SparkFun Forums

[Myxo]

Hello!
New to the forum so go easy.
I tried posting this to a reddit AskPhysics forum and just ended up getting some low effort responses, so I wanted to share my thoughts with some other people that might have some more valuable feedback

What i'm trying to do is measure the velocity of air coming out of a vent after its passed through a series of filters. Originally I had planned on hooking up this item:
https://moderndevice.com/product/wind-sensor-rev-p/

Which is a thermal anemometer. However, when I hooked it up via my redboard ( I used a regulated 12v power supply), the data I was getting back from it was far too noisy to be useful.

So I was thinking more on this issue today and I drew up this sketch:


Which led me to here:
Prandtl tube
A site from NASA outlining how a Prandtl tube works.
The basic idea here is that a Prandtl tube takes two pressure measurements, one in the direction of flow and one not, and converts that to air speed.

So i've already gone ahead and sent the cart before the horse and bought two barametric pressure sensors that I can hook up to a redboard, but my conundrum here is really about physics.
My thought is that one pressure sensor can be mounted inside the box, and one outside, and I can use the two measurements in terms for solving the bernoulli equation. I'm stuck however, on if the sensor that is placed inside the box needs to be facing the flow of air. My thought is that it shouldn't matter; the static pressure of the air in the whole chamber should be the same until it leaves the chamber.

Thoughts? Considerations?

[Cannibal]

Hi there,

I haven't done anything like this before, but when using barometers/pressure sensors for altimetry the flow of air over the sensor is a nuisance:

Imagine a pressure sensor inside a sealed vessel - it will measure some value of pressure. Now imagine you have put a fan inside that sealed vessel that is blowing right into the aperature of the sensor. Has the pressure in the vessel changed? No. Has the measurement changed? Yes, it has gone up (and become noisier).

I recommend trying to get your hands on some foam material similar to a wind guard from a microphone - it will serve as a mechanical filter for turbulent air (noise) and will greatly reduce the bulk flow rate directly on the transducer that could bias the pressure measurement.

Things might be different with a pitot tube, as it looks like you want the second pressure cell to be directly exposed to that bulk flow of air.

A practical problem you will run in to is that the individual sensors will likely have a bias offset you will have to null out, and potentially they will have a time varying differential bias that will make this thing hard to keep synched up unless you can enter periods of known zero air movement and do a quick online re-calibration.

Best of luck and keep in mind that I haven't done this before and could be way off base with my assumptions about what is 'desired' in the measurement.

[jremington]

The air pressure won't be constant anywhere in the box. It will be approximately equal to the ambient pressure at the entrance and exhaust ports, and vary within the box (lower than ambient close to where the fan is "pulling" and higher than ambient close to where the fan is "pushing").

To measure air speed just use an anemometer, like these cheap ones: https://www.dx.com/p/smart-sensor-1-5-l ... eter-17461

[Myxo]

I haven't done anything like this before, but when using barometers/pressure sensors for altimetry the flow of air over the sensor is a nuisance:
Imagine a pressure sensor inside a sealed vessel - it will measure some value of pressure. Now imagine you have put a fan inside that sealed vessel that is blowing right into the aperature of the sensor. Has the pressure in the vessel changed? No. Has the measurement changed? Yes, it has gone up (and become noisier).
I recommend trying to get your hands on some foam material similar to a wind guard from a microphone - it will serve as a mechanical filter for turbulent air (noise) and will greatly reduce the bulk flow rate directly on the transducer that could bias the pressure measurement.

So would the foam fundamentally alter the measurement of the sensor? I'm trying to think of this conceptually, and I don't imagine that it would, or I could correct for it by putting the same foam over both sensors (I think). Either way, that is a very good idea.

Things might be different with a pitot tube, as it looks like you want the second pressure cell to be directly exposed to that bulk flow of air.
A practical problem you will run in to is that the individual sensors will likely have a bias offset you will have to null out, and potentially they will have a time varying differential bias that will make this thing hard to keep synched up unless you can enter periods of known zero air movement and do a quick online re-calibration.

Right. So i've already bought the sensors, they should arrive in a few days. Then i'm planning on building a cardboard mockup with a small fan and testing the principles first. I can then try it facing the direction of airflow, and facing across the direction of air flow. I may even buy a kestral or its equivalent so I can get a calibrated measure of airspeed.

Best of luck and keep in mind that I haven't done this before and could be way off base with my assumptions about what is 'desired' in the measurement.

Hey no, absolutely and thanks for the input. Like I said, I already bought the sensors and already have the other gear to play with this, so I hope to have some preliminary data to work with maybe as soon as the weekend.

The air pressure won't be constant anywhere in the box. It will be approximately equal to the ambient pressure at the entrance and exhaust ports, and vary within the box (lower than ambient close to where the fan is "pulling" and higher than ambient close to where the fan is "pushing").

To measure air speed just use an anemometer, like these cheap ones: https://www.dx.com/p/smart-sensor-1-5-l ... eter-17461

I'm aware that the air speed won't be constant any where in the box; I'm actually considering installing a series of these in 2 dimensions to account for that. The series of filters i'm using should bring the flow to as close to laminar flow as one could expect. It's a 12 inch*12inch by 24 inch filter the air is passing through. In fact what this is for is a laminar flow hood i'm building, and I'm trying to do with this part of the project is two and half things:
1: Have a record of the total volume of air that has passed through the filter during its use lifetime
2: Have a digital display out of the current airspeed, the number of hours the unit has been in use, the current amount of time the unit has been running for, etc..
2.5: Possibly for dynamic control of air speed, but its probably too much effort. I have a control box that can adjust the airspeed coming from the blower, but it doesn't read in volume of air per unit time. It just goes from low to high. I want to regulate the flow at 0.65 m^3/ second. I realize it may be "nuking it", but I think it would be cool to use a rotor to regulate the air speed dynamically.

A cup anemometer is not going to be a working solution. I don't want any moving parts in the work space, and I need the sensor head to be flush with the surfaces of the interior of the work space. I'm building the front of the working space (where the sensor head will be) so it can collapse down and fasten to the sides of the instrument for storage.
Thank you for your considerations.
-Myxo

[jremington]

I'm aware that the air speed won't be constant any where in the box;

I was referring to a pressure gradient.