Mavic drifted out of control

[BudWalker]

It looks like auto takeoff, so that presumably explains the absence of throttle.

But, it appears it ascended, just barely. The vps didn't show any ascension. Looks like it may have scooted, dragging one or more of it's feet. The eventLog stream shows it didn't think it was airborne until 12 secs.

@gulfrider can you describe more of what happened in the first 12 secs?

 

[BudWalker]

It looks like auto takeoff, so that presumably explains the absence of throttle.

Was there an indication of auto takeoff?

 

[sar104]

Was there an indication of auto takeoff?

I understood this to be the RC instruction for auto takeoff:

[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff​

Manual takeoff just reports as:

[L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF​

I have to admit that I haven't confirmed that however.

 

[BudWalker]

I understood this to be the RC instruction for auto takeoff:

[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff​

Manual takeoff just reports as:

[L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF​

I have to admit that I haven't confirmed that however.

I think

[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff

is just your normal, manual takeoff. I don't know what

[L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF

is.

 

[sar104]

I think

[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff

is just your normal, manual takeoff. I don't know what

[L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF

is.

"[L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF" is what my M2 reports on takeoff.

 

[BudWalker]

"[L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF" is what my M2 reports on takeoff.

The M2P, P4Pro and MA report

L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF

on manual launch.

But, the MP reports

[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff

on manual launch. I looked at one my MA .DATs and a few others.

I don't recall exactly but when auto takeoff is used there is an explicit mention of it in the eventLog stream.

 

[BudWalker]

Here's an example of a P4 using auto takeoff
[L-FLYMODE][Ctrl<5>] REQ_APP AUTO_TAKEOFF ctrl_auto_takeoff
whereas a manual takeoff (i.e. directed by the RC) looks like
[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff

I'll bet that if you did an auto takeoff with the M2 it would say something like
L-FLYMODE]MR_REQ_APP_ASST_TAKEOFF

 

[sar104]

The M2P, P4Pro and MA report

L-FLYMODE]MR_REQ_R_RC_ASST_TAKEOFF

on manual launch.

But, the MP reports

[L-FLYMODE][Ctrl<2>] REQ_RC_COMMAND ASST_TAKEOFF ctrl_asst_takeoff

on manual launch. I looked at one my MA .DATs and a few others.

I don't recall exactly but when auto takeoff is used there is an explicit mention of it in the eventLog stream.

OK - well that seems conclusive. So it is back to the question of what induced takeoff

But, it appears it ascended, just barely. The vps didn't show any ascension. Looks like it may have scooted, dragging one or more of it's feet. The eventLog stream shows it didn't think it was airborne until 12 secs.

@gulfrider can you describe more of what happened in the first 12 secs?

Looking at the log - it was a throttle takeoff - just a very tentative one:

 

[BudWalker]

But, it appears to have moved during those 10 secs before the throttle input. See my post #39.
Mavic drifted out of control

It also appears to have ascended a little during those 10 secs. The magYaw changes indicating a change in the geomagnetic distortion. It would have had to have been a small ascension as the vps data remained constant.


Looking at this some more it appears to have moved in the 6 secs prior to motorStart. But, the IMU data has it stationary during those 6 secs. So, I guess that leaves GPS location as drifting. But, the FC had set navHealth to 4 and was using GPS for navigation.

 

[sar104]

But, it appears to have moved during those 10 secs before the throttle input. See my post #39.
Mavic drifted out of control

It also appears to have ascended a little during those 10 secs. The magYaw changes indicating a change in the geomagnetic distortion. It would have had to have been a small ascension as the vps data remained constant.
View attachment 59037

Looking at this some more it appears to have moved in the 6 secs prior to motorStart. But, the IMU data has it stationary during those 6 secs. So, I guess that leaves GPS location as drifting. But, the FC had set navHealth to 4 and was using GPS for navigation.

Yes - there was pretty clear evidence of GPS drift.

 

[gulfrider]

That's OK. Since yesterday I was trying to figure out why the MP lifted off without control inputs. Didn't even think about looking at the mag data.

Appears the MP just barely lifted off and then became a victim of it's mag problems.

Bud Walker/Sar104

Thank you for looking at the link and giving me your analysis. The MP was launched at the hood of my Jeep. From there it started drifting as you all have mentioned. Curious about what you mentioned on magnetic interference (hopefully I said that right). the location was in the woods and there were no power lines or strong antennas close.

So now, should i recalibrate the compass or compass and IMU? What other things would you recommend i take a look at before i take it to flight again. The only damage i can visually see is the small chip on one prop and that’s it.

 

[rickw]

might have been the metal hood on your jeep!

 

[Meta4]

The MP was launched at the hood of my Jeep.

Never launch from a steel or reinforced concrete surface.
It's just asking for trouble.

So now, should i recalibrate the compass or compass and IMU? What other things would you recommend i take a look at before i take it to flight again.

Don't do anything to the IMU and you probably don't need to do anything with the compass either.
Go and launch from a clear area well away from cars, steel and concrete.
Launch and put the drone 6+ feet in the air to observe how it hovers.
If it holds position without slowly spiraling, it's good to go.

 

[rfc]

But, it appears to have moved during those 10 secs before the throttle input.
It also appears to have ascended a little during those 10 secs.

The OP reports he took off from the hood of his jeep. I tried this ONCE (and never again). Just the motor start provided just enough lift, perhaps combined with the vibration of the props to start sliding forward down the gentle incline of my Saab, and then right off it (by which time I applied full up throttle).
If the OP's Jeep hood were similarly inclined, could that explain the first 12 seconds?
What would happen if it found itself with motors started, and 3 feet off the ground? Similar perhaps to a handheld launch. Would it just drop?

 

[sar104]

Bud Walker/Sar104

Thank you for looking at the link and giving me your analysis. The MP was launched at the hood of my Jeep. From there it started drifting as you all have mentioned. Curious about what you mentioned on magnetic interference (hopefully I said that right). the location was in the woods and there were no power lines or strong antennas close.

So now, should i recalibrate the compass or compass and IMU? What other things would you recommend i take a look at before i take it to flight again. The only damage i can visually see is the small chip on one prop and that’s it.

Assuming that the hood is steel - that's the explanation for the magnetic distortion. Steel is ferromagnetic and easily magnetized. As others mentioned above - never launch from a steel surface.

 

[gulfrider]

Assuming that the hood is steel - that's the explanation for the magnetic distortion. Steel is ferromagnetic and easily magnetized. As others mentioned above - never launch from a steel surface.

Thank you so much y’all. I’ve learned my lesson. Hopefully the damage to the drone prop hitting the tree didn’t do much damage. I can only Google so much and your analysis has been extremely informative and helpful to me.

 

[BudWalker]

Bud Walker/Sar104

Thank you for looking at the link and giving me your analysis. The MP was launched at the hood of my Jeep. From there it started drifting as you all have mentioned. Curious about what you mentioned on magnetic interference (hopefully I said that right). the location was in the woods and there were no power lines or strong antennas close.

So now, should i recalibrate the compass or compass and IMU? What other things would you recommend i take a look at before i take it to flight again. The only damage i can visually see is the small chip on one prop and that’s it.

Just to be clear. Did the MP actually move/drift in the first 12 secs - from motorStart to when you first applied throttle?

 

[BudWalker]

The OP reports he took off from the hood of his jeep. I tried this ONCE (and never again). Just the motor start provided just enough lift, perhaps combined with the vibration of the props to start sliding forward down the gentle incline of my Saab, and then right off it (by which time I applied full up throttle).
If the OP's Jeep hood were similarly inclined, could that explain the first 12 seconds?
What would happen if it found itself with motors started, and 3 feet off the ground? Similar perhaps to a handheld launch. Would it just drop?

Shouldn't just drop. There have been reports of starting the motors while on a ship at sea. When the ship dropped the AC responded by becoming airborne. I.e., it tried to maintain it's initial altitude when the ship dropped out from under it.

 

[gulfrider]

Just to be clear. Did the MP actually move/drift in the first 12 secs - from motorStart to when you first applied throttle?

Yes it did. It did not hover at all like it’s supposed to do.

 

[sar104]

Yes it did. It did not hover at all like it’s supposed to do.

But that's not the question being asked - which is did the aircraft move before you instructed it to take off? It's not supposed to move, or hover, after motor start but before takeoff.