Re: [Paparazzi-devel] Airspeed Sensor

[Michal Podhradsky]

Hi Alonso,

Barometer on Aspirin (v.2.1 and higher) is MS5611 and it has a temperature sensor too. So if you need, you can just read your temperature straight from the baro.

M

On Wed, Aug 28, 2013 at 4:25 AM, Nicolas Quendez <address@hidden> wrote:

Hello,

Compressibility effects in air appear at high velocity only (near MACH 1), so for our classical models, you can consider that air is incompressible (approximation is really good below mach 0.3 : 100m/s).

Stall speed is linked to lift drop at high angle of attack, mainly due to airfoil behavior at high angle of attack. Lift is air density dependent, so yes, is linked to air pressure, and air pressure can be linked to air temperature (ex: ascending flows due to locally temperature increase enables gliders to climb). So you can just calculate the stall airspeed difference at different pressures using the air density difference at different pressures  (http://www.engineeringtoolbox.com/air-temperature-pressure-density-d_771.html).

But what is important, is that Pitot tube (like the sensor you have) is not really measuring airspeed directly, but Dynamic Pressure (Total pressure minus Ambient Pressure, so it is already air density dependent). And lift is directly proportional to dynamic pressure. So stall speed given by your sensor should not depend on air density. The result is that the airspeed given by the sensor is not 100% good (because it depends on air density at the moment of the measurement) but it gives you the good dynamic pressure which is directly linked to stall speed...

I am not sure to be very efficient to explain this... 

Nicolas

Le 28 août 2013 à 00:36, alonso acuña a écrit :

I read elsewhere that in order to get true airspeed with a sensor like the Eagle Tree one needs the pressure diferential (reading from sensor), ambient pressure (one could get from barometer on IMU hopefully) and ambient temperature. I have no way to get temperature with my Aspirin so that won't work. A sensor like the one you recommended seems much better. 

On the other hand the only reason I am interested in airspeed at the moment is to prevent stalling at landing. I was wondering if someone knows if stall speed at a normal landing angle of attack is related to true airspeed or just having CAS would be fine? In other words does stall speed at landing vary much with temperature and or ambient pressure?

I found this page useful to understand what was meant by CAS and IAS http://en.wikipedia.org/wiki/Airspeed and this one about stalling http://en.wikipedia.org/wiki/Stall_(flight)

On Tue, Aug 27, 2013 at 7:31 AM, Tobias <address@hidden> wrote:

Hey guys,

I spend some time with the eagle tree to get it working....

so I can confirm that there is a big problem with the offset at startup, thats why we have the calculation and no fixed values - however if you wane check your sensor at an pressure difference (maybe the equivalent for 15m/s for example) you will notice that the drift and noise are well in an acceptable range, just seems to be an issue with offset at startup...

just an idea for the calibration - some years ago I also started with calibrating my sensors on a car, but there are two things you should have in mind
1 you will probably stay inside the boundary layer of the car which will effect you measurement (hard to overcome when your not at the very front of it...)
2. even if you are able to calibrate it perfectly you will get IAS when mounted to your plane...

sounds good, but its not since you will have to correct this for the mounting error to get CAS (which is actually named as IAS in PPZ)

so what I do, is just checking the GS compared to IAS(ppz) in a constant circle at constant altitude.
IAS should be exactly in the middle between GSmax and GSmin
if it is, you probably calibrated it perfectly and therefor already corrected for any mounting error (you got CAS!)
if not adjust the scaling factor
thats an quick and easy way and is working just fine

just a tip, I changed to this sensor which has a temperature sensor and i2c, the code is already in v5 - looks very promising
http://www.amsys.de/products/ams5812.htm


hope that helps ;-)

Tobi
_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel