Glidecam design -- I'm doing it wrong

[Pete]

Hey everyone, I have an engineering question for y'all!

So for the past few days I've been attempting to redesign my steadicam. As you may already know, I've had problems with my merlin-styled steadicam ( http://www.indietalk.com/showthread.php?t=33955 ), so I tried to redo it styled like a Glidecam. Based on the product images...

...and demo videos on Youtube I concluded that the gimbal/joint area consisted of a handle that twists and pivots side-to-side, an up-down pivot point on the shaft, and a bearing that allowed the whole handle to spin freely around the shaft.

So that's exactly what I tried to do --

The problems with this DIY glidecam are severe. If one joint moves too far in a single direction (namely the side-to-side hinge on the handle), it will lock, causing the whole steadicam to fall basically sideways. It's extremely wobbly even when I try to hold it still -- as if my hand's movements are magnified rather than minimized by the apparatus.

All of the individual components/joints are fine, so I don't think that's the issue -- I'm almost certain it's a design issue. So that's why I came here -- I'm really frustrated and can't figure out why it isn't working, and my only logical thought is that my initial design was fundamentally wrong.

Thanks for any and all help!

 

[cameronchapman]

What kind of weight/balancing system are you using? That could definitely affect how the whole thing is moving.

 

[Pete]

What kind of weight/balancing system are you using? That could definitely affect how the whole thing is moving.

Just like the glidecam, the main shaft has a telescoping bottom so I can lengthen or shorten it (the shaft is actually one leg of an easel). On the very bottom, there is a board that extends out sideways with weights on the ends, much like here:

On the top is a place to mount and adjust the position of my camera.

I haven't tried to mount the camera or adjust the weights too much yet, as I don't trust the steadicam to hold it until I can get it to stay straight on its own.

 

[CamVader]

How are you attaching the camera? The Glidecam has mounting holes for the approximate front to back and side to side center of gravity, you then fine tune this (f to b, s to s) with knobs that give you microadjustments. If you don't have a way to do these microadjustments, you're SOL.

You have to have the unit's handle mounted on a c-stand to do this right. The camera has to be perfectly centered and balanced before you mess with the drop time.

Edit: you have to have the camera mounted before any adjustments mean anything.

 

[Pete]

How are you attaching the camera? The Glidecam has mounting holes for the approximate front to back and side to side center of gravity, you then fine tune this (f to b, s to s) with knobs that give you microadjustments. If you don't have a way to do these microadjustments, you're SOL.

You have to have the unit's handle mounted on a c-stand to do this right. The camera has to be perfectly centered and balanced before you mess with the drop time.

Thanks for the reply, but I'm not even to the point of microadjustments yet. The camera, once mounted on top, can be moved left/right and forward back, so when I do need to fine-tune the balance that won't be an issue. Neither is the drop time -- as it stands (without the camera), the contraption is bottom-heavy, but it will still wobble like crazy and the joints will still lock.

What I'm really wondering is if this is even the right approach to mimick a Glidecam, or if they actually operate differently and I just failed to see that difference. If it is the right approach, then I'm in a bit of trouble :/

 

[CamVader]

Thanks for the reply, but I'm not even to the point of microadjustments yet. The camera, once mounted on top, can be moved left/right and forward back, so when I do need to fine-tune the balance that won't be an issue. Neither is the drop time -- as it stands (without the camera), the contraption is bottom-heavy, but it will still wobble like crazy and the joints will still lock.

What I'm really wondering is if this is even the right approach to mimick a Glidecam, or if they actually operate differently and I just failed to see that difference. If it is the right approach, then I'm in a bit of trouble :/

It doesn't matter what it does without the camera.

Just so you know, you're talking about rough adjustments, not the required microadjustments. If you stick the camera on there and it tilts in any direction, it's out of balance. Correct on problem at a time, and that can only be accomplished if you can lock the plate in a rough balance and then microadjust front and back independently from side to side.

 

[SinEater]

I'm not understanding the point if the side to side hinge at the top if the handle. What happens when you get rid if that?

 

[Pete]

I'm not understanding the point if the side to side hinge at the top if the handle. What happens when you get rid if that?

I just tried that in fact by clamping the joint down. It's still amazingly wobbly and the shaft still spins off-kilter. The only thing it really accomplishes is now the joint won't lock at 90 degrees and pull the steadicam sideways.

It's supposed to be the third axis of rotation to separate the hand's motion from the steadicam's motion.

 

[David.rhsc]

I just tried that in fact by clamping the joint down. It's still amazingly wobbly and the shaft still spins off-kilter. The only thing it really accomplishes is now the joint won't lock at 90 degrees and pull the steadicam sideways.

It's supposed to be the third axis of rotation to separate the hand's motion from the steadicam's motion.

It might be that having that pivot point that low on the handle is causing part of the problem. The handles usually pivot on that axis at the point where they join with the center gimbal. On yours essentially where the 90 degree bend in your wire handle meets the u-bolt. You want that to be a pivot point rather than where you have that small bearing. In order for the whole thing to level out you want all of those pivot points to be in the same place.

Unless I am reading the pictures wrong, in which case, carry on.

 

[Chrisk]

I really don't like that joint on the handle either, cant see how that can do anything apart from let the whole thing fall sideways.
Hard to tell from descriptions but I'm guessing the balancing is way out, your centre of gravity is in the wrong place. If it tips a bit and the top isn't balanced that can cause a spin as the heaviest part heads downwards. As for the wobbling, could that be play in the joints?
If you manage to get a video up I'm sure everyone would be able to give much more helpful advise.

 

[gpforet]

There is a few problems with your design. The most detrimental being the lack of a true gimbal joint on the vertical shaft. The purpose of the gimbal is to locate the x&y pivot (fulcrums) in as close proximity to each other as possible. In this way, angular forces are more easily communicated and distributed. From what I can see of your design, there is several inches of space between your x and y pivots. What I can't tell from your pictures is the distance from the pivot to the camera plate. This joint should be up high, near the camera which helps reduce the amount of counterweight required. Think of a see-saw. Fat kid on one side, skinny kid on the other. By moving the fulcrum closer to the fat kid, the easier it is for the skinny kid to get his side to come down.

The number of variables in your design is mind-blowing from a weight distribution and calculation perspective.

There are many people getting very usable results from both the glidecam and merlin style stabilizers. I recently purchased the Longbow U3x (merlin style), and spent many, many hours setting up the initial balance. Then practiced with changing lens and rebalancing. Within a week or two I had more clear understanding of what I needed to do to bring the system into balance and the fine adjustments needed to keep it in balance. Even properly balanced, technique is very important and it's taking lots of hours of practice to get usable shots. But the more I use it, the less time it takes for me to get a usable shot.
Even steadicam operators have to go thru a week long training session to learn the basics of balance and use. Then many, many hours of practice to get usable shots. On a recent job, we had a steadicam operator and it took him over an hour to set up and balance the rig, and then 20-30 minutes for him to practice the walk-thru, first just using his fingers to point where the camera will be framed, and then a couple of walk-thrus wearing the rig. And even after that, it was 3 or 4 takes before we got a usable 20 second shot.

Like most things in filmmaking, it ain't as easy as it looks.

 

[Pete]

@The last three posts: Thank you! That's exactly what I've been looking for.

The extra length on the bolt that the handle attaches to was just in case I messed up with the initial attachment -- turns out I didn't so I just left it long. Hopefully by this weekend I'll have a modified version to test out.

Thanks again to everyone who responded!

 

[Pete]

Alrighty, I had some time today to do yet another re-design of my steadicam. This is actually steadicam v4.0 for me -- the first was based on the design here: http://www.yb2normal.com/DIYsteadicam.html. The second was based off a Merlin (details at http://www.indietalk.com/showthread.php?t=33955), and the third was based off of a glidecam, as detailed above.

So this is my fourth, and I can proudly say that it's finally starting to work!

It's based again off of this guy's design: http://www.yb2normal.com/DIYsteadicam.html

I went back to the gimbal design being completely around the shaft rather than spread-apart. It seems to do much better than the Merlin-based design at containing the swinging because the entire length of it is longer, and it's certainly better than my Glidecam design above. I will be sure to post some sample video once I get the steadicam well-balanced, but in the meantime, here are some photos of the finished design...

The gimbal includes two outer pipes which rotate along perpendicular axes. The inner pipe holds a bearing, which spins freely around the shaft. I hold onto the outer pipe.

The bottom has rotate-able weights that are spread out from the center of the shaft. This allows minor balancing issues to be fixed at the bottom rather than from the camera mount.

The camera mount can slide along the slot in the metal in one direction, which can all slide in a perpendicular direction. It's crude, but it gives me full flexibility in the x-y- direction on top with the camera, and allows multiple different cameras with different centers of gravity to be mounted easily.


So there you have it -- my day's project (but really my year-and-a-half project) is finally starting to pay off. Any other suggestions to continue to improve the design/functionality are of course welcome!

P.S. Also, what do you think of the pictures' quality in this thread, from a photography standpoint? I haven't put much thought into framing, and I didn't bother with retouching on my computer -- they're all just me holding the steadicam against the wall with one hand and holding the camera with the other hand. I'd like to hear some ways to improve the quality of my photos as well as video, but like I said I haven't put much thought into that aspect of the camera.

Thanks!

 

[wheatgrinder]

great progress..

I like this shot a lot.. looks very cool and sci-fi ish..