(moved) SteadyCam Mark II

by MarcusWolschon, published

(moved) SteadyCam Mark II by MarcusWolschon Feb 5, 2011


Use This Project

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag.

Print Thing Tag


Thing Statistics

10033Views 503Downloads




Changes from Mark I:

  • handle below gimbal-assembly
  • handle offset to be on the axis of the center of gravity
  • handle free to yaw due to 2 608 ball bearings.
  • attachment of side-handle still possible and compatible with Gimbal Mark I
  • Gimbal Mark I can be upgraded by replacing "grip_v07" with "grip_v08"+"handle_v08". "inner" and "outher" are basically unaffected


for details, version history and assembly-instructions.

All Apps

Upgrade this Thing with Thingiverse Apps

No results.

Can't wait for it to be finished.

this thing sucks, not only do all the parts not print correctly the author of the thing is avoiding the questions and bringing up irrevelent arguments.

avoid this at all costs. i regret giving it 2 stars at this point and will never print anything from this person ever again.


here is how the lower 608-bearing hole looks in the latest version (2011-02-05) of the design.

I'll upload it once that 1.5mm thin wall has been found.

The arms of "grip" are 7mm thick and should support enough weight at
gt;=20% rectangular or hexagonal infill.

The sides of "inner" are 5mmm. Easily enough to hold both halfes together.

The thickness of "outher" is 66-57=9mm. It's holes are in fact designed as 8.5mm in Mark I and Mark II. I had to make them larger. I do not know why but printed with skeinforge it fits well.

the 2 ridges are a result from designing the walls of the back of the bearing with a wall thickness of 1mm that tapers down to nothing. when designing things the first thing that needs to be considered is what is the object going to be manufactured on. if its a home built 3d printer i recommend keeping the thickness of walls to at least 2 mm

Doubling the wall thickness on the indicated place in the part "inner" would have no structural effect.

It does not matter even of the piece would break there as the required pressure would still be build up by the 4x M5. 1mm is okay as it can be printed easily (I did it with a printer that has not been calibrated in half a year) and is not subjected to stress in a direction where the thickness matters.

Why would the 8 corners need to be cut? They never touch anything when the part is turned
around the axis given by the 608-bearings inset into it. That is the only direction "inner" ever moves with regard to "outher". Did you find that is touches "grip" at extreme angles maybe?
Then I guess "grip"
should be made a but longer to accomodate that.

The "tapering into nothing" works perfectly fine in Skeinforge and I can print it just fine.
If your software produces artifacts like the ridges on your photos then your software is plain broken.
Skeinforge is the de-facto standard here and produces
working g-code down to the thickness
of 1 wall-with and ignores the part that is too thin. (depends on the calibrated wall-thickness
and is thus not a feature of the design.) The feature works as an end-stop-marker when inserting the 608-bearing after cleaning the printed part and has no structural
No problem there.

I am printing with quite thick 0.4mm layers and a 1.5 height-to-width ratio. A good balance
between speed and details. (0.25 x 1.5 works fine, less then 0.25mm requires advances tweaking of the temperatures and the time the head stays in the same place but is possible.)

2mm as a minimum feature size is very poor even for a diy-printer. 1mm is quite okay.
Search for things marked with the "single walled" tag in this site.

If you think otherwise about the issues you mentioned, feel free to publish an improved design.
I can export the Alibre -files into a number of v
olumetric cad-formats if you need one.

i still dont understand your arguments. tapering to nothing might work sometimes on some machines, but it will print badly when using a home made 3d printer all of the time. what is the point of consciously making that decision in the design process? i sliced this in skeinforge and it produced the two rings on the inside that needed to be cut out. skeinforge added this exactly where it tapered to nothing. that is not a coincidence. running a file through netfabb will not always make it perfect.

it is very clear to me that you have not actually printed this or you would have instantly found out that the corners of the inner part prevent 50% of the motion that is possible with a vertical center column installed. you mention that both inner and outer parts carry over from your original des
ign and can be used. i can guarantee that you will have to trim that part to make this work to its fullest potential when or if you ever end up printing it.

its a shame that you dont understand that things need to be designed for the machine it is manufactured on. you might be able to get away wi
th the choices that you made but there is no benefit to them and the chances that it will cause problems are very great if not guaranteed. a 1mm wall is ok if its on its own and supported on both sides, a 1mm wall tapering to nothing then a gap then to a tapering 1mm wall will not print correctly ev
er. printing a 1mm stand alone post like on the bolt hole will most likely not print at all with a dc extruder, a good stepper with reversal settings tuned in will sometimes have trouble with it, there is no reason it cant be 2mm, it will print correctly for most people and be stronger and still fi
ts in there

I went out of my way to be harsh and criticizing towards you to because of the comments that you have made towards others on this site doing the same, every one of the issues that i brought up is real and discovered from actually printing the items multiple times, i dont think you do th
at when criticizing others. i know for a fact that i have made more of these than you have and i know for a fact that there are problems with the files that you continue to deny. i dont own or know how to use 3d software other than sketchup so uploading a new version is not an option for me. i ho
pe that you can see from my behavior that if everyone was combative, argumentative and generally awful to people this place would not be as awesome as it is.

Marcus exists to thrash around and feel superior. Since he is consistently rude and cannot handle when others reciprocate I recommend simply ignoring him. Most others seem to. Perhaps eventually he will improve or leave.

At least I'm commenting with my own name.

I received some more or less constructive criticism, tried to reproduce the issue, failed and asked and asked again for the details needed to handle it.

Marcus your a great designer and please dont let these people get you down or discourage you! Thingiverse is excellent because of you!

You're joking right? He is the greatest troll on the site and on the side he makes things to destroy camera equipment.

The anonymity wasn't intentional, simply the product of commenting via cellphone.


where would the edges around the 608-bearing ever collide with anything?

You can see the axis of allowed movement clearly. (Just imagine the missing pole in the middle as an additional constraint.)

"might work sometimes on some machines, but it will print badly when using a home made 3d printer all of the time."

You are contradicting yourself.

I just ran it again on Skeinforge version 11.01.03 to check.
No ridges. (screenshot of skeinview uploading...)
If your Skeinforge triggers a bug, we need to report it. It wouldn't be the first bug I found, reported and got fixed in it.

I don't see any updated AD_PRT -file from you on how to make
this better.

The back of the 608-holes is not tapered but cut to measure into a round part thus making a vertical strip in the middle of the hole go all the way through and the edges not go all the way through (since the part I cut into is round).
The edges provide a very nice end-stop when
inserting the 608.
Please propose an improvement that works equally well and does not trigger the bug in your software.

What parts is colliding with what if turned around what axis if you don't cut the corners you mention? You have the original design-files, you have a free 30 day version of the
program used to design it. What parameter of what rounding do you think needs to be changed down to what value?
Since I can't reproduce this, I can't determine how much to cut off on "inner" or how to change "grip". Feel free to upload an improved version or your own design if you don't like it.

Well, the corner you had to cut is a misunderstanding of the axis each part has to move in. You should have a look at the technical drawings to see what axis each part moves in (only 1 axis per part) to understand that "inner" never turns around that axis with regard to "outher" but only perpendicular to it. So that cutting was not required.

The ridges are a glitch in your software. They don't show in the parametric design, not in netfabb and not in Skeinforge. The STL is checked for errors in Netfabb and looks good so far.

As far as the place with the thin wall goes, I would like to know where in what object that is to correct it (a
fter all, Mark II is a work in progress). There are minor changes in sizes in Mark II compared to Mark I to give a tighter fit with less play. You did not answer that one yet.

make one and you will see that the part needs to be trimmed

Tell me the place where to trim or give an improved file and I'll do it.

I'll not be with my printer for at least 2 weeks probably 3 weeks.

i tried printing this last night and was wondering why you added unprintable 1/8" overhangs where it seems that they dont add anything to the design. its almost like you put them in so the part had to be printed with support just for the hell of it. i dont know if the stl is messed up somehow or if skeinforge choked on it or what but i have a very well tuned in profile with a stepper extruder that prints just about anything that i throw at it. wasted a whole night trying to print the part that bolts into the Y part. have you printed this design?

also why is there a hole on both arms of the Y part? nothing seems to bolt to it and it seems like it would weaken the parts

1/8 " = 0.3cm is no problem in terms of overhang. I've had overhangs up to more then 2 cm.

Your temperature is probably too high.

In the 22mm hole for the lower 608 bearing there should be a small ring to make it easier to print.

I printed the lower part of the Y (aborted later due to Z-axis failure) and all the upper parts with my RepMan withthout support and there where no major problems.

The holes are described in the original design. They are for the optional side-handle.

PS: I'm filming fine with the printed Mark I design (mark II is still a work in progress as noted above).

there is something wrong with the stl files. I will take pictures tonight. what i got out of my printer is not the same as what the file looks like in REPG. the part that attaches to the Y part ended up with 2 flanges on the inside (the overhangs i was talking about) that need to be cut out or the inside part wont fit. (they dont show up when i look at the file in repg or pleasant3d) and the inner most part filled in the inside with flat layers covering the whole part about half way up the print (holes in the file?) there are also some design issues on that one. some of the wall thicknesses are less than 1mm when there is no need to make them that thin, there is plenty of space to add a mm or 2 to the wall thickness that would make the printing easier, mostly around the 4 small bolt holes, there is nowhere near enough material between the bolt hole and outside of the part.

I'm currently on vacation for a few weeks with internet in the range of bytes per minute. (if there is radio reception at all)

I'm eagerly awaiting your photos to improve the design.

I'm not sure what part you mean with "the part that attaches to the Y part". It could be the handle or "outher". There are only minor changes to the dimensions of "outher". I did not print the handle yet after my Z-stage jammed.

i was not going to print the handle, i can get a dowel for about 50 cents and have to try to do a 4 hour print. its the "outher" mine printed with 2 ridges on the inside about .75 cm from the top and bottom. they prevent the 2 part piece from fitting in.

On "outher" the back of the 608-bearing-holes are quite thin but print fine here on the Repman (32mm/s not very detailed) and I can't see it preventing anything from fitting anywhere. A simple thingle- or double-walled part of the design. No major overhang.

What part are you refering to as "the 2 part piece" and fit "in" where?

are the bolts supposed to thread into the plastic?

gt;are the bolts supposed to thread into the plastic?

Yes, they are.

It works very fine in multiple designs here and makes the gimbal much smaller and thus stronger (leverage).

why are the holes 8.5mm when they are supposed to hold a M8 bolt threaded into them. they should be 7.5mm or even 7mm if you want to properly thread them first.. on my prints the bolts slip right into the holes with a half a mm around them just like you designed it.. just about every part in this design has problems with it.

the inner part does not spin in the outer part because it is too big (the corners of the cube need to be shaved off, the bearing holes are oversized in the design by .5mm so the bearings fit sloppy, the outer part needs a half an hour of dremeling it because 2 flanges on the part (I have no idea w
hy you put them in)

the site wont let me upload images right now for some reason.

http://grab.by/99Puhttp://grab.by/99Pu here is one of them

and the trimmed inner part

there is also not enough material holding the 4 bolts that hold the inner together. I had to change that on the print that i made.

that is just poor design not taking into account what
this is designed to be printed on. 1.2mm of abs is not very strong and incredibly difficult to print even with a stepper extruder and a tuned in machine. I doubled it and it still just barley enough. if you tighten the bolts that will break before you know it.

gt; http://grab.by/99Puhttp://grab.by/99Pu

There are not supposed to be any bolts on these nuts and the nuts are supposed to be way shorter.

The 1-walled back-wall of the 22mm holes are not printed. What extrusion-width-over-layer-thickness and what layer-thickness did you use for skeinforge to remove that?
gt; http://grab.by/99PDhttp://grab.by/99PD

Your solid layers don't look very solid and the infill does not reach terribly far into the shell.
gt; http://grab.by/99PDhttp://grab.by/99PD

I have no idea where that detail is supposed to be in the object.

gt; http://grab.by/99PDhttp://grab.by/99PD

Your solid layers don't look very solid and the infill does not reach terribly far into the shell.

that is not the point of that image, i had to cut the corners off the part so it would fit in the ring and spin. that is why the holes are there on the corners

gt; i had to cut the corners off the part so it would fit in the ring and spin.

Why did you not tell me then? I can't magically induce what you want to show in a photo.
As far as cutting these corners goes....WHY?
That part will never spin across that axis. It only spins around the axis of the 2 empty holes of
In fact the way too long bolts on the photo go
right through "inner" perpendicular to it's only axis of movement and prevent it from spinning at all.

the part does not move unless the corners are cut.

gt; http://grab.by/99Puhttp://grab.by/99Pu

There are not supposed to be any bolts on these nuts and the nuts are supposed to be way shorter.

The 1-walled back-wall of the 22mm holes are not printed. What extrusion-width-over-layer-thickness and what layer-thickness did you use for skeinforge to remove that?

when you design a hole for a M8 bolt and sizxe the hole to be 8.5mm there is no chance whatsoever that it will thread in. the size of the bolts is irrelevent as are the nuts on the bolt. the photo was uploaded to show you the 2 ridges on the inside of the part that should not be there and had to
get dremeled off

The 2 ridges are not a part of the design. (In the original Alibre files, the shape is a cut-through-everything) I don't know why they are there.

I just checked "Steadycam_Gimbal_Outher_v08 (repariert).stl" with Netfabb Studio and there

are no such ridges in the STL. Could be a glitch in whatever software you are using

(from the screenshot it's no software I'm familar with.)

here is another problem with the outer part bearing, its the same on both sides..


it tapers down to nothing. no printer can print that.

gt; http://grab.by/99QWhttp://grab.by/99QW

It actually worked with 3 solid layers.

First layer has 4-5 lined curling up and easily broken away, second and third massive layer are okay.

There is an update with an inner ring to break away later. I though I had uploaded that already.
I'll upload it when I'm back.

Would look better if designed in 3DTin ;)

Inner and Outher are working perfectly. They have trivial changes to Mark I, so they don't grind on the bearings but have slight contact to the (turning) inner ring of the bearing with no play..

I'm still using the ones from Mark I with a bit of dremel-work myself.

I'm currently printing Grip and Handle.

A 3d-pdf and technical drawings will follow after I printed and tested this.

I expect this to happen today or tomorrow.

Like the other similar designs,utterly pointless. What's it for, a cell-phone camera?

If you would have bothered to actually read the website, you would know that the intended cameras are 500-1000g Bridge- and Micro-4/3 cameras for hobyists with a budget of 100eur.

It works quite well here with a 540g Panasonic FZ100 bridge-camera + Sennheiser MKE 400 directional microphone. No comparison to filming fullHD by hand even while standing still. Let alone while moving.

It already proved to be a working and highly useful addition to my camera for filming at hobby- and vacation -level.

For semi-pro you'd probably rather use a larger camera and a machined aluminium rig instead.

For your next vacation- or family- video this is a decent and cheap tool.

At just 20% rectangular infill Mark I tested well at 5Kg total system weight for 1g of horizontal acceleration (higher loads not tested. I'm not doing braking-tests here).
The Onyx itself weights about 2.5Kg, the rest is camera-weight and g-force.
No additional counterweights require
d for slightly larger cameras.
(More then enough reserves for increasing the leverage instead of the counter-weight.)

Ability for improvement of results:
gt;20% infill and going hexagonal instead of rectangular the carrying-capacity can be increased drastically.

In Mark I you can always increase the capacity with an M8x70 nut and a large washer through the handle.

Of cause you can also just go to shapeways
amp;co and have it printed in metal instead of ABS.

Next design:
Mark II is supposed to hold more weight/more g-forces with the same amounth of material.

Your case:
For your cell-phone or compact camera I suggest you look at UAVG-NP and other electromechanical devices instead. A 2.5Kg rig for a 150-300g camera is way too unpractical.