Water Level Detection

[Boncuk]

Bubble tube level gauge:
Got to disagree with you about air tight, if you have small leaks in the bubble tube, as long as you have bubbles coming out of the tube from the bottom, you will still have a good reading until the air flow is so great as to create pressure drop from to much air flow. That's one of the other nice things about bubble tubes.
Kinarfi

Sorry, but I didn't mention a bubble tube in any of my posts.

Any static pressure sensing device works only reliably if it's absolutely air tight.

A bubble tube is not a sensor, just a device to prepare for sensing. Making a sensing element using a bubble tube requires sensors along the tube either working capacitive or magnetic (e.g. floating magnet and hall sensors along the tube), increasing the bill of material considerably. The more accurate you want to measure the more sensors you'll need.

Addionally it will require an aquarium air pump at least.

A sealed system measures water heights accurate to a Millimeter, depending on the resolution of the A/D converter. At 12bit conversion the lowest digitized value will equal 1.2207mm for a total of 5m water column, which gets down to 0.076mm using 16 bit A/D conversion.

Using a common MCU with 10 bit A/D conversion the accuracy will still be 4.88mm per bit.

Boncuk

 

[premkumar9]

Yes!


If you are interested in a PCB containing a TL074N wired as instrumentation amplifier with a gain adjustable from ~50 to ~5000, and and offset compensation circuit please PM me. I'll send you the schematic and PCB layout (single sided) via email. Board size is 2.275X2.1 inches.

Boncuk

From the catalog of MPX2050 you sent, it is seen that the max pressure that can be applied is 200kPa. 20ft of water gives 161kPa.(calculated using **broken link removed**) This may give a O/P of 0.8 x 161 = 128.8 mV if sensitivity is taken as 0.8mV/kPa. But the full scale span is given as 40mV. Does it mean that P1 - P2 is to be limited to 40/0.8 =50kPA. In that case what can be the maximum depth I can measure with MPX2050?
Pls correct me if the above statements are wrong.

 

[Boncuk]

Hi premkumar9,

the datasheet says 200KPa is the maximimum pressure to be applied to the sensor without damaging it. Remember, also semiconductor type pressure sensors have a kind of diaphragm which might burst at excessive pressure.

A friend of mine ruined a 50Pa sensor inadvertantly connecting compressed air (8bar, approximately 800KPa) to one port. The diaphragm just said "good by, cruel world" with a faint click.

A 50KPa sensor will be far beyond measuring range at this high pressure. The 50KPa sensor can safely measure water levels of 5m height resulting in 49KPa.

In other words 1mm of water height equals 9.8Pa. (For a good estimate just use a multiplication factor of 10 when converting mm of water height to Pa.)

50KPa is the full measuring span of the MPX2050DP where it has an output voltage of 40mV.

Since the output voltage increases linear all you need is an amplifier to make the signal either drive a Voltmeter (displaying water height in m/ft or any dimension you want) or use it for A/D conversion in an MCU.

A good instrumentation amplifier is e.g. a BURR-BROWN type INA111. It is programmable for gain factors of 1 to 10,000 and has a very low offset of 500µV and an offset drift of 5µV/deg/C max. You can purchase it in an 8pin plastic DIP or 16pin SO-16 package.

If you don't like the lowest 500µV offset voltage you might connect an OPA177 OpAmp for offset compensation.

I'd also plan for an input zero setting circuitry since at 0 pressure the output voltage of the MPX2050DP will be somewhere between -1mV and +1mV. You can't trim that offset by offset trimming of the amplifier.

Before you purchase a relatively expensive INA111 (approximately US$10) you could try using a TL074. Three OpAmps make an instrumentation amplifier and the fourth one can be used for offset compensation. (The TL074 is 15Cents)

Calibration of the circuit is easy. Use the PU-hose you want to use anyway. Mark it at several distances and dunk it water (the other open end connected to the pressure sensor) and measure the output voltage of the amplifier. Doing five different measurements at equal distances, e.g. 1 .. 2 .. 3 .. 4 .. 5m water depth and note the measured values. At 1m the readout from the sensor shoud be 9.8KPa which corresponds to a sensor output voltage of 8mV, stepping up 8mV with each m of increasing depth.

Calculate the gain resistor of the amplifier according to your requirements and measure the output voltages the same way. They should be linear as the sensor output is.

Regards

Boncuk

P.S. If 50KPa does not suffice you might select an MPX21000DP (100KPa) or an MPX55000DP (500KPa)

 

[premkumar9]

Hi premkumar9,


In other words 1mm of water height equals 9.8Pa. (For a good estimate just use a multiplication factor of 10 when converting mm of water height to Pa.)


Boncuk

Than you for your detailed explanation. The calculation table I sent is wrong? It gives 150kPa for 5 Ms of water height.

 

[premkumar9]

Yes!

If you are interested in a PCB containing a TL074N wired as instrumentation amplifier with a gain adjustable from ~50 to ~5000, and and offset compensation circuit please PM me. I'll send you the schematic and PCB layout (single sided) via email. Board size is 2.275X2.1 inches.

Boncuk

Please send the details to me.

 

[Boncuk]

Mine says: 5m of water height equal 0.49 bar which equals 49,000Pa or 49KPa.

I don't know which table you are referring to. Please post a link.

 

[premkumar9]

Mine says: 5m of water height equal 0.49 bar which equals 49,000Pa or 49KPa.

I don't know which table you are referring to. Please post a link.

**broken link removed**

 

[Boncuk]

Please send the details to me.

If you forward your email address I'll certainly do.

I've changed the design to compensate for sensor input offset voltage ± 1,5mV (zero setting). I couldn't avoid making the board design double sided.

 

[premkumar9]

If you forward your email address I'll certainly do.

I've changed the design to compensate for sensor input offset voltage ± 1,5mV (zero setting). I couldn't avoid making the board design double sided.

premkumar9@gmail.com

 

[Boncuk]

on the way.

 

[premkumar9]

**broken link removed**

Is this wrong?

 

[ericgibbs]

Is this wrong?

hi,
It looks wrong to me.!

Some of the conversions are 'wacky'.

 

[premkumar9]

on the way.

I have another doubt. If we fill a water tube with water, close one end keeping the other end at the bottom of the tank, lift the closed end to a height of 15Ms, then what I thought was the water level inside the tube will go up to 10 Ms max only (as a water barometer). Can you pls tell me what is wrong in the above statement?

 

[tytower]

I saw one in a shop today for $189 ,a floating transmitter and a small receiver . No idea how it works but they are expensive arn't they

 

[ericgibbs]

I have another doubt. If we fill a water tube with water, close one end keeping the other end at the bottom of the tank, lift the closed end to a height of 15Ms, then what I thought was the water level inside the tube will go up to 10 Ms max only (as a water barometer). Can you pls tell me what is wrong in the above statement?

The air pressure at sea level is about 14.7lbs/inchsqrd, which roughly equivalent to a water depth of 32ft, ie: at 32ft the pressure is 14.7lbs/inch sqrd above that at the surface.

If you fill/immerse a tube completely with water and seal the top end and then raise the tube from the water, no water will run out of the tube until the end of the tube comes out the water.

If you have a tube full of air, with one end sealed and immerse that in water, a little water will rise in the tube until the air is compressed to ambient 14.7lbs/sqin.

You cannot compress water.

 

[Boncuk]

I have another doubt. If we fill a water tube with water, close one end keeping the other end at the bottom of the tank, lift the closed end to a height of 15Ms, then what I thought was the water level inside the tube will go up to 10 Ms max only (as a water barometer). Can you pls tell me what is wrong in the above statement?

If you lift the filled tube 10m above tank level you get the equivalent suction value of the water column being trapped in the hose, provided the lower end of the hose is still under the water level.

But why would you want to build such a difficile measuring device?

Stick the pressure tube inside the tank after it's connected to the sensor. That's all what is required for exact measurements. If you are worried about leakage in the PU-hose you might connect a three way valve and blow air through the tube from time to time until you observe air bubbles coming from the bottom of the tank.

Just observe that the hose is always closed overhead the tank before dunking it into the water, either by the connected sensor or a shut-off valve. Not observing this will make the water rise inside the tube to reach tank water level and the resulting air pressure will be zero.

To find out more about the reliability of PU-tubing I suggest to have a look at Festo Deutschland - Home. Search for PU4. You'll find technical desrciptions, similar to an electronics part datasheet.

The weakest member in a pressure line is not the hose, but the connectors. Using a inner diameter calibrated hose for a clamp-on (quick connector) connection is the worst choice. Using an outer diameter calibrated hose the connection is 100% airtight and leakproof.

Especially in semiconductor manufacturing and packaging processes pneumatic tools are used. They are faster than any stepper if speed is an issue and strenght (power) depends on three factors: applied air pressure, piston diameter and amount of air to fill a cylinder within a certain time.

Festo also offers pneumatic tools interfacing perfectly with MCU control.

I can tell from personal experience that Festo products are very reliable. e.g. electromagnetic air valves. Within 18 years I never experienced a burnt up coil which did using Webber valves combined with other malfunctions such as cut off rubber sealings caused by sharp edges of the port they had to close.

Boncuk

 

[Boncuk]

I saw one in a shop today for $189 ,a floating transmitter and a small receiver . No idea how it works but they are expensive arn't they

For "cheaper" sensors check out Furness controls.

 

[premkumar9]

If you fill/immerse a tube completely with water and seal the top end and then raise the tube from the water, no water will run out of the tube until the end of the tube comes out the water.

Sorry, I am still confused. What is the difference between this and a water barometer where level will raise up to 10 Ms only?

 

[Boncuk]

Basically there is no difference, except with the water column inside a tank and the hose reaching to the buttom you will measure pressure, with a lifted hose you will measure vacuum. Pressure will vary with water height, but vacuum won't vary since the reference point is outside the system.

Think about a drinking straw. What happens if you dunk it in any liquid and close the upper end with a finger and then raise if off the glass? Nothing will happen! Not even a droplet of liquid will escape because it is held by suction (caused by the closed top) and capillar force. You can move the straw to high altitude and the true negative pressure won't change. You might refill the glass or empty it completely. The drinking straw and the vacuum in it won't change a bit, because it has lost it's reference point.

Raising a hose outside the tank to measure the same (negative) pressure as it is with the hose submerged in the tank is just impracticable. Additionally the result is only valid for a "snapshot". Changing water levels won't change the measuring result anymore.

How would you raise the water level in the hose if you have lifted it in a tank half way filled? Increasing air pressure as a result of higher water pressure is the most uncomplicated way to use pressure sensors for water level (height) measuring - and it happens fully automatic.

What prevents you sticking the hose into the tank?

 

[tariq7868]

i am also thinking of making water tank level indicator... but i want to make it as cheap as possible... so am thinking of making switches like in the attached pic. ie water will make the ball attched to the lower strip float and make contacts (simple ?) ..which can b placed at different levels... how about tht??