Help: Drowned Mavic Pro

[sar104]

I just cannot believe that my Post would get you so angry that you would resort to SEXUAL TALK , that is low and you got your Followers doing the same thing . My post might not have been clear and perfect to you but what you did is SHAME FULL .

SAR - My post in no way deserved any SEXAUL or Abusive comments , you get that .

Congratulations - you made it onto my rather short ignore list.

 

[MavicFT1P]

lol, love it

 

[Bigbird48]

sexual talk were was that sorry no rain coats for me either. Guy just pushs the wrong buttons weres the ignor button

 

[Bigbird48]

YEP: I read it, , saw it, pondered upon it, saw the hard breaking, searched my vast video library of trips over the water when the breaks were applied and came to the conclusion that it may have been Magnetic interference that caused the EMERGENCY BREAK to happen.

You thought it may have been a firmware issue and I believe that it is possible it was Magnetic Interference.

Watch in this video as Magnetic Interference stops the Phantom 4 Pro cold from moving forward by putting on the breaks.

Phantomrain.org
Approved Vendor
Gear for flying in and out of the storm.

theres a thumbs up button but no down so I made my own LOL

 

[Bigbird48]

If your angry with me brother, dont resort to SEXUAL ABUSIVE POSTS. I have reported this nonsense.

 

[brett8883]

Ok there’s is still one mystery here and that why the act of braking would cause it to go beyond the tilt limit and I have a theory, which is just that, not a conclusion because I don’t know that we will ever know.

When originally looking at this I noticed that the pitch excursion happened during an abrupt throttle up input so my theory has this in mind.

I did some tests today and something I thought was interesting is that the emergency brake only applies to pitch and roll maneuvers. It’s does NOT apply to a throttle input. If throttling up without any attitude inputs from the remote nothing happens. This is true even in windy conditions when the aircraft IS tilting to maintain position.

I did another test to see if when inputting both throttle and attitude with the remote if it would also stop the throttling. It does not. The emergency brake seems to be independent of the throttle control. When doing this test I was cautious and did this at low speed and it thought my Mavic Pro flipped at LOW speed. It was very unnerving even though I’m fully comfortable hitting the emergency brake at full speed in sport mode no throttle.

It doesn’t seem far fetched to me that under the correct circumstances the throttle up command at just the right time during a braking maneuver can cause g forces greater than the motors are able to counter act causing the tilt error.

To demonstrate this effect put your hand out so it’s in the forward pitch state and then quickly change to the reverse pitch State while at the same time raising it up. You’ll feel the whiplash created by this maneuver more so then with just a rapid opposite pitch alone . During an emergency brake we would expect the rear motors to be slowed and the front motors to speed up but we don’t see this. We see all 4 motors speed up. However, we do see the braking in that the front motors are faster than the rear for a short time. We then quickly see the rear motors abruptly go up to max output and the front motors to slow abruptly. My theory is that the drone was put into a maneuver that it lacked the physical ability to recover from and the flight controller stopped the motors. We go see the pitch begin to correct itself but only after the motors are cut. Possibly if there was enough altitude the motors could have possibly been turned back on after the flight controller dumped some inertia by allowing an altitude drop but unfortunately there wasn’t enough altitude and it hit the water.

Again just a theory like the firmware is to blame is a theory but I think there is some evidence when you look closely at the timing of the motor speeds braking maneuver, pitch changes, and RC inputs.

 

[BD0G]

I think this thread just put an end to my full speed Sport Mode Pause Button Tail Stands

 

[Bigbird48]

yep I did it a couple time thats enough

 

[Simmo]

How can 'the pitch begin to correct itself after the motors are cut'?
What would do that? (Gravity??)

 

[sar104]

Ok there’s is still one mystery here and that why the act of braking would cause it to go beyond the tilt limit and I have a theory, which is just that, not a conclusion because I don’t know that we will ever know.

When originally looking at this I noticed that the pitch excursion happened during an abrupt throttle up input so my theory has this in mind.

I did some tests today and something I thought was interesting is that the emergency brake only applies to pitch and roll maneuvers. It’s does NOT apply to a throttle input. If throttling up without any attitude inputs from the remote nothing happens. This is true even in windy conditions when the aircraft IS tilting to maintain position.

I did another test to see if when inputting both throttle and attitude with the remote if it would also stop the throttling. It does not. The emergency brake seems to be independent of the throttle control. When doing this test I was cautious and did this at low speed and it thought my Mavic Pro flipped at LOW speed. It was very unnerving even though I’m fully comfortable hitting the emergency brake at full speed in sport mode no throttle.

It doesn’t seem far fetched to me that under the correct circumstances the throttle up command at just the right time during a braking maneuver can cause g forces greater than the motors are able to counter act causing the tilt error.

To demonstrate this effect put your hand out so it’s in the forward pitch state and then quickly change to the reverse pitch State while at the same time raising it up. You’ll feel the whiplash created by this maneuver more so then with just a rapid opposite pitch alone . During an emergency brake we would expect the rear motors to be slowed and the front motors to speed up but we don’t see this. We see all 4 motors speed up. However, we do see the braking in that the front motors are faster than the rear for a short time. We then quickly see the rear motors abruptly go up to max output and the front motors to slow abruptly. My theory is that the drone was put into a maneuver that it lacked the physical ability to recover from and the flight controller stopped the motors. We go see the pitch begin to correct itself but only after the motors are cut. Possibly if there was enough altitude the motors could have possibly been turned back on after the flight controller dumped some inertia by allowing an altitude drop but unfortunately there wasn’t enough altitude and it hit the water.

Again just a theory like the firmware is to blame is a theory but I think there is some evidence when you look closely at the timing of the motor speeds braking maneuver, pitch changes, and RC inputs.
View attachment 70151

Interesting tests although I don't completely understand your description. Were these simulator or flight tests? And I don't know what you mean when you say that you found that emergency braking doesn't apply to a throttle input. Can you elaborate?

Anyway - part of this explanation is demonstrably what happened - the emergency braking during full elevator and full throttle did put the nose up at 75° and the FC did stop the motors due to the excessive attitude. Those observations are in the data and not up for debate.

So the two unresolved questions are (1) why did the firmware, in calm conditions in P-GPS mode, allow the aircraft to reach an attitude that exceeds its own threshold for motor stop and (2) why does the firmware stop the motors in flight due to attitude? (1) suggests that some of the gain values are set too high, although that attitude, per se, is not really a problem - it's perfectly recoverable. These aircraft can recover from inverted and tumbling falls. In other words the attitude achieved in that maneuver is only a problem because of the motor stop function - hence my suggestion that this is a firmware issue - a bug or perhaps, more accurately, a feature with unexpected and undesirable consequences.

There are good reasons for the FC to kill the motors on rollover on the ground, but no good reasons to kill them in the air.

 

[sar104]

How can 'the pitch begin to correct itself after the motors are cut'?
What would do that? (Gravity??)

Actively it cannot after motor stop, but the FC was already starting to correct the pitch and the aircraft was rotating forwards when the motors were stopped (orange trace below), so it simply continued that rotation back to approximately level.

 

[BudWalker]

Ok there’s is still one mystery here and that why the act of braking would cause it to go beyond the tilt limit and I have a theory, which is just that, not a conclusion because I don’t know that we will ever know.

When originally looking at this I noticed that the pitch excursion happened during an abrupt throttle up input so my theory has this in mind.

I did some tests today and something I thought was interesting is that the emergency brake only applies to pitch and roll maneuvers. It’s does NOT apply to a throttle input. If throttling up without any attitude inputs from the remote nothing happens. This is true even in windy conditions when the aircraft IS tilting to maintain position.

I did another test to see if when inputting both throttle and attitude with the remote if it would also stop the throttling. It does not. The emergency brake seems to be independent of the throttle control. When doing this test I was cautious and did this at low speed and it thought my Mavic Pro flipped at LOW speed. It was very unnerving even though I’m fully comfortable hitting the emergency brake at full speed in sport mode no throttle.

It doesn’t seem far fetched to me that under the correct circumstances the throttle up command at just the right time during a braking maneuver can cause g forces greater than the motors are able to counter act causing the tilt error.

To demonstrate this effect put your hand out so it’s in the forward pitch state and then quickly change to the reverse pitch State while at the same time raising it up. You’ll feel the whiplash created by this maneuver more so then with just a rapid opposite pitch alone . During an emergency brake we would expect the rear motors to be slowed and the front motors to speed up but we don’t see this. We see all 4 motors speed up. However, we do see the braking in that the front motors are faster than the rear for a short time. We then quickly see the rear motors abruptly go up to max output and the front motors to slow abruptly. My theory is that the drone was put into a maneuver that it lacked the physical ability to recover from and the flight controller stopped the motors. We go see the pitch begin to correct itself but only after the motors are cut. Possibly if there was enough altitude the motors could have possibly been turned back on after the flight controller dumped some inertia by allowing an altitude drop but unfortunately there wasn’t enough altitude and it hit the water.

Again just a theory like the firmware is to blame is a theory but I think there is some evidence when you look closely at the timing of the motor speeds braking maneuver, pitch changes, and RC inputs.
View attachment 70151

Emergency braking happens because the front sensor detects an obstacle. It's not related to control inputs.

Unfortunately, the tablet .DAT doesn't contain the Obstacle Avoidance data so we can't see that data for this incident. But, the .DAT on the MP does contain that data. It's
OA:frontDistance

 

[Simmo]

Congratulations - you made it onto my rather short ignore list.

Me too!! Thats three!!

 

[razoraerial]

Ok there’s is still one mystery here and that why the act of braking would cause it to go beyond the tilt limit and I have a theory, which is just that, not a conclusion because I don’t know that we will ever know.

When originally looking at this I noticed that the pitch excursion happened during an abrupt throttle up input so my theory has this in mind.

I did some tests today and something I thought was interesting is that the emergency brake only applies to pitch and roll maneuvers. It’s does NOT apply to a throttle input. If throttling up without any attitude inputs from the remote nothing happens. This is true even in windy conditions when the aircraft IS tilting to maintain position.

I did another test to see if when inputting both throttle and attitude with the remote if it would also stop the throttling. It does not. The emergency brake seems to be independent of the throttle control. When doing this test I was cautious and did this at low speed and it thought my Mavic Pro flipped at LOW speed. It was very unnerving even though I’m fully comfortable hitting the emergency brake at full speed in sport mode no throttle.

It doesn’t seem far fetched to me that under the correct circumstances the throttle up command at just the right time during a braking maneuver can cause g forces greater than the motors are able to counter act causing the tilt error.

To demonstrate this effect put your hand out so it’s in the forward pitch state and then quickly change to the reverse pitch State while at the same time raising it up. You’ll feel the whiplash created by this maneuver more so then with just a rapid opposite pitch alone . During an emergency brake we would expect the rear motors to be slowed and the front motors to speed up but we don’t see this. We see all 4 motors speed up. However, we do see the braking in that the front motors are faster than the rear for a short time. We then quickly see the rear motors abruptly go up to max output and the front motors to slow abruptly. My theory is that the drone was put into a maneuver that it lacked the physical ability to recover from and the flight controller stopped the motors. We go see the pitch begin to correct itself but only after the motors are cut. Possibly if there was enough altitude the motors could have possibly been turned back on after the flight controller dumped some inertia by allowing an altitude drop but unfortunately there wasn’t enough altitude and it hit the water.

Again just a theory like the firmware is to blame is a theory but I think there is some evidence when you look closely at the timing of the motor speeds braking maneuver, pitch changes, and RC inputs.
View attachment 70151

Thank you for getting out of high school and back to original post.

 

[razoraerial]

I think this thread just put an end to my full speed Sport Mode Pause Button Tail StandsView attachment 70152

And see I heard about that just didn’t have the guts to try it. Lol

 

[sar104]

Emergency braking happens because the front sensor detects an obstacle. It's not related to control inputs.

Unfortunately, the tablet .DAT doesn't contain the Obstacle Avoidance data so we can't see that data for this incident. But, the .DAT on the MP does contain that data. It's
OA:frontDistance

It often seems strangely random what is and is not in the DAT files - the mobile device DAT has the OA_emergBrake flag but not OA:frontDistance.

 

[BudWalker]

It often seems strangely random what is and is not in the DAT files - the mobile device DAT has the OA_emergBrake flag but not OA:frontDistance.

Those come from different record types in the .DAT. The tablet .DAT doesn't have the record type containing the frontDistance data. The fact that they both have the same prefix is due to the labeling that DatCon does.

 

[brett8883]

Interesting tests although I don't completely understand your description. Were these simulator or flight tests? And I don't know what you mean when you say that you found that emergency braking doesn't apply to a throttle input. Can you elaborate?

Anyway - part of this explanation is demonstrably what happened - the emergency braking during full elevator and full throttle did put the nose up at 75° and the FC did stop the motors due to the excessive attitude. Those observations are in the data and not up for debate.

So the two unresolved questions are (1) why did the firmware, in calm conditions in P-GPS mode, allow the aircraft to reach an attitude that exceeds its own threshold for motor stop and (2) why does the firmware stop the motors in flight due to attitude? (1) suggests that some of the gain values are set too high, although that attitude, per se, is not really a problem - it's perfectly recoverable. These aircraft can recover from inverted and tumbling falls. In other words the attitude achieved in that maneuver is only a problem because of the motor stop function - hence my suggestion that this is a firmware issue - a bug or perhaps, more accurately, a feature with unexpected and undesirable consequences.

There are good reasons for the FC to kill the motors on rollover on the ground, but no good reasons to kill them in the air.

Well maybe this is common knowledge to everyone but myself but I would have assumed the emergency brake would stop both horizontal AND vertical flight. But it doesn’t. It only stops horizontal flight. If holding horizontal position but gaining vertical height pressing the pause button doesn’t do anything. The AC continues to gain altitude. Even if moving both horizontally and vertically the emergency brake will only ignore RC commands for horizontal movement while allowing the vertical movement to continue. This I believe is a design flaw.

In this case we see the AC continues to increase height even as emergency brake is applied. My tests were in real life not the sim where I pushed the throttle up and pitch forward sticks at the same time and then pressed the emergency brake. This was actually hard to do without a third hand and so likely why it hasn’t come up before. The Mavic’s reaction to this was unnerving and I could see if going at a higher rate that the whiplash and resulting g forces could cause the attitude to exceed the limit. And possibly due to the g forces the gyros might temporarily be unable to determine which way is the ground much like the rotational force of a rollercoaster is greater than the force of gravity when going in a loop. This leads to the roll over error even though there is no roll over.

Again don’t think it can be proved with the current evidence but it’s an attempt to explain the unexplainable at this time.

 

[BudWalker]

Well maybe this is common knowledge to everyone but myself but I would have assumed the emergency brake would stop both horizontal AND vertical flight. But it doesn’t. It only stops horizontal flight. If holding horizontal position but gaining vertical height pressing the pause button doesn’t do anything. The AC continues to gain altitude. Even if moving both horizontally and vertically the emergency brake will only ignore RC commands for horizontal movement while allowing the vertical movement to continue. This I believe is a design flaw.

In this case we see the AC continues to increase height even as emergency brake is applied. My tests were in real life not the sim where I pushed the throttle up and pitch forward sticks at the same time and then pressed the emergency brake. This was actually hard to do without a third hand and so likely why it hasn’t come up before. The Mavic’s reaction to this was unnerving and I could see if going at a higher rate that the whiplash and resulting g forces could cause the attitude to exceed the limit. And possibly due to the g forces the gyros might temporarily be unable to determine which way is the ground much like the rotational force of a rollercoaster is greater than the force of gravity when going in a loop. This leads to the roll over error even though there is no roll over.

Again don’t think it can be proved with the current evidence but it’s an attempt to explain the unexplainable at this time.

I don't quite understand. By emergency brake do you mean the pause button? The emergency brake in this incident was applied by the FC which seems to be an attempt to stop forward movement as fast as possible. I don't think the effect is the same as pressing the pause button.

 

[brett8883]

I don't quite understand. By emergency brake do you mean the pause button? The emergency brake in this incident was applied by the FC which seems to be an attempt to stop forward movement as fast as possible. I don't think the effect is the same as pressing the pause button.

In the firmware the pause button is called “emergency brake” and yes it has the same effect like you said “to stop forward movement as quickly as possible” which I think anyone who has ever tried it can verify is exactly what the pause button does. But it does not stop vertical movement which as a result caused this crash. If it stopped both vertical and horizontal movement this crash wouldn’t have happened because the aircraft would have ignored RC commands to ascend which is what I believe and I think @sar104 is saying is what caused the excessive pitch angle because it was trying to brake as well as ascend at the same time.