# Pressure sensor for water depth

i am trying to use a pressure sensor to pressure the depth a ROV is in water, this would be between 0-3m. I have bought the following sensor : http://uk.rs-online.com/web/p/pressure-sensors/7191134/
but i think i may have gone wrong. Firstly i know it is an air sensor but the water ones are just so expensive, i was thinking of attaching some tubing to the port and coiling it around the ROV so that it had access to the water but water never got in and came in contact with the sensor. I'm not sure if this would work or if there are any other better ways of doing it? Secondly i am confused on what the pressure will be underwater, i used two different calculator to tell me the depth at 1 metre and they said: 101.337 Kpa and the other 111.377 Kpa. Not sure where to go with this really, i need to pressure the depth and ideally i would want to use an air pressure sensor (for cost reasons), just not sure a) how to waterproof it and b) if it would give accurate readings underwater.

Have a look in book "Practical Arduino" where they do just what you are after.

Air pressure sensor underwater not a good idea but would be OK if you could really get things water-tight. That means everything needs to be IP68. You could use a dip tube and keep the sensor above water.
Make sure you have the right type as most of these come in three forms - gauge, differential and absolute.
You can get hydrophobic in-line filters (used to protect vacuum pumps from water) which let gases through but keep the water out - what you don't want is liquid or vapour phase water getting at the expensive silicon membrane. If everything is totally air-tight, then a tube at depth (several metres) won't let water creep up the tube.

What you are trying to achieve is to measure the difference in pressure between the pressure at the surface and the pressure at depth - this is what a differential sensor does. Off the top of my head, a column of water 10-metres high exerts a pressure of about 1-bar, or 14-psi - I don't work in Pascals.

pressure = depth x density x gravity

density of fresh water = 1000 kg/m^3, salt water you'll need to look up.
so pressure/depth = 1000 x 9.8 = 9800 (in pascal (newtons per square metre) for fresh water

your solution with the hose will work if hose is very pressable.
and the other port should be in a steady hose so not compressable.

sam-r-90:
Secondly i am confused on what the pressure will be underwater, i used two different calculator to tell me the depth at 1 metre and they said: 101.337 Kpa and the other 111.377 Kpa.

Saltwater or fresh? It makes a difference.

Sorry just seen the replies:

Delta_G:
Saltwater or fresh? It makes a difference.

It will be freshwater, for testing it will be a big fish tank. If i get past that maybe then a pond.

shooter:
your solution with the hose will work if hose is very pressable.
and the other port should be in a steady hose so not compressable.

I have got an absolute pressure sensor so it only has one port.

[/quote]

tigger:
Have a look in book "Practical Arduino" where they do just what you are after.

Air pressure sensor underwater not a good idea but would be OK if you could really get things water-tight. That means everything needs to be IP68. You could use a dip tube and keep the sensor above water.
Make sure you have the right type as most of these come in three forms - gauge, differential and absolute.
You can get hydrophobic in-line filters (used to protect vacuum pumps from water) which let gases through but keep the water out - what you don't want is liquid or vapour phase water getting at the expensive silicon membrane. If everything is totally air-tight, then a tube at depth (several metres) won't let water creep up the tube.

What you are trying to achieve is to measure the difference in pressure between the pressure at the surface and the pressure at depth - this is what a differential sensor does. Off the top of my head, a column of water 10-metres high exerts a pressure of about 1-bar, or 14-psi - I don't work in Pascals.

Thank you for your answer. Unfortunately i need the sensor to be underwater, it will be in a waterproof case so that is why i thought of a tube out to the water from the box.

I have a few more questions. So i am thinking of going down the route of my coil idea or thinking about it if the tube is sealed around the pressure sensor then i don't really need to coil it around the ROV i could just has a short bit going from inside the box to the water. Water shouldn't go up the tube if it is sealed well enough round the sensor should it?

Secondly, how do i work out the resolution of my sensor, i.e the error band of measuring a depth?

As i will only be testing in water unto 2 or 3 metres deep i need to make sure it gives relatively accurate readings over that range. (i assume i just set it to zero at the surface then measure the difference to my depth).

sam-r-90:
I have a few more questions. So i am thinking of going down the route of my coil idea or thinking about it if the tube is sealed around the pressure sensor then i don’t really need to coil it around the ROV i could just has a short bit going from inside the box to the water. Water shouldn’t go up the tube if it is sealed well enough round the sensor should it?

Secondly, how do i work out the resolution of my sensor, i.e the error band of measuring a depth?

Even if the water did make its way up the tube you’d still be OK in the short term. The sensor contains a protective gel to prevent liquids/etc. from touching the sensor directly. Personally I’d fill a segment of the tube with a waterproof grease for that extra margin of safety.

The only way to work out a pressure<->depth correlation is to calibrate it; so much depends on the flexibility of your tube and the ROV as a whole that it’s impractical to have any expectation of what the readings will look like. Just set the ROV to 50cm depth, take a reading, set it to 1m depth, take a reading… etc. From there you can use something like MultiMap to extrapolate for the in-betweens.

Going from zero to 3m depth the absolute pressure will range from about 101 to 130 kPa, assuming you're close to sea level. Your sensor has an operating maximum of 115 kPa so it doesn't have enough range for a 3m depth. You'll max out at about 1.5m. And since you'll be using the very top portion of its range the sensitivity will be something like +/-15cm.

Maybe you can push it over its stated limits and get something meaningful out of it. But if it were me I'd look for a different sensor.

You might also need to compensate for the change of water density with temperature if the water can change its temperature. Between 4°C (highest density) and nearly boiling water the density changes appr. 30%. (see http://en.wikipedia.org/wiki/Density#Water).

Dande80:
You might also need to compensate for the change of water density with temperature if the water can change its temperature. Between 4°C (highest density) and nearly boiling water the density changes appr. 30%. (see http://en.wikipedia.org/wiki/Density#Water).

You're off by an order of magnitude.