Planning to build a stabilizer

[Jim Yang]

I'm pretty confident that I can build a stabilizer of decent quality...

I'm not sure of the design I should go for though. I've put alot of thought to the I style fersus the > style.

The question I have is about the gimbal. Should I go for a 3 or 2 access gimbal? So basically it would rotate on the X and Y and also the Z.... would the Z rotation be necessary and would it cumbersome?

[Ken Tanaka]

Jim,
Do a Search on "stabilizer" and you'll find pages of threads with tips as well as links to various home-brew designs.

[Jim Yang]

I've done that. No one really mentions whether a 2 or 3-axis system is better.

[Charles Papert]

Jim:

For the system to work, it must have 3 axis rotation. The classic design is to have a "planetary ring" design for the pan and roll axis (I am using names rather than "x" and "y" to avoid confusion), in that an inner bearing race circles the center post for pan, and an outer yoke contains bearings at the attachment point to allow for movement in the roll axis. The tilt axis bearing is contained in the yoke. In the unique case of the Steadicam JR, which has an inter-gimballed handle, all three axis have the same center with pan in the middle and tilt and roll emanating outwards from there (offset 90 degrees from each other). It's a great design but one that would be pretty complicated to build yourself!

In any event, a two-axis gimbal would severely limit the isolating effect and make it impossible to achieve properly stabilized images. The reason for this is that the motion of a human body will always involve compound axes and whichever axis is not isolated by the stabilizer will directly transfer into the photography.

I hope this helps, if not I will try to clarify further. There's a good description of the Steadicam principles at http://www.howstuffworks.com/steadicam.htm.

[Ken Tanaka]

Ahh, I was hoping you'd have a chance to chime-into this, Charles. Nothing better for this thread than the voice of a veteran professional Steadicam operator!

[Jim Yang]

<<<-- Originally posted by Charles Papert : Jim:

For the system to work, it must have 3 axis rotation. The classic design is to have a "planetary ring" design for the pan and roll axis (I am using names rather than "x" and "y" to avoid confusion), in that an inner bearing race circles the center post for pan, and an outer yoke contains bearings at the attachment point to allow for movement in the roll axis. The tilt axis bearing is contained in the yoke. In the unique case of the Steadicam JR, which has an inter-gimballed handle, all three axis have the same center with pan in the middle and tilt and roll emanating outwards from there (offset 90 degrees from each other). It's a great design but one that would be pretty complicated to build yourself!
-->>>


I have not doubt that the machining shop here will be able to make parts to my specification. I'll just need a drawing! Good thing I'm an engineer....

[Charles Papert]

Jim:

Reverse-engineering will be probably be your best friend--just ensure that you are copying the best product!

If you do decide on the "<" design which will allow for an inter-gimballed handle, you are talking about much, much smaller components in the gimbal which requires spec'ing out just the right bearings. The "I" design is much easier to build but is more tiring to operate due to the leverage involved in an offset gimbal. Either way, keeping the stiction in the gimbal to an absolute minimum will likely be the make or break in the results, followed by rigidity (as much as possible to avoid vibration), weight (as light as possible to allow for maximum shooting comfort) and flexibility of design (accomodating varying configurations of accessories or even different cameras).

Good luck!

[Jim Yang]

So when I do eventually build this device what will be the most optimal way to balance it?

I'm making it so the gimble assembly will be adjustable along the main shaft. I'm also going to allow for 2 axis adjustment of the camera on the top (i.e. forward, back, left, right)

Should I place the gimble axis in the center of the pipe then attempt to balance the GL2 with weights or should I get approximatly the weight of the camera in weights then adjust the gimbal to suite?

[Charles Papert]

Jim:

I've discussed this very thing recently in another thread, but I can't get around to finding it so I'll post it again.

Keeping the weight to a minimum is very important to a handheld stabilizer as it can be very fatiguing to operate. To that end, my recommendation is that you load the camera with every accessory you might possibly use, then slide the gimbal up towards the camera platform, stopping short about 3/4" inch. Then turn the whole assembly sideways so that the center spar is horizontal. Experiment with different weights at the bottom (in this configuration, it might be easiest to hang them off the bottom with string or in a plastic bag) until the rig levels off, i.e. hangs horizontally. You can hold the rig from the gimbal while you do this, or simply place it on a fulcrum at the gimbal, whichever is easier. This will help you determine the absolute minimum amount of weight needed to do the job. You can see that the lower the gimbal, the more weight is needed to achieve the same balance.

Once you have mounted that weight, you will need to raise the gimbal slightly to achieve the ideal balance (slightly bottom-heavy), which is found by again turning the rig sideways to horizontal and allowing it to fall back to vertical. A 2 to 3 second drop time is optimal. The slower the drop, the harder it will be to tune the x and y axis perfectly, incidentally.