Sheesh, you people call yourselves engineers, yet you can't even interpret a customer's simple request? Let's break it down:
He wants to build a pitot tube anemometer based microcontroller. In other words, he wants to build a processing element that functions by measuring gas flow, presumably some kind of pneumatic Analytical Engine. This fact appears to have been provided to the audience merely to show how clever and original the OP is, and that he likes to "think outside the box", since he has not asked for specific help with it.
He is asking how to build a pressure sensor. In other words, he wants to see an indication if there is pressure. I suggest a very thin-walled metal sphere, completely evacuated. If there is pressure, the sphere will implode, thus providing the desired indication. The sensor can be calibrated by making the walls thinner or thicker as appropriate. I suggest rotary molding might be an easy way to produce the desired sealed shape.
(BTW, I work from "specs" like this every day, except that mine mostly have down the bottom some language that says "must be fully backwards and forwards compatible with some other non-speced product originally designed in 1978").
Basically, a tube, vent or whatever will have air driven into it if it faces into the airflow (that's what a pitot tube is). It will have air 'sucked out' of it faces away from the airflow. So, they design the static vent to be at right angles (as close as possible) to the airflow.
--
Paul Hovnanian mailto:Paul@Hovnanian.com
------------------------------------------------------------------
Answer: Eight. Twelve if the light bulb is cross-threaded.
OK, there's something I've wondered since I've learned about both the pitot tube and the Venturi effect - howcome the Venturi effect doesn't suck all of the air out of the static ports, or do they compensate for that, and have been doing so since before I was a gleam in Mom's eye? ;-)
A 'pitot tube', and an 'anemometer', are two very different bits of kit. All a pitot tube involves, is a tube extending beyond the local skin effects, heading into the incoming air flow. It is used in conjunction with a 'static vent', which is sited into a point on the structure, where the air should be close to the local ambient pressure, and you can then use the pressure difference between these two points as a measure of speed. The 'Freescale' MPX4100, and similar sensors work fine for this. An anemometer, instead has an impeller being moved by airflow, to measure speed. The big advantage of the latter, is that it is not directionally sensitive (depending on the design of the impeller). The downside is it involves moving parts. Designing a good pitot, is more complex than it sounds...
errr, well yes and no. The most common version of an anemometer is as you describe (precisely named a 'vane anemometer'). There is another version that is based on a pitot tube. See:
formatting link
And for the original post, I suggest you contact Dwyer Instruments and ask for their technical document on Pitot tubes. It contains a lot of precise theoretical presentation of the Bernoulli effect and how Pitot tubes work, a discussion on air flow measurement using differential sight indicators, and, guide lines for measuring flows in square/rectangular ducts, circular ducts, etc with respect Reynolds numbers for flows (basically if the flow is laminar or turbulent.)
A direct answer to the question is to construct a simple differential pressure cell (look up differential pressure cell theory if you don't know that they are) by attaching a strain gauge across a thin mylar membrane in a sealed vessel and attach one side of the vessel to the measurement port and one side to the reference port. You will need to calibrate flow in terms of resistance (actually change in resistance as a function of flow/deflection). This approach may not be very sensitive but it would demonstrate the principle. If you want to make your own strain gauge you could try using the circuit fix up sticks that dispense a thin layer of conductive material on a tightly drawn piece of plastic bag. I am pretty sure that as the bag deflects the resistance of the material will change - but you would have to experiment.
If you want a lab quality instrument, the differential pressure sensors can also be purchased (look at Motorola, omron). Check out the article here:
formatting link
As far as the comment of 'working from specs'...one very important piece of information that is missing is: What is the range of velocities that you are interested in measuring???? One needs that to determine the best applicable technology (a pitot tube won't measure velocities of
1f/s, for example, but a vane type anemometer will).
ANother approach is to use (or construct) an air velocity gauge, similar to:
formatting link
and using opto couplers + LEDs - position them along the proper graduations of the air velocity gauge. The poster did not specify the resolution or accuracy that is desired. Clearly this latter method is not offering a high degree of resolution....but hey, could be cheap and effective depending on time and money... John
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.