My Phantom flew away

So last night was pretty windy out at Patrola Park. I was doing okay, then decided to take it up above the trees, and off it went and out of sight.

I tried to get it back and I still has some video feed, I don't know if it was just low on battery or what, but it couldn't seem to fight against the wind at this point and drifted away until I lost video signal.

I flipped it to GPS, and gave it some altitude, but it looks like it just couldn't hold and kept going. I regain video for a few seconds and tried to turn it back around, but the feed cut out again, and it was gone.

So as I came to terms with the situation, I turned of the controler with a last ditch hope that it would Return to home (not quite fathoming the strength of the wind.) Wainint there , hoping to hear the sound of the Phanom as it came home. Nothing.

Definitely a heart sinking feeling.

We drove around, but theres fields and bushes all over the place. After about 5 mins or so of looking, the prospects we're pretty much nil. I figured I'd post here and maybe do some flyers in the area in case someone found it.

I drove to that section of road where it ends at the old train crossing on the south side. scanning a nearby field and then looking across the tracks at the old mill.

wait...what? Wooo hoo I see it tipped up on it's side! A very lucky day.

Here's the video, theres a section where the gimbal freaks out because the battery was running low. And I compressed some time when it was just sitting there for several minutes. Enjoy.

 

Thats some crazy footage the gimbals going nuts im glad you found it!

 

We'll have to stop calling you "Jonus" and start calling you "Lucky". Quite a video!

In the first part of the video, I was blow away with how well your gimbal was working. You could see the frame going through all kinds of wild gyrations and the video was really holding solid. Very impressive performance from that gimbal for sure.

Glad the story ends well. Think I'd stay home when the wind is blowing that hard next time. You couldn't get this lucky twice!

...Tiger

 

I wish a had FPV gear on my Quad when we had all that rainy weather the clouds where amazing !

 

Well looks like the Quad is in find shape, however something has gone amiss with the gimbal might take some time to figure out whats up. Basically it's just flipping irradically, and sending the basecame software for the gimble 100-'s of errors per second. Guessing something to do with the Accelerometer.

Wast interesting is you notice there is a spot in the video where the Phantom seems to have totally lost it's heading and goes into an fast spin. Now the gimbal has its own redundant accelerometer , but I wonder if something odd happened on that flight. It wasn't long after that spin that the gimble starts to freakout.

I dunno, hopefuly I'll be able to debug whats happening.

 

@Jonas - Wow! What a great video. Not great that it flew away, I'm glad you found it!

It looks like it was auto-landing with a battery low condition since it descended nicely.

I saw a gimbal do weird things once when the front axis was spun around 360 degrees. I don't know if it was because it was turned or because the wire wrapped around.

 

Yeah the gimbal thing is weird. My 1st inclination when I saw the video was that it was just running low on power. However so far I haven't been able to get the gimbal to act correctly since. And its not like it crashed,so its a mystery why I'm still having issues on the front. Its throwing a boat load of ICS errors and wigging out pretty much like in the video. I've tried calibrating it, I should check the voltages and maybe separate it off the main battery and see what it does.

 

Have you fixed your gimbal? I'm interested in hearing what you've found out.

 

not yet. I've checked the incoming voltage and that looks okay. I need to do some more debugging around the i2c sensor. I'm not seeing any physical damage to it, the cable or the main board. I need to verify the able has continuity and go from there.

 

Is the i2c sensor the one that is on the back of the camera? If you unplug that sensor do you still get a stream of errors?

 

So did some more testing. Yep the errors still happen even if the IMU is disconnected.

Something else I've noticed is in the configuration software (SimpleBCG) sometimes it will show the battery as 0volt when its not working, but when I do get a correct voltage showing up the sensor appears to be showing the correct values for Pitch Roll and Yaw if I turn the gimbal motors off. But if I turn motors on the I2C errors come back.

 

How do you make it get a correct voltage?

How many volts is the Phantom battery? 11.1 or 14.8? Does the DYS gimbal connect directly to the power source?

 

There is a power out for the Video Transmitter that it's sharing. I'll try to rig up a separate battery to test it out. Just short on those little red connectors.

Also gonna try disconnecting one motor at a time and see if its an issue with one of them.

 

That was incredible! I'm glad to hear and see that she landed herself safely. Amazing!

I will look at wind with new respect. As soon as you got above the tree line, it just took off like a jet stream. Wow.

 

One of my errors was thinking 5 satellites was enough to provide home lock. 6 is the bottom limit.

So once I lost it over the trees and my video feed died, I was pretty much done without that home lock.

 

Gimbal update.

Rule out power supply as I attached it to a sperate battery.

Also, tried unpluging all 3 motors. That stops the i2c errors. Pluging in ANY of the motors will cause it to go wiggy and throws a bunch of errors.

So Im at a bit of a loss. and thinking perhaps its the Gimbal board since it doesn't really makes sence that all 3 motors would fail in some way at the same time.

 

Have you found out what was broken?

 

Well still working on it. I took the gimbal off and it looked like some of the wires we're kinda smashed (not so bad they we're broken) but kind compressed between the gimbal and the Airframe.

It seemed to be more functional after that. Im still getting some freakouts now and then (i2c errors) but mainly happens with some Pitch extremes. So still isolating whats up, but I think the board is fine, and Im thinking about locating some Ferrite rings to see if that will mop up the rest.

Replacing the wires on the motor, while doable would be unpleasent, since they don't just connect to the motors, but are attached and sealed in with some goop.


I took some footage at the Redwoods, but I was still kinda finicky and would wiggout and I'd have to bring it back for a hand tweak.

Nice thing about the ol Zenmuse, it just worked

 

Sounds like you are making progress. Do you think it's a motor wire, the ones soldered inside the motor? The ones on mine seem to be silicone flexible types so I don't know where one would get them.

I like the Zenmuse also, but I don't like that they made the suspension wrong. I guess some re-engineering might not be too difficult.

 

Jonas - Noticed your comment on ferrite rings. If you think that RF might be getting into the gimbal and causing all the trouble, why don't you make a couple test flights with the transmitter disconnected. That's a total elimination of RF!

If you do look for a ferrite solution, be aware that most of the common products are of the wrong permiability to work at your transmitters frequency.

...Tiger