# Calibrating a pressure sensor using water

1. Aug 12, 2014

### themonk

So I have built an incomplete rig to measure the pressure of a setup I have in a lab. The pressure sensor is hooked up to electronics and I have not used the electronics (basic National Instruments electronics) and am not sure if did the wiring correctly. The setup I have basically has a bottle of water higher up (about 2 m up and can hold maybe a litre of liquid) and then a tube coming down to the pressure sensor. That's how I have it set up but there might be something else I have to do.

I remember making these in a basic physics or chemistry lab a long time ago but am not sure how I should calculated pressure due to height difference of the water. It will give me a reading, but how will I know if it is a correct reading.

Edit: also if this is not the right place, let me know. I could post it in homework, though technically it's not a homework problem!

2. Aug 12, 2014

### TechFan

which sensor are you using?

3. Aug 12, 2014

### themonk

It's a differential pressure sensor.

MPX5100DP by Freescale is the actual sensor being used.

4. Aug 12, 2014

### Integral

Staff Emeritus
About all that can be said is: Neat.

We need more information if you want any help. A wiring diagram and a link to a data sheet rather than a picture would be helpful.

I just did a calibration run on a pressure sensor, seems that we are off by ~200m at 6000m.

5. Aug 12, 2014

### themonk

I can fix the electronics' settings but I was wondering more about the water system setup. Correct me if I am wrong, but basically using the height difference you would know what a difference in pressure should be and then you can check what the reading is.

If I raise the water container (or add more water) the pressure reading would increase, if I am not mistaken.

If I had 1 L (about 1 kg) of water 1 m higher than the sensor, I would get a higher pressure reading than if I had it 0.5 m from the sensor.

Looking more at the equations - with P=Force/Area = 9.81 m/s / "A" where I am not sure what "A" would be, since the bottle exit is a small hole to the sensor, which would measure the difference between the pressure from the room to the pressure coming from the bottle. Perhaps area would be the inner diameter of the tube?

Edit: or perhaps P=Energy / Volume = Potential Energy (grav) / 1 litre = mass*g*height / 1 litre = 1 kg * 9.81 m/s2 *1 m /1 L = 9.81 kg m2/(L s2 ) = 1.423 PSI

Last edited: Aug 12, 2014
6. Aug 12, 2014

### TechFan

datasheet:http://cache.freescale.com/files/sensors/doc/data_sheet/MPX5100.pdf
-page 5 warns parameters are specified for dry air as media. Other types of media may affect its performance. You are using water...
-page 5 shows the voltage output vs. pressure. That if the sensor is Ok at the accuracy stated by the manufacturer, should be better than a calibration performed using an elevated water bottle as standard. You can simply try your experiment and observe the results are within expected values. For approximate calibration you can use a pressure gauge and a manual air pump. Make sure the gauge is calibrated and within specs. That should give you an idea if the sensor is working fine. Do not surpass the max Pr stated in the data sheet, it may damage the sensor. For precise calibration later on you will need a standard several times better than the sensor you are calibrating.

7. Aug 12, 2014

### jim hardy

Pressure = density X gravity X height
and this is why as an electrical type i pity mechanical types - they have to deal with so many more unit conversions than we electricals.

I went through school on English units and still don't handle Pascals at all well.

Pressure of a column of fluid is
Pressure = Density of fluid X local gravity X height , and you have to sort out the units - slugs per cubic foot, pounds-force, et al . I usually wound up with pounds-force per square foot.

For most work , the approximation "a column of water at 68F that's 27.7 inches tall produces one psi." is plenty accurate.
http://en.wikipedia.org/wiki/Inch_of_water
You can correct for temperature, local gravity, and buoyancy of air on your water column if you are being extremely precise. But with that sensor, i doubt it.

You'll want all the air bubbles to come out of your little plastic sensor so put it with fluid connections in vertical plane.
if you're lucky water won't hurt the sensor element .

Deionized water at known temperature makes a great local pressure standard. We used a 300 inch column for calibrating feedwater flow meters, it's much more readable than a pressure gage..

If you're near the equator gravity is ~0.2% less than standard.
The easy way to keep your thinking straight is to imagine a cube that's one (whatever unit you like) on each side filled with water (or any other fluid).
It has a bottom area of one square unit.
You know from its density how much one cubic unit of it weighs.
That much force is applied to its bottom area.
That's the pressure it makes, F/A . (Archimedes)
end edit

old jim

Last edited: Aug 12, 2014
8. Aug 13, 2014

### jim hardy

ps be sure to measure from the center of the sensing element in your transmitter , not the top or bottom of its case.