I’ve been running numerous grayscale tests on cardstock using Lightburn instead of Luban.
I’ve ran a speed test vs power and settled on S800 at 15-85% power with .18mm Line Interval (distance between two lines). This gives me a good dark and an acceptable white spread.
On the test matrix speed 800mm/min does not burn through even at 100% power and I’m using 85% max.
So it baffles me when I’m now burning through my image with 85% power.
Why would a solid square at 85% not burn through, but an image that occasional has 85% max does burn through?
The bottom image on the right burnt through at S800 P15-85 LI-0.18
I then reduced the power to 15-75% and increase the LI to 0.2. A little less burn through but stilll.
The little black square between the two images is at a solid 85% again, just to test that 85% does not burn through. It did not.
What do you guys think is causing the burn through at proven safe values for my cardstock?
Just to clarify, you’re using grayscale mode in Lightburn? Since the snapmaker firmware halts motion before changing laser power the travel speed will not be anywhere near 800mm/min, as changing from 75%->75.1% will cause the toolhead to stop and start moving again. Since your actual travel speed is far less than 800mm/min that’s what is causing it to burn through.
Your test rectangle is not changing from 75%->75.1% every 0.18mm like images do so it is running at the true 800mm/min except for the left and right edges where it accelerates and decelerates, which is why your edges are blackened on the rectangles.
I’d recommend not using grayscale mode unless you want to run custom firmware that enables GRBL-esque inline laser power control.
1 Like
Brent,
Yes, I am using “Grayscale” in Lightburn.
Your response makes perfect sense. I have gone on to read many of your other posts on the subject and agree with your conclusions.
I can’t wait for SM to include the Marlin in-line power adj. methods in their firmware.
For other’s reading this, Here’s another sample of my tests.
The top one I bumped the speed from 800 to now 1000mm/min and set power up to 100% instead of 85%.
It was the worst yet! It makes sense in light of what Brent said. Since it is stopping the laserhead motion with each power change, it is now at full 100% power at each new start up (speed 0 through 1000) of each pixel.
The second image down is a slight workaround,
I took the image into gimp, and indexed it into 10 separate colors (shades of grey). Then place each shade on it’s own layer, and exported each layer into lightburn. When they are stacked in Lightburn you have the full image.
I then adjusted the power level of each layer to my best guess.
So in essence, the laser head would go over the whole image 10 times. Each time at a specific power and speed, to avoid stopping due to power level changes.
It almost works…
The full 85% power black is great! it is not punching through and solid shading.
Lesser power areas where the Laser head has to turn on and off more frequently, I can see light through it’s little holes. … Not sure what is causing that since the power should remain the same.
I’m going to try to burn just that layer and see what’s up. My guess, is there is some slight overlap of two layers.
Thanks again Brent for this reply and your other posts.
1 Like
Just as a fun side note, I do have this working in my custom firmware, and I run at 0.1mm steps. However I’ve found that the touchscreen cannot feed gcode fast enough to the machine, so it has to be run directly from Lightburn over USB. I think Luban over USB also works.
Except for also there’s a bug somewhere that I’m currently trying to chase down that results in the Lightburn playback breaking randomly throughout a job.
If I can get all of that fixed I’ll be submitting a PR to the repository.
The best I can get now is with a slight jitter with playback if running from the touchscreen. But it mostly works.
2 Likes
