Cheapest way to measure water depth

[yalew]

I am interested to measure the water depth in a tank above a reference point. The water depth or pressure varies above the reference point from 0 to 2.5 meters. How can this be done in a cheapest way without buying expensive piezometers and data loggers?

 

[Warpspeed]

O/k being serious here. A pressure transducer is a good solution, but the cheaper ones will need some sort of amplifier to give a signal of usable amplitude. There may still be problems with long term accuracy and drift, and the results may not come up to expectations.

A very simple, low cost, and accurate depth measurement can be obtained with a float, a drum, and a ten turn potentiometer.

The required travel is 2.5 metres, the circumference of the drum would be 250mm, or around 80mm diameter. Don't laugh this is an entirely practical solution.

The typical linearity of multi turn potentiometers, even the very cheap ones is excellent, and if you wrap several turns around a grooved drum there will be no slippage. The voltage can be scaled to read directly in whatever engineering units are most suitable.

I have had great success with the plastic coated very fine, stranded stainless steel wire sold at fishing supply shops. This is used by fishermen to prevent fish from biting through the last few inches of line. This wire is light, flexible, incredibly strong and absolutely will not stretch.

Find someone that can turn up an suitable drum on a lathe with about a fifteen turn helical groove that will seat the wire. A light counterweight on the free end, and you are in business.

 

[itsmoked]

I think we have demonstrated the huge range of possibilities here for measuring 'water level'. Now I think it is time for yalew to get-off-dead-center and start being specific to the point that useful discussion can commence.

To wit,
Define:

a) Pressure/depth?

b) Closed tank or open?

c) What data rate do you actually need? You realize all this data needs to 'handled', so never exceed what you 'need'.

d) What is your actual budget? "Too expensive" is not a budget.

e) Are there any threaded ports in the tank below the lowest required depth?

f) What is the tank made of?

g) In what format do you need the data?

h) If you are going to run the entire experiment using a PC, why would you want to add the complication of having to time stamp the data so it has relevance once you bring it to the PC?

i) What power is available?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[yalew]

I would like to thank you every body again for contributing alternative solutions and suggestions. To put it in a nut shell, I am interested in the following alternatives so far:
1. Using pressure sensor as suggest by Riches … thanks Rich! I would be glad if I can buy a pressure transducer with cost of less than $20. Rich can you give me a link or example how I can use the pressure sensor and the data logger together to get the pressure which I want.

2. Using float, drum and potentiometer as forwarded by WarpSpeed. Could you please explain a bit about the practicality of this method?

I will explain the problem as follows for those who are not clear about the problem and need some specific things. Thanks itsomked for your interest and question

1. Accuracy – the set up is not necessary to be very accurate. There is a tolerance up to 10 cm of water.
2. The tank doesn’t have specified control valve to measure. It has some porous side and bottom parts i.e. there are small openings at the bottom even below the reference point.
3. What I want is exactly the variation of water depth in the tank with time at a certain reference level in the tank. If I get the water pressure at a certain point in the tank (net water pressure or to gether with atmospheric pressure), it is simple to get the water depth … elementary fluid mechanics for static instants ?
4. The tank is open tank … just like a pond.
5. Required data … time series of water level verses time for other model input.
6. Budget …. Budget …. I would be glad if I can buy the pressure sensor under $20 as suggested by Riches.
7. The tank is made of plastic
8. The final data format should be in tabular form --- Time against water level in two simple columns.
9. The time series is an input to simple water balance model which accounts the fluctuation of water level in the tank.
10. Itsmoked I couldn’t get your last question…. About power…

 

[Cardoso]

My first thought was the float, with the drum, potentiometer and string as well...

is it possible to weigh the container? Don't laugh, I've seen it in 30 tonne tanks so it would be an option here!

Why not try to wires along the side of the tankwithout touching each other and the tank, but not far apart and see if the capacitance between them changes with the water level. You would still need a bit more than an amplifier to get the signal though

Two flowmeters, one on the input and the other on the output... depending on the flow I think you could manage some cheap flowmeters... :|

 

[OperaHouse]

The name I have always used for that is a PRINCO probe and it helps to have a metal tank. In your case a wire like a 300 ohm TV twinlead would be a good choice if you insulated the end of the wire with something like liquid tape and molded it into an epoxy weight. This would be the timing cap of a 555 timer. Then you measure frequency or convert frequency to DC volts. I think this is beyond what you want to get into unless you can find an old ALBIA DM-8 capacitance module on ebay, this had a DC out.

Now really, what is your budget. I hope you are not wasting our time on a $20 budget.

 

[Warpspeed]

Practicality of the method ?

It could not be simpler.

If your PC has a games port, the potentiometer could quite likely be wired straight to the connector. Just write a simple program in Basic to read the games port, scale the data to suitable units, and print data plus time at the required measurement intervals to the screen, or printer.

A very few dollars and an afternoon, and several cups of coffee should have it working.

Rather primitive, but entirely practical.

 

[richs]

Hiya-

Weeellll, although the game port *MIGHT* work, when I played with it a number of years ago, I was kind of dismayed as to how far the drift was on the one that I played with. It might be worth a shot though.

Here's a link that might help explain the process. Of course you will have to replace the "lab"
equipment with some of the cheaper methods mentioned here.

Hope that this link helps. You might have to paste it together:

http://ceeserver.cee.cornell.edu/eac20/cee331/Assignments/CEE331Lab1_06.pdf#search="bubbler depth"

It should give you most of the information that you need. A good insturmentation amp is also required to get the output of the transducer to a "useable" level for data logging.

There is a pretty good writeup of insturmentation amps at microchip.com, however a quick google search might be better.

Hope that this helps!

Cheers,

Rich S.

 

[LionelHutz]

For the data acquisition?

Try

http://www.dataq.com

DI-194RS - serial starter kit 4 x 10 bit, +/- 10V, <240hz sampling rate with software and cable for $25

DI-148U - USB starter kit with 4 x 10 bit, +/- 10V, <240hz sampling rate with software and cable for $50

DI-148U-SP - same as above but with 2 x pot inputs.


OR

http://www.measurementcomputing.com

USB analog I/O modules from $110
modules with others like 4-20mA from $200