the earlier link was a lot of help for that), the rest of my UI is served up via a web browser.
I also had encountered some trouble with variance of the distance measurements. At times it was up to ~30mm. My solution was to take a bunch of individual measurements (10/12 or so), remove the outliers, then average whatever was left. There’s still a little bit of wobble in the resulting measurements. But I think it’s acceptable.
I used a 2.2k and a 3.3k ohm resistor as a voltage divider off the Echo output of the sensor. To take its 5volts down to the 3volts that the Raspberry Pi prefers. This bit was important so as not to damage the GPIO.
Hi Jon and John,
I had started this project a while ago, and then I was side tracked. I have gone back through my backups, and found I still had the code files, but not the data files.
Let me know if you still want these and I’ll upload them?