What kind of content would y’all want to see in a 3d printer engineering book chapter on designing Deltas?
“the math is crazy complex” hahaha
Incidentally, I already have 9/10 chapters written for this book, Deltas are the last chapter. Already got all the 3D printer basics like stepper motors/drivers, bearings, frames, belt drives, screw drives, etc.
Motion calibration:
- make sure homing movements work
- set Z=0 where the nozzle touches a business card on the printer with the nozzle in the center of the heated bed
- make sure that when you move the nozzle up 10" (254mm) that is actually moves that distance
…
or some such crap like that. I found out first hand that you have to calibrate the vertical movement before calibrating the rest of the movement.
There are of course other steps that I omitted and measuring the diagonal rods and your horizontal spots between the effector and the shoulder joint of the diagonal arms is important.
Man, what are you doing when you build your deltas that you’re not getting steps/mm right? Lol, actually, don’t answer that 
I will definitely add a basic manual calibration procedure.
Just make sure that the Z movement calibration needs to come before trying to fix any convex/concave issues or they will be wasting their time, please. It was my stupid mistake to not think about it soon enough when I changed firmware. Any stupid mistake that I make, I like to share with others so they will not make the mistake. Maybe that is backwards. I just figure success or failure or mediocre results all have value as long as someone learns from them. Knowledge is power. yada yada yada
It is difficult to have more than one nozzle on the effector without one of the nozzles dragging on the print.
How firmwares differ in their handling of arc segment processing.
The importance of little to no slack in components. Even the littlest misalignment causes artifacts that are difficult to track down.
The importance of having a completely equilateral triangle. The resolution changes and moire artifacting that comes from those resolution changes in the X/Y plane.
It is important to have your towers parallel to each other and perpendicular to the bed and ceiling (since the endstops are on the ceiling).
Good stuff guys, thanks for ideas. Keep’em coming
@Whosa_whatsis do you mean anything aside from “segments vs non-segments”? What comes to mind is that RRF doesn’t use segments, Replicape uses a fixed segment length, and everybody else uses a segments-per-second rate. Anything else you can think of?
That’s exactly the kind of thing I had in mind. That and the pros/cons of each of those approaches.
Ok, got it.
Hmm, it occurs to me that I have no idea how MachineKit does motion interpolation for Deltas.