Crash test

Well I feel better now. I'm always a little nervous to fly a new ship until the first crash. You never know how she'll deal with adversity.

I was flying FPV in loiter at ~15m when it suddenly pulled hard right and back making a nice arc into the ground at about 20kph according to the telemetry log. The current jumped from the ~30A normal to over 60A in the few seconds of the death plunge. When I got there the motors were spinning and wouldn't react to transmitter commands including disarm. All the props were broken of course.

Looked through the logs a bit, but I can't tell what happened. If anybody cares to take a glance and give me some insight, I would be very grateful. I don't mind when things go terrible if I can at least understand why it happened.

The damage seems to be limited to a couple of cracked arms and a few other frame bits. The GoPro was fine thanks to the waterproof housing. Not even a broken antenna on the 3dr radio or video Tx! I'm taking it all apart tonight and checking the electronics, they got covered in snow, but no physical damage that I can see so far. I'll upload some pics once Iphoto stops crashing every time I try to get them off my camera. There must be weird solar weather or something going on today, I've never had a problem getting pics off of my camera before, really strange...

I'll have to upload my tlogs from the windows machine, just a minute.

: )

2013-01-17%2015-24-08.zip

Joshua,

I've been looking at your logs for a bit and I haven't determine exactly what's happened but here's what I've found so far in anycases. It doesn't seem like a GPS issue. The kmz file produces a nice track that looks pretty close to what we see in the video.

As you say, the clock-wise spinning motors are spinning about 150 faster than the counter-clockwise motors meaning that you've there's some imbalance with the props or motors. It's a significant difference and I think it's a factor in the crash.

Another thing to look at is the nav_roll/pitch_roll vs (actual) roll and (actual) pitch. If these diverge wildly then it's normally a failed motor or prop but in this case we don't see that (so it at least wasn't a complete failure of a motor or prop). In graph below we have the autopilot commanded roll, the actual roll and the pilot's input (but it's a raw pwm signal so you can't directly compare the angle). You can see that the actual roll follows the auto pilot roll which means the autopilot wanted that lean angle of nearly 40 degrees. The autopilot asked for that at the moment that the pilot let go of the roll and pitch sticks. It was very aggressively tring to get back to the loiter point which was 5m away.

When it did this, even though it gunned the motor it was unable to maintain it's heading or altitude which is a bit weird. It's also weird that it did not move back towards the loiter target...like it was confused about it's direction.

hmm...this is a tough one.

You need to be a member of diydrones to add comments!

Join diydrones

Comments

T3

Richard Boyhan January 19, 2013 at 7:14am

Joshua, does iMovie 11 have the ability to remove shakes and jitters from a video? Also what kind of Gimbal were you using? It seem to be doing a really good job. I'm not very happy with my $100 gimbal. It's to slow and the angle of movement is limited. How many batteries do you use? It seems from your early thread you were using 2 or 3. I also don't think you had a power failure. When mine browned out, all flight data stopped dead and it dropped like a rock. I was powering my APM with a separate BEC and was also powering my gimbal servos and it was not large enough to handle the load. My flight lasted 30 secs. What was really embarrassing was I was showing off my Hex to some RC helicopter jocks at an RC park. My 30 secs of fame. LOL
Quadguy January 19, 2013 at 6:34am

@Leo, How much current does servo draws? Arducopter ESC is able to supply current upto 2 Amp.
Crashpilot1000 January 19, 2013 at 6:30am

Perhaps a magnetometer issue?

Brown out seems unlikely since the whole system has own lipos and regualtors for everything.
Quadguy January 19, 2013 at 6:27am

@Randy, I just check nav_bearing and yaw angle . It looks like they are going in opposite direction, could this be cause of problem?
Leo Hogg January 19, 2013 at 5:54am

I had the same problem with my quad after installing a camera gimball. I think the servos drew to much current from the power module causing the autopilot to brownout and the quad to lose control. The Vcc log showed some transient under and over voltage situations before the crash.

I would recommend that you analyze the Vcc graph on the Tlog. This may reveal if it was an autopilot brownout.
Developer

Randy January 19, 2013 at 5:28am

Joshua,

I've been looking at your logs for a bit and I haven't determine exactly what's happened but here's what I've found so far in anycases. It doesn't seem like a GPS issue. The kmz file produces a nice track that looks pretty close to what we see in the video.

As you say, the clock-wise spinning motors are spinning about 150 faster than the counter-clockwise motors meaning that you've there's some imbalance with the props or motors. It's a significant difference and I think it's a factor in the crash.

Another thing to look at is the nav_roll/pitch_roll vs (actual) roll and (actual) pitch. If these diverge wildly then it's normally a failed motor or prop but in this case we don't see that (so it at least wasn't a complete failure of a motor or prop). In graph below we have the autopilot commanded roll, the actual roll and the pilot's input (but it's a raw pwm signal so you can't directly compare the angle). You can see that the actual roll follows the auto pilot roll which means the autopilot wanted that lean angle of nearly 40 degrees. The autopilot asked for that at the moment that the pilot let go of the roll and pitch sticks. It was very aggressively tring to get back to the loiter point which was 5m away.

When it did this, even though it gunned the motor it was unable to maintain it's heading or altitude which is a bit weird. It's also weird that it did not move back towards the loiter target...like it was confused about it's direction.

hmm...this is a tough one.
Joshua Ott January 18, 2013 at 9:52pm

Looking at this log I see that ch 3&4 output is consistently higher than 1&2. This seems like a constant yaw correction to the left. Does this look strange? I'm very conscientious about getting all my motor shafts vertical.

I looked at "magfield" and it is moving around for sure, but I didn't notice any big spikes when the motors first come online or before the crash. Not sure what normal looks like for this.
Joshua Ott January 18, 2013 at 9:46pm

Thanks Richard, I'm just using iMovie 11 that came with my computer. It works well and the learning curve is friendly.

My first career was in film production back in the late '80s and through the '90s, during that period non-linear editing systems like Avid were just coming into the industry. Editing was always something that one paid a very pale person to do while you sat in a chair behind him waiting for what seemed like hours between each little change. Sometimes he would have to go back behind all the stacks of equipment and move a bunch of plugs around like an old time switchboard telephone operator.

Now I sit in my pajamas drinking a beer while throwing together a quick edit of the days flying robot camera footage. So awesome to be alive now that the future is finally here!
T3

Richard Boyhan January 18, 2013 at 8:24pm

Very cool video! Sorry about the crash. Really like the slo-mo at the end. What video editing software are you using?
Joshua Ott January 18, 2013 at 4:01pm

Greg, The APM gimbal control is such a pleasure to use, it was very easy to get that part of my system working, plus the interface on mission planner really makes it simple to mess around with. Thanks!

I'm looking at work being done with direct drive brushless systems, I assume the APM gimbal controls will work just as well talking to an ESC as they do a servo?

of 4

This reply was deleted.