-
colemanjj posted an update 5 years, 5 months ago
@neilh20,
I’m interested in your mention of working on a ” Water Depth probe using gas sensor”. I’ve been testing a water depth probe made with a atmospheric pressure sensor (BME280) in mineral oil and encapsulated in a section of PVC with a flexible diaphragm at one end. It seems to be working but I need to calibrate etc. I haven’t seen this idea proposed anywhere and feel like I’m out on a limb with this experiment. Any thoughts?
john
Welcome to EnviroDIY, a community for do-it-yourself environmental science and monitoring. EnviroDIY is part of WikiWatershed, an initiative of Stroud Water Research Center designed to help people advance knowledge and stewardship of fresh water. New to EnviroDIY? Start here
Activity
Hi John – feel free to email me direct on neilh at biomonitors.com as I get it quicker.
comments: generally pressure sensors are pretty easy to understand and calibrate for a water column – but the devils in the detail. What range (eg 0-10′) and accuracy do you want.
1) for the BME280 it hasn’t got an easy mechanical port to isolate the pressure on – but sounds like you are referencing to absolute – and then probably taking the I2C wire out.
I was choosing the MP3V5010 with a dual port – but requires a port to atmosphere.
2) You may want to consider whats the accuracy of the measurement you desire.
3) Calibration is placing it a water column, noting the results, and then changing it. Typically for calibration it might only be a couple of depths. Initially though you might want to do it for 10depths to check linearity
I’ve got the calibration OK and I’m using another BME280 for atmospheric compensation. Both BME280s use I2C.
What I’m unsure of is submersing the BME280 in mineral oil. It seems to work but I haven’t heard of anyone putting the BME280 in mineral oil, or any pressure sensor in mineral oil for that matter, to isolate it from the water. Any thoughts?
The dual port MP3V4010 looks interesting. Maybe I’ll get one and try feeding it mineral oil in one tube (sealed underwater) and atmospheric air in the other. I don’t want to use air in both tubes because the effect that temperature would have on the air pressure in the tube under water.
BME280 – I don’t know. Its a materials/chemistry issue – how does mineral oil react with the materials in the sensor. Probably easy to test – but how to do longer term reliability testing.
Years ago, a Motorola/Freescale product “Pressure sensor” engineer suggested I use a high molecular silica gel for interfacing to its air sensor diaphgram. Silica gel thats not reactive – I’ve got it somewhere if you are interested.
As you identify, the issue is how to have a filler/gel that doesn’t change its volume with temperature. In addition, the gel transmitting the pressure from the water to the diagraphm 1) needs to not dissolve into the water 2) be sufficiently viscous/mobile to transmit the pressure from 0C to 20C.
If you do look at the MP3V5010 I can share the analog amplifier and interface circuit I’ve had good success with.
BTW – whats the range of water depth do you want to measure, and to what accuracy. That defines a lot of parameters.
neilh, Thanks for your thoughts on this.
I’m looking at small rivers and streams. So depth change is generally less than 10 feet, in most cases max change of 1-5 feet. I’d like to get less than 0.1% accuracy and +_ 0.1 inch but I could live with as much as +_ 0.5 inch if I had to.
I’ve been running the BME280 in the mineral oil for about 8 weeks now and getting +_0.12 inch (0.3%) accuracy over, independently measured, 36 inches of water depth change. That’s probably as precise as I can measure given the crude system I’m using for independently measuring the water depth.
that sounds pretty fantastic. Of course doesn’t indicate long term reliability and the issues are mechanically how to connect the water pressure with the mineral oil. I’d be interested in how you are thinking about doing it in a steam with the logger connection is say 20′ above the point of measurement. But its giving me someother ideas. I’m primarily interest in how to measure the low flow side accurately +/-0.005ft/+-/2mm – and the high end over a couple of feet can be more like +/-0.05′ or ~0.5″