QTPlasmaC with fourth axis - rotary

LesNewell wrote:

just out of interest how do the A axis velocities compare to the first cut?

Interestingly it is a little faster but still a long way from what it should be. Rapid moves are fine but coordinated moves are ordinary to say the least.
 

I will try to set up a Linear A axis next.

Here is a short video showing what I mean by G93. Perhaps sheetcam already exports code this way, I am not sure. But if nothing else, he has a good explanation of the problem and the solution.

This run was with the G93 conversion, apart from a 2.7 second longer run time the numbers are almost identical to the standard G94 run.
 

EDIT: Changing the A axis to LINEAR made no difference to either the G94 or G93 figures.

I am convinced it is the lack of look ahead in the trajectory planner for axes other than X, Y, or Z.
This seems to gel with 3D printer configs using the Z axis for the extruder and the A axis for the cartesian Z axis.

The issue with the one line look ahead is usually jerky movement (the shorter the segments the jerkier). If your runtime with g93 is longer than with g94 then I would suspect something went wrong with the calculation of the F values.

snowgoer540 wrote:

Here is a short video showing what I mean by G93. Perhaps sheetcam already exports code this way, I am not sure. But if nothing else, he has a good explanation of the problem and the solution.

 

SheetCam already compensates for angular feed so this should not make any difference.

If your runtime with g93 is longer than with g94 then I would suspect something went wrong with the calculation of the F values.

Actually, looking more closely at the original GCode using G94 it does indeed seem to adjust the feed rate in relation to the angular movement. So a conversion to G93 might not give the advantage that is seen with other Gcode.
The thing that I find a bit odd here is that with G93 code I never seem to have an issue with slow motion but more with jerky movements as the planner seems to accelerate/decelerate for (each?) segment. Given the rather large feed rates I wonder if maybe there is an issue with max acceleration of your axes.

On the sim I have tried accelerations for the A axis ranging from 1000 to 50000. 
In the second video when doing the second cut you should be able to hear the odd vibrations each time it moves the A axis around the corners of the tube.
Tomorrow I will do a comparison on the second cut with and without the A axis.

In the second video when doing the second cut you should be able to hear the odd vibrations each time it moves the A axis around the corners of the tube.

I presume you mean this one:


It is not clear to me why it is much faster up until 0:40 and then really slow after that. What changed in the GCode between the two cuts?

Yep, that is the one.
 I don’t think there is any real change. It may be there is less X motion in the second cut so the segment lengths are shorter?

The segment lengths will make a big difference. I have found this with cutting a hole that was specified with lot of very small segments instead of longer arcs. If the acceleration of the machine is not set supper fast it becomes even worse with the small segments