Starling Position and Manual Mode dysfunctional
-
Hello,
My name is Ivy and I am using the Starling drone for research at my university. So, whenever I would arm it in position mode, it would immediately take off and hover 2 inches above the ground-would not get any higher than that-and hover to the left and would not receive any input from the commando 8 controller. Then, if you put it in manual mode and arm the drone, it would rise 2 inches off the ground and do an aggressive flip and crash-again not receiving any input from the controller. I recalibrated the drone through QGC and recalibrated the commando 8 controller and it would still veer to the left. Before, it worked as normal where you arm it and just the propellers would spin and received input from the controller. Before I flew the drone I looked for physical damage to the propellers or motors and there was not any. During the flights where it kept flipping I was checking the propellers and it was fine. Then after many flips and crashes one of the propellers tore a little bit on the front right motor. But the drone was veering to the left and flipping way before the propeller tore. I honestly have no clue how to fix the reason its veering to the left and immediately taking off when armed. I ordered new propellers but I don't think that is the solution for what's going on any help would be great!
Thanks,
Ivy
Here is the tear in the front right propeller
Here is the left side of the drone
-
Hey!
Here is a link to videos of the drone veering to the left when flying. The new propellers have arrived and I will be replacing all the current ones. I don't think that will fix it though. A colleague of mine checked the amps of the motors and found that the back right motor from the camera was producing lower amps than the others. We hypothesized that the reason the drone kept flipping to the left is because the right motors were overcompensating maybe? I have no clue any advice would be appreciated.
1 - .05 amp Camera right front
2 - .031 amp camera right back
3 - .046 amp camera left back
4 - .045 amp camera left front
Starling Drone flight video: https://photos.app.goo.gl/RxEaN8YGxXUdh1j87
Thanks!
-
@ijones2 Can you attach a log from the manual mode flight? (Perhaps with the messed up propellers so that you don't break the new ones)
-
Hey!
Here is the flight log in manual mode. I also uploaded the videos of the drone in Manual mode they are really short simply because it would flip violently then I had to stop it really quickly. Also on the last attempt at the drone in manual it flipped then 2 of the propellers snapped off. So I have to change those along with the torn one. I hope this helps!
Thanks
-
@ijones2 It's best to get manual mode working well before trying any of the more advanced flight modes such as VIO position hold. The log shows that after you arm with the left stick down and toward the center in manual mode the stick goes back to center position. You need to keep the throttle all the way down and then slowly increase to takeoff. Also, for testing in manual mode, try changing the parameter EKF2_EV_CTRL to 0 in case bad VIO data is corrupting the state estimation.
-
@ijones2 But the log also shows RPM being commanded to 10K even when the throttle is all the way down so that is confusing. We're still investigating.
-
@ijones2 said in Starling Position and Manual Mode dysfunctional:
Before, it worked as normal where you arm it and just the propellers would spin and received input from the controller
So the drone did work as expected in the past? Was there anything obvious that happened before it stopped working? Does it have the same SDK installed that it was shipped with or has that been changed?
-
@Eric-Katzfey said in Starling Position and Manual Mode dysfunctional:
You need to keep the throttle all the way down and then slowly increase to takeoff.
Yes but after arming the drone, the drone would no longer receive any input from the controller. That was what I was originally doing but once I and few of my other colleagues gave it try, moving the sticks in every direction, the drone was not responding. Even when we would try to unarm it right stick down the drone was not responding.
-
Yes the drone operated normal. I got to a point where I would arm it in position mode have it hover 7 feet, and then perform figure eights. In practicing landing it would the floor but the floor in my lab is foam matting which is better than hitting the concrete. To be honest it has 2 or 3 rough landings but after checking after each landing, I nor my colleagues inspected any physical damage. The SDK has been updated 1.3.3. it originally had 1.1.3. I think until I updated it last month.
-
@ijones2 Did you have any good flights after updating to 1.3.3?
-
Yes!
I had several good flights with 1.3.3. I also had good flights with 1.1.3 as well.
-
@ijones2 Might be worth a try then to do a fresh installation of 1.3.3 and see if that helps. Perhaps a file got corrupted somewhere and a fresh installation will clear it out. Can you also inspect the motors carefully? It's possible some debris got into one of them and it's causing it to occasionally get hung up. Really not sure what's going on but I think the fresh installation of 1.3.3 would be a good start.