12

The gyroscope (not to be confused with the accelerometer or GPS) in my iPad (3rd generation) works wonderfully both when standing still on the ground as when moving at high speeds in a car on the highway.

However, as soon as I try out the gyroscope while in flight on an airplane, the gyroscope returns non-sensical data!

Does anyone else experience this? I have tried replicating the airplane conditions on the ground (high temperatures, vibration, driving as fast as possible) but the gyroscope keeps working while on the ground.

Why does the gyroscope not accurately work in the air?

Gerry
  • 14,551
austin meyer
  • 161
  • I like your question, though from some quick Googling, it might be hard to formulate a concise answer that's appropriate for the Q&A style of this site. There are plenty of detailed articles on the web though about the gyroscope in the iPhone/iPad and about MEMS gyroscopes in particular. Still, I'll be looking out for the answer to this question :) – Gerry Aug 31 '12 at 21:07

2 Answers2

4

OK I THINK I FIGURED IT OUT! It was my fault almost entirely, but I have an answer that may be useful to others. Here is the deal:

1: I set the gyros to update at 60 hertz 2: On the ground, the app RAN at 60 hertz, so grabbed one gyro update per frame 3: In flight, my app did MORE WORK (!!!) to display what the airplane was doing, and thus fell to less than 60 hertz update rate of the app! 4: At this point, the update messages from the gyro queued up and the event-reader got behind... WAY behind, so I was looking at real gyro data from EARLIER IN THE FLIGHT!

So, the REAL lesson here is that setting the update rate of the gyros to a value greater than the frame-rate of the app results in event queing that has you running behind.

If anyone knows how to soak up ALL the gyro events EACH frame of the app, rather than just reading one, then that would APPEAR to solve this issue... whew! Interesting!

austin meyer
  • 161
1

If the iPhone uses a gyroscope based on tiny parts inside vibrating (sensing changes in vibration under acceleration), it's possible that all the extra vibration in an airplane in flight swamps out the detectable signal with noise. To test this theory on the ground you'd put your phone in a "mechanically noisy" environment with similar characteristics. A vibration table might work, or it might be too strictly periodic (too close to sinusoidal, not broadband enough) to be a good simulation. Maybe hold it against the side of a power drill while trying to drill through a heavy board? Attach it to a bike frame and ride over gravel?

http://en.wikipedia.org/wiki/Vibrating_structure_gyroscope
http://www.ifixit.com/Teardown/iPhone-4-Gyroscope-Teardown/3156

Matt B.
  • 111
  • I should point out that I was, indeed, HOLDING THE IPAD IN MY HANDS and rotating it in flight... It seems that that would isolate the gyros from the vibrations of the airplane. The flight test was done with the ipad held in my hands in the cockpit of a light airplane, the ground-test was done with the ipad held in my hands in the cockpit of a sports car with similar noise and temperature and vibration characteristics as the airplane. – austin meyer Sep 01 '12 at 14:30
  • I just exchanged my ipad3 for another one at the phone-store.. I will fly again today with the replacement pad and let you know what happens! – austin meyer Sep 01 '12 at 14:31
  • One more thing to point out... I DID, INDEED, hold the ipad pressed against the aircraft instrument panel a number of times during the flight, so aircraft vibrations were indeed transmitted to the pad... BUT, the gyros were non-functional for ALL of the flight, even if the pad was only held in my hands, but the gyros returned to normal function upon landing... this question is especially interesting to anyone that wants to use an ipad as an aircraft instrument panel in flight... the possibilities are endless... – austin meyer Sep 01 '12 at 14:33