Hi Everyone,

Very weird crash today. I have 12-15 flights all flawless before this incident. When it occurred I immediately thought I lost a servo - but upon inspection both servos were fine. Watching the video (below), I did not input the left roll that began the spin. I tried to switch to 'manual' but that made no difference; it's like I had no attitude control what-so-ever.

-APM 2.6 Plane 3.2.3, GPS with external compass
-3DR power module
-APM airspeed sensor

Here's a video of the crash - winds were about 10-15mph hitting the aircraft from the right side. I left the sound in, so maybe lower your speakers before clicking the link:

Attached are both the telemetry log from ground station & dataflash log.  

Any help would be greatly appreciated! Trying to rule out the autopilot as the cause before venturing back out. It almost looks like a stall, but I'm doing at least 20mph when the spiral begins.. I've flown this plane at 10-15mph without any issue.

Views: 1676


Reply to This

Replies to This Discussion

What type of plane is that?
I'm on my phone so can't veiw the logs, but looking at the video it looks like a classic tip stall to spin. In FBW the autopilot will react opposite to what is required to stop a spin. The left wing was very close to stall and started to lose lift so it dipped. FBW dutifully attempted to lower the left aileron which increases the Angle of Attack (AOA) to create more lift on that wing, and the opposite on the other wing. Since the AoA was already critical, it caused the left wing into a deep stall. It simutaneously lowered the AoA on the right wing, preventing it from also stalling. So now you have one wing deep stalled and the other one flying. To make things worse, the throttle went to max creating torque and gyroscopic effect which increases the rate of rotation and flatten out the spin making it hard to recover. The proper recovery is to quickly switch to manual, cut the throttle, push the nose DOWN, and if you have the skill (few do) apply rudder opposite to the spin direction. If you have a flying wing (especially the X8)you may need to put the stick full nose down and roll into the spin to get full down Elevon deflection.
Why did you stall at a higher speed than expected? It could be:
1. You hit some turbulence
2. You were flying in a slip because your yaw control is not tuned, or you have a wing with no rudder.
3. Load factor (pulling more than 1G)

Did you have the new Stall_Prevention code activated? It's a new parameter created by Tridge to prevent this scenario.
Is your ARSP_FBW_Min set too low?
Hi iskess,
Thank you for taking the time. This is a Mako flying wing (no rudders) http://www.readymaderc.com/store/index.php?main_page=product_info&a...

The aircraft was autotunded, which really made it fly quite nice. Also the airspeed was tuned as per the wiki. The FBW min is set to roughly 21mph which has been more than enough (maybe until now?) - this would be the first tip stall I've had on this frame. I did not activate the stall prevention code but I will now. I hope the logs substantiates your initial diagnosis; I'd much rather it be pilot error!

just a correction - double checked my params and STALL_PREVENTION was enabled (set to 1) when this crash occurred. 

I've had this happen more than once in the past (before stall prevention was added).  If you switch to manual fast enough you can recover.

Since you had it happen with stall_prevention enabled, you either didn't have your airspeed sensor well calibrated or your min speed is set too low.  The airspeed sensor is not super accurate, especially at low speed so you need to give it a few MPH of margin.

Hi Scott,

Thank you for your input. Are you aware of a good way to check if the airspeed sensor is badly tuned? I ran through the calibration as directed; on the ground it shows a value of 3-4, and if I blow on the tube it does increase..

I attached a graph from the crash showing altitude, GPS speed & airspeed. To an untrained eye (me) it looks ok? I also increased my FBW min speed as well..

It appears that just before your altitude peaked your airspeed dropped to 9 m/s. Since your airspeed is reading 2-3 m/s when zero, it seems to be reading faster than it really is. It could be that you were actually flying 6-7 m/s when the stall occurred. There can be a bit of airspeed noise when the airspeed is close to zero, so I can't definitively say what the speed was.
Did you do a preflight calibration before the flight? I do this calibration before every flight to be sure the airspeed is zeroed out after the APM warms up. Do you have the digital or analog airspeed sensor?

hi iskess,

I did not perform a preflight calibration - on the field I'm using a tablet for telemetry; im not aware of a way to do this calibration with the android app (haven't found it).. the wiki only gives instructions for mission planner.

Is there any way to enable this calibration on boot every time? Seems to be a requirement for safe flight.. i'm surprised it doesn't do that by default..

So the bottom line is, on the ground the airspeed should read 0? or is there a range that is 'ok' like 1-2?

I'm using the analog airspeed sensor. I didn't know I could use the digital one with the APM - is it worth it in your opinion?

The preflight calibration can't be automatic because it requires you to place your finger over the pitot tube to stop any wind. I use the analog with much success. I don't see more than 1 m/s at rest in any of my planes. If you zero it out it is very accurate. I don't know anything about Droid Planner.

I also use the analogue one but I never fly with a groundstation, so instead of doing that calibration I made a cover for the pitot tube out of heat shrink. I just leave the cover on while plugging in the battery, once everything's settled and ready for flight I take the cover off.

This has worked well for me.

I tried to run the preflight calibration, and I cannot get the airspeed value to be less than 2.8. It is fluctuating between 2.8-4.2 while sitting on the ground. Could it be a faulty airspeed sensor? 

Hi Mark,

I also keep a foam cover over the pitot tube when powering up - but as iskess mentioned i guess it doesn't make a difference as the calibration has to be done manually anyways?

if I cover the hole on the end of the pitot with my finger the airspeed value jumps to 30, not sure if that's normal. 

Reply to Discussion



Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service