I need some help, I have had some issues where my quad has been repeatedly flipping over mid-flight. Once the quad is on the ground a very strange sound is emitted. I have checked my motors and ESC's and they are all perfect. I am thinking that it might be a glitch. I have a videobelow which continently catches it flipping. Luckily All I have broken so far is a few props and some nylon bolts.
Please help me!
Please look at my new discussion, the problem continues!!
the latest code has had us spending about a month bullet proofing the "DCM" implementation. A few bugs were found that could cause this to happen. They are very rare but certain vibrations and frame setups could induce it. We feel very confident the next version - now 2.5 will solve your issue. It also flies even better!
I expect the release to happen in the next day.
I had a bit similar issue. Perfect hover and by applying yaw one could make it flip.
Changed the ESC --> no improvement. It was always the motor 1 that shut down and caused the flip.
Tired with handheld and could reproduce the issue.
Then I switched motors 1-2 (the opposite ones). All of sudden motor 2 was shutting down --> motor was an issued.
Changed the faulty motor and all worked smoothly.
Please test handheld and try to reproduce the flip. If it is always the same motor that shuts down, then try to switch Esc and motor and retest. You will find the faulty part.
Calibrating esc trottle could be done as well...
This was how I solved my similar issue. A faulty motor (running smooth on normal load by by applyin yaw caused higher load and shutdown) Very similar situation than yours.
I suspect a hardware issue. Maybe you haven't initialised your ESCs so an engine is shutting down when your throttle gets too low? Is it always the same engine that stops turning (i.e. does it always flip the same way?). If you move the ESC to another arm does the problem move with the ESC?
Maybe try putting something heavy on the quad to force it to work harder (i.e. keep the motors spinning faster). If the problem doesn't happen then you know this theory is correct and it's time to check (or reinitialise) the ESCs.
check connection between motors and ESCs, it should be soldered directly, do not use connectors.
The reason I am posting now is that I have tryed moving ESC's and motors. It still happens.
@Randy I have calibrated all my ESC's separately with the TURNIGY programming card, they are all set to shut down at the lowest possible battery voltage setting, the soft start is off and the brake is off.
@heikki.Tukiainen I am assuming it is the APM because it has been flipping on different sides.
@jasonshort I will load up v2.5 as soon as it comes out and I will feedback to you guys.
@Duy Pham I am quite sure my connections are solid, I would solder them but I don't really want to :), if I load v2.5 and the problem continues I will solder them.
Thanks to all
PS I just wrote that entire reply on my iPod, what a mission!!
Craig, can you post the .log file of that flight?
I don't have telemetry, I might be wrong but from what I understand I could only get the .log if I have telemetry, if I can get it, them sure.
The .tlog's come from telemetry but the onboard .logs are recorded every flight, instructions here: http://code.google.com/p/arducopter/wiki/AC2_Logs
Craig, I strongly recommend you solder your motor-esc connections. I used to think like you - solid connections checked with a volt-ohm meter. But after three unexplained flips at rather high altitude, I soldered them and have gone through a hunderd battery cycles since with no issues. This may not be your problem, but its an easy thing to try. I have also found that having to solder the connections to check or replace a motor or esc is not as big deal as I thought it would be.
As I said before, I'm going to try v2.5 for a while just with some low altitude flights then I will solder them. Thanks for the insight, was your doing the exact same thing?
I see that there is a strong consensus for soldering the connectors between ESCs and motors in the forum, but I have not bothered either. Why? Two reasons :
1) I tend to crash from time to time due to the nature of this sw, it is unstable. The crashes could very well destroy a motor, or needing a replacement of an arm etc which is more difficult with soldered connectors.
2) I have flown 3 different helis 2 x 450 size and one 500 size for a couple of years, all having more vibration than I guess an arducopter will produce, and NEVER had any issue with the connection between ESC and motor.
If anyone could explain to me that there is a REAL reason for soldering the connectors, I will do it at once. As the situation is today, it seems to me that this is more like a "couldn't hurt" measure, rather than a "must do" (like for instance threadlocking metal screws, which I certainly do, and I think is this is not suggested in the wiki manual either, which I in my opinion should be there)
I would be happy to be convinced otherwise, but as of today I am not,
Hey - I'm not here to convince you of anything. I am just saying that I had several crashes exactly like in the above video. In between each crash, I loaded a different release of the software thinking it was a software issue. After the last crash, I tested the copter on my test bench and luckily witnessed a buzzing, non functional motor. After wiggling a connector that I thought was bullet-proof, the motor came to life.
The other point to make is that if your soldering slkills are anything better than beginner, soldering and unsoldering motor wires to perform maintenance is a snap - takes only slightly more time than the connectors do (remember, it takes time to solder connectors on new components too).
But the best reason to solder is peace of mind - to be sure connectors are not a problem and thus avoid chilling events like that in the above video if the connector fails. There's high current and vibration acting on the connectors. I too thought the connectors couldn't be a probelm at first.
Just trying to share my experience. My feelings will not be hurt if you ignore my advice.