Page 1 of 2 12 LastLast
Results 1 to 15 of 19

Thread: Yaw Instability

  1. #1 Top |
    Uses Prop guards
    Join Date
    Apr 2014
    Posts
    18
    Post Thanks / Like
    Rep Power
    0

    Yaw Instability

    I tried posting this over at the other site- it was buried in 10 minutes.

    I tried searching- that was a chore with no results.

    Perhaps there is some help here.

    As many of you may know, I have managed to get my retracts operational- the Arris is also working beautifully. The Phantom takes off, and holds it's position as well in GPS mode as it ever did from day 1... and it flies forward and backward and side to side fine... but the slightest yaw input will cause the Phantom to start a slide in the opposite direction of the yaw and only recover once stick input is returned to center. It will slide over 3 to 8 meters before recovery.

    I have tried many variations of gain settings, but seem to not be finding the answer... compass is fine- motors are stock, and worn, but they are working... it is isolated to a yaw setting or gain or something I am missing.

    Any advice, suggestions, or links to reading is appreciated.

  2. #2 Top |
    Respected Pilot Pull_Up's Avatar
    Join Date
    Mar 2014
    Location
    South Oxfordshire, UK
    Posts
    151
    Post Thanks / Like
    Rep Power
    8
    It's an obvious one, so I'm a bit loathe to mention it a I'm sure you've covered it, but have you tried an all-new set of original props?

    The other thing is to try and see if one motor is running rougher than the others or is noisier or harder to turn by hand. Might be worth changing the bearings on all the motors which is not very expensive and as long as the windings aren't damaged they are the only parts that really wear out...

  3. #3 Top |
    Respected Pilot Shrimpfarmer's Avatar
    Join Date
    Apr 2014
    Location
    Sussex England
    Posts
    250
    Post Thanks / Like
    Rep Power
    8
    Does the yaw slider move nicely when watching it in the assistant?


    Sent from my iPhone using Tapatalk
    I fly Phantom 2 with Gopro 3 Black, Zenmuse, Imersion RC600, Black Pearl Monitor.

  4. #4 Top |
    Uses Prop guards
    Join Date
    Apr 2014
    Posts
    18
    Post Thanks / Like
    Rep Power
    0
    I have tried 3 sets of props and settled on the one set I have always used as control- they are a very nice carbon fiber set that are balanced and veritably the most reliable.

    Now- as for the motor or a motor being the culprit- I was not really focusing on that possibility- I figured it was too easy of an explanation... but it is most likely the cause as I have been hard on this original set, and there is one that does make a little more noise than the others- but it did when I got it, and I chalked it up to the way ccw motors sound different than cw... or mine did from day 1.

    So I suppose I will move my gains back to what they were, and do the t-motor swap (I have a set on the workbench) and see if that solves my problem.

    Oh- and pullup? Never pull up on the punches if I am being obtuse... my forum feelbads are pretty non existent until we move into the realms of politics and religion... I get a hardon for those topics, but that is why I try and chill in the multirotor menagerie- it is therapy.

    Quote Originally Posted by Pull_Up View Post
    It's an obvious one, so I'm a bit loathe to mention it a I'm sure you've covered it, but have you tried an all-new set of original props?

    The other thing is to try and see if one motor is running rougher than the others or is noisier or harder to turn by hand. Might be worth changing the bearings on all the motors which is not very expensive and as long as the windings aren't damaged they are the only parts that really wear out...
    Quote Originally Posted by Shrimpfarmer View Post
    Does the yaw slider move nicely when watching it in the assistant?


    Sent from my iPhone using Tapatalk

  5. #5 Top |
    Flight Ready Visioneer's Avatar
    Join Date
    Apr 2014
    Location
    Louisville, KY, USA
    Posts
    60
    Post Thanks / Like
    Rep Power
    8
    Hiway - I'll be interested to see your results. I have the same issue of drift when yawing, though not quite as severe as you report.

  6. #6 Top |
    Uses Prop guards
    Join Date
    Apr 2014
    Posts
    18
    Post Thanks / Like
    Rep Power
    0
    Quote Originally Posted by Visioneer View Post
    Hiway - I'll be interested to see your results. I have the same issue of drift when yawing, though not quite as severe as you report.
    Well- with only 2 batteries to test before the rains came, here is what I have found out:

    The t-motor swap (they were here in house and planned for later) was a chore as I was distracted on at least 3 of the 4 motors during soldering... and since none of these damn motor manufacturers think it necessary to color code wires, or even make a better system or documentation to describe the mod, I ended up having one motor turning wrong and had to go back and fix it. (this really peeves me- there is no good excuse for not putting in color coded wires- none- not even cost as it is nominal at best)

    So- rant out of the way- I am flying again, and the yaw shift is dramatically less (I say less because due to my specific set up with the retracts, there is going to be some shifting as the COG is not like it is stock, and I reset all gains to stock for the test) *also- the Arris CM3000 is pretty heavy up front

    I would say that with a tweak or 2 on the gains, I will be as stable as it can be. I also had 2 props with worn holes so they would move slightly on the motor shaft- hence the testing done had lost that control factor from before and a new set of CF vision props (balanced very close) were used- they seem to offer better lift than the 9in CF original style props did- but I loved those old props- they were so silky.

    So pullup and Shrimpy were on the money with their suggestions as far as I am concerned- all personal modifications aside on my Phantom. Those stock motors are very weak, and the bearings are poor... I did not know how shot they were until making the swap- I imagine they were causing more issues than I know at this point.

    Rebuild, or replace your motors- especially if they have been abruptly stopped a few times as mine were from various incidents of abuse.

  7. #7 Top |
    Respected Pilot Pull_Up's Avatar
    Join Date
    Mar 2014
    Location
    South Oxfordshire, UK
    Posts
    151
    Post Thanks / Like
    Rep Power
    8
    Looking good, then. Let us know how you get on with the T-Motors over time as well.

  8. #8 Top |
    Uses Prop guards
    Join Date
    Apr 2014
    Posts
    18
    Post Thanks / Like
    Rep Power
    0
    The T=Motors so far are giving me a total of 6:30 before it has to come down... I know there is 45sec more before it will descend- maybe a minute, and this is all on the factory voltage settings which I believe is @3.47 or thereabouts- I could push it. This is with the Maddog 2700's.
    On the Zippy 2200's I get 5:15. Stock I will not bother with as it is a dud unless you are flying the bare bones stock version. The Dronefly 2200 that came with the Phantom is actually a decent bat and offers about 6:00 before the Phantom force lands. All Bats are @12.6 on fireup. I plan on testing with a separate volt meter with audible I use on my mini to double check- but my multi meter and the software assistant jive- and my multi-meter is a good one- in fact, I have an array of pretty high tech meters from my tip and ring, and coaxial tech days... just don't know if they will take a charge and the adapters are scattered (I am a total wreck and disorganized)
    The T-motors are way better sounding, aand you can feel a pretty significant increase in power- you hear it, and fell it in the sticks if that makes sense. The craft is much more responsive... much more. Stock gains made it feel far more spry than it did brand new with no weight on it... and the resistance when I put it in gps mode and tugged on it to pull it out of position was surprising- it pulled back! I was used to it being easy and not giving a struggle... the T-motors will fight you.
    All in all, I would say that if you are really in love with your Phantom, then they are a must. Hindsight being 20/20 and all things considered with what I have learned in the last 6 months, and I would drop the Phantom, and save money for a real multi-rotor as the Phantom just has too many limitations, and is basically an entry level offering as it is designed, but not any sort of platform for true aerial photography. Barely hobbyist- but I have become jaded since I have dabbled with other flight controllers and experienced what a truly tuned quad can do. I mean- for example- my mini can carry the same gopro- offer everything that the Phantom can do except carry the gimbal (do not discount the stable footage from the rubber dampened vibration isolation system those little things have- it is very well thought out) and stay in the air for 10 minutes... 12 without the gopro and just fpv with the ccd. It is smaller, faster, more stable and more precise, and it can do footage the Phantom could never accomplish... and I spent less than half building it than the Phantom cost.

    I think the Phantom is wonderful- but if this is more than a passing fancy as I have discovered in myself, then save the cash and put it towards any number of options that only cost a little more, or less than the Phantom, and require a lot less tech knowhow than you might think... or I just find this stuff easy.

  9. #9 Top |
    Uses Prop guards
    Join Date
    Apr 2014
    Posts
    18
    Post Thanks / Like
    Rep Power
    0
    I wanted to post an update on the fine tuning, and how another local Phantom user had the same issue, and I corrected it. His issue was different, and worse, but the effect was the same- he would yaw- never touching the right stick, just a tad, and the Phantom would spin the proper direction but make a sudden and seemingly uncontrollable backward sweep- given enough time and no winds, and it would recover, and stabilize in place.

    I noticed when I test flew his Phantom, it seemed to track and fly backward pretty well- the yaw was still odd and needed compensation, but not nearly as bad. I landed it, flew mine and showed off the retracts, and then I asked him to calibrate his quad (we had a lappy on site) and calibrate his compass. I noticed when he did his compass dance, he bent over and spun the Phantom in place near the ground (I have always stood up and turned in a circle- bad leg) but I noticed when he did the 2nd axis he was pointing the led (the back) down and not up. I asked him why he did that... he said he always did the compass calibration like that. Now while I had no documentation with me, I was sure the nose/front was supposed to be pointed down- and that is how I calibrate my compass... so I asked him if he would turn it over and calibrate it... which he did, and suddenly his Phantom stopped all poor yaw behavior. It seemed to need a few minutes to fine tune itself (Naza learning maybe), but by the end of his battery it was solid and steady... he was extremely grateful.

    My quad needed one last adjustment- the vtx removed. It was a heavy 600mw Boscam and I took it off to use on another application and test flew the Phantom today- the first few seconds were instantly very tight- I charged up a Maddog and took it back out an hour ago- very gusty winds, with the retracts and the arris operational and the gopro filming... I would not do much flying as the winds were about 17 to 20mph... and too heavy to fly but perfect to test... and the Phantom was excellent... worked it's ass off staying in place... but it would yaw on it's axis and blow as you would expect it to.

  10. #10 Top |
    Respected Pilot Shrimpfarmer's Avatar
    Join Date
    Apr 2014
    Location
    Sussex England
    Posts
    250
    Post Thanks / Like
    Rep Power
    8
    That is very interesting and significant information Highway. I can't remember how I calibrated mine as I have only ever done it once. Worth keeping in mind when we see posts from people with this problem. Thanks for updating your thread.
    I fly Phantom 2 with Gopro 3 Black, Zenmuse, Imersion RC600, Black Pearl Monitor.

  11. #11 Top |
    Respected Pilot Pull_Up's Avatar
    Join Date
    Mar 2014
    Location
    South Oxfordshire, UK
    Posts
    151
    Post Thanks / Like
    Rep Power
    8
    Interesting. When I took the P2 loaded with H3-3D, GoPro, AVL58 tx, iOSD Mk2 and FPV hub I did it's first calibration in the way you usually do - standing up, holding it in front of me (easier to see the lights). It didn't hover as well as I expected, and worse than could have been just down to the firmware (the latest of which does seem a little less tight in the hover). So I calibrated again, but this time in the "bend over and spin it on the spot" way - bingo, all sorted. I only calibrate the compass if I add/remove hardware, travel more than 100 miles or so from last calibration or when I've updated firmware.

  12. #12 Top |
    Flight Ready Visioneer's Avatar
    Join Date
    Apr 2014
    Location
    Louisville, KY, USA
    Posts
    60
    Post Thanks / Like
    Rep Power
    8
    I have followed with interest Hiway's extraordinary efforts to stabilize his Phantom. My new symptoms are similar but not identical. In any event I'm not convinced that we should have to go to the extremes Hiway's gone to "settle" a Phanton Vision, especially one like mine that is completely stock..

    Given many posts here and elsewhere indicating success (more or less) with the latest upgrades I went ahead and upgraded all the way to 3.2, and have the latest android & iPad apps.

    Here's how my P2V has started acting. (Stick & transmitter calbrations completed post upgrade, IMU checked OK)

    It does a good job of maintaing a hover. It flies forward/backward, left/right, and up/down in response to my inputs, and stops movement almost immediately when the input stops.

    If in a hover and I yaw 1/8 of a turn to the right, it stops as soon as my input stops, but then it flies gently about 6-8 feet in the direction of the right front arm (post yaw), and then slowly returns back to the start point. There is no additioanl yaw introduced. If I yaw 1/4 turn to the right, it reacts in the same manner but flies 10-15 feet before returning.

    The exact same behavior occurs if I yaw to the left, except that it flies in the direction the left front arm.

    This behavior is repeatable and occurs whether there's no wind or a light breeze. It is not simply drifting during yaw as some forum posts have explained.

    The post yaw "flights" are gentle (any tilt is scarcely visible) and apparently involve equal parts forward pitch and right (or left) roll as it follows the direction a front arm is pointing.

    If I anticipate this "flight of fancy" by holding in backward pitch and opposite roll I can hold the Phantom in place, but if I release before its "free time" is up, it will still fly per the above description, just for less distance. It's almost as if the yaw is triggering a set time period of pitch & roll after the yaw completes, one that is proportional to the length of the yaw input. I wasn't in a location where I could do a 180 yaw to see how far it flies afterward (assuming it will fly farther), but I'll probably give it a try when I can.

    I've seen speculation regards bad motors/bearings/ESCs but I'm not sure how that could apply in a situation where the abnormal behavior is in different directions depending on the direction of the trigger event.

    About all I can think of at this point is go back to a previous version and see if this behavior goes away.

    Thoughts?

  13. #13 Top |
    Respected Pilot Shrimpfarmer's Avatar
    Join Date
    Apr 2014
    Location
    Sussex England
    Posts
    250
    Post Thanks / Like
    Rep Power
    8
    This must be so frustrating. Its akin to having a model without gps lock. You have to do all the work to retain position instead of being free to concentrate on the images. Have you approached DJI ? A decent video of the behaviour superimposed with film of your stick inputs may embarrass them into action. I hope you get it fixed.
    I fly Phantom 2 with Gopro 3 Black, Zenmuse, Imersion RC600, Black Pearl Monitor.

  14. #14 Top |
    Flight Ready Visioneer's Avatar
    Join Date
    Apr 2014
    Location
    Louisville, KY, USA
    Posts
    60
    Post Thanks / Like
    Rep Power
    8
    Quote Originally Posted by Shrimpfarmer View Post
    Have you approached DJI ? A decent video of the behaviour superimposed with film of your stick inputs may embarrass them into action.
    I had the same thought though I plan to play with it a bit more first. The fact that this odd behavior is (so far) completely predictable and repeatable just strikes me as odd and, honestly, a bit intriguing (as long as I don't have a specific use planned). Of course, it does make me a bit leery as I wonder what other quirks may suddenly arise.

  15. #15 Top |
    Administrator Frank Adams's Avatar
    Join Date
    Mar 2014
    Posts
    1,581
    Post Thanks / Like
    Rep Power
    10
    Hiway, I really would like to see your copter. Do you have a pic you could share?

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •