Water Tank Level Meter
I need some ideas on how to go about this project.
What I want to do is;
Measure the water level on my water tank and display the level as a percentage on a 3 digit 7 segment LED display.
And ideas?
How deep is the tank? (Max measurement value)
Well its a 4500L (1000gal) tank.
I haven't measured its height, but its around 1500mm high.
All i could think of, becasue i want a 100 point measuring system (to count in increments of 1 to 100), was using 13 Parallel to Serial shift registers all going to a PIC.
13 PISO shift registers would give me 104 input points.
Ignoring the top 4, i've got 100 inputs that can easily be displayed as a percentage.
Tank half full= display 50%
I dont want to drill into the side of the tank to get contact will the water, so i need some sort of rod with all the contacts on it.
Then i have the issue of wire being submersed in water. which means i need thin stainless steel wire, lugged onto a screw on something like a plank on wood that stands verticall y inside the tank.
and 100 cables comming out to the shift registers, then 14 wires to the pic (13 data, 1 clock).
There has to be a better way.
I thought of a float on a string that moves up and down with the water that turns a rod with a hall effect counter on it, but that opens the door for a lot of error, and everything has to be precise....
You could use ultrasonic ranging with a PIC
If the tank were plastic, one of these and a clever float/magnet arrangement could be worked out http://www.sparkfun.com/commerce/pro...oducts_id=9072
Mike 2545
back to the ultrasonic ranging..... Does it bouce back off water?
I wish you were not set on not drilling, because differential pressure is one of the most accurate and straightforward ways to measure level. The head pressure on the floor of the tank correlates to the depth of the tank very well. Of course this isn't really something for home-made projects.
But FYI a great company that I deal with quite often called Omega engineering (Sensors, Thermocouple, PLC, Operator Interface, Data Acquisition, RTD) has a lot of free engineering tutorials on various instrumentation techniques including level measurements. You might look at their site for ideas.
Another possibility would be sound waves directed at the surface of the measured liquid.
Bouncing IR off of it might also be feasable, but in that case I would bounce it off the surface at an angle and measure the diffraction with an opposing sensor.
If there is even a slight temperature differential between the liquid and the air in the tank, thermistors are pretty sensitive to sudden changes in temperature, so an array of thermistors (or even thermocouples) along a pvc pipe or something might work well, with each one monitored for sudden step changes as the liquid subsides.
The thermsitor/thermocouple idea might work better with a digital counter as you suggest using.
Hope one of these is a possibility for you.
I bought some of these a while ago...
Ultrasonic Sensors
I think its about time i learnt how to use them. But i have no idea where to start.
Can anyone direct me to a link that can teach me how to use them?
A capacitive sensor could work OK.
Two insulated wire, spaced say 10mm apart will form a capacitor, use that capacitor as one of the frequency determining components in an oscillator.
When the space between the wires is full of water, there will be a change in capacitance and hence a change in the frequency.
Measure the frequency and relate it to the water level.
JimB
Experience is directly proportional to the value of the equipment ruined.
Hi,
here is an approach to measure tank level using air pressure.
This circuit uses a Motorola MPX2050DP differential pressure sensor.
Measuring pressure has the advantage that the sensor can be placed far away of water. It just requires a 6mm (outer diameter) polyurethane hose (as used for pneumatics) to be inserted to the bottom of the tank.
To compensate for possible air losses you might connect a three way valve to blow air through the hose until bubbles rise. This can be done at monthly intervals. The sensor can withstand a pressure of 200KPa without damage.
The circuit uses a precise instrumentation amplifier (INA111) and the necessary circuitry to compensate for sensor offset (sensor offset at zero pressure ±1mV).
For amplifier offset there is an output offset correction circuit using an OPA177GS wired as unitiy gain amplifier.
The amplification can be varied in a wide range with amplification factors of 49.5:1 up to 5000:1 (sensor output voltage 40mV at full range and a linearity error of ±0.25%) and be adaped to any pressure, not to exceed 50KPa (range 0-7.5PSI)
The board supplies its own negative voltage by using a MAX1721 voltage inverter.
The PCB including the sensor has a size of 1.85X1.32inches.
Boncuk
Proper Planning Prevents Piss Poor Performance
I just came up with a new one.
How about a float, fixed in the bottom of the tank, attached to a strain gauge (which can be in the bottom of the tank or even above the tank to keep it dry).
The higher the water level, the more force trying to move that float upward... So that is measured on the strain gauge.
The underground tanks of gasoline stations .... I think that may be 'petrol' in some countries .... use a vertical resistance rod, going from the top to the bottom of the tank ... with a float that rises and falls with the fluid level.
You might be able to duplicate this sensor function by getting a fiberglass, or other non-reactive rod, and wrapping it in a close pitched coil of nichrome wire.
You would have to select a nichrome wire gauge that would give you a reasonable resistance range.
You would have to devise some sort of float that would slide along the coil surface.
Nichrome is inert to some degree, so corrosion should not be a problem.
Also, you would need to design something for the readout. Do you have to have a numeric output, or would an led bar gauge be acceptable?
You only need one insulated wire. Use the water as the other conductor.Originally Posted by JimB
VEGACAP | Level Indication, Capacitive Level Probes for example.
Reply With Quote
