Calculating volume of fluid discharged

[user_88]

If your number .... 100 mL .... or 100 cubic centimeters .... per actuation ...is correct, then the ultrasonic T/R might not be the best choice. A graduated cylinder, or anything of such a small dimension might have peripheral interference from the container walls ....The idea is to detect a distinct echo pulse from the fluid surface .......
A larger diameter cross sectional area would be better....have more of a chance of working properly. ...Apparently there is a dispersion of the ultrasonic wave from the transmitter ..... at some acute angle.

If you were to place a transmitter at some point above a 1 cm diameter cylinder, not within it, and generate a pulse, it may be that there would not be sufficient energy in the reflected wave to be picked up by the ultrasonic receiver.

 

[user_88]

This is more like an industrial project for research purposes. The solution needs to be able to withstand quite caustic environments, aka ammonia. In terms of accuracy, the change in liquid would be in the neighborhood of 100 mL per feeding session. If the cross-sectional area of the bin is small enough, the change in distance would be quite noticeable. What do you think based on this new information that I have missed out earlier?

It may be that you want to find a means to determine the height of the fluid column by measuring the pressure at the lowest point .... the base of the fluid column. The pressure of a fluid column is equal to:
(fluid density) * (g ... a gravitational constant) * (height)
or P=ρgH.
The great thing here is that there are a wide variety of pressure sensors to choose from. You should be able to find one that will have enough 'resolution' .... smallest measurement interval ... to be able to detect the 100 mL increments that you anticipate. You will have to decide what form you want the data to take .... Voltage A/D conversion .... or whatever, but that should be a trivial problem.

n.b. : technically, the first sensor parameter that you will want to consider is the 'resolution' or the smallest increment that can be measured.
The term 'accuracy' is related to the consistency of repeated measurements.

 

[Roff]

I just realized that ssylee has hijacked mersenne31's thread. I wonder what happened to him?

 

[ssylee]

It may be that you want to find a means to determine the height of the fluid column by measuring the pressure at the lowest point .... the base of the fluid column. The pressure of a fluid column is equal to:
(fluid density) * (g ... a gravitational constant) * (height)
or P=ρgH.
The great thing here is that there are a wide variety of pressure sensors to choose from. You should be able to find one that will have enough 'resolution' .... smallest measurement interval ... to be able to detect the 100 mL increments that you anticipate. You will have to decide what form you want the data to take .... Voltage A/D conversion .... or whatever, but that should be a trivial problem.

n.b. : technically, the first sensor parameter that you will want to consider is the 'resolution' or the smallest increment that can be measured.
The term 'accuracy' is related to the consistency of repeated measurements.

Sorry for getting "accuracy" and "resolution" mixed up. For accuracy, of course I would want the system to be as consistent as possible. For the resolution, the change in liquid could go anywhere between 20 mL to 200 mL, but for operational purposes, the smallest liquid mass change should be about 10 mL.

 

[user_88]

Sorry for getting "accuracy" and "resolution" mixed up. For accuracy, of course I would want the system to be as consistent as possible. For the resolution, the change in liquid could go anywhere between 20 mL to 200 mL, but for operational purposes, the smallest liquid mass change should be about 10 mL.

Just as an initial approximation...say you have a cylindrical container of
1 cm diameter, and using water as the fluid of interest:
Then a 10 mL change in the volume of the container would equal about 10 cm of vertical height change. The pressure change due to this same height change would be in the neighborhood of 1 kN, or 0.01 bar, or 0.15 lb/sq. in., or equivalently, about 4 in. of water.
This would be the approximate resolution to look for in a pressure transducer.
....just using water in the column.

Do you have a value for the density of the liquid that will be used?

 

[ssylee]

The density of the liquid is remarkably close to water, so assuming that it's water would be fine. Are you suggesting a pressure sensor to sense the pressure change due to the change in the amount of liquid? The container used is around 150 L, while the bin is around 25 cm high, so the cross-sectional area is quite large.

 

[user_88]

The density of the liquid is remarkably close to water, so assuming that it's water would be fine. Are you suggesting a pressure sensor to sense the pressure change due to the change in the amount of liquid? The container used is around 150 L, while the bin is around 25 cm high, so the cross-sectional area is quite large.

... Given this information, the question is: what will the change in vertical height of liquid in the bin be, if the minimum liquid volume amount is removed from the container?

I am guessing that the minimum Δh of the fluid will not be a very great quantity.

Consequently, the ΔP of the fluid at the bottom of the container will hardly change at all...

So, the pressure sensor method is not likely to work.

.... Going back to the ultrasonic transducer, I have doubts about using any sort of home made / microcontroller unit, due to the lack of resolution....
As I recall, the home brew version could only measure to within a few centimeters ....Maybe a commercial ultrasonic level meter could give the desired results for your application.... would cost more though.

 

[user_88]

Peristaltic Pump

There is a type of low flow rate pump that works by repetitive squeezing of a flexible tube. The tubing comes in various types .... depending on what chemical resistance is required. You would have to set up some sort of counter mechanism to measure amount of fluid delivered ... and some way to turn the pump on and off... Not sure about the economics ....
Welcome: TAT Pumps Peristaltic Pump Manufacuring and Sales

 

[ssylee]

I actually got some new information about the apparatus. Even though the tank is 150 L, the ideal flow rate is actually around 1 to 2 L/min. So the change in liquid volume would be much larger than I thought. Given the new information, I guess the change in liquid level would be much more significant that ultrasound may fit as a suitable solution now. The current apparatus uses a pump to pump the liquid out, with the time that the pump is on translates into the amount of liquid consumed.

 

[user_88]

I actually got some new information about the apparatus. Even though the tank is 150 L, the ideal flow rate is actually around 1 to 2 L/min. So the change in liquid volume would be much larger than I thought. Given the new information, I guess the change in liquid level would be much more significant that ultrasound may fit as a suitable solution now. The current apparatus uses a pump to pump the liquid out, with the time that the pump is on translates into the amount of liquid consumed.

To use either the ultrasonic or the pressure method, you have to have a significant change in the height of the fluid level. Even a Δh resulting from 1 or 2 liter effluent from the container may not be sufficient for these methods to work...If your value of 25 cm for the container wall height is correct, then for 150 L of fluid, you would have about 0.6 m² of container area....quite a large area, considering the height of the container wall.

 

[Lead Head]

Just throwing this out there, perhaps an automotive fuel sending unit? I'm not sure how one of those would hold up in ammonia, but from the few I've taken a part, they seem they could read with a decent resolution.

 

[Pepsiiuk]

This is more like an industrial project for research purposes. The solution needs to be able to withstand quite caustic environments, aka ammonia. In terms of accuracy, the change in liquid would be in the neighborhood of 100 mL per feeding session. If the cross-sectional area of the bin is small enough, the change in distance would be quite noticeable. What do you think based on this new information that I have missed out earlier?

Hmm, what are you feeding ammonia? Or was ammonia just an example?

(This is a bit off topic)
If you are using ammonia, how are you planning on keeping it liquid as it boils at -33C (ish) if I remember correctly or are you just relying on the pressure in the tank to keep it liquid? I’m just wondering as you want to measure the change in liquid volume.

(Back on topic)
I know someone has mentioned the capacitive method but I’m not sure about the resistive method. Just like the resistance of a wire depends on its, length, cross section and resistivity of the material etc. Have the apparatus measure the resistance of the fluid across the cross section which will change according to the volume in the tank.

This might have a few implementation issues, especially if it is a metal tank and/or with a non-uniform cross section but it could work.

Mat

 

[ssylee]

Hmm, what are you feeding ammonia? Or was ammonia just an example?

(This is a bit off topic)
If you are using ammonia, how are you planning on keeping it liquid as it boils at -33C (ish) if I remember correctly or are you just relying on the pressure in the tank to keep it liquid? I’m just wondering as you want to measure the change in liquid volume.

I'm not feeding ammonia. It's just that whatever method I use needs to be able to withstand the ammonia gas in the atmosphere in case it condenses on the surface of the meter/sensor.

(Back on topic)
I know someone has mentioned the capacitive method but I’m not sure about the resistive method. Just like the resistance of a wire depends on its, length, cross section and resistivity of the material etc. Have the apparatus measure the resistance of the fluid across the cross section which will change according to the volume in the tank.

This might have a few implementation issues, especially if it is a metal tank and/or with a non-uniform cross section but it could work.

Mat

 

[hypnopotamust]

Hi,

I'm brand new around here but I feel your pain. I run the blending/receiving area for a chemical manufacturer and we receive, store,blend and transfer quite a number of different fluids, mainly oils.

We use a few flow meters for estimates but thats about it. Obviously temperature is going to affect your readings unless that is accounted for. The meters we use can be calibrated, but if we send a different product through one then all bets are off.

Peristaltic pumps would be an option but it sounds like your flow is getting up there for that. In my opinion, any method that relies on measuring volume is prone to variations. We now do all our batch blending by mass on load cells. A pound is a pound no matter what the temperature is. I don't know if your process requires that level of accuracy but in my opinion, you can't beat load cells.

To tell the truth though, due to price, I arrived here looking into methods to determine fluid levels in storage tanks using ultrasonic range finding. I'll not get off into that now though. I can't say what the cost of a "legal for trade" flow meter for your process is but we have about 20 scales that are ISO certified and good to go and the cost isn't too bad... as long as people don't tear up your scales.

Measuring both volume and mass have their pros and cons so I guess it depends on your requirements...measuring from/to single vessel or multiple? Stationary or portable? Dedicated product or diverse? Level of required accuracy? Constant flow or start and stop?

You did ask for volume and if I'm preaching to the choir, we'll blame it on me being new here but I just want to remind you that the coefficient of thermal expansion can pose a problem for some processes.