The other day I was flying my IRIS in loiter mode with gusty winds and a Contour roam cam attached to one of the forward legs, after few minutes the IRIS started drifting away while pitching up and down in a seesaw motion. After failing to gain any kind of control I switched to RTL but it still did not respond, it then started to drift towards power/telephone lines so I tried to send a land command but that didn’t appear to help. If I remember correctly I believe I then switched back to loiter mode to send it over the lines however when it cleared them it quickly lost altitude and crashed (lucky no damage). Very odd, any ideas why it happened?
Did you leave your Contour camera’s wifi switched on? Done any modification to your iris? Checked the logs of this flight? It would be good you attach both telemetry and pixhawk logs when raising a question like this.
The Contour Roam does not have a wifi option or include any other type of radio. As for logs I’m still a novice at understanding/reading the pixhawk logs. The pixhawk log is attached with OP and I use droidplanner for telemetry, I’ll try to find the matching tlog.
Well I’m not sure I know how to interpret your pixhawk log… What’s strange is that at some moment you trigger a RTL and the IRIS just continues its course and does not come back to your take off place. You also lost telemetry connectivity.
From the telemetry log I can see also how your IRIS keeps on pitching up/down continuously, it seems to amplify even over time. Are you sure your contour cam was properly attached and not just a bit loose so that it introduced vibrations ?
It seems my post was moved to this forum; no notice, a little confusing.
Moombe: thanks for looking at my logs and yes the camera was very secure. In fact it did not budge after the crash. And yes the pitching did seem to amplify over time; the winds were gusty but I have flown in worse.
IRIS is trying to pitch back 30 degrees for most of this flight. It is not achieving this. The IRIS must have a really bad CG and this is the root cause of the crash. It is also underpowered - if you weren’t 5500ft up in the high desert in the heat of summer, it would almost certainly have worked out better. I don’t think you can fit bigger props onto IRIS, but you might try fitting higher pitch props.
That said, the copter still has some amount of throttle range to play with - front motors are sitting at maximum, but rear motors are not at minimum. I’m also not sure why the huge oscillation is there. It should be keeping its attitude at the expense of throttle, which would result in a relatively gentle level descent to the ground. I would like to understand why it doesn’t do that and I’ll work on that.
Jon: Thanks for your analysis and interest; I did not think about air density, I normally fly in San Diego but I can see how that would stress the IRIS. The pitching up/down only became noticeable about 3-4 mins into the flight. Another note, the IRIS would struggle to keep altitude during strong gust (I assume the barometer was being affected). Also any idea why the IRIS did not respond to an RTL? And why did it rapidly lose altitude at the end? Thanks again.
What is the weight of the Contour camera (including anything you used to mount it)? The IRIS is a little bit underpowered as is so adding weight + altitude only makes it worse.
Did you retune your gains (manually or via autotune) after you mounted the camera?
Something is really messed up with the logs. Modes are completely different between the tlog and the dataflash log. Per the dataflash log, you were flying around in LAND mode, but the tlog reveals you were in LOITER. tlog doesn’t agree that you were in DRIFT, and it never missed a heartbeat message around the time the df log says you were.
Anyway, other than that strangeness (are you running stock firmware or have you made modifications?), I’d say the copter is behaving normally for an incredibly underpowered quad with a big weight on the front.
moombe: The camera with mount weighs 180 grams and I did not retune the gains, I’ll have to look into that some more.
jschall: Completely stock firmware (v3.1.5) with no modifications; IRIS flew normally for the first few minutes, it did not appear to be underpowered. I flew only in loiter mode since it was gusty that day, the only other modes I used were RTL and LAND when it started to drift away and after it did not respond I switched back to loiter then it crashed. I would really like to understand why it did not respond to an RTL or a LAND?
moombe: The camera with mount weighs 180 grams and I did not retune the gains, I’ll have to look into that some more.
jschall: Completely stock firmware (v3.1.5) with no modifications; IRIS flew normally for the first few minutes, it did not appear to be underpowered. I flew only in loiter mode since it was gusty that day, the only other modes I used were RTL and LAND when it started to drift away and after it did not respond I switched back to loiter then it crashed. I would really like to understand why it did not respond to an RTL or a LAND?[/quote]
It did respond to RTL, but it was just too underpowered to maintain its demanded pitch angle at that point. If you’d yawed it around, it would’ve flown toward you - it wouldn’t have had any choice.
Sorry I’ve been telling you this and that, but the modes in the bin file are wrong for some reason. The tlog almost certainly tells the real story, which is that the crash was in RTL mode.
Ultimately the problem is that IRIS is severely underpowered in that configuration, particularly with a bad CG.