An airspeed device for flight computers

 An airspeed device for flight computers

Preamble

This project started off both as a project for its own sake, and also a way of providing airspeed data to gliding flight computers. The ultimate goal being to better allow the flight computer to calculate wind direction and strength.

Currently (April 2022) XCSoar uses airspeed data (has done for a long time) to optimise its wind calcs, however XCTrack does not. As wind direction is key for assisting hang gliders pilots to land with the least fuss after an inland cross country flight, XCSoar has become my preferred flight computer. Although I do appreciate XCTrack's no fuss approach to logging flights and its live tracking to XCContest.

2023 Update: My preferred flight computer app is now TotalVario ("TV"). The developer of this amazing app and I are collaborating to meet the original goal of developing an instrument that can provide wind direction and speed as rapidly as possible. TV has many other features that are too numerous to list. Go here to see for yourself: TotalVario

Sensors

A good place to start when looking at DIY airspeed (and many other flight electronics for that matter) is the RC Flight Controller websites (eg: ArduPilot). For a long time the predominant sensor was a TE MS4525DO. This sensor worked ok, but had to be zeroed prior to each flight. It was also relatively inaccurate at low pressure differentials, ie: low airspeeds.

Then the Sensirion SDP3x ('DP' being: differential pressure, 'x' being: '1' or '2') entered the market. This sensor uses hotwire flow measurement technology which provides 0DP (ie no airflow) always equalling 0pa - no preflight zeroing required. And as a bonus the device's output vs. pressure difference is greatest at smaller pressure differentials - translation it is accurate at lower airspeeds. Lots of info here (scroll down to tech docs):  https://sensirion.com/products/catalog/SDP31/

There are two SDP3 models that I was interested in: SDP31 & SDP32. These are identical except for their measurement range.

SDP32: 0 to 125pa. Or equating to: 0 to ~50km/hr.
(the sensor I used as I could not at the time get an SDP31)

SDP31: 0 to 500pa. Or equating to: 0 to ~100km/hr.

Sensirion Sensor Testing

I have built an SDP32 airspeed device and tested it by mounting it on my car's wing mirror. I then compared it to both my car's speedometer and the GPS speed on my phone running Google Maps. Although a tad tricky to carry out the testing I'm happy to say my rig seemed to be accurate to a couple of km/hr over say 20km/hr and even better than this at lower speeds. I say tricky to test, because GPS speed below about 15km/hr is pretty flakey. And I would say the same for most cars' speedometers as well.

The Pitot

Of course, having a nice DP sensor is one thing, but we need to make a device that produces the DP, proportionally to airspeed. Googling found quite a number of sites with measurements for making a Pitot tube. I finally settled on the below set of ratios. This because the RC guys had found it to work and it was also nice and small.

For my Pitot I used a 'large pipe' with an outside diameter of 5mm (aluminium). And a 'small pipe' with an outside diameter of 2mm (brass). Total length was 75mm.

The tip was made from 3D-printed PET. I also 3D-printed a jig for accurately drilling the static-pressure holes as these were 0.5mm in diameter so pretty fiddly!

The end result:



 A Device Enclosure

As I have a 3D printer I used this for the production of an enclosure. The print material (filament) used was PETG as this prints well and is very resilient.

For designing the enclosure I used OnShape. This is online CAD software and if you are happy for your designs to be public, completely free. My designs can be found here:OnShape
Please feel to use, however with all the usual caveats that I'm not responsible for what you do with it 😏

Here is a transparent view:

The long tailpiece allows the print to be mounted into a drill press and spun so it can be more easily sanded off uniformly. The tail is then cut off with a hacksaw blade while it is spinning.

The hollow tail is where I added a Sensirion SHT41 temperature and humidity sensor. Interesting to see what the environmentals are while flying. Both XCSoar and XCTrack have settings that allow the display of outside air temperature ("OAT"). When/if I build another version of this device, I would place the SHT in the nose in order for it to read changes in environmentals faster.

2023 Update: Please go to the link below and look for the latest design. I'm tweaking often! I am also now using ASA, simply for aesthetics - it can be vapour smoothed.
Link: OnShape