iNavFlight
Airspeed inaccurate using Matek Airspeed Sensor ASPD-4525 and INAV 8.0.1
Current Behavior
Airspeed is very noisy before the flight and during the flight, completely inaccurate. Always below ground speed no matter the heading. Brand new pitot tube. Unmodified. Clean. Air-tight tubing. Note that my pitot tube is all the way at the front of the plane without any aerodynamic side effects of the airframe or airfoil.
Steps to Reproduce
- Plug in Matek Airspeed Sensor ASPD-4525, static tube lower port, dynamic tube higher port 1.1 Select MS4525 sensor or "auto" in inav airspeed sensor.
- Boot the aircraft. Airspeed is always auto calibrating and noisy.
- REMOVE AIRSPEED SENSOR SOCK 🤡
- Fly the aircraft in the wind and against the wind. Or full speed. Airspeed is always much lower than ground speed, also at speeds faster than 100 km/h.
Expected behavior
Airspeed should, at constant throttle, be rather constant. While the ground speed should match wind speed and airspeed combined. Either added or deducted depending on wind direction. My airspeed is always under-reporting. Funny part is that, when blowing into the pitot manually, the airspeed seems plausible.
Suggested solution(s)
Manual calibration of airspeed sensor? Possibility to calibrate for ZERO reading with blocking the holes and a calibration factor for the measured speed?
Additional context
- FC Board name and vendor: SpeedyBee F405 Wing
- INAV version string: V8.0.1
Senzor MS4525 is marked in inav as noisy, https://github.com/iNavFlight/inav/blob/master/src/main/drivers/pitotmeter/pitotmeter_ms4525.c#L147
if yopu block both holes, how it looks like in inav configurator -> sensors -> enable airspeed. is line stable or noisy?
I notice your process as listed does not include "remove the sock". Do you have a piece of heat shrink or whatever as a sock on your pitot when you boot? It self-calibrates on boot.
This would be a good place for a "remove before flight" tag. :)
Senzor MS4525 is marked in inav as noisy, https://github.com/iNavFlight/inav/blob/master/src/main/drivers/pitotmeter/pitotmeter_ms4525.c#L147
if yopu block both holes, how it looks like in inav configurator -> sensors -> enable airspeed. is line stable or noisy?
My reading is constantly noisy between 50 and 150 units. Only if wind blows into the pitot tube, the reading shoots up to lets say 1000 and the reading before that looks "flat" in comparison to that for the next few seconds until the graph resets.
I notice your process as listed does not include "remove the sock". Do you have a piece of heat shrink or whatever as a sock on your pitot when you boot? It self-calibrates on boot.
This would be a good place for a "remove before flight" tag. :)
I had my pitot protection on the last time. It did not change anything about the observed behavior. The airspeed reading is still completely nonsense during flight. Will test soon while driving a car and holding the pitot tube out the window.
I notice your process as listed does not include "remove the sock". Do you have a piece of heat shrink or whatever as a sock on your pitot when you boot? It self-calibrates on boot.
This would be a good place for a "remove before flight" tag. :)
Already one step ahead of you :)
It seems the openings for static pressure are too close to the zone of high pressure at the nose. I would change the pitot-position.
It seems the openings for static pressure are too close to the zone of high pressure at the nose. I would change the pitot-position.
To this point, the following wind tunnel image may be interesting.
Versus
It looks like with a pitot mounted on a blunt nose, one would want it to extend at least 1.5X the height of the nose.
The nose of the aircraft is typically the best spot to mount a pitot tube because the sides of the fuselage and also the wing have many different pressure zones and turbulence. Maybe my static pressure is too close to the nose of my plane. For testing purposes, I extended the tube by 20mm.
The nose of my plane is self made 3D printed and modified to be sharper than the stock ZOHD nose. I have no ports and no webcam in front, so no turbulence. The nose is inspired by sharks with a wide leading edge and an angled upper and lower arc.
Old pitot position:
New pitot position:
The static port definitely looks too close to the nose on the first image. It may still be too close on the second image.
It's not just the physical nose that it needs to be away from. It needs to be in front of the boundary layer that is generated around the nose, in this case.
The nose of the aircraft is typically the best spot to mount a pitot tube because the sides of the fuselage and also the wing have many different pressure zones and turbulence.
Yeah you'll typically see them mounted coming off the nose or coming off the leading edge of the wing on certified aircraft. You'll ALSO typically see a rounded nose on aircraft that have them mounted there. Then the pitot is mounted ahead of and below the nose:
--
Possibly the WORST place for the static port (other than inside an engine) would be at the stagnation point, the area of no flow, ahead of the nose or leading edge:
I connect the power supply to the room and the speed readings are about 12 km/h. If you reset the sticks, the airspeed is about 0-3 km/h. Could there be a problem with the first initialization?
Update after long bad weather period:
I tested the new pitot position. Indeed my static port was too close to the boundary layer of my airplane. The new position fixed the reading. It is now accurate. Thanks to everyone for the input! Would never have guessed the boundary layer to be this thick at the slow speeds my ZOHD Dart XL is flying.
I connect the power supply to the room and the speed readings are about 12 km/h. If you reset the sticks, the airspeed is about 0-3 km/h. Could there be a problem with the first initialization?
How do you reset the sticks? My understanding is that at slow speeds the relative pressures are very sensitive which causes the bad 0 readings. But I'm interested in how you re-initialized it. There could be some heat-up effect after first initialization?
12 km/h offset seem to be a lot. Please calculate the differential pressure and look for the impact of this (very low) pressure on the usual flying speeds. It's almost imperceptible.
How do you reset the sticks? My understanding is that at slow speeds the relative pressures are very sensitive which causes the bad 0 readings. But I'm interested in how you re-initialized it. There could be some heat-up effect after first initialization?
I go to the OSD menu and reboot the flight controller there