Repository navigation
Water Tank Monitoring System on Solar Power #13354
Replies: 4 comments 23 replies
|
Since IP44 is "only" splash-proof, I'm wondering if you do something to manage or measure humidity inside the box. Seasonal and daily variations might lead to condensation under some circumstances. |
I just followed your suggestion and added a BME280 to the box. Now we can monitor environmental conditions and see if there is a humidity problem. |
|
Unfortunately, I had to learn that the BME280 probably needs some air flow in order to produce sensible humidity readings. Once you close the lid of the box humidity maxes out at 100% very quickly although the box is completely dry inside. The other readings are stll worthwile, however. |
|
I actually did a small test with a BME280 in a closed jar, without airflow. Nothing weird about humidity readings, just kept the same quite plausible level. |
|
It may takes time... |
|
Your sensor may be busted. What vic posted about (on Discord) was it very quickly rising to 100% when the box was closed. My test was for nearly 2 days. My current outdoors BME280 (in a plastic enclosure with slits for ventilation) gives plausible numbers, currently mostly above 90% (I'm not expecting great accuracy at high humidity). While it sometimes flatlines at 100% for hours, the surrounding plot points does not make it look like it was stuck, merely maxed out, possibly not really able to distinguish differences at the high humidity level. My previous outdoors BME280 was housed in a way less reliably protecting it from becoming wet. It failed (producing too high temperatures), and when inspecting, I found that one pin was so corroded that it fell off. Currently the replacement is paired with a BMP280, as I assume that both would not fail the same way :-) |
|
very interesting is there a wiring diagram and code available |
|
There is not much code involved, just Tasmota. (And the typical node-red backend plus some php web graphics using RRDTool) The wiring is so obvious that I did not even draw a diagram.
That's pretty much it. |
|
Hi @vic42, I have a similar setup, I'm facing an issue though, it's really warm where I live, and I started noticing short readings between specific hours, I checked the sensor, and there seems to be condensation on the lens of the Vl53l1x, I assume because of water vapour that rises and hits the tank lid, and that's where the sensor is placed. |
|
You should avoid the sensor as such getting wet, possibly by enclosing it to avoid condensation directly on the breakout board. If enclosed, one technique to reduce condensation effects would be the same as for a car window, as adding a bit of soap tends to avoid condensation droplets. Of course, heating or a wiper is also possible, but may not be feasible. |
|
@sfromis thank you for the reply, the pcb and the sensor are all waterproof, I've designed a custom water-tight enclosure and used a 3D printer to make it, and all the gaps are covered with silicon. |
|
Hi Youssefff1, I do get these odd readings once in a while. I assume reflections from mist in the tank. Currently I'm filtering in software but I'm thinking about modifying the driver in order to reduce sensitivity of the sensor to focus on the strongest echo. I currently lack the time to work on it.
I used an ultrasonic transducer (sealed and water tight) for another tank that doesn't exhibit this behavior. But the downside of that one is a dead zone of ~300 mm (typical for single transducer sensors). |
|
Thank you @vic42, |
|
@vic42 what's interesting is that we are both seeing the same effect (around the same time of the day), which I assume is when the sensor is cold and the water is getting warmer, that supports condensation being the main culprit and not water mist, otherwise we would see false readings throughout the day, and more in the afternoon vs only for a few hours in the morning. I assume the ultrasonic transducer didn't exhibit such effects because of the large sensor detection area, I doubt any single droplet will block the entire reception area at any given time. I'm considering a physical cover that opens just before taking the reading, but this comes with it's own complexities. Btw did you see this effect in colder times? Since you had the sensor installed before I did. |
|
Ventilation could be a solution which also comes with its own complexities |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Solar Powered Tank Monitoring System
Purpose
The system was created to monitor the water level of a cistern that supplies fresh water from a well to the house and, to communicate through MQTT to the well pump & valve controller in order to maintain a desired water level.
Design
The cistern's access hatch is covered by a steel lid that has a 20 mm dia hole at the center. The basic design idea was to place an autonomous waterproof box on top of the lid and shoot a laser through a window at its bottom to the water surface, in order to measure the water level. I decided to use a Wemos board and chinese modules because I had only very basic soldering equipment available on-site.
Dimensions of the tank:
Hardware
Materials I bought from the local hardware store
The electronics I had brought from my lab for the project
Upgrade
Build
Operation
I was prepared to send the ESP into deep sleep to save power but surprisingly the dimensioning of the solar system never allowed the battery voltage to drop below 4 Volts so far, while the ESP is still running at full steam with a
TelePeriodof 60 seconds.The diagram shows a week's worth of battery voltage readings.
Conclusion
This device works surprisingly well. I think this design could serve as a template for many outdoor sensing applications. Being the first of its kind, it is built like a mash-up, proof of concept thing, built from the modules randomly at hand. The next one will definitely receive updates to the circuitry as part of a clean design. It remains to be seen how the system will survive the particularly rainy winter time at southern Greece.
All reactions