Solo Flyaway

[jasonhr13]

And here is the log files
logs.zip

 

[Ian [P13]]

then show that it switched to land.

Which is what I said it would do, land is a controlled descent for speed so wouldn't hit the ground with force, so it's a surprise to see the shell broken, even if it landed sideways and flipped. Given land kicks in at 10% the only reason that immediately comes to mind for the shell being broken is if the battery didn't have enough charge to complete the land.

 

[Dirby]

Wow that was a Hard Hit.

 

[BobG]

Lot of weird things in that log. Before takeoff it says "leaning" but allows you to arm. At some point it goes into RTH with still a 3D fix. It rises to 89m you get a failsafe message and then something called "position horizontal variance" and Solo switches from RTH to Fly: Manual (alt hold). Then another error, "Velocity variance." and finally "failsafe." Throughout it says you have a 3D fix.

 

[Ian [P13]]

Those will be EKF errors looking at the names (off the top of my head), if GPS is above 6 it'll still show as having GPS but it needn't necessarily be accurate, likewise a compass issue could cause the EKF to start a failsafe too. The EKF is an algorithm to work out if the sensors are providing valid data or not, if not it goes to Fly:manual and where appropriate to land. It can happen simply because the GPS was crap so you can't read too much into the errors. Mission Planner is going to better support analysing EKF errors in future.

Although it's NOT an indicator used in flight for failsafe it's worth looking at the HDOP. The value actually used to figure out the GPS is inaccurate is sAcc but you can't see that...

 

[jasonhr13]

Which is what I said it would do, land is a controlled descent for speed so wouldn't hit the ground with force, so it's a surprise to see the shell broken, even if it landed sideways and flipped. Given land kicks in at 10% the only reason that immediately comes to mind for the shell being broken is if the battery didn't have enough charge to complete the land.

The solo definitely wasn't low on battery. When I recovered the drone. the battery has been flung off of it. But I clicked the button on the battery and it had over 70% charge left. I took off from the ground with 100% battery and the drone crashed about 2 min later after take off.

 

[z3r01]

wow

 

[Retired2Fly]

Lot of weird things in that log. Before takeoff it says "leaning" but allows you to arm. At some point it goes into RTH with still a 3D fix. It rises to 89m you get a failsafe message and then something called "position horizontal variance" and Solo switches from RTH to Fly: Manual (alt hold). Then another error, "Velocity variance." and finally "failsafe." Throughout it says you have a 3D fix.

Solo will allow you to arm if it is leaning relative to the horizon. The second logging event labeled solo.tlog.1 in the provided zip file is the actual flight in question. There are a lot of compass errors in that flight which is usually indicative of a failed ESC/motor. The Pixhawk II in Solo tries to recover as best it can but to no avail. The Pixhawk II then loses connection to the solo TX so yet another failsafe scenario must be evaluated. The FC then enters failsafe RTL but then loses GPS almost immediately so yet another decision to make. The only other option available to the FC is to initiate the land failsafe. Actually that is some very clever programming which in real-time is very difficult to execute flawlessly! Bottom line the Pixhawk II chose the least of several bad scenarios to attempt to bring down solo as safely as possible. The visual flight path in MP correlates solo's .erratic flight behavior. At least no one was injured so I would call the Pixhawk II's decision making in this un-winnable scenario a success.

 

[jasonhr13]

Solo will allow you to arm if it is leaning relative to the horizon. The second logging event labeled solo.tlog.1 in the provided zip file is the actual flight in question. There are a lot of compass errors in that flight which is usually indicative of a failed ESC/motor. The Pixhawk II in Solo tries to recover as best it can but to no avail. The Pixhawk II then loses connection to the solo TX so yet another failsafe scenario must be evaluated. The FC then enters failsafe RTL but then loses GPS almost immediately so yet another decision to make. The only other option available to the FC is to initiate the land failsafe. Actually that is some very clever programming which in real-time is very difficult to execute flawlessly! Bottom line the Pixhawk II chose the least of several bad scenarios to attempt to bring down solo as safely as possible. The visual flight path in MP correlates solo's .erratic flight behavior. At least no one was injured so I would call the Pixhawk II's decision making in this un-winnable scenario a success.

That's pretty cool to see. So do you think that log would backup the case that 3dr would replace the drone? The imGe doesn't show it but the flight path continued in that line until t took a sharp u turn at the back of the track and I found it crashed by the tree line in the baseball field by the basketball courts.

 

[jasonhr13]

I only know that because of the recovered gopro footage.

 

[Retired2Fly]

That's pretty cool to see. So do you think that log would backup the case that 3dr would replace the drone? The imGe doesn't show it but the flight path continued in that line until t took a sharp u turn at the back of the track and I found it crashed by the tree line in the baseball field by the basketball courts.

My guess is that if you sent 3DR the same logs they will determine it is a failed motor pod. Did you notice any unusual noise when you armed solo? Also the last log shows it may have bounced and then continued flying ~54m which may account for the location near the basketball court when you plot the last log. The other logs show you were having some pre-arm issues.

For future reference you might want to find a safer place to fly than a crowded neighborhood next to a descent size highway. It is quite plausible in this out of control scenario solo could have landed on that road between you and the baseball field and caused an accident or worse. When flying always assume the worse can happen so you are always prepared. Anyway lessons learned.

 

[Jubalr]

My guess is that if you sent 3DR the same logs they will determine it is a failed motor pod. Did you notice any unusual noise when you armed solo? Also the last log shows it may have bounced and then continued flying ~54m which may account for the location near the basketball court when you plot the last log. The other logs show you were having some pre-arm issues.

For future reference you might want to find a safer place to fly than a crowded neighborhood next to a descent size highway. It is quite plausible in this out of control scenario solo could have landed on that road between you and the baseball field and caused an accident or worse. When flying always assume the worse can happen so you are always prepared. Anyway lessons learned.

I agree with the above; First the area you were flying in. It's a pet peeve of mine, but to see people that step out in their front yard in a crowded neighborhood, especially when still learning the equipment.

Regarding the logs, I too see a motor problem. Not enough for one to quit, as it would have hit the ground a lot sooner. But I see where the Solo was drawing as much as 35a! That is quite a load considering it wasn't climbing. Typical hover is in the 13-17a range. And it was spiking like crazy between 15-30a.

If it hit the ground hard while trying to land, it probably means there was fair amount of wind. Without GPS it was landing in Manual and would have been traveling at roughly the speed of the wind when touching down.

 

[Lamda]

That's pretty cool to see. So do you think that log would backup the case that 3dr would replace the drone? The imGe doesn't show it but the flight path continued in that line until t took a sharp u turn at the back of the track and I found it crashed by the tree line in the baseball field by the basketball courts.

My first Solo had a failed motor mid-flight after about 10 flights. Flipped instantly and crashed. Logs showed similar behaviour, with position errors etc. due to the unbalanced flight. Solo switched to Alt-Hold once inverted. Not a ton of damage to the vehicle - landed in sand along a beach. After 3DR analyzed the logs I was offered to be shipped new motor pods to do the repairs myself or to send the bird back in for them to repair/replace as needed.

 

[jasonhr13]

I agree with you all, the location wasn't a great choice. I did not fully think through the worst outcome and expected a smooth flight. So that is a lesson learned and the good news in all this is that no one was hurt and it hit a open field.

There actually wasn't a lot of wind. Obviously more up high then down where I was but it was a pretty calm evening during the flight. But I am very glad for all the information you guys have given me. I am looking forward to hearing back from 3DR about this.

 

[ChrisM101]

There seems to be a problem with the latest software though. I have a friend who just got his second Solo. He stated he was flying out and at about 1000ft said his video in app and control was lost, but he could visualize it, It did not RTH? it sat there hovering for like 20 seconds so he starts moving toward it until he can get it to respond so this takes another 10 or so seconds. 30 seconds with no control, yet no RTH. That makes you think that GPS is lost and its hovering. but honestly in an open area far above structures, why are people losing GPS...? There is still a hardware or Quality Assurance issue here.

 

[brunoferreiro]

auch

 

[GNTLMN_Z]

Any word from 3DR if this is going to be covered by them or not?

 

[rotorwash]

I agree with you all, the location wasn't a great choice. I did not fully think through the worst outcome and expected a smooth flight. So that is a lesson learned and the good news in all this is that no one was hurt and it hit a open field.

There actually wasn't a lot of wind. Obviously more up high then down where I was but it was a pretty calm evening during the flight. But I am very glad for all the information you guys have given me. I am looking forward to hearing back from 3DR about this.

Unfortunately, there are A LOT of people that have this same mind set. A lot of people ignore safety procedures until something like this happens to them.

 

[jasonhr13]

Unfortunately, there are A LOT of people that have this same mind set. A lot of people ignore safety procedures until something like this happens to them.

3DR told me I would hear back today around 4pm PST. so we will see what they say and I'll update here

 

[ChrisM101]

3DR told me I would hear back today around 4pm PST. so we will see what they say and I'll update here

This reminds me I need to email them again about where mine is....