Kinematics for KFLOP linear and circular

Moderators: TomKerekes, dynomotion

AndreyP
Posts: 7
Joined: Wed May 29, 2019 6:34 pm

Kinematics for KFLOP linear and circular

Is there kinematics for KFLOP? (Note KFLOP may have this functionality but its called something else).
We have another controller that is a TRIO and they have some functionality that offers linear and circular kinematics such that you can make lines and circles without doing point to point gcode and reduces the gcode required dramatically plus since it is formula based the movement is in smooth paths and not jagged point to point movement.
If KFLOP does not have this any ideas of how it can be implemented?
If this sounds interesting to you please join this thread and let's figure this out.
Cheers.

TomKerekes
Posts: 591
Joined: Mon Dec 04, 2017 1:49 am

Re: Kinematics for KFLOP linear and circular

Hi AndreyP,

I think the term "kinematics" normally refers to something else such as a robot arm where the XYZ position is controlled by changing angles of the arm links. You might see our wiki on kinematics.

What you seem to be describing is simple linear and circular interpolation of multiple axes. GCode can do this easily with G0/G1/G2/G3 linear and circular motion commands.
Regards,

Tom Kerekes
Dynomotion, Inc.

AndreyP
Posts: 7
Joined: Wed May 29, 2019 6:34 pm

Re: Kinematics for KFLOP linear and circular

Ah right yes this is true what you are saying. However I guess what I am meaning is we need a way to edit the "G" in the gcode because those are specific to gantry xyz machines. We have an 8 axis cable robot that connects to a single head unit and that moves in 3d space. When say a G1 command is used "The program specs the start and end points, and the control automatically calculates (interpolates) the intermediate points to pass through that yield a straight line (hence "linear"). The control then calculates the angular velocities at which to turn the axis leadscrews via their servomotors or stepper motors." Wikipedia. "linear interpolation is a method of curve fitting using linear polynomials to construct new data points within the range of a discrete set of known data points. " Wikipedia.
So for our machine it does not know how to move properly in between points.

So how do the delta printers handle this? They needed to take the gcode produced by the slicer and do some magic to reinterpret the code to their non gantry style system. We need to do the same kind of method maybe.

An example of our format is:
F90
G1 X0.2921 Y0.2906 Z0.2974 A0.2842 B0.2922 C0.2905 U0.2978 V0.2845

Since the G1 does not move our head properly in straight and level lines we have to put in tons of points really close together and for a simple cube we're over 400,000 lines of code plus the real path of the head is not correct.

We have to put in a "G" of some sort and is mostly G1 so we need to redefine or make our own "G1" and call it say G200 which is not used -I think. We have already made an linear interpolation of sorts so maybe we could that for the G200.

BUT after reading more and especially seeing the example video of the 6 axis cable robot and how they are using the KFLOP kinematics (not clear how though) it would seem that is the way to go. It seems the kinematics can sit in between the slicer gcode and the controller and with relatively few lines do the conversions and calculations required. If you have some experience with cable robots and have discovered using the kinematics is the way to go can you detail or especially provide some example code for the different stages required to go from regular sliced gcode file to final execution code?

TomKerekes
Posts: 591
Joined: Mon Dec 04, 2017 1:49 am

Hi AndreyP,