Flow meter to differentiate between pumping air / water

Hello,

I'm looking for a straightforward, simple and cheap solution to detect if i'm pumping water or air through a small translucant silicone tube. This to cut off the flow when air starts passing through the tube instead of water. I am limited to using sensors on the tubing itself.

I've been looking at IR and ultrasonic options but I wonder if the following is viable too (Have no flow meters to try it with)

The question I have is if a simple inline flow sensor like the YF-S201C would give me less pulses when air is going through the tube than it would when water is passing through? Like that i could stop the pump after x seconds of no water passing through.

Thank you for your thoughts on that :)

T

Hi,

Is the water clear?

There is a definite optical difference between water or air in most tubes. With water, there is a more pronounced lens effect, I believe. That might be enough with a simple LED/optical sensor.

One flow meter technology uses self-heating of a thermistor. When the thermistor is cooled by more flow, the current flowing increases.. I bet that the difference between water and air would be very pronounced.

How long is the tube? What is the pressure?

Can you insert a small thermistor through the wall of the tube and RTV/Silicone it in place??

Let us know what works!

tomdierckx:
I’m looking for a straightforward, simple and cheap solution …
The question I have is if a simple inline flow sensor like the YF-S201C would give me less pulses when air is going through the tube than it would when water is passing through?

It’s very likely, and the obvious place to start. You say nothing about the pump, or what it does and how it relates to the sensor, but it is probably the most important consideration. If it is centrifugal, you can be pretty sure its capacity to pump air is somewhat limited, and the characteristics of the flow sensor may hardly matter. If it is positive displacement, they do, but it is still not likely to be a problem.

I can’t imagine throwing extra hardware at the problem could ever be a good idea. It is easy enough to put a condition in a flow meter programme, and adjust it to your needs. And that may be all you need, as it can enable you to be aware of and address the problem while still pumping water. I use <1 l/min as “no flow”. In your case, you may find a 50% reduction in normal pumping capacity, tells you that you have a problem, or are just about to.

@Terry

Yes the water is clear. The thermistor seemed like a good idea, but I have to take into account the ambient outside temperature and the temperature of the liquid. They can vary a lot so i'm a bit afraid that the margin for error would be to big. Although I wasn't planning to, i'm going to test the optical sensor too.

@Nick

I have a diaphragm pump and if you think i'm going to have a drop in flow of 50% or less between air and liquid, i'm definately going to test it. The flow rate through the tubing would be about 1L/m.

Thanks for the responses.

Measure motor current and compare the values for water and air. There should be a difference.

A water filled pipe is heavier than an empty pipe. Rest a section of pipe on top of a load cell.

tomdierckx: I have a diaphragm pump and if you think i'm going to have a drop in flow of 50% or less between air and liquid

No, I'm saying a 50% drop in water flow should be measurable and cause for alarm, i.e. you know there is a problem before you start pumping air. Again, this depends on the pump, and how you measure it. I imagine a diaphragm pump can be very variable, depending on how it is operated but this is probably not a problem if you are measuring average flow. A spring diaphragm pump like an AC fuel pump can have a pretty steady flow because it is oversized and the rest of the system regulates it.

You can speculate all night about how the sensor measures airflow but, despite all the vagaries about the pump, it is still probably less inclined to react with air, and that should be measurable. So try it...