# Finding the depth of a shallow body of water?

Hey guys, I'm currently working on a project that involves measuring the depth of a small body of water. We're building a small, self-guiding OverSandVehicle to find water, measure it's salinity, temperature, and depth. I have the salinity and temperature figured out, it's the depth that's stumping me. I've been thinking about having a small pole that is braced against a roller move downwards, and turn the roller. There will be a sensor of some sort on the roller and the brace that keeps track of how many rotations the roller makes before it stops, and then we can use that, along with how many revolutions it takes to reach the ground to figure out the depth. The thing is I do not know what kind of sensor to stick on the roller, that will make the counting possible. Can somebody please point me in the right direction?

I'm having a hard time visualizing this. Your vehicle will sit on the sand/bottom of, say, a rain puddle. And it will be tall enough that the body is above the water surface. So you want a sensor to measure the water depth that the wheels are sitting in?

How about a Ping style ultrasonic sensor looking down. You measure the distance down to the water surface and subtract from the known height of the vehicle?

Unfortunately the OSV will not be in the water. The water will basically be in a small (around 4 inch diameter) bowl that has been placed into the ground. One thing I found that seemed like a good idea was using some sort of pressure sensor to calculate the difference between the atmospheric pressure and water pressure to figure out the depth of the water. However I have not been able to find the proper sensors to do this so far.

How deep are we talking? Is it safe to assume that the surface of the water is at the same level as the wheels -- ie: fixed/known relative to the vehicle? Or do you have to measure that also?

If you can envisage sticking some sort of probe down into the liquid, what's the issue with just judging how far you have to stuff it down? If you use a roller running against a chopstick to slide the chopstick up and down, then count how many steps of your stepper it takes to reach the bottom.

gardner: How deep are we talking? Is it safe to assume that the surface of the water is at the same level as the wheels -- ie: fixed/known relative to the vehicle? Or do you have to measure that also?

If you can envisage sticking some sort of probe down into the liquid, what's the issue with just judging how far you have to stuff it down? If you use a roller running against a chopstick to slide the chopstick up and down, then count how many steps of your stepper it takes to reach the bottom.

i agree, if you have a system capable of placing a pressure sensor at the bottom of the liquid in the bowl, you have a system capable of figuring out where the bottom of the bowl is

1) How deep is your water likely to be? 2) How accurately do you need to measure its depth? Would accuracy to 0.1" be close enough?

water sensor on a stick. move to sense water. then extend to the bottom. count stepper steps, turns of a screw or angle of a crank arm

mmmmmmm746: One thing I found that seemed like a good idea was using some sort of pressure sensor to calculate the difference between the atmospheric pressure and water pressure to figure out the depth of the water. However I have not been able to find the proper sensors to do this so far.

No experience with this, but I don't see why a standard atmospheric air pressure sensor with a thin tube connected to it can't measure shallow water depth. ~10 meters of water is the same as atmospheric pressure. Atmospheric pressure varies with ~10%? A rubber tube and an air pressure sensor is also used in washing machines, to fill the macine to pre-defined levels. But these devices work with switches... Leo..

Hi, This may help.

https://ccollins.wordpress.com/2013/12/11/how-to-measure-air-pressure-with-arduino/

http://learn.parallax.com/pressure-sensor-arduino-demo

Tom...... :)

Lower one of these on a stick into the water:- https://www.sparkfun.com/products/10221

I'm a bit pressed for time, so I'm going to go with Grumpy Mike's solution. Thank you guys so much for all of the feedback. I was searching for a sensor like that earlier on, but for some reason I could not find it. Thanks again guys :D. I will also read through what you posted tom, even if I am not using that method I really want to know how it would work.

You never did say how deep your water might be, or how accurate you expected the measurement to be. Those liquid level sensors only measure water depth greater than 25mm. If you needed to measure 5mm of water, it would not give you useful info.

Good luck.

gardner: You never did say how deep your water might be, or how accurate you expected the measurement to be. Those liquid level sensors only measure water depth greater than 25mm. If you needed to measure 5mm of water, it would not give you useful info.

Good luck.

I asked him that because I had the idea of using stripboard to measure the depth. 0.25" to about 3" could be measured that way with an accuracy of 0.1" The resistance between submerged strips would be far less than the resistance between unsubmerged strips. Just detect where the resistance between the bottom strip and the first unsubmerged strip is high (almost infinity).

You only need one pair of contacts. Just connect a long arm to a servo with the contacts at the end; rotate the arm until it detects water. A little trigonometry gives you depth from the angle of the arm.

Or use a linear servo.

Chagrin: You only need one pair of contacts. Just connect a long arm to a servo with the contacts at the end; rotate the arm until it detects water. A little trigonometry gives you depth from the angle of the arm.

I can see that that method will detect where the top of the water is, but how can it detect how deep the water is? You've not shown a way of knowing where the bottom is. I'm not assuming that his 'puddles' are all dug to the same depth, as you appear to be.

Lots of workable suggestions here...

My immediate thought was a 'clunky mechanical float' as mentioned in that resistive sensor at spark fun.

Second thought was rotating a rod down until it hit ground (under the water), where it would then start to lift the unit, i.e. a motor current/strain-gauge/whatever sensor could be used to provide a threshold (you set accordingly). The actual value is just a lookup vector of calibrated depths away (or you could do the actual trig and kinematics of the arm and use a function).

Henry_Best: I can see that that method will detect where the top of the water is, but how can it detect how deep the water is? You've not shown a way of knowing where the bottom is.

Well, yeah, that's a good point :-[

Personally I would have went with Wawa's suggestion. And while rethinking the idea of how to implement moisture contacts my concern would be residual water on the probe messing up future readings -- it makes the pressure sensor that much more appealing.

Chagrin:
Well, yeah, that’s a good point :-[

Personally I would have went with Wawa’s suggestion. And while rethinking the idea of how to implement moisture contacts my concern would be residual water on the probe messing up future readings – it makes the pressure sensor that much more appealing.

Not if you arrange the strips correctly. Each strip, except the bottom one, only going 1/3rd of the way across. Alternate strips starting from different edges. The middle third not having any copper at all. Mask all the remaining copper and spray the board with water repellent. Scotchguard or enamel paint?