Don explains the theory behind the depth sensor for the Riffle

by laurenrae | 24 Nov 15:24

What I want to do

In New Orleans, we have major issues with flooding and stormwater runoff, which contributes to the over-pumping of our soils to keep us dry. We want a create a depth sensor that would allow us to measure the depth of the water table around green infrastructure projects such as rain gardens. We came up with a working prototype that could be buried in the ground.

There are lots of other uses for a digital depth sensor and data recorder. We could put these in rain gauges, streams, and ponds. They could record changes in depth from tides at coastal restoration projects. We could do sooo much! But first, we're going to have to do a lot of field testing.

My attempt and results

We applied this theory (video link: http://youtu.be/-yEr4WnvxFc) and built a working digital depth sensor. Check out more about our experiment in Scott's research note: http://publiclab.org/notes/eustatic/11-23-2014/riffing-on-the-riffle-the-youtube-files

Questions and next steps

We think that temperature might effect the readings we get from this sensor. We need to test this further and develop a calibration formula to compensate for changes in water temperature.

Why I'm interested

I'm very interested in developing low-cost water quality monitoring tools. Conventional tools are extremely expensive. The more money we have to spend on sensors, the less we can spend on construction of projects. I want to help create a tool that will demonstrate how well green infrastructure stormwater management projects work, or alternatively, help designers build better projects if what they've built isn't working as expected.

That is really an interesting idea to measure water height with capacitance and a 555!

Continuing to use the same oscillator circuit for depth and conductivity is super awesome! I support whole heartedly.

In the 555, pulse time in stable mode is equal to t=RC*ln(3), where R is in Ohms, and C in Farads. assuming a minimum listenable frequency of 10,000hz, and a capacitance of 0.4nF (0.00000000004F), I calculate we'd need a 2275598Ω resistor, which is pretty close to 2.2MΩ. That is a standard part! not bad.

this is based on this probe listed in scott's note, right? http://njhurst.com/electronics/watersensor/

clever design. I'm really into running the frequency into a listenable wave. I'm realizing the 10000hz, which is listenable, won't work over the phone. 2000hz would though, which would require a 10MΩ resistance.

Is this a question? Click here to post it to the Questions page.

Yeah, this is related to the "Riffling on the Riffle note." The Arduino recorded the wave for us. We just need to figure out a formula to covert pulse data to a depth measurement. We did the conversion for temperature, but not for depth.