Close Call

[GlassKnees]

Wow, I somehow managed to avert disaster, though I don't know how. I decided to take out my modified Elev-8 out for a flight around the golf course, flying autonomously using waypoints I uploaded to the APM controller. I had previously uploaded a new version of firmware, then checked parameters and tweaked the compass declination a bit in order to get the aircraft heading to agree with my iPhone. Satisfied, I set up my ground station outside, fired up the quad and sent it on its way.

Almost immediately, I noticed something was wrong as it seemed to reach the first waypoint but continued on without loitering as I had expected it to. Then I heard a voice from Mission Planner say loss of GPS. I had already switched to RTL, but given that I had no (valid) GPS it was on its own, so I switched back to manual mode (Stabilize), but then Mission Planner said "No R/C". Crap. The aircraft is flying on its own and I have no control!

It seemed to respond to yaw commands, but I could not throttle down the aircraft even though I had my throttle stick completely in the down position. It was up pretty high and it was difficult to discern its orientation. Somehow, I seemed to regain control - throttle was responding and I managed to lower it and once I determined its orientation, I was able to bring it back close to my house, but I was not confident enough to attempt a landing on the pad I set up, so I managed to bring it down on the golf cart path behind my house - it settled down gently on the cart path and the motors shut down - whew!

I tried to play back the telemetry file but it doesn't seem to work, so all I have is the dataflash file to try to figure out what went wrong. Really scary flight, but at least it ended okay..

 

[GlassKnees]

Wow, I somehow managed to avert disaster, though I don't know how. I decided to take out my modified Elev-8 out for a flight around the golf course, flying autonomously using waypoints I uploaded to the APM controller. I had previously uploaded a new version of firmware, then checked parameters and tweaked the compass declination a bit in order to get the aircraft heading to agree with my iPhone. Satisfied, I set up my ground station outside, fired up the quad and sent it on its way.

Almost immediately, I noticed something was wrong as it seemed to reach the first waypoint but continued on without loitering as I had expected it to. Then I heard a voice from Mission Planner say loss of GPS. I had already switched to RTL, but given that I had no (valid) GPS it was on its own, so I switched back to manual mode (Stabilize), but then Mission Planner said "No R/C". Crap. The aircraft is flying on its own and I have no control!

It seemed to respond to yaw commands, but I could not throttle down the aircraft even though I had my throttle stick completely in the down position. It was up pretty high and it was difficult to discern its orientation. Somehow, I seemed to regain control - throttle was responding and I managed to lower it and once I determined its orientation, I was able to bring it back close to my house, but I was not confident enough to attempt a landing on the pad I set up, so I managed to bring it down on the golf cart path behind my house - it settled down gently on the cart path and the motors shut down - whew!

I tried to play back the telemetry file but it doesn't seem to work, so all I have is the dataflash file to try to figure out what went wrong. Really scary flight, but at least it ended okay..

I decided to re-calibrate the compass and radio. Then I checked the value of the radio throttle failsafe setting - satisfied that it is below the minimum throttle value, I cleared the logs and took it out to a field for a test flight. The flight went well - in Loiter, it held position and RTL worked fine. I am guessing that doing the firmware upgrade messed with some of the calibration settings.

 

[RENOV8R]

Loss of control is pretty scary, glad you were able to regain control and get your bird back. I know you've put a lot of hours into it!

 

[LoneRCRanger]

Sounds like it was almost a SIP event.

 

[GlassKnees]

Loss of control is pretty scary, glad you were able to regain control and get your bird back. I know you've put a lot of hours into it!

Thanks! I brought out my laptop to the field for the second flight and when I reviewed the telemetry, I saw a voltage failsafe was triggered shortly after takeoff. I had almost fully charged the 4S lipo and so I was surprised at this. I checked the voltage record on the data flash log and saw then when I took off, the voltage dropped to around 14 V which is my failsafe trigger. I flew for about four minutes or so without any problem.

For some reason, I was not able to specify an action, such as Land or RTL when the voltage failsafe is triggered, so I just got a warning on Mission Planner. This morning, I calibrated the voltage sensor so that Mission Planner reads the same voltage as what I read with my multimeter, but I suspect that perhaps the lipo is failing to hold its charge and its dropping too much - may have to experiment.

I captured some FPV video of some of the original incident - video quality is terrible possibly due to a cheap camera but also, lousy reception from the aircraft to my monitor. I may post it - if I do, I will provide a link....

 

[GlassKnees]

A new development. In trying to resolve the voltage failsafe matter mentioned in my last post, I attempted to run a test in my basement - flip the propellers so when I throttle the motors up, the props will push the aircraft downward; this way, I can run a power test safely in my basement. After arming the aircraft, I throttled up but the number 1 motor failed to spin. I tried several time, but no joy.

Not having a spare Tarot motor, I decided to swap the #1 with one of the others and run a motor test - again, the #1 motor (the motor attached to the #1 boom) failed to run. This means either a bad connection or an ESC problem. This is bad news because it means I pretty much have to disassemble the aircraft. So, over the next few days, I took it apart and removed the ESC's. Immediately, I noticed a problem.

Back when I first built this aircraft, I was concerned about voltage spikes from the side of the ESCs receiving main power, I decided to install 470 microfarad capacitors across the power and ground leads, but for some reason there wasn't one on the #1 ESC. My memory fails me but I suspect that I only had three capacitors and so didn't obtain and install a forth - DOH!!



So, I think I will replace all four ESCs. I don't think that I wll need the capacitors as they go inside the body of the aircraft and the power/ground leads are relatively short (shorter than they would be if installed out on the booms underneath the motors. I may document the rebuild process.