PDA

View Full Version : Gimbal Design



MMaister
12-10-2011, 08:11 AM
Hi pit builders,

I would like to show the gimbal layout I am currently working on.
I use the principle of the real aircrafts "feel and centering unit". Ball bearings are rolling on CAM plates (one for each axis) as the stick is moved. The CAMs forcing the bearing holder arms up and down. These arms lifting the extension springs.

The good thing is, that with different profiled CAMs I can produce different stick characteristics.

I will also use rotational fluid dampers (one for each axis) which will damp the movement of the stick to get a nice smooth realistic feeling.

http://farm8.staticflickr.com/7003/6486479749_f187209139.jpg (http://www.flickr.com/photos/22174211@N02/6486479749/)
gimbal (http://www.flickr.com/photos/22174211@N02/6486479749/) by csaba.moharos (http://www.flickr.com/people/22174211@N02/), on Flickr

Any suggestions and comments are welcomed:)

Westozy
12-15-2011, 07:29 AM
Nice work, I can see how it operates, it should be good!

Westozy

Ronson2k9
12-15-2011, 08:36 AM
Looks pretty interesting. You may want to add a physical stop to your cams so the stick can't be forced beyond them. You could add a bolt beyond the spring that passes through a hole with the lock nut on the other side. This will let you adjust the amount of rotation per axis. The cam node has to be a lot less. As you move the stick this also rotates the node. That angle gets steeper the more you rotate so you have to subtract that from the angle. That is the height of the node and the angle as it arrives at the top (extent of rotation). The angle in the valley of the node needs to be greater.

A few other things..

1 - There should always be tension in the springs (always under load) This will insure there is no play when the stick is in neutral position.

2 - To increase longevity in the arm and bearings a second node and longer pin to bracket your lifter bearing might be wise. As you lift the arm there is going to be torsion on the arm this torsion over time will ware the bearing faster then normal use.

A second cantilever might make the rotation and return force a bit more smooth. The double rotation you have currently multiplies the angles and there by makes it more rigid. You can tweak the angles to make them less rigid but it's always compounded. So you will be moving in increments of 2. In essence it's one pulley when if you had a turnbuckle it could smooth it out more.

As you have the I'm thinking Sketchup model you could play with the rotations and see what the angles are doing. Especially look at the top of the rotation. You will find that as your cam (node) nears the top it's bearing and the surface it rides on will be quite close to parallel with your direction of movement.

MMaister
12-15-2011, 09:45 AM
Looks pretty interesting. You may want to add a physical stop to your cams so the stick can't be forced beyond them. You could add a bolt beyond the spring that passes through a hole with the lock nut on the other side. This will let you adjust the amount of rotation per axis. The cam node has to be a lot less. As you move the stick this also rotates the node. That angle gets steeper the more you rotate so you have to subtract that from the angle. That is the height of the node and the angle as it arrives at the top (extent of rotation)

Thank you for your valuable input.
I also thought about screws to limit the rotation, I will place them to stop the rotation between +/- 20deg.
I am not 100% sure that I understood correctly what you wrote about the "node". Do you mean that the distance between the rotation axis and the deepest point of the CAM is too long? Should I place the CAM`s deepest point as close as possible to the rotation axis to achive a linear movement?

Right now I am working on the compression of the design. I have to use more compact springs and shorter CAM arm to reachthe desired size of the mechanism.

Ronson2k9
12-15-2011, 11:13 AM
I had drawn up a node for you but this could be better. Christmas so working fast..

5909

Ronson2k9
12-15-2011, 02:12 PM
Changed it for the better I'm thinking..

5912

On the left you see the new node. This one the arc is so that cam arm can roll smoothly over the surface. The one on the right shows what the arc would be if the node was circular. Giving you the distance traveled by the arm between the red and outer edge of the node.

This design will let you have nice free movement of the arm (not much binding if any) other design could bind near the top of the movement..

I added a bit of divot to the top center. Not enough to hold the stick in place as it is the point where the force restricting movement is the least.

It took me a bit to realize that the outside of the node had to curve out from the fulcrum as the other way makes for weird forces. This creates an easement into the force. Depending on spring force perhaps enough to return to center but you have to think of the amount of force you will use to extend spring. Some experimentation could be required.

MMaister
12-15-2011, 03:46 PM
Hi Ronson,

thank you for the energy and time you put into this:)

I see the point. The goal is to make a cam what will produce a linear spring expanding compared to the stick movement.

The problem is that to achive compact size I need to use short springs with huge pull and short travel. Therefore the CAM profile will be short and quite flat. I started to create a new profile. As you said the big question is that the spring will be able to come over the radius and center the stick. Now I am experimenting with 610N springs. These are pretty strong, and should give a ca. 3,5kg max deflection force on the 530mm long stick.