Hex Troubleshooting

Log Analysis -- deprecated Copter Logs -- deprecated
1

First, sorry if this is long winded.

I’d like to get another set of eyes on these logs from my Hex. It just gets stupid sometimes. I can almost repeat this with some pretty hard input. But not always.

Specs of the copter:
Pretty Straight up DJI F550, All DJI components, all 3DR components (APM, APM Power Lead, 3DR external GPS compass, and 3DR telemetry radios).

APM is mounted, or was with small ‘gel’ pads. Vibs, while not ideal, are in the acceptable range, at least I think they are. I suspected Vibs get out of control during hard movements, and just moved it to the 3d printed earfoam platform with the exact same results.

The best way I can describe this, and what leads up to this is that waypoint transitions are often BRUTAL. Which seems to make, it or can cause it to lose it’s mind. I’ve gone over all of the connections and they seem OK. Flies fine in stability mode, it’s been auto-tuned a few times. It loiters and alt holds. I personally can’t make it crash in loiter, not the way it does.

Yesterday I set up a pretty aggressive auto route, that included lots of 180 degree changes. It made it through those, then circled twice, and right before RTL fell out of the sky.

The log that is attached is a short flight where I took it back up about 10m shook the sticks a bit and got the same result. Thinking this is vib related, but I don’t know, or something bad happens when it gets in it’s own prop wash.

Fortunately nothing broke. I moved the APM to the new anti-vib platform and used the same mission. This time, about 1/2 through the zig zag test, it literally fell out of the sky during transition. Fortunately it caught itself before it hit the ground.

I’m hopeful someone can look at the logs and say: That right there.

2

Here’s a google Snapshot, showing the mission, to test for vibs.

The log below is the log from the first run, where it crashed either transitioning to the final waypoint before RTL or transitioning from that waypoint to RTL, I’m no 100% sure.

It also notes where during the second run, it tumbled and caught itself. 10 less meters of altitude and it would have been bad.

-MD

3

Your copter’s roll and pitch are suddenly diverging from command, which is caused by hardware failure. You may have motor sync issues, you may have a loose connection, or a loose prop.

4

I see that but don’t understand it.

The logs are awful similar, in the Auto-crash and the manually induced crash in stability.

This is pretty basic stuff hardware wise, nothing exotic, I’ve had a motor/ESC sync issue with another quad (motor/ESC) and know what that looks like. I’ve also double checked all the connections, and the plugs from motor/esc and there is no movement between APM and ESC Connectors.

My gut says it’s not hardware but I could be wrong.

I’ve truncated the log to the crashing part, it’s attached below along with screen shots of the pertinent parts.

5

If it isn’t hardware, it could be a tuning issue or MIN_THR too low, such that a motor is stopping. Have you calibrated your ESCs?

6

Yes, ESCs’ have been calibrated, but I’l do that again. I will also look at min throttle, but this isn’t happening in low/no throttle situations.