Personal View site logo
3-axis stabilizers ala FreeFly MOVi or ZenMuse - DIY and commercial
  • I am interested in discussing possible DIY setups for an affordable ($500 or less) motor-based camera stabilization system. Most likely this will use a combination of gimbals, gyros, brushless motors and some controller. However, I think there is potential of at least discussing the combination of of rack pinion gear system instead of gimbals for stabilizing 3-axis movment (X, Y, Z).

    TOPIC RULES:

    1. Don't put people down for asking basic questions. I am new to this knowledge area too (with no experience using RC multicopters).

    2. Encourage creative thinking. Perhaps stabilizing cameras handheld requires/deserves a different approach than tech used on multicopters.

    I am reserving this post as a summary of "good/feasible" ideas.

  • 458 Replies sorted by
  • This post is reserved for examples of more expensive solutions.

    FreeFly MOVi http://www.personal-view.com/talks/discussion/6607/freefly-movi-very-expensive-handheld-stabilized-system-/p1

    ZenMuse

  • This post is reserved for possible solutions.

    Other DIY Attempts/Projects

    http://www.iso1200.com/2013/04/are-you-ready-for-diy-movi-2-axis.html?m=1

    From that article, as an example of 2 axis (pitch and yaw) stabilization... kinda slow response:

    http://forum.freeflysystems.com/index.php?threads/converting-my-cs-gimbal-to-a-brushless-gimbal.1257/

    DIY guide for simple bgc - controlling pitch, yaw and roll

    http://www.simplebgc.com/eng/downloads/

    AlexMos Brushless Gimbal Controller

    https://viacopter.eu/multirotor-shop/camera-gimbals

  • Proposal: Need to stabilize LINEAR velocity NOT ANGULAR velocity

    I see that flight stabilization requires a focus on eliminating pitch, yaw, and roll.

    How GYRO sensors work - angular velocity/motion

    http://www5.epsondevice.com/en/sensing_device/gyroportal/about.html

    http://www.futureelectronics.com/en/Search.aspx?dsNav=Ny:True,Nea:True,N:645

    Or.....! What if accelerometers were used on X AND Z AXIS?

    Consider the following alternative idea: for handheld camera movements I'm not rolling to the right or left suddenly like in flight (roll). Also, sometimes I want to pitch up or down (pitch, i.e. tilt). I admit, smoothly panning handheld (i.e. yaw) is difficult to do without some kind of stabilization (maybe a shoulder rig is sufficient). Overall, with handheld camera movements micro jitter is the most distracting feature I want to eliminate (not pitch, yaw, or roll).

    So I'm thinking, instead of counter-rotating the camera to neutralize pitch, yaw, or roll - what if the camera was only stabilized linearly on the x and z axis (left to ride and up to down)? The goal would be small linear adjustments, with each axis controlled by a rack gear attached to a brushless motor reading a linear motion sensor?

    Thoughts on this? In your opinion, would pitch and yaw be necessary to stabilize too?

    Imagine a small, quiet version of this linear brushless motor, but on x and y axis for camera stabilization:

    http://au.answers.yahoo.com/question/index?qid=20080319082534AAtrClv http://wiki.answers.com/Q/How_to_convert_Angular_velocity_to_velocity

    Info on G force http://www.gforces.net/a-discussion-on-typical-examples.html

    Accelerometers for sale, organized by 1, 2 or 3 axis capabilities, plus g force rating: http://www.futureelectronics.com/en/Search.aspx?dsNav=Ny:True,Nea:True,N:634

  • I'm not at all familiar with this specific tech, but I've done robotics stuff before. The way that these copter stabilization gimbals are designed effectively allow them to correct/stabilize for pretty much any type of movement AFAIK (roll, pitch, yaw, and various planar shifts).

    Honestly, the main attraction of the MOVi is that it can handle heavier/bigger cameras. Personally if I were to get a system (or make one) similar to the MOVi in function, I would certainly want it to handle a heavier setup, as I'm not a huge fan of sticking a GH2/NEX with a little pancake lens on the thing and running around.

    Motors are the key to these systems. You need exactly the right ones in order for them to be strong, quiet, extremely accurate and fast (the last two things especially, because if they are not fast and accurate, they will just move jerkily or wobble as they try to catch up).

  • It is really the angular stabilization that does the heavy lifting in keeping an image looking stable. A tiny motion of a portion of a degree is a visible change, especially with a longer lens. It is the uncontrolled angular turning of the camera that make the shooter look amateurish. It makes the footage look like it was shot handheld with a very lightweight camera. The linear movement does not look bad, and is much smaller on screen. If you are shooting something 20 feet away and lift the camera up an inch or so, it will be imperceptible. Ditto with moving the camera towards the subject by an inch or two. Any small truck, dolly, crane/elevate moves (made when the 3 rotational axis movements are controlled) are very minor, and even when visible tend to NOT be unpleasing to look at. As a side benefit in the CMOS space, they also tend to not generate skew and jello.

    No stabilizers really attempt to suppress the linear motions, especially the front/back and left/right motion. There is some smoothing out of up/down motion coming from a Steadicam arm, but it only stops fast motion and converts it to slower motion, in a sense.

  • Thanks for your comments @B3Guy and @neededandwanted - you both make a lot of sense.

    Hey here is an interesting thread about a DIY Movi project (not sure his goal is for flying or handheld): http://forum.freeflysystems.com/index.php?threads/converting-my-cs-gimbal-to-a-brushless-gimbal.1257/

  • Why re invent the wheel. The overall structure of the Muvi is pretty good. What would the optimum size be for a GH-2/3? How about someone specing out dimensions for a Gh-2/3 version. All you need are proper motors on axis and the controller. I like the side handle idea the Muvi uses as apposed to a top handle.

  • Quick dumb question - is this a Steadicam killer?

  • @Riker, i think so :-) I would call it "robo steadicam" :-)

  • Have there been any rumors or noise from other companies trying to produce a Movi type clone? Everything I've seen so far DIY has looked messy and not very helpful in terms of spreading information. The Zenmuse video looks promising but how would one go about constructing that for themselves. Any step by step instructions available?

  • This the closest diy MovI project I can find, but its not step by step, yet. Its by Bill Collydas: http://forum.freeflysystems.com/index.php?threads/converting-my-cs-gimbal-to-a-brushless-gimbal.1257/

  • Thanks, will be nice when we have more Movi-esque options on the market.

  • there is a 2 axis gymbal for the GoPro (and further down: NEX). http://fpv-community.de/showthread.php?23292-FANmuse-Gimbal-for-Gopro first: Its in german though... second: the NEX is much much more lightweight tahn the GH2...

  • Looks like Zenmuse is making a GoPro size gimbal, mini-Movi for the Hero3. No specs on their web site yet, but video says price is going to be sub $1k usd, so that means $999 :P. Wonder if a GF model could fit on it?

    EDIT: Footage of unit starts around 2:30.

  • @CRFilms there is almost no chance that a gopro gimbal built specifically for that camera is going to be able to handle any larger cameras. Have you held a gopro? those things are tiny and weigh almost nothing, so the gimbal motors do not need to be very powerful. That is where the MOVi is unique. It is beefy enough to handle a cinema type camera setup. There are already plenty of gimbals out there that could handle a go pro or even a GH2 with a pancake/small electronic zoom.

  • That's looking good VK, that basic rig without mulicontroller is exactly what I need.

    Amazing how quickly this has escalated, a few months ago I posted here and in Multicopter forums requesting a handheld stabilising rig based on the motor gimbal system and it aroused no interest, just a few months later, we now have working rigs for sale and it seems like it is going to be a huge new tool for film making and event shooting.

  • @Rambo , you are right, no more stabilizers, balancing, weights etc ... This is the ultimate tool needed long time ago ... If i get hold on this, it will be used 80-90% in my projects :-)

    Hope they will work on supporting GH3+12-35 = 900 grams or above ...

    I think in short time we will see a lot of these :-)

  • @Rambo , you are right, no more stabilizers, balancing, weights etc ... This is the ultimate tool needed long time ago ... If i get hold on this, it will be used 80-90% in my projects :-) Hope they will work on supporting GH3+12-35 = 900 grams or above ... I think in short time we will see a lot of these :-)

    Absolutely, I've disbanded my self build project simply because in a very short time someone/manufacturer will have exactly what I want very cheaply.

    I see miniature versions if these being used on racing motorbikes and other horizon lock uses, especially with GoPro type POV cameras.

    Only problem I have, is all the years I've spent developing balance skills and stabilisation techniques sitting on the back of a jetski or boat, now is replaced with a simple cheap electro-mechanical device costing peanuts that anyone can use....sigh..bloody robots....I'm obsolete.

  • I'd like to try to build a handheld stabilizer for my GH3+14-42 X Lens (roughly 705grams)... I saw some week ago the Vincent Laforet film shot with the MOvi system...and it looks so fine.. Unfortunately the MOvi it's a little out of my budget.... just a question of a couple of zeros... :-) I'm not working in the cinema...i'm only a photo and video lover hobbyist.. Some days ago i saw here on PV forum some info about this low budget kind of gear... of course low budget in relationship to the MOvi one.. I start to search some info and i found some useful info on this forum :

    Simple AlexMos brushless gimbal from Viacopter: http://www.rcgroups.com/forums/showthread.php?t=1815204

    Open Source Brushless Gimbal: http://www.rcgroups.com/forums/showthread.php?t=1829058

    Differences between controllers: http://www.rcgroups.com/forums/showpost.php?p=23983141&postcount=347

    I think it's a good starting point to go more in depht on this interesting argument..

    Hope it could be useful to other people. Have a good day.. Windsurf.

  • Here is the FAQ being sent out by Relentless from the video posted by Vitaliy when one inquires:

    FAQ:

    What is it called? It is called the DEFY gimbal. It is a 2 axis stabilized gimbal for handheld, multirotor, jib or any mountable surface. It comes with a handheld rig and base platform for mounting on many applications.

    • when is it available? The first runs will be very limited to the first 20 pre-orders. We are building 20 only before it goes into larger production which will cause a delay to catch up to orders. The first 20 will target to ship in July.

    • What cameras will it carry? Right now our rig is made for the black magic pocket cinema camera or the sony nex series (or similar). This is popular for its weight and size and it will be part of the first 20 runs. COMING very soon will be sizes to cary a DSLR like the 5D Markiii. We are currently not planning on carrying a RED (even though we shoot with RED), but may have opportunities soon to do so.

    • will it be available around the world? Yes. We could ship anywhere as the user would pay for shipping as long as there are no restrictions to the country we are planning to ship.

    • Can I pre-order? Yes. Very soon, we will have a stand a lone website to pre-order the gimbal and find more information.

    • where can I see footage of the gimbal? We will email you a link very soon for all the sample footage as well as post footage on our site soon to view.

    Let me know if you have any further questions and let me know how I can help. Thanks again for your intrest!