SparkFun Forums 

Where electronics enthusiasts find answers.

Have questions about a SparkFun product or board? This is the place to be.
By jayjay
#6726
Hello Guys

I am looking to build a simple water level detector and exploring the option of a suitable method. The objective is to measure peak level. The sensors need to be around sense around 2-3m range.

1.Use Ultrasonic sensors, but do they go that far?

2.Use infrared tx/rx pair, probably need to factor the reflective nature of the liquid?

3.) Use a pressure sensor with a tube, the ammount of water in the tube affect the pressure of the sensor.

Any suggestions, ideas on which might be the appropriate method?

Thanks
Jay
By LopeD
#6759
is the liquid conductive? i thought of a ghetto method. float an electrode on the surface, and put an electrode on the bottom, then have a lookup table for resistance and water level.
i thought of another ghetto method.
have a long rod that can reach from the bottom of a container, to the top, from some mounting point. at the mounting point you attach the rod to a potentiometer, on the end of the rod you put a lil air bubble thing. lookup table.
By jayjay
#6762
Well the liquid is conductive, but I am not measuring it in a tank, the set up is more like lake/river.

Cheers
Jay
By dpaton
#6770
Have you thought about a capacitive sensor?

The theory is that the level of water that covers a pair of isolated plates (capacitor) will influence it's capacitence, and change the operation frequency of an oscillator. A F-V converter and a zero-and-span amplifier combine to offer a nice linear output of water level. You could also use a small microcontroller to read the frequency directly and output a calibrated analog or digital signal.

I've built the sensors from conformally coated PCBs, and from mylar and plexiglass. I liked the ease of the PCB models better, but I had more success with the mylar/poly/plexi versions in terms of durability and calibration reliability.

-dave
By Norman
#6772
You can purchase time domain reflectometry liquid level sensors from a variety of place.

Cheaper ideas, maybe float a stick on some type of empty bottle inside a pipe with holes. You could measure the the height by either putting holes in the stick and using IR sensors, or something like that.
By riden
#6778
I can think of a few factors that could influence the approach taken.

1) Is it permissible to have a sensor come in contact with the liquid whose level you're measuring? If you're measuring water levels in a lake, for instance, it would okay to insert a sensor in the liquid. However, in a food processing plant it probably wouldn't be acceptable or, at the very least, you would need to be very careful of what you put in the liquid.

2) Is there a lot of movement (i.e. waves, sloshing) that could have an impact on your measurements?

3) How much do the levels vary (millimeters, inches, feet, yards, etc.)?

Ultrasonic ranging will work in the range you indicated but I'm not sure how well the water will reflect ultrasonic pulses and what effect waves and/or sloshing will have on your readings. My second idea would be to float a reflector on the surface and use a laser measuring device to calculate the distance from the reflector to the laser device.

The nice thing is that ultrasonic and laser tape measure devices are cheap and you could do a quick proof of concept using parts from Home Depot or your favorite hardware store. When the concept is proven, you could substitute the device with one that can interface to a processor for recording your measurements.
By SOI_Sentinel
#6782
One note on the above post: I have yet to see anything that actually uses a laser for rangefinding for less than $190US except maybe on Ebay. All those "laser tapemeasures" you normally see for less than $50 at a home improvement store simply use a cheap laser diode to provide a fixed target point, they still have an ultrasonic transducer right next to the LED for the actual distance.
By wiml
#6823
You could use a capacitive sensor like the q-prox or an equivalent handmade one to detect whether there's water on the other side of a tank wall at different heights. If you're measuring a lake, you could put them on the inside of a sealed plastic pipe that's standing in the water.

this is similar to dpaton's idea except using multiple "digital" capacitive sensors instead of one linear-readout sensor. I guess it depends on whather you want to know the actual level or just whether it's above or below certain thresholds.
By transcendentnb2
#6825
Two different "easy" solutions that I can see, but I don't know how much each will cost:

1) Specific bands of IR will bounce off water. This is how aerial mappings of sea depth is done (one bounces off water, other bounces off the ground... they compare the two). That could be mounted on a pole facing down. I haven't used these, so I can't point you in the direction of where to buy it. You could also use the other band and have a floating sensor pointing down, but it would either have to be battery powered and wireless, or tethered. In each case, having a non-submerged sensor would be easier and safer. Wave bounce compensation would be done in software.

2) Even though ultrasonic sensors don't do well with water, you can just have it bounce off a floating device. Rig up the ultrasonic sensor at the top of a pole, and have the floater attached to it, yet free enough to move up and down with the water. Bounce reduction from waves can be done in software, or changing the "RC" of the floater (add more weight to dampen its movement).

The last ultrasonic sensor I used I got through my university. It had a range of 7ft, which should be enough. Use math to figure out the water depth based on the pole height (obviously).

Capacitive sensors would require too much calibration, and dielectric property of the water changes based on other factors, such as pollutants, natural organism count, a duck, etc... plus, it's not KISS.
By jayjay
#6830
Thank you guys for all your valuable input, I just bought a water level reference kit from Freescale, it uses a pressure sensor for detecting the water level. I will play with it for now to see if its usuable. I will also evaluate other methods suggested to see which is the most ideal one. Once completed, I will post some results here :-)

Cheers
Jay