Intermittent high frequency vibration problems with our Movi MR

Discussion in 'MōVI MR' started by Sam Leggett, Feb 25, 2014.

  1. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,351
    Likes Received:
    1,160
    Hi Adam,
    I suspect that the other place that might benefit from some foam is either side of the camera -- the idea is to reduce/eliminate any chance that the camera's mass is acting like a counter-balance and zigging when the MōVI is zagging (if you see what I mean). So, as you say, anything that you can do to make the MōVI+camera appear like a single mass would be a British Imperial Good Thing (which, as you know is slightly larger than the U.S. Good Thing).

    Cheers
    Andy.
     
  2. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Wow the Red really fills up the Movi! I can see why you would try that foam now.

    On another note, how flexible are all your leads going from the camera? Does the Movi have to exert more force when the leads have to twist or move much? We managed to get all our downlink equipment mounted on the tilt bar with the camera so there are no links to the rest of the gimbal.

    I have never flown a Phantom, but we are thinking of getting a Blade 350QX as a reconnaissance aircraft. I have flown fixed wing for ages but cut my multirotor teeth on a Hubsan X4 for £30 they are great fun! Are the results really that good from the Phantom? Apart from the curved horizon that apparently you can get rid of the stuff I have seen looks good.
     
  3. Adam Etheridge

    Joined:
    Aug 23, 2013
    Messages:
    52
    Likes Received:
    9
    Hi Guys

    So we've done some more tests today (v foggy in Bristol today!)

    NB: GoPro was hard mounted to the airframe, as you suggested Andy...


    8 x Soft Dampeners

    GoPro / Epic Movi - split screen
    http://exchange.aspectfilmandvideo.co.uk//8xSoft-split_screen.mp4

    Slow motion (200fps) of the rig
    http://exchange.aspectfilmandvideo.co.uk//8xSoft-rig_slow_mo.mp4



    8 x Hard Dampeners

    GoPro / Epic Movi - split screen
    http://exchange.aspectfilmandvideo.co.uk//8xHard-split_screen.mp4

    Slow motion (200fps) of the rig
    http://exchange.aspectfilmandvideo.co.uk//8xHard-rig_slow_mo.mp4


    Looking at the GoPro footage - although the airframe is stable and holding position well… there's loads of jello effect on the GoPro image, which to me (because of the rolling shutter) suggests very fast high frequency vibrations??

    To my eye the 8 x soft dampeners look better - and they don't appear to be bottoming out on the slow motion clip?

    We found we have to bring the pan stiffness down to almost 30% before we stop getting oscillation, but then at this level the stabilisation is rubbish!

    Sam - do you find it ok to get good pan balance on your MR? ours is slightly front heavy but we can't move the pan knock any further back to compensate as it hits the rear leg!

    Will definitely try more foam down the sides. Moving the Video TX to the tilt is a good plan too… so many thanks to try :/

    Adam
     
  4. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Hi Adam,

    Not had time to look at your footage yet, but yes we managed to balance our pan - and that might be you problem. I think you might need a longer carbon tube at the top. For our initial trials just to get it balanced we added weight to the back leg - not great for battery duration but it did the job... Then we changed our lens to a shorter one and we could nearly balance pan without any weight - but not quite - to balance we shuffled our horizontal bars going from the roll bar to the tilt pivot back slightly... only took about 2mm - but you have to check your camera can still tilt as much as it needs without hitting the roll bar. Ours balances in every plane and I think that is the secret to high gains and good results.

    Amazed that the soft ones work best! Your stuff must weigh more than ours! But if not flying fast - and highly tilted over and not flying in turbulent air I guess the soft ones would work best. Its when its bumping or tilting a lot that it will happen. (I think)

    Anyway I will look at your clips soon...

    All the best,

    Sam
     
  5. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,351
    Likes Received:
    1,160

    The Phantom 2 Vision seems to be a fun copter to fly -- and very good for reconnaissance to show a DP what shots are feasible. The flight times of 25 minutes per battery also make it much easier to work with an even rehearse shots before you lob a "heavy" into the air. The original lens sucks in terms of distortion and softness around the edges. I've ordered up a flat, albeit narrow FOV lens from Ragecams.com so we'll see if that improves things. This includes the means for interchanging lenses so that I can still go back to a wide lens if I need to.

    It's not in the same league as the Cinestar, but it's a useful "toy" tool for pop-up, see-what-there-is-to-see flights...

    It's dead easy to fly too -- but you have to make absolutely sure that you have a good GPS lock before you take off otherwise there is a risk of invoking the "Go Home" (to China) feature. :rolleyes:

    Andy.
     
  6. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Hi Andy,

    Thanks for the info on the Phantom 2. We are in the process of writing our "Operations Manual" a requirement from the CAA here in the UK - and I have included details for the Blade 350QX as I was thinking of having one of them for the purposes you mention. We already have a go pro hero 3 and all so it might help keep costs down. We were thinking of flying it in risky radio interference areas before the big one... but now you mention sizing up and showing shots too - that makes a great deal of sense - not running down the big lipos for a shot that doesn't work, and 25 mins is a long time! that's great! We get 12 mins with Movi and C300 with 25% battery left which is nice and safe.

    We have wookong gear in our Cinestar and to date its been brilliant... but the go home to china function is always at the back of my mind! Our "Waypoints" set up frequently places our rig in the Gulf of Guinea - which is one of the reasons we haven't used it yet! Dead keen to get good at it though. Want as many strings to the bow as poss!

    Thanks again, its great learning all this stuff!

    Sam
     
  7. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Ok, we have just got back from another show reel shoot - we have enough shots for the show reel (thankfully as we had to pay - charity building) BUT our "perfect" settings from last weekend were not so perfect today! The footage was pretty good, but the Movi looked as if it was fighting itself a bit during shots. The frame was a bit fidgety. The weather conditions were very similar - but a lot less windy and turbulent so we are at a loss. We were using all the same gear - all exactly the same as the week before. We reckon we got a minute at best out of 45 mins of flying time. Last time everything we flew well enough was good enough quality... that is what we want - and what we know is possible!

    The underlying issue is that the cameraman cannot notice the imperfections in the shot on a smaller monitor HD or not. Its only when you get it on the big screen that you can see these things... which make it hard to use for work with confidence.

    Once again, any thoughts would be greatly appreciated.
     
  8. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    910
    Likes Received:
    472
    Sam - I just watched the video. Can you do 2 things to help me trouble shoot?

    1. Remove the big rubber tires from the landing gear.
    2. Hang the copter with your current settings and lightly push agains the pan axis (tip of the landing gear) See if you feel any oscillations. If you can video that as well for me to watch that would be great. Provide as much detail as possible so I can try and help. My gut is that you are getting oscillations on the pan axis and my initial thought is there is slop somewhere in the pan system or that the wheels on the end are setting up an oscillation.

    Thanks!

    T
     
  9. Adam Etheridge

    Joined:
    Aug 23, 2013
    Messages:
    52
    Likes Received:
    9
    Hi Sam

    We tried moving the horizontal bars back about 10mm to compensate for the pan balance a few weeks ago, but unfortunately its just not enough and any further starts inhibit the ability to tilt the camera as you say.

    We've just ordered a longer carbon fibre tube from someone on ebay. I'm surprised its even necessary to do this, given that MR is advertised as being suitable for the Epic - its not like we have a particularly unusual rig... anyway!

    Sounds like you've had a bit of a backwards step… We've found that warp stabiliser does a fantastic job at repairing our judder-y footage, but when you're paying for a premium product thats not really the point in my view - especially when other people seem to be getting great footage without any post stabilisation. It also forces you use a higher shutter speed to reduce motion blur. Hope you're showreel shoot wasn't entirely wasted.

    I agree its virtually impossible to see imperfections in the shot on a small monitor. We've recently been taking a Teredek Bolt and a large 17" panasonic monitor to our tests, which works well. You can definitely see the judders / fidgets, and it allows you to change the stiffness settings while its still in the air. Although we're limited to a very short (10m) range on the HD teredek.

    Adam
     
  10. Adam Etheridge

    Joined:
    Aug 23, 2013
    Messages:
    52
    Likes Received:
    9
    Thanks Tabb - I think this may be directed at me rather than Sam, but will try both of those and let you know...

    Thanks,
    Adam
     
  11. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Hi All,

    Ok Yes Adam we have had a backwards step :(... but I will come to that in a minute...

    I just watched your video and I think you are getting similar kind of wobble to us. I have noticed our wheels and undercarriage wobbling and so will remove the wheels and test again. I was also thinking of sticking something between the two struts that make a leg as there is plenty of flex there. Apart from the legs the super slo mo looked wicked and showed no nasty stuff.

    I know the wheels look clumsy but I like a bit of give with all that £££ landing on tarmac! But I will remove them for a test for sure. I would really like some moon-lander style shock absorbing legs ;)

    Our findings from today...

    Today we filmed the two most spectacular locations we have yet visited: Lancing Chapel and Brighton's West Pier. Fortunately we did get enough high quality stuff for our show reel and don't need to go back. BUT two big problems still exist:

    1. The wobble.
    2. Horizon drift - a newcomer.

    1. Tabb I will most certainly do you requirements and let you know, however, we have had some silky smooth joy coming from our movi the way it is... everything is exactly the same as when we last flew it - yet the results today were worse. Its very strange. Also we have a low quality downlink for long range... but we land and watch the footage in a very good quality small monitor - neither can pick up the wobble we get - until we get home on the big screen. Then we see it. The thing is we don't want to kit out the truck with a huge TV for analysis. We want to be able to say confidently that if I fly right and Adam sets up and points the camera right then we get the shot, and at moment we need two go's or so to get it. That is the core of our problem. We know the Movi can do it, but we need it to be repeatable.

    2. Horizon drift. We only noticed this today... filming at the sea... when filming the horizon it was great, then we fly over the subject and rotate as we go and as we tilt up you see the horizon behind - and all to often that horizon is off. If you don't show it you can't tell, so its not the end of the world, but can it be fixed?

    I really love all this gear, and fly for 10 mins out over the sea 250m out was exhilarating. My confidence is building with the airframe and the shots looks stunning, I can't wait to post the show reel... but we just need it to be replicable.

    I will put up a vid of our footage later... and will do as Tabb says soon too.

    Thanks everyone for you help,

    Sam
     
  12. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,351
    Likes Received:
    1,160
    Sam: How frustrating....

    If I were standing in front of your kit, here's what I'd be double checking:

    1. Balance, balance, balance....

    2. All cables disconnected and reconnected -- vibrations can work them loose and Firchau's Law states "If it flies, it vibrates." :)

    3. Settings : suspend the copter by a cord from the battery plate, the better to simulate flight conditions -- if need be, remove the props and spin up the motors, then try the same moves that you did out over the Brighton Pier (I know it well...) Note: you might need to do this EDIT: outside to get a good GPS signal.

    Some additional questions: Were you using a Tablet/PC to connect up to the MōVI before the flight? The problem being that gravity and acceleration are indistinguishable so when you did what I call the Gomez shot (the laterally offset approach to target, tilting down as you approach, panning as you go by, and tilting up once you've gone by) (my camera op's last name is Gomez) the accelerations can alter the "gravity vector" (aka "which way is down?").

    My understanding is that the MōVI should be able to null out some of these effects based on GPS data, which raises the question: did the MōVI have GPS lock -- and the only way you'll know that is if you were linked to it via Tablet/PC.

    All of the above said, I defer to Tabb's superior knowledge on this... :)

    Andy.
     
  13. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Hi Andy,

    Thanks for your reply, wow - now were are getting down to the nitty gritty... more learning - but that's a good thing!

    Balance was definitely spot on... we really have been practising that one... but we will try all the other things and let you know soon. Not now though... I got up at 0445 this morning and its now 2215 - a full flying day and flight planning! Never really thought about the GPS lock... the Wookong had great GPS, I do know that!

    Can you go out every time and know you will get great results? That is our goal. Its all so easy in handheld! The Movi is so tolerant then!

    Anyway we will crack it!

    Thanks again,

    Sam
     
  14. Chris Fox

    Chris Fox Active Member

    Joined:
    Apr 14, 2013
    Messages:
    463
    Likes Received:
    86
    Hey Sam, I don't think the learning ever stops!!

    Thanks for posting the 200fps footage. I have been seeing similar effects with my setup on occasion, some times our footage looks just perfect, and then sometimes we get a vibration in a similar fashion ... that is imperceptible until you watch it on a large screen.

    We have just been using the standard landing gear extensions, and have been thinking that the vibrations may come from them, as we most often see the vibrations when trying to do fast forward flight with no pan input by the cam op, if the cam is panning while flying it seems to reduce the issue.

    Cheers

    Chris
     
  15. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,351
    Likes Received:
    1,160
    G'day Chris:
    From what you say, I'm starting to wonder whether the relative wind blowing through the gimbal is generating vibration? Could it be the slipstream that triggers vibrations at some natural resonant frequency? (I'm starting to speculate at the same level as the theory mongers for MH370 here!)

    The clue was "fast forward flight."

    Andy.
     
  16. Chris Fox

    Chris Fox Active Member

    Joined:
    Apr 14, 2013
    Messages:
    463
    Likes Received:
    86
    G'Day Andy,

    Quite possibly, I'm thinking of trying a flight with out the landing gear attached, as it would reduce the surface area that could be effected by the relative airflow, the only issue is that with the X8 config, I really have limited any options of attaching landing gear to the booms, and would need to muck about with retracts .... hand launching is not an option, only a lunatic would put their hands or any part of their body near the x8 to launch it without landing gear ....

    In slow moves, even descents we get the tripod in the sky effect that we are all chasing. Just not (like Sam) 100% every time ...

    Where has MH370 landed? I wonder if anyone saw it on JORN .....

    Chris
     
  17. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,351
    Likes Received:
    1,160

    In general I can get usable footage every time I go out, but there's always something that needs to tweaked, cajoled, or have obscenities yelled at it just to make it work right.... :rolleyes:

    But I had to work my way through dealing with tuning issues initially -- and just when I thought everything was OK, I'd occasionally see a quick burst of buzzing/vibrations and have to crank the gain down a bit more....

    Andy.
     
  18. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,351
    Likes Received:
    1,160
    I can't say a blame you there, and nor can I really condone the idea that I've seen done, which is the take off using a helper holding the copter high over their heads, and similarly, a slow vertical descent such that the helper grabs the booms directly overhead.

    I have done motor tests that way with a trusted friend, but that did not involve any release/capture of the copter so the test was relatively safe because the booms were held throughout.

    Interesting that it works fine on slow moves....again, makes me wonder about the slipstream...

    No data, I'm afraid. Feel badly for the families and friends...
     
  19. Sam Leggett

    Sam Leggett Member

    Joined:
    Dec 26, 2013
    Messages:
    58
    Likes Received:
    7
    Hi All,

    Chris:
    Ah that was Adam - another chap who posted all that slo-mo and split screen footage, but it seems likely we have the same problem now! I am going to get those wheels off asap as I have seen the gear wobbling about before - originally I was going to put struts between the two sides of each leg, but wheels off will come first now. Perhaps both will be tried!

    We have also noticed that when the pan is busy doing something it generally behaves much better! Setting up a shot - at hover and moving gently into position often looks worse than the actual shot - which is kind of good I guess!

    Andy:
    Great to hear you can get great footage reliably nearly all the time... that is the holy grail. It will just make our confident in our product when we get to work.


    I think we will get to the bottom of it, but like you say the moments when you get those tripod shots are tremendous! Sometime I really think the TV has frozen when at a hover... and the descending shots when the copter is struggling in its own prop wash but the footage is rock steady blows me away every time! As good as a jib... faultless! Other times though it just fidgets around, not really high gain vibration or low gain slop... just a bored fidget!

    Anyway thanks for all your thoughts, I'm looking forward to more testing and getting to the bottom of it.

    Sam
     
  20. Chris Fox

    Chris Fox Active Member

    Joined:
    Apr 14, 2013
    Messages:
    463
    Likes Received:
    86
    Whoops my bad, thanks go to Adam. And to Sam for pointing out the error of my ways. I'm that bad with names when they are written in front I me, but I'm even worse in person .... Doh!

    Apologies all around
     

Share This Page