I was hoping to get some resolution on this topic which is now closed due to the lack of responses. Hopefully the developer has settled into his/her new location and can provide some insight.
Previous post below:
Again the issue is that Lightburn will incorrectly convert vertical height to a rotary distance resulting in a non seamless etch around a cylinder. I need to know if there is a software fix planned, Or if I need to look at difference hardware/controllers, or if I am just missing something.
I can have a look at this, but if you’re able to see the GCode output, and the last line is just short of where it should be, a stop-gap would be to just tweak the image height until the last A value is one step shy of 360 degrees.
Depending on the interval value chosen, it might not be possible to get a perfect ‘first line meets last line’ engraving of an image, but you should be able to get within 1/2 of the interval value chosen, either over or under.
I’ll have a look at the generated code and see if I can figure out what’s happening.
Yes that is what I have been doing. I edit my graphics height value to be a little over or under the circumference in 0.1mm increments in order to have the Gcode A-values add up as close to 360 degrees as possible. It’s tedious as I have to count the the values in the code manually and further if I change any other parameter like the width of the object then then the A-values will add up to something different.
Another issue I have, is with dither patterns like Jarvis where the corners of the image are not interpreted uniformly. Below is an image where I tweaked the Gcode to minimize the seam. Near the ends the dot pattern trails off.
The issue with the dither patterns is not going to be easily changed - Error diffusion is directional, and goes top-down, left-to-right, and the generated dot patterns will never wrap perfectly as a result. Atkinson dither is less sensitive to this, and might help. Ordered dither will produce visible patterning, but it will be very consistent across the entire image, and should wrap seamlessly, as will newsprint.
LightBurn emits all scan moves in relative coordinates because the GCode is generally much smaller, and therefore takes less bandwidth to stream. There is no way at present to disable this, and I might even have to change the code itself to handle non-relative scanning moves. Vector cuts will always be absolute when using Absolute Coords mode, but User Origin and Current Position modes also use relative moves, again because the code is smaller. It also makes re-positioning a job trivial by editing the GCode - you only have to edit one or two lines, instead of all of them.
Yes we have good success with newsprint. A’ll give atkinson a try.
I do have a quite a wishlist for what I am doing but having the rotary produce a seamless dither (getting the Gcode to add up to 360) is the most important for me.
I saw a new version was recently released. I tested the Gcode output again but the result is still the same. Have you had chance to review the issue that the A axis moves do not add up to 360 even when the vertical size of the object is equal to the object circumference?
