I want the exact profile (wrt to time) of my effector motion.
-
Any way to get this from the gcode or other method? It would help a lot knowing what my actual jerk,accel, and velocity emerges from a move. My theoretical upper speed on my delta (setting step rate of Duet at 50Khz, and .0003" step resolution) gives me an upper end of 380mm/sec, WAY faster than I can go. I just updated my machine with lighter parts and want to see the difference more than just "it prints faster".
(it occurs to me I can probably write a python script to analyze the gcode)
edit: there is no timing information in gcode so I guess that won't work since the time between each step is unknown) -
Slicers have 'virtual' printers than generate print time estimates, Cura's used to be hard coded (based on an Ultimaker), I have no idea if that has changed (but it accounted for jerk, acceleration etc).
But for now, the simulate function in RRF should give you a close approximation. At least I have found it to be reasonably accurate.
Assuming your settings are within the limits of the machine, you can have a reference piece of gcode that you run to test the impact of the settings. -
thanks for the reply!
Here is a web based analyzer that is accurate enough to include motion parameters like accel. https://www.gcodeanalyser.com/Otherwise, I guess I'm trying to get a real-time plot because I want to know WHEN I'm hitting limits. The weight of the mobile parts and the dynamic torque of the motors determine when those limits are reach but that isn't practical at all (and even harder with a delta).
But I it seems like without an external way to track the tip there is no way to know. And of course if you could do that, then you can close the control loop and get rid of calibration problems (almost) completely.A lot of concepts from the original reprap release were fine for the early days and cheap builds but as performance is increased some of those elements need to be revised. Closing the loop to get cheap but precise nozzle placement would be a big improvement and just eliminate most calibration issues. Easy to do with 3 encoders on a cartesian, much harder on a delta. Sometimes I wonder if my delta is worth the trouble. There's a reason most printers aren't deltas.
