Weird laser cut behavior

[dfasdf (json jj)]

Here is the SVG source file and the actual rendering, what exactly caused this result

[dfasdf (json jj)]

[berainlb (PY)]

There are a few things that aren’t making sense to me about the pictures. Can you share the .lbrn file for the two images? Will help provide some context.

[dfasdf (json jj)]

I’ve carved part of it. I had this fault on my computer with version 1.2.04, version 1.1.02 does not have this issue. This is the file saved in 1.2.04
1.lbrn2 (70.5 KB)

[dfasdf (json jj)]

There is no problem if i Open the lbrn file with 1.1

[berainlb (PY)]

You’re saying you only get these issues with 1.2.04 but not with 1.1.02? That’s odd. Possibly a regression with the Marlin device profile?

Can you generate gcode from 1.1 and post here?

I looked at the design and saw that you only had line operations. I don’t understand why you’d get fully burned objects in that case.

But from what you’re saying I suspect a change in the gcode being generated.

[dfasdf (json jj)]

version1.1.gcode.lbrn2 (13.3 KB)
can not upload gcode ,youcan rename it

[berainlb (PY)]

I suspect you generated the gcode from a different file than the .lbrn file that you uploaded so makes direct comparison difficult.

But from a general comparison it seems to me the major difference seems to be that feedrate for rapid operations may be specified differently.

Example:
From 1.1:
G0 X4.907 Y9.219 F4000

From 1.2.04:
G0 X15.126 Y2.767 F0

Also, can you confirm that you have set S Value Max to 100 in Device Settings? I had assumed this was 255 generally for Marlin machines but that may be incorrect.

[dfasdf (json jj)]

here is my settings

[berainlb (PY)]

I can see that rapid feedrate is taken from the “fast whitespace scan” speed of 4000 mm/min in your Device Settings.

In that case I don’t see any other major difference in the gcode. Can you reinstall 1.2.04 and generate gcode from there? Maybe I’m not getting what you’re getting.

[dfasdf (json jj)]

1.2.gcode.lbrn2 (399 Bytes)
i just dram a retangle then 1.2 generate this gcode ，and Engraved some completely different figures

[berainlb (PY)]

It’s more useful if you can generate the same gcode for both versions so that they can be directly compared. Otherwise it’s just speculation as to what the cause is.

Can you generate gcode for the same .lbrn file in 1.1?

At minimum it looks like you’re using a START and END user script in 1.2?

;USER START SCRIPT
M3 S0
G21
G90
G1 Z15 F1500
;

;USER START SCRIPT

;USER END SCRIPT
M3 S0
;USER END SCRIPT