I tried to create a really powerful quad with HK components and an APM 2.5. It looks like I have created an unflyable monster:
The first flight attempts were a disaster. Upon just touching the throttle it jumped in the air and flipped on its back faster than I could react. It wasn't damaged, so I tried again as slow as possible correcting pitch/roll tendencies before take off. But as soon as it is in the air it's uncontrollable because it's reacting very very fast. It just takes 1/8 throttle to take off. After 3 take off flips it was damaged and I had to stop.
I have calibrated the ESCs and everything is basically working as it should (rotation direction, roll/pitch/yaw etc.). What I noticed was, that one prop was accelerating slower to the idle rotation speed after arming than the others.
Here is my setup:
- Frame: Turnigy H.A.L. (Heavy Aerial Lift) Quadcopter Frame 585mm
- Engines: 4x Turnigy Multistar 4830-420Kv 22Pole Multi-Rotor Outrunner
- ESC: 4x TURNIGY Plush 40amp Speed Controller
- Batteries: 4x 6S 2200mAh
- Props: 12x6 (for now)
My guess is, that the stock PID values are wrong for such a setup. I could also increase the distance between the motors by using the booms from the Hexa HAL Frame.
Any guess what could be wrong, or what a conservative PID setup would look like?
Thanks,
Erik
Replies
I had this exact problem with my Turnigy HAL Hexacopter. Everything seemed fine but as soon as I took off it would flip fast and hard! After 3 days of going through everything and finally setting up a rig to stop the crashes it hit me. I had plugged in the servo wires from the escs to the board in a counter clockwise orientation instead of clockwise. So simple!!! Such mysterious symptoms.
Set your rate-p for both down to 0.0650. Try again. If it takes off but feels to sloppy slowly raise up until it feels solid.
Very good!
HERE is how to obtain and analyze the logs. There is a very good video down the page.
-=Doug
Congratulations on getting to this point, now the real work/fun begins.
1) Make/rig some form of tether mechanism to allow about 16 inches/ 35 cm of vertical space. I used two lengths of cord, one on each opposite leg, tied to a heavy item. The idea is to prevent an out-of-control event but allow some vertical lift for low hovering.
2) Confirm your configurations:
a)Frame type loaded into APM via MP
b)Radio calibrations
c)Motor wiring/positions via a CLI command test found HERE.
3) Find and post your tlog here. It will tell us your parameters and give us a picture of the flight event.
4) Confirm your radio is in ACRO (airplane) mode, not HELI.(What radio TX are you using?)
I do not think your arm length, motor selection, props, etc are the problem.
This type condition usually results in a configuration issue of another kind.
Do not mess with your PIDs, yet.
Let us know...
-=Doug