Wondering if anyone can point me in the right direction to approach what I want to achieve. I have done some Arduino projects, but by no means I consider myslef to be an expert....
I would like to make something that measures air volume. What it comes down to, is measuring the time it takes for a certain amount of air, in this case 0.25 liter to go through a certain area size of cloth, 38.5 cm2.
(This is to measure the porosity of a fabric, like parachutes, sailboat spinnakers, hot air balloons, paragliders, etc.)
Are there any sensors out there that enable to measure air volume?
What other approaches could there be?
My first idea would be to use pressure sensors. Using the Bernoulli Equation you can relate the pressure difference across your cloth to the velocity and flow rate through it. There are certainly some constraints and assumptions that may or may not have to be considered, but it could give you could comparisons. Pressure sensors are remarkably precise and easily interfaced with if you are familiar with the basics of serial communication.
Personally i think the pressure method is your best choice and also the simplest. I presume, you're not looking for accurate readings, but just the relative permeability of the cloth.
If this is so, then a tube with the cloth in the middle, pressure sensors on both sides near the cloth and compressed air on one side. The other end closed.
If you apply a finite amount of pressurized air on one end and constantly measure both sensors you should get a curve of pressure equalization together with the time.
I'm not sure how absolute the readings would be but you'd get a good idea about the permeability. But still, you'd have to consider the air volume & pressure, tube diameter, cloth porosity etc. to get decent readings.
What @dave-in-nj says sounds reasonably practical.
You probably want to have one reference cloth (or maybe a clean metal mesh) as a standard and compare the others to that. Relative values would be much easier to acquire than absolute values.
I reckon you should be think of mass-flow rather than volume flow as volume will be affected by all sorts of things. For example with a more dense cloth the pressure difference required for a given flow will be greater which means the temperature on the high pressure side will be greater (unless you have a very sophisticated constant temperature system) and the temperature drop through the cloth will be greater due to the greater pressure drop.
If you are not familiar with the subject you probably need to get thermodynamics and fluid mechanics textbooks - which are not light reading.
There may also be useful information in HVAC textbooks - forcing air through ducts - and it may be more accessible (though I have never read any).
In re-reading your origional post, you want to measure the flow rate, not volume.
if you have a piston with a displacement of the volume you want to measure, you can measure the pressure, then by keeping pressure at a constant, move the piston to displace that volume.
measuring time would be the variable that would then be able to be plotted.
I agree with Robin 2 that thermodynamics can come into play. however, SAD or Standard Air Density of 29.92" in of Hg at sea level at 70°F and 50% RH is a rather elusive thing. using a reference as he mentioned would be the most practical method of calibration. metal cloth or an ASME nozzle ( a calibrated hole in a plate)
I believe my previous version would be easier to create and maintain as well as allow easier operation.
the released gas should not be allowed to directly impact the sample, it should be diffused to that any jet would be eleminated before the air touched the sample. a simple baffle plate would suffice.
0.25l of air is not a measure of the amount of air at all. 0.25l of air at a certain pressure will correspond to a certain mass of air. Volume is not used as a measure because volume is never really constant. There's law of conservation of mass, never a law of conservation of volume simply because volume changes all the time. For your application, you should consider the use of a weighing scale. Here's a bigger picture:
Make a weighing scale with a load cell (or a few load cells).
Place the cylinder of compressed air on the weighing scale.
Measure the output of the load cell.
Open the cylinder valve and start the timer at the same time.
As the air is released you should see the output of the load cell change.
The mass of air you want to dispense/release should correspond to a delta (load cell output).
Monitor the output to see when the delta is achieved.
Switch the valve and the timer off.
Depending upon your setup, improvements can be made to ensure all the air has gone through the fabric and accuracy is improved to the max. possible extent. FYI, load cell was just one idea, you could use whatever suits your pocket/needs
Hi, you could calculate your volume by multiply your area by your velocity. You would first have to have an orifice (ie. a known hole in a plate, or your test patch size) Then calculate your velocity based on pressure drop across that area. You would require two pressure sensors or either side of your known area and you would require some math. All things that your uno could easily handle
Calculation for volume = Vol = AreaxVelocity
Calculation for Velocity = Vel = 4005 x (sqrt) Delta P
ACremers:
Hi, you could calculate your volume by multiply your area by your velocity. You would first have to have an orifice (ie. a known hole in a plate, or your test patch size) Then calculate your velocity based on pressure drop across that area. You would require two pressure sensors or either side of your known area and you would require some math. All things that your uno could easily handle
Calculation for volume = Vol = AreaxVelocity
Calculation for Velocity = Vel = 4005 x (sqrt) Delta P
4005 is only the correct number for a Pitot Tube. a Pitot tube uses static pressure and total pressure or velocity pressure.
orifices will all come with their own calibration number as should any other pressure type devices.
mass flow sensors are more accurate at low flows and pitot tubes are more accurate at high flows.
as an example NASA uses Pitot tubes on air planes. and with their budget, they can buy anything they want.
You need a calibrated point in your setup. I'd suggest a piston setup pushed by an actuator moving at a given speed to to calculate the exact volume moved during the test. The size of the test orifice will determine the flow rate thru the test fabric orifice. Sensitive pressure sensors could be put on both sides of the fabric test orifice to determine the resistance to flow thru the material, which should be a function of the fabric porosity. The tough part is to start with a calibrated standard for the measurements.
dave-in-nj:
Look at the fan laws. that will help you use pressure to calculate flow.
I would measure flow of the compressed air,
let the compressed air enter a chamber
the chamber would have one end open with the cloth to be tested
increase pressure to some level. measure the flow into the chamber.
change your cloth, increase pressure to that same level. measure the flow.
if you know the flow into the chamber and the resistance, or pressure drop across the cloth, you can determine the resistance.
This is also a good idea. The only problem i see is that a duct anemometer sensor is hard to come by. I'm guessing he wants to plot the data trough his arduino. Pressure sensors are somewhat easily obtainable.
But otherwise i agree. Your method would produce cleaner results.
OK, the trick here is how to arrange for either a constant volume, or a calibrated volume. Big pistons and such are probably going to be inconvenient. An air pump, such as in this instance an aquarium pump - would be more manageable.
So you need a reference. The "standard" reference in this case, is called a pneumotach consisting of an array of screens (fine metal mesh) as a "calibrated" impediment to the flow with a pressure (and temperature) sensor on each side which you would place in series with the airflow to your test material, having a pressure sensor on each side of that also. Clearly, one pressure sensor is in common.
In terms of calibration, to the extent that this is required, you may wish to servo control the pump (easier if it feeds a large reservoir such as a plastic drum) to deliver a constant pressure. In the more general sense, you are comparing the pressure differential of your test material to the pressure differential of your pneumotach.
More practical might be to make ones own calibration standard. Use a consistent air source and develop a standard using fabrics of varying desired qualities as a base line. Then one could monitor the DP across the fabric and compare this to the DPs of the various that were previously developed. The last time I tinkered with air flow was when the fuel injection mass air flow monitor in my truck started failing. Interesting how they operate in vehicles.
zoomkat:
So how do you maintain constant flow from the fan? How do you know if the flow is constant without measuring it?
What I had in mind in Reply #6 is to arrange for the temperature and pressure upstream of the test piece to be constant. Then measure the pressure downstream of the test piece and compare it to the pressure downstream of the standard test piece.
I think that is what is in @Paul__B's post - though I think "solved" may be ambitious.
It also sounds like it is consistent with your Reply #16 - which I assume doe not involve a piston.
Robin2: The OP seems less interested than we are (as usual).
Sorry guys, I've been away from a computer a couple of days. So, certainly not less interested!
I'm overwhelmed with the responses so far, and my brain is going in overdrive right now.
Certainly different angles to it, and some things to consider. Certainly for what I thought was possibly going to be a relatively easy thing... I may have to do a degree in fan-science first now.
If it is of any help visualizing, the machine that is currently pretty much the standard for paraglider cloth porosity is the JDC:
This machine works by having a bellows, which is raised and therefor air expelled from it, then the cloth is put over it, a slider removed, which starts the timer, and allows the bellows to hang freely, therefor the vacuum creating suction on the cloth. When the bellows has reached the bottom, indicating it is full with 0.25l of air, the timer is stopped.
This machine is ridiculously expensive, and I thought that there must be another way, using arduino, and possibly do away with the bellows setup.
Thanks so far. Keep up you thinking caps...
Robin2:
I think that is what is in @Paul__B's post - though I think "solved" may be ambitious.
No, jut practical - relating to the things most readily available.
(Noting the last post as I type this,) you want to do away with the bellows which is the equivalent of the large piston others have suggested. OK, a fish-tank pump would be more practical (and cheap).
You simply arrange for the pneumotach to be in series with your test mounting, presumably receiving the air flow before it. You require three pressure sensors - one before the pneumotach, one between it an your test mount and one after - which is presumably the local air pressure anyway, given that it is in otherwise still air. It is essentially ratiometrically comparing the pneumotach to the test mounting.
You might wish to introduce various compensations including temperature, but what I describe certainly sounds more readily and accurately calibrated than what you have now.
I might be missing something but from the looks of it the bellow thingy might be quite easily replicated.
Make a cylinder and a weighted piston. Make a release mechanism (solenoid) for the piston and a contact sensor on the bottom. Fix a piece of cloth on the intake side. Program your arduino to write the release time and end time.
That's pretty much it.
It might get a bit trickier if you don't have any machine tools, but even then a similar system could be fashioned fairly easily.
