Vertical Float Sensor with 4G Telemetry

This video summarizes various improvements on remote flow detection that I'll be testing over the following weeks including (1) use of float vs. conductivity sensors in acidic environments; (2) use of a Boron (4G) modem as a replacement for the Feather FONAs; (3) use of Adafruit IO vs Thingspeak for logging data remotely; (4) use of webhooks in the Particle IO ecosystem for sending text messages in response to flow events. This video is mostly focused on (1) with future chapters providing more details on items 2 - 4.
Note: In this demonstration, I am using a Particle Argon for testing. The code and demonstration works identically on both the Argon or Boron, but I have poor cell-phone coverage where I live so development takes place using an Argon (over WiFi). The field deployment that will be shared in the next video will be using a Particle Boron for telemetry using the identical code.
Background videos:
Flow detection: • Adruino - Flow Detection
Remote sampling part 1: • Integrating an Arduino...
Remote sampling part 2: • Remote Environmental M...
Links for PVC fittings not found in local hardware store:
1/2" x 3/8" PVC Schedule 40 Spigot x Female Bushing: www.supplyhous....
3/4" x 1/2" CPVC CTS 90° Reducing Elbow (Socket): www.supplyhous....
Search eBay for float sensors and waterproof DC barrel jack hookups

[@michaelgraff6978]

I’m a sensor nerd. Hobby not professional. I do have a lot of experience with microcontrollers, LoRa, and other low power networks. For the simple switch system you are using, I’d see if it’s possible to use a digital input and if there is an internal pull up or pull down configuration for the pins. Most of the micros I’ve used have these, and that will eliminate a component and some wiring. The other nice thing about digital is you might be able to sleep longer. For example, you could sleep for an hour unless you get an interrupt triggered by the pin change interrupt. This would allow you to do wellness check ins but only care about changes when you care about them. For example, once flow is detected switch to a sampling strategy but without any flow poll extremely infrequently.

[@ModestMaker]

Agreed - I have played a bit with digital interrupts as summarized here: kzbin.info/www/bejne/fmOTgnh5Za2aqs0 . Definitely something I want to revisit.

[@otaku2dforlife761]

You just need to make leads on your fork. Censor that are long so you can tie that inside the Box so they don't get corroded you may. Have you know A45 foot lead with an end on it? But that's easier to deal with with the zip ties. Then it is you know have to keep you know trying to fix what you're dealing with. So hard shotter that I'd done and then make the connectors so they fit inside the container. Not to mention that they do make a 2 part Russian or Apache gel type weather proofing substance. So? When you you can turn around and mix some p*** in there and it makes a gel and then you can clean that out later. You can find my amazon

[@ModestMaker]

Thanks very much for the suggestions and ideas.

[@tomran247]

Thank you for your excellent video! High quality work as always. Have you considered an arrangement supplying a divided voltage to your analogue input through a series of resistors to measure the height of the flow? A depth gauge with exposed contacts at regular intervals would sense the water depth and presence. It should be a reliable mechanism, inexpensive and provide some useful additional data regarding flow depth and duration.

[@ModestMaker]

That is an excellent idea- a yard stick would make a great prototype! Let me see what I can come up with.

[@tomran247]

Thank you! I've used an arrangement like this in the past. Some attributes to consider are that the fluid resistance varies depending on dissolved salts and pH and that corrosion between contacts can be reduced by switching the sensing current off when not in use. I tried to gather useful data regarding the pH but it was generally a fraught process because of the constant need for recalibration. It may be worth exploring this idea further if its of interest. I dealt with the issue of variable fluid resistance by using a relatively high sensor voltage of 12v through a good regulator to keep the input voltage level as static as possible. Then I programmed fairly broad tolerances in to the analogue read of the micro to accommodate fluid resistance variation so the level increment output was fairly consistent. The sensor was switched on and off by a triac to minimise corrosion but its a good idea to let the sensor stabilise for a short while after turning on to prevent erroneous readings.