How to measure water level in a tank?

[dk99]

how to measure water level in a tank??

How to measure water level in tank in the cheapest way?
The first thing that came to my mind is pressure sensor... but it's ruled out because it's expensive.
Can somebody suggest a cheaper mechanism?
And it is preferred to have a low power mechanism...

 

[FredGarvin]

A stick.

 

[brewnog]

Can't beat Fred for simplicity, but may I suggest a sight gauge?

 

[minger]

A stick.

haha, the first thing I thought of was a ruler.

 

[S_Happens]

The best method depends on a few things. How big is the tank? What type of tank is it? Does it need to be a permanent real-time indication, or only at intervals (we have some inventory tanks that we only check once a month, so a simple tape measure with weight is sufficient)? There are more questions, but hopefully you get the picture that it's all in the details.

Like mentioned there are different types of direct level indication such as a dipstick/floating gauge, sight glass, etc. which are much cheaper than a sensor/transmitter.

 

[Gokul43201]

A suggestion from a complete outsider to this field (therefore more of a question to the insiders): if you needed real time, continuous monitoring of the level (rather than just spot checks, which could be done with a measuring stick taped to the inside wall of the tank), how about a simple cylindrical capacitor built out of copper tubes sitting vertically inside the tank, whose capacitance is naturally a function of the amount of dielectric (water) filling the space between the electrodes?

Or for a discreet solution (depending on the accuracy required, this could be more than sufficient), I suppose a series of magnetic floats would work just fine.

 

[FredGarvin]

Both of those are pretty standard from my experience. The discrete sensors can get you into a little bit of trouble if you are not careful with how you deal with levels in between sensors. We had a test cell tank overflow because of that reason (coupled with some control logic that had issues).

 

[minger]

The OP hasn't gotten back to us yet; maybe we're thinking of the wrong kind of tank. Perhaps he's thinking of a military tank going through some body of water, in which case I would say the best measurement tool is the wetness of my socks.

 

[dr dodge]

it is real hard to beat pressure sensors for long term reliability, accuracy, and ease of installation. our depth sensors just hang in the fluid. there may be simpler ways (stick) but the stick looses its cheapness, when the stick reader is attached. there is also the possible risk to the stick reader faiiling in, and if he needs to read said stick multiple times a day, climbs an top of a tank, has to monitor deadly chemicals, or gets paid $30 an hour, then the pressure sensor gets more cost effective pretty quickly. real time stick reading over short distance is pretty cheap (shouting reading) compared to cabling, but they are only good for ear-shout distance (unless shout-repeaters are used). float assemblies and any other products that have moving parts need to be cleaned and maintained, where the presssure sensor just sits there and reads. in deep or irregular tanks the design and installation is simpler than a float system. the variable capacitance idea will have problems where the fluid is not consistantly homogenous, or is dirty or corrosive. the pressure sensor is much more forgiving.
with the good old 4-20ma sensors, ease and accuracy of calibration is pretty hard to beat, too

many people are replacing various other systems with depth sensors because their economy proves themself in the long run
I am a little biased because we make them
but I can also see the merits of the stick and sight tube

dr

 

[Doug Huffman]

Tap the tank wall and listen for the change in the sound.

 

[dr dodge]

watch for sweat on the tank

dr

 

[Doug Huffman]

Hat tip to Dr. Dodge.

 

[S_Happens]

Pressure (or most likely differential pressure) sensing is by far the most common way of monitoring levels in industry. Dr Dodge brings up a lot of good points, although the $30/hr technician (even if he does "strap" a tank) is doing much more than a single level measurment.

Like I said, it depends on the size, type, and application of the tank and whether the level indication needs to be monitored remotely or locally. I've dealt with everything from a simple "sight glass" rigged up from plastic tubing all the way to radar (and nuclear sources, but not in any water service).

 

[dr dodge]

differential pressure is actually more correct.
but then there is no such thing as "non-differential" pressure, because it always has a reference
weither its absolute (ref'd to 0 pressure) gauge or negative gauge (ref'd to atm), or high line delta P, its always got to refer back to "defined zero"
This is the hardest thing to get across to people during training
example of actual phone conversation, repeated at least once a month:
customer:I need to go to -14.7 psi
but your atmospheric pressure is 14.5 so you can't do it
do I need a bigger vacumn pump?

(dr dodge picks up hammer and starts beating himself in head until hammer is broke)


dr

 

[dk99]

I'm sorry for replying so late... the major problems involved are that the tank is pretty small... and i the content of the tank is not always water so i couldn't consider the option of using that water as a dielectric...(because the contents are different... and I'm only concerned with the level in the tank). The reason for not considering conventional options is... we don't want to power the mechanism continuously... lots of stuff involved... that's why i wanted to know if there is a general mechanism... which could be modified according to our needs :)

 

[FredGarvin]

Is the tank pressurized or is it an open tank? Have you thought of a basic float lever system?

 

[S_Happens]

If you don't give any details then you won't get a good answer.

By "not always water" what do you mean? Will the density be significantly different? If so, then you're limited to a direct level measurement that will be unaffected by changing composition (like a float as mentioned or sight glass). Again, does it need to be remotely monitored and a signal sent somewhere or can it be a completely unpowered indication that somebody can simply look at. What size is small and what is the tank being used for? Luckily Fred already repeated my question about type of tank and hopefully it will be answered.

but then there is no such thing as "non-differential" pressure, because it always has a reference
weither its absolute (ref'd to 0 pressure) gauge or negative gauge (ref'd to atm), or high line delta P, its always got to refer back to "defined zero"

Semantical, but when I say differential pressure I'm talking about a sensor with an actual DP cell that is referenced back to another point in the process/equipment not simply the calibration range (which is certainly important) of a single pressure sensor. We typically use DP sensors for levels because our references are almost never atmospheric or constant.

 

[sophiecentaur]

There are systems using sound.
Take a vertical tube, open at the top and extending to the bottom of the liquid. The column of air will resonate at a fundamental frequency (easily measured) depending on the level of liquid. Or you can measure the time for an ultrasonic pulse to bounce back from the liquid surface. I think you can buy systems based on these methods.
It depends on how much money you want to spend.
You can have a float on a string wrapped round a shaft. The shaft can turn a potentiometer or optical encoder - or just work a needle on the end of the shaft.
It really depends on the actual application - as do all engineering solutions.

 

[zgozvrm]

You say the tank is "pretty small." Depending on how small it is, you could mount it on a scale and determine the volume by weight (assuming you know the density of the liquid).

 

[MThornton]

run a small air-compressor into the tank vent for a specific time (ie 10 sec). The air-pressure rise over that time will be inversely proportional to the volume of the air-space in the tank, and directly proportional to the fluid volume.

 

[eindon1]

Very simple: (Takes longer to explain than to do) Measure and cut a piece of 1/2" lightweight(thin wall) pvc the length of the tank or where you want the "full" level to be. Glue & cap this pipe at both ends. Cut another slightly larger pvc pipe at least half of the measurement of the capped pipe, make sure the capped pipe will fit into this pipe. Drill holes into the larger pipe every 2" or so. Attach this to the tank so it does not move. Insert the other smaller capped pvc pipe into the larger pipe. It Floats - Duh.

To measure gal - divide the gal by the number of inches and you have the number of gal per inch. Use a marker for your specification. Example: 55 Gal tank that is 44" at full level gives me 1.25 gal per inch ~ Every 4" is 5 gallons. Mark the float accordingly.

 

[Hologram0110]

If the tank is for liquid, one easy way to do it is to use a glass (or other clear material) pipe on the outside of the tank. Here is how it works:

Connect the clear pipe at the bottom of the tank on the outside. Since the liquid can flow into the tank it will arrange itself so that it has the same head pressure at the bottom of the tank. If the pressure at the top of the tank and the tube is the same, this will show the height of the liquid in the tank. This is really easy if both of them are atmospheric you can just have them both open to the atmosphere. If they are pressurized then you need to connect the top of the pipe to the top of the tank.

Alternatively, but more difficult, is cap the end of the pipe so it is air tight. The liquid will still rise to compress the trapped air but you now have to account for the difference in gas pressure acting on the liquid so it isn't as easy.

If the shape of the tank is irragular, you can then calibrate the height of the tank with volume of the liquid either by measurement or calculation. Just take a sharpie and draw some lines on the level gauge.

This basically amounts to making part of the tank clear so you can simply see how much is left.

 

[dr dodge]

to make a simple site tube, all you need is clear tubing, and set it up like a siphon, with an extra loop at the end. no plumbing required

dr

 

[Hologram0110]

to make a simple site tube, all you need is clear tubing, and set it up like a siphon, with an extra loop at the end. no plumbing required
dr

While simiple, wouldn't this require you to fill the tube each time the fluid level reaches the ends of the tube and let's air/gas in? This would make a good short term solution though.

 

[dr dodge]

as long as the "sump end" never uncovers from the water, It will continue to function. If the tank were completely emptied, it would need to be primed each time

and its not ends (pl) as the other end has to be open to atmosphere to work

dr