Freesteel Blog » Decoding the airspeed probe with a wheatstone bridge

Decoding the airspeed probe with a wheatstone bridge

Sunday, January 11th, 2015 at 9:10 pm Written by:

By not including a datasheet with their airspeed probe, Brauninger/Flytek gave me the pleasure of two successful days of hacking involving an oscilloscope and much experimentation to work out its parameters and build a circuit to exploit them.

I bought this thing as an optional add-on to the Flytec 6030 (which I’ve never got to grips with) back when I had more money than sense. I wouldn’t have got it for the purpose of reverse engineering like this because I couldn’t do electronics then, and anyway I’d have rated the chances of success as quite low.

Nevertheless, by applying various voltages and different directions and blowing on the propeller to get a response, I established that if you apply a positive current on the tip of about 1Volt (and ground the other connection), the device exhibits a resistance of between 11200 Ohms and 12000 Ohms, depending on the position of the blade.

This was a job for a Wheatstone bridge:
wheatstonebr

You can actually see the voltage differences (in millivolts) over 1/12 of a turn of the propeller:
proplowvolt
prophighvolt

At about 1Volt across the terminals you get this half-sin signal as each blade passes the stem:
oscil1v

If you put too many volts across it, there’s a problematic wobble on the floor voltage:
oscil3v

The reason the wobble is a problem is that I then pass the signal through an INA125 OpAmp using the wiring diagram copied from here (because so little of the datasheet made sense), and these irregularities would get stretched up to make an ugly indeterminate signal.

But if you get the volts right, then it produces these beautiful LED illuminating square waves shown in the video below (listen for my blowing on the prop above the noise of the office):

And finally there’s the finished circuit that can act as input to the IRQ pin of an arduino so it can easily count the number of pulses per second:
finishedcircuit

Now all that remains is to find a way to calibrate it.

This is just another tiny step on the way to building my portable high frequency logging device for the discovery of things.

If you’re doing this for yourself, it might be worth looking at the weatherflow device which works with a smartphone.

More notably in the I-simply-can’t-keep-up department filed under discoveries-that-I-made-30-seconds-ago, there’s also the AS Sensor for XC Soar android app, which first requires an adapter plug to route the signal into the microphone ring of the phone where it presumably analyses the pulses using the audio circuitry!

So much for doing something new.

As you can see, I’m going down this micro-controller route rather than attempting to master the awesome-does-everything Android operating system and all its amazing sensors. I wonder if this is an error. Who knows?

In such an interconnected world, I think it’s necessary to invert the question, and assume that you can never trust an idea until someone else has had it first.

In which case for the idea I’m labouring under, I have to thank Mr Untuckable who posted on 21 April 2008 in a discussion about thermal rotation:

A track logging feature ? Show if the airspeed is greater than the circling track speed…

And that could be made possible by the new Adafruit Ultimate GPS (on order) which can read absolute position at the rate of 10 times a second and send it down a serial line.

1 Comment

  • 1. mike replies at 13th January 2015, 2:21 am :

    there is nothing like thinking of a good idea then googling it and they already make it.

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <blockquote cite=""> <code> <em> <strong>