EC45EN Brushless Gimbal Motor with Encoder
  • so I am trying to set up a gimbal with these motors ,EC45EN

    the gimbal worked very solid without encoders and normal tiger motors
    now I converted it and trying to tune it , since already 3 weeks
    it seams it has lots of power and works beautiful as hand hald gimbal
    but the big problem I have is in fast flights, it starts to shake badly

    I tried all firmwares , all kind of different tunings and especially also tried to maximise power and P to get the motor to lock better
    I had the feeling that the encoder firmware does not lock the axes tight enough ..since it almost uses no power if it is in the desired position as seen in the monitoring tab, yes it boosts it up once there is force on it but I it look like this takes too long time
    it seams to be much more locked with super high pid tuning but the results are still much worse then before
    I also start to wonder if these motors are bad for aerial use? they have this unusual magnetic ring like in maxxon motors but that may cause this unwanted play they have?

    the also dont unusual little amount of windings
    maybe the bigger sizes of theses motors are better?

    any ideas very welcome
    if somebody has a setup that works very well I would at least I would now there is achance it could work on my side
    btw, it looks like the HDAir Gimbal also uses them
  • sounds like too much vibration from copter making it's way to the BGC.
    you can tune out vibration
  • that was my though too, and I checked this

    -vibes are limited pretty much as good as it gets (perfect props and and good working dampening system
    -same copter handles a normal setup absolutely perfect in high speed

    I might l do a test with an air blower to really proof this
  • no good reslults with a leaf blower
    I will have to compare this to other setups I have
  • same camera on 8bit board gives better results with the leaf blower

    and now that I compare this holding it by hand and compare it against each other
    without encoder it feels that it locks the angle tighter

    so my question now , are the motors or encoders I have not good for my aplication or does the encoder fw have a problem , or maybe both?
  • I am quite happy how this encoder gimbal is working http://levitezer.com/styled-2/blog-2/files/f03a914b964f341256efb82b3866d93b-18.html At the video on this link the follow and RC parameters needed still some tuning.

    This one is with GB54-1 motors + encoders, plenty of power. it is for BMMCC sized cameras http://levitezer.com/styled-2/blog-2/files/d154897449315a7e7acfdb0e704167c7-21.html , handheld, but work good on high speeds.
  • its a bit hard to see because the video is very compressed
    but it almost looks like you have some tiny shakes also , and your camera might handle them better
    is it a gopro on medium wide settings?
    also a gopro in the housing is more aerodynamic then a photo camera

    I know I am looking into details, I tect my camera mostly in 70mm tele
    its very easy to get rolling shutter in 4k
    but it worked on 8bi witout encoder, thats why I am lost right now

  • At 0:10 there is some shakes on the BMMCC video as I was riding ATV and holding the gimbal on my shoulder, and back wheels where sliding a bit ;)

    70 mm is of course more challenging. Is it same setup otherwise, same multicopter, propellers, vibration dampers, gimbal, other than the motors. Is the gimbal as solid as it used to be?

    Especially the camera balance is very important at longer lenses. And it makes a big difference how smoothly the axis move, any binding is bad.

    I think the 8 bit has less accurate stabilisation, but that could mean also smoother stabilisation on difficult conditions, it does not try to correct the errors as fast and powerfully. Maybe tuning the encoder gimbal for a bit looser settings could help on your case. Try higher D and LPF filter, could be good to try also the 2,58 FW, it has improvements on filtering and tuning.
  • thanks Garug, allways apriciate other ideas

    I did try the new fw and also and pretty much maxed out D (between 200 and 255)
    it felt like raising P was locking the gimbal better
    I also tried low power and high power
    but all this didnt help at all

    balance is good, those motors dont have cogging so its easy to balance them
    and it works very good handheld...viloent shakink is totally fine
    but in the end the outside forces are the ones that making the problemes

    maybe I shouldnt have used any filters and less power?
    I had notch filters 150hz 10 and 80 on all axis
    no lpf filter , this was causing mostly very low frequency vibes

    but in the end I really doubt those motos, if there is anybody using them with good results I would be hapy to hear about it

  • If you already have that high D, maybe you should try IMU high sensitivity.

    With the 2.58 I do not need any notch filters. The auto tuning sets automatically the LPF filters.

    I would try just lower PID, sounds like the condition changes causes vibrations, see on real time data that when you knock the gimbal, the vibrations dismiss soon. Do not use adaptive tuning.
  • yes I used adaptive tuning on basic settings, I already thought about disabling it, probably dosent make sense with encoder anymore?

    high sensitivity , yes I tried this on older gimbals, could make sense, would have to try this

    I didnt like the lpf filters the autotuning came up with
    it resulted in ringing when touching the axes...the notch fitlers were giving good resluts straight away

    but now there is also the d-term lpf...also this one worked better to turn off and have the notch filters instead I felt...but I tried a lot of different combinations just trial and error style...but no light on the tunnel

    thanks for the help!