Continued from: http://bradisdrab.wordpress.com/2013/04/30/cambot-pt4/
I haven’t been 100% satisfied with the movements I’ve been getting from the Pan/tilt head so far. Its promising, but there is some play in the parts which introduces some counterbalancing and wobble. That and the motors turn a bit more quickly than I’d like, which makes small, controllable, incremental, movements a bit tough.
Time to take it apart and rebuild.
I couldn’t get good results with my hand drill. It was just too sloppy, I needed a better way to do it. I wish I had a milling machine, a full machine shop would be amazing, but I can make due, and make whartever I need.I tried pricing out mini milling machines, but they are just too expensive, instead I looked into small drill presses. I found one for a good price and Kyle and I put it together with a vice and the cross slide and rotary table, I had been using as the camera mover previously. I ended up mounting the whole thing to a metal stool I found so it can be wheeled around the shop. Totally custom and it works great.
Thankfully getting better tolerances from this drill press. The right tool for the job. Cambot is reassembled, now with everything fitting together much tighter, and swapped out the motor in the pan for a slower one.
The basic blocking for Cambot’s pan and tilt is finished and working, so I’ve started planning and working out the mechanics for the next two axes. The first of which is pictured above, the rack-focus. This is likely the easiest of the axes; simply a pinion gear mounted to a mini robotzone gearmotor, which will fit with the len’s focus wheel’s teeth. I mounted this motor to a little arm attached to the 15mm rails on the P/T head, so it will remain adjustable for different lenses as well as remianing in position with the camera.
Next comes the Jib, it needs to be able to crane up and down. The Jib is well counterbalanced so won’t need a lot of torque, but with nothing there at the moment it tends to drift slowly.
I spent awhile thinking about ways to hold that axis, and looked at a lot of options, eventually deciding on a timing belt. Also I learned that parts can be ordered in bulk to save money from ebay, so I picked up a bunch of timing pulleys, and bearings. Unfortunately they weren’t as easilly matched as the parts I got from Servocity, so I had to do a bit of work on them to get them assembled.
I fit two idler pulleys and one to drive the belt. Looking pretty slick now, just needs a motor.
With 4-axis nearly in play, and two more on the way, I’m going to have to start thinking about how I’m going to control this thing. My current testing methods the MX2 and can manage 2-axes at a time
This new MX3 from dynamic perception looks interesting, it can control 3 axes, and includes a MoCobus port which could be very promising in the near future, afterall I’ve been loving the Mx2.
My other option is to customize something from scratch using an arduino. I’ve been pouring over tutorials and it actually sounds totally do-able. I’ve been calling that “The Magic Box”, basically a controller that does anything I want/imagine, from stop motion, to live control and follow modes, with programmable control of dynamic lighting, and the ability to import/export 3D camera moves from the computer. One can dream.