Hi, My goal is to create a camera operator that replicates shots that I demonstrate to it. It needs to go 1-1.5 metres a second - bonus if it can go up to 3 m/s. For that in my research and measuring the area required, I need to create a 3 axis system at least. The X axis needs to be 7.2 Metre X 2 Y axis at 3.53 metre X 1 Z axis at 2.34 metre X 1 The camera equipment (approx 3-4kg) sits on a moving carriage with a breaking motor on a stable Z-axis . It would be belt-driven with stepper motors on the other two axis. For the long length, I would join two rails with flat butts together with one inversely positioned so the gaps are not the same. I would use a displacement sensor to collect position data. Now I need help, I am not sure if any of this is right for this length and weight. I have discussed this with a couple of other people who say that the load on the x-axis might be too much for the belt since the length is long and the speed is this fast and. That instead I should use steel ropes and a pulley mechanism. It would be a bonus if I could have panning and tilting. What are your thoughts? I would deeply appreciate your help with planning this out. Where possible please include the components needed and their configuration. Any additional help is also appreciated.
I've thought about one of these too. They're complex, though, I've never been able to justify the time or cost I'd have to put into one. It's robotics, really, more than CNC. 1) You're gonna need servos, not steppers. A) they're fast enough to operate the system as you want, B) they have built-in encoders that you can use for teaching orientation data, no sensors required. Servos are expensive- think $400 per axis (including rotary axes) at the lower end. 2) This is oriented like a standard gantry router system, it seems, with X and Y axes reversed. I'd counterweight the Z-axis, rather than adding a brake. Any rotary axes too, or at least fluid-head attachments for training. 3) Not sure your butt-rails-together plan will work super well, unless you build some amount of shock absorption into the system (maybe have the camera itself on some kind of shock mount). 4) I'd probably do some kind of pinch-pulley system where the motors pull themselves along. Maybe rack and pinion. Something where the motors are moving, anyway, to minimize static line. Very wide (30mm+) belts might be ok at those long distances, if it suitably tensioned. 5) Teaching the moves might be tricky. You may need to write some kind of software that reads the servo encoders while the motors are unpowered, then plots, fits, and converts to g-code those points, so you can send them to a regular 5-axis control, like LinuxCNC.
Hey Rob thanks for your time and inputs! I would love to get this done so I'll put my time and pay for the costs, what I could use is some consultation and teaching. Would you be interested in coming into a compensated agreement for consultation and teaching to help me complete this?
