EN CN
Many problems w/handheld rig
  • Wow, where do i start??

    Right now I am trying frimware 2.40 B6 and I cannot get the yaw to respond properly. It just doesn't activeate and when it does it doesn;t do what its supposed. Iv'e tried 2.3b5 and the yaw works right. I have no clue what is going on. 2.3b5 is OK but I cannot get a stable roll it is OK but not like some of the results ive been seeing here with 2.4 b6. I get minor bumps/shakes in roll on 2.35b that I cannot manage to work out.. I can't walk very vast at all, forget about running.

    I'm using an rctimer legacy with 4S. Motors are 5208 75T roll, 5208 75T pitch, 200T Yaw. I have these all set at 22 poles. My rig is balanced perfectly where I put it on rotation it stays, no fally upside down rolling to one side or the other, so not much work to push around my T4i with 11-16mm tokina

    23 b5 was the closest I've come to a stable performance. anything in 2.4 has been difficult especially the latest version.

    Why would version 2.4 be so difficult? I'm not getting this at all. Hwelp would be appreciated.

    TIA


  • Here are specifics:
    rctimer legacy
    4S
    5208 75T roll, 5208 75T pitch, 200T Yaw
    Canon T4i with Tokina 11-16mm lens, perfectly balanced all around.

    So far my best settings on firmware 2.3 b5:
    http://static.rcgroups.net/forums/attachments/2/3/0/6/4/a6647243-138-bgc1.jpg
    http://static.rcgroups.net/forums/attachments/2/3/0/6/4/a6647244-165-bgc2.jpg




    Anything in 2.4 just doesn't work. I can not get the yaw to respond in the later firmwares.

    Please let me know how this looks. Maybe I'm being to picky but I think this could be better.



  • I changed to 2.4 and got the yaw to work by clicking on the External FC gain Auto button. Give it a try.
    By looking at the footage, normally we will feel the walking motion when there is no subject in your frame. I feel you should increase your P for your pitch, this should improve on some of the walking motion.
  • I changed from 2.3 to 2.4 and cracked my head on the tuning. From the helpful forumers here i gathered that for the lastest firmware. "I" must be very low @0.01 and "D" must be high about 30 and above. Give these values a try too.
  • ill try the auto gain idea though. hopefully that works. yeah i used low I settings but yaw just spun freely or accasionally got power but didn't respond to input or oscilated. is 2.4 a more stable firmware if you can get it to work?
  • this was my tuning sample



    I tried the steadican shuffling on the later part of the video and I tried very hard to reduce the the walking motion in PID.



    Then followed by a shoot with a kid. horizontal drifts because I didn't calibrate the sensor well. now awaiting 32bit controller here.
  • seems a bit smoother
  • I did some tweaking. I adjust Pitch to P-23, I - .01, D - 13, Power 180. It seemed to kill the up down movement. But now I have a rill issue. Again, I am thinking it is IMU related as everything I try does not fix it. My settings are pretty close to the same as the settings from my original post but I turned down P and D as roll started getting vibes about 2 min in of use with the settings from the screen cap earlier in thread.

    Here is raw clip from new settings:

    [url]


    Here I did a quick edit, color job, and cinematic crop on the same test footage. I also put some post stab on very last shot. I actually prefer the vibes in comparison but the warp look would be cool for a surreal floaty look if needed. Just put it together to see how it edits and what I need to do better with movement for an actual gig with this thing.

    [url]
  • yup can see your up down motion became better. Its normal that we see less walking motion when we run but the roll becomes unstable and shakey. I was told by a forumer here there is more to PID then just tuning them. We have to make sure the structure of the gimbal does not introduce additional vibration and the balance of the camera have to be perfecto.

    Thats why when you see the more expensive kits, their structure and fixtures are more sturdy, more firm. I am still learning.
  • Boom, fixed it! near perfect now!
  • That is really a lot better then your first video here. What have you done this time round?
  • increased D on pitch and roll. Now my question is why does the gimbal shake out of control if I go past 30 degrees on the pitch? is it not meant to go more than 30 degrees?
  • I have this issue too. I need help on this too. But if I use follow mode on pitch, my gimbal would not shake and can follow 180 degrees.

    My guess is our yaw is not balanced properly. Maybe someone here can help us.
  • i read somewhere it was the nature of the alexmos board. so you can put just pitch on follow? im sure thats what I need to do. but does it effect stability in normal movement.? yeah, my yaw is perfectly balanced and when I go to far with pitch it is the roll that gets wacky not pitch or yaw.
  • Ok maybe 32bit can help with 2 imu. I heard its the 8bit too but I couldn confirm. Yup its my roll that gets wacky too.
  • Garug, informed me on rcgroups that his goes 360. so i have no clue, there is so much info out there about these boards, but not really enough and not organized.
  • rctimer 200T is not 22p enev if the label seem to said so
    just count the poles.
  • hard to count installed. auto count says it is 14 but not sure why that'd be. I tried both both settings seem to work fine.
  • ooh says 12 poles
  • wingbreaker you have to check a box on follow mode tab which says "estimate frame angle from motors"
  • Kikiju that option is in 2.4. I can get fallow mode to work in 2.4, but not 2.3. But in 2.4 I cannot get my yaw to work right at all.
  • I got follow mode to work right in 2.3 but 2.4 yaw still is an issue. but I am happy with 2.3. maybe when I upgrade to 32 bit things will work out.
  • Yaw motor (200T) from rctimer legacy gimbal is 14 Poles.
  • Wingbreaker have you tried to power cycle the board before testing it and after finishing the tuning?
  • what do you mean by power cycle. if you mean disconnect turn of then turn on before connectng, yes. yaw catches. but it will not engagfe right away when connected to gui. so it is difficult to tune yaw, luckily yaw is not the hardest of the 3 axis to tune.