Freesteel Blog

Sensors in the sky

Thursday, April 28th, 2016 at 12:04 am Written by: Julian

Well, the hang-gliding has been quite lovely. But the logger appears shot to pieces.

I still don’t know what I am doing when I get to cloud-base, which is probably why I plummeted out of the sky shortly after this picture was taken.

As soon as I took off, the barometer stopped communicating most of its data.

This is the device I lavished so much time on isolating it from the rest of the electrical circuit and arranging for it to bitbash the information back through an interrupt pin.

It could be some timing issue, or whatnot. No way to debug it. Luckily I got myself a Bluefly vario which does the same thing of reading a MS5611 barometer on a tight 50Hz loop and transmitting it back to the main board. In the Bluefly’s case it’s for the purpose of running a Kobo/XCsoar system. I’ve just given up on the one I built as it’s too inferior to simply running XCSoar on the phone where I’ve got colour and more or less know how to use it.

Luckily the Bluefly also sports a GPS and works through a serial port, so I’ve yanked off the Adafruit GPS breakout board and bodged the wires to insert the Bluefly in its place.

Then there was a small matter coding it up using a complex state machine to program the GPS module through the Bluefly pic processor to get it to read at 10 times a second.

But then the BNO055 orientation sensor played up and decides to shutdown at unexplained moments for unexplained periods of time.

The white vertical lines are 10 minute intervals, and there is a green dot for every successful orientation reading, with y-value proportional to the time since the previous reading, so I’m getting gaps in the data of half an hour in flight.

I’ve produced a reset timer to try and start it off again if no data comes through for 20 seconds.

Anyways, here is one of those nerve-wracking close encounters with the ground during the flight.


You only deserve to get away with one of these low-saves on a flight where you should be staying up. I got a total of 45kms downwind to Church Stoke. It then took 4 hours of local bus rides to get back to Llangollen, half an hour to walk up the hill, and an hour to drive off to get my glider. Then I drove to Sheffield to stay with someone for the night. Tried to go out flying the following day and it rained. Meanwhile, distance records of 300km had been set on that day. “Piano, piano,” Adrian said in Italian. Little by little.

Meantime, here’s the altitude vs humidity graph (green for up motions, red for down motions); vertical lines are 100m, horizontal lines at 10%. Max altitude was 1600m. Oddly the humidity didn’t get to 100% even though I got above the clouds in places (though not in them). Cloudbase seemed to have various altitudes. Or the accuracy of the device is shoddy.

We can do the same with temperature (horizontals are degreesC, the red line is 0degC)

Zoomed in, with points where the glider was climbing in green, and descending in red.

This does potentially give the first evidence for the thermic air being half a degree warmer — if we believe it. It could also simply be sensor lag, as the ifferent temperatures match if shifted 50m in altitude, which provides an average of the previous 100m, which would have taken about one minute to move through. This would be quite a lag, so maybe I believe it. The signal is quite noisy, possibly due in part to the inaccuracy of GPS altitudes, but unfortunately I’ve got no barometric pressure records to correct it.

Anyway, here is the track from the start of the flight with north to the left:

There was a north wind blowing at about 7.4m/s. This is the trace if you subtract this wind vector:

This has me flying against the wind for a long while (which was when I was ridge soaring on the Llangollen hill), then drifting with the wind on the first thermal (which means staying in place on this diagram), flying directly cross wind and then downwind a bit (in humongous sink) to get a second thermal that I chased upwind for part of the way, until finally a gradual heading downwind through four crappy thermals down to the ground.

This shows that first big thermal up to cloudbase followed by that really vicious period of descent of between 3 and 4m/s for 3 minutes where I basically dumped 600m before I knew what was going on (vertical lines are minutes, horizontal lines are metres per second(m/s) gradients).

There’s actually a bit more to the causation of what went wrong at this point on this flight. I’d thought I’d be smart and try to aim for a series of waypoints, the first one of which was Moelfre Gyrn about 11km away at the yellow circle in the bottom right.

As soon as I got up high I aimed for it reasonably convincingly. By keeping the arrow on the flight computer aiming for it, I was actually heading directly cross wind (according to the compensated wind drift). In the diagram the blue dots are points of descent of between 2 and 3m/s, yellow is between 3 and 4m/s and red is greater than 4m/s going down fast.

From the end of leaving the first thermal at the top of the picture in the space of 2.7minutes I dropped 209m and covered 2.4km, which is not so bad. Then in the next 0.9minutes I dropped 175m covering 0.9km. So I panicked and turned downwind and dropped 313m in the next 1.6minutes covering 1.9km. The altitude was now 856m, and the Moelfre waypoint was 510m high. I dumped my stupid plan and headed towards sunshine and lower ground, where I should have gone in the first place.

Plotting what I think is the glider orientation on that descent (with an 11m long whisker) it seems to indicate that I was actually pointing upwind for some of it. The whisker shortens in the greatest descent areas (red dots) as though the nose is pointing more downwards. Was I trying to speed up and get away? Doesn’t work so well on an intermediate glider like mine.

Below is the point where I was going downwind and suddenly found some lift to start circling in.

This does suggest I could have a downwind detector that says to turn left or right according to which way the nose is pointing in relation to the GPS direction of travel.

I was still intending to find a way to calculate the local wind direction from the data available, but I can’t do it with this flight as I forgot the windspeed meter.

To recap, the whole point of this project is to find out what can be done with a set of the latest sensors, and whether they can help do something interesting, such as get me to have more and better hang-gliding flights. That’s a pretty good motivation to keep at when I am pounding my head against this wall. Many sensor projects go nowhere when they face the reality of deriving anything actionable from the data collected.

In my case the events take place over minutes and hours, and the actions of where to fly to next have my ongoing undivided attention. This is quite different to most sensor monitoring projects where you lay out some sensors in a building or a town, record stuff over a couple of days or months, and then you really don’t have much option to do anything with it because it’s now old and anyway doesn’t pertain to something you can very much control.

Even with this advantage I’ve still not got very far. The ideas are presently getting thin on the ground.