Any water measurement sensor output linear voltage

Thread Starter

fiatuno

Joined Jan 28, 2011
47
Hi people,
I am working on a water level project to check, how much water I have in my tank and I had the idea of using something like the float in the below link, which is usually used for fuel tanks in cars.

http://cgi.ebay.co.uk/ws/eBayISAPI.dll?VISuperSize&item=170817020653

My main problems are that I don't know if I can use such thing with 5V or what Output would this give me. I don't want to use sensors which have a digital output and this is why I am coming up with such idea as I need a linear voltage output, similar to the LM35 like in the below link graph.

http://www.cmb.ac.lk/wp-content/uploads/science20.jpg

I also tought of hooking up a float to a variable resistor, but I don't know If I will be getting a linear graph as someone has told me that, thats not linear even when using single turn potentiometer.

Thanks in advance.

Regards,
FiatUno
 

evilclem

Joined Dec 20, 2011
118
What about a pressure transducer? Many of them are linear and water height is a unit of pressure anyway. The Honeywell 26PC series is just one example of a transducer that would work.
 

Thread Starter

fiatuno

Joined Jan 28, 2011
47
What's the description and dimensions of you tank?

ken
The tank is a plastic tank is very similar to the one in the below link attached:

http://www.plasticboxshop.co.uk/ekm...lastic-storages-boxes-and-clear-lid-801-p.jpg

as regarding pressure transducers to tell you the truth I haven't ever encountered them and I will read some more data on them even though the data I have read, can't seem to understand much as from what I understood I need to pass the water from this sensor to know what is the volume. I can be wrong. Thanks in advance.
Regards,
Fiat Uno
 

KMoffett

Joined Dec 19, 2007
2,918
Be aware that if the tank has sloped sides, water level does not translate directly to volume...if that's what you really want.

Ken
 

Thread Starter

fiatuno

Joined Jan 28, 2011
47
Be aware that if the tank has sloped sides, water level does not translate directly to volume...if that's what you really want.

Ken
Yes I know, but I will be giving only approximate values, so that should not be that much of a problem for me, as this is only used for monitoring purposes to tell you the whole truth. So that is why I had the idea of using a variable resistor, or a fuel level sensor like I have listed earlier.
 

Thread Starter

fiatuno

Joined Jan 28, 2011
47
Can you explain what you mean by approximate? Would 5 levels be enough?...or 10?..or 100?
Ken
10 levels would do great for me as I have a 20 characters in each line of the lcd. So it would do great if i put the bar in the center of the lcd. By approximate I mean that in the programing I will make some if statements that will set ranges. for example, 0-100mV I will do 1 box. and continue like that.
 

Thread Starter

fiatuno

Joined Jan 28, 2011
47
I found a very good video about pressure sensors, but would that need to have a fully sealed tank with no air present in it? as this would be a problem for me as I have an inlet in the tank. Would appreciate if you have an example of how the atmosphere should be to make it wrkable please. Thanks in advance,
Regards
FiatUno
 

Thread Starter

fiatuno

Joined Jan 28, 2011
47
what type of pressure sensor would I be needing as I entered the rs site and found loads of types. I am aiming to have not a system which is very expensive if possible. My tank will be sealed to tell you the truth as I don't want to contaminate the water, but I am not sure if when I put on the lid, there are no leaks of air. Would greatly appreciate if you know of anyone who did something similar, the programing is not the problem for me and the circuitry as I am in an electrical course. But the mechanical is a bit of a problem for me.
 

wayneh

Joined Sep 9, 2010
17,168
You don't want a sealed tank! If you're worried about contamination, look into traps or filters for the air entering/leaving the vessel. A winemaking or beer making shop will have a number of small, DIY options.

Inadvertent failure of venting on industrial tanks causes them to implode when emptied. They're not meant to withstand atmospheric pressure of 15psi on the outside (ie. vaccum inside). A very expensive boo boo. :(

Sorry I've never used a pressure transducer so can't advise. BUT, you'll want one capable of covering the full range of pressure you anticipate for your tank. The pressure due to water head is about 2psi per foot of depth on top of atmospheric pressure (14.7psi at sea level). For a small vessel, you'll need a relatively more precise sensor since your pressure range is not so big. "Gauge" pressure refers to pressure above ambient and is what is normally measured, "absolute" pressure includes ambient atmospheric.
 

KMoffett

Joined Dec 19, 2007
2,918
Last edited:

Thread Starter

fiatuno

Joined Jan 28, 2011
47
I would need to have one which has a 10mV per change and this doesn't seem to have it. Anyone know of something like what i need please and which can handle such small distances please. Thanks in advance and hope there is a solution as I seem to have come to a dead end.
Regards,
Fiatuno
 

wayneh

Joined Sep 9, 2010
17,168
I finally looked at your "tank". I didn't dream it was that small. :eek: Just put it on a load cell. That's how almost all tanks are monitored in the industrial world, until they get too huge.
 

svdsinner

Joined May 17, 2011
39
10 levels would do great for me as I have a 20 characters in each line of the lcd. So it would do great if i put the bar in the center of the lcd. By approximate I mean that in the programing I will make some if statements that will set ranges. for example, 0-100mV I will do 1 box. and continue like that.
And easy, cheaper solution if you don't mind "levels" is to take a ribbon cable, and cut it diagonally so that the longest conductor hits the bottom of the tank, and each successive conductor stops at the next "level" you want to measure. Put 6-12v voltage into the longest conductor and then measure which conductors get signal.
Two ways to do this:

  1. Measure each line individually (would probably need a micro-controller with 10+ free inputs to be feasible)
  2. Tie each conductor (except the first) to the next with a known value resistor. Measure the resistance to get the water level. (Empty = infinite resistance, first level = sum of all the resistors, second level = sum of all but the first resistor which is shorted out by the water, etc.) Depending on how you measure the resistance, you can use identical resistors for each segment, or use logarithmic resistance values, to make your measurement the most reliable.
Note: If you want low cleaning/maintenance intervals, you should use AC for this, and put some thought into wire types since some metals corrode more in water than others.
 
Top