Hi all, I’m looking for a bit of info on delta kinematics:
What’s the relationship between the size of the effector and machine accuracy?
A lab I’m helping out with is trying to do a special sort of paste extrusion. Naturally you want a stationary bed for that application, but they’d prefer not to use a custom gantry to get that. Deltas have stationary beds, but their effectors are maybe a bit small for a prototype paste extruder.
If they bought an Eris and just replaced the effector with one perhaps 50% larger in diameter, would accuracy go to shit? Also worth noting they’ll be printing crazy slow.
If its crazy slow the best way I think, is an overconstructed system with a big moving head, were the one motor drives the axis and on that axis is the other axis with the extruder, like a gantry style cnc.
I dont think delta is the way to go.
I am not sure size really matters, it would matter if something were not right like a set of arms to long, then the angle of the offset would be less with a wider effector, but that is not the right way to go about mitigating issues of accuracy. If you mean what effect mass has on accuracy, it is the same as any printer, more mass equals lower speeds or resonance. More stuff, more stuff to bounce around messing the print up. With belt drives you can’t really stop resonance from happening except lowering the mass. Delta is very good at that.
Also, tuning a filament profile is totally independent of the effect of the object’s mass on linear motion. Resonance is contained to the linear motion, the object being printed does not move only the hotend can cause surface issues. Until one masters a delta and is comfortable with it, one really doesn’t get how much that matters to fine tuning a machine. One less HUGE variable. Not having to consider the object’s mass on print quality is really nice. If tuned well with no corners lifting and a little Z hop, a tiny base on a large object is totally OK as well.
I like my Delta’s a lot more than my cartesian. They are way faster way due to less mass moving in the linear motion, more accurate at a given speed.
Spare parts for one axis work on all 3! that last one, really nice I am finding. Same bearings and axles not only easy to replace but ease to troubleshoot as if it goes wrong on one it can be corrected there and everywhere else preventatively.
Direct drive is easier, but it is not the future of FDM with multi materials and completed objects with multiple parts and functions right off the printbed with no assembly. I am glad to be getting a handle on bowden tubes right now, seems like good timing.
Multi material setups demand a bowdan so some delta users have a head start understanding it much better. With an extruder mounted on a stepper enabled carriage tracking the Z of the effector the bowden can be shorter than a core X/Y.
Accuracy in a Delta is related to the length and angle of the arms. So by increasing the dimensions of the effector you are making the angle of the arms steeper. This usually increases the speed and reduces the accuracy. But it depends on the relationship between original size and new size. With a 50% increase in diameter, you will lose some accuracy. Quick fix is to shorten the arms (if that can be done) by 10% or so. (This is a guestimate, not the result of calculations)
Ok thanks all, shortening the arms makes sense to me…