I am trying to make some large laser engravings that need to cut for quite some time. When I step away from the laser I hit the pause button and the machine seems to continue to cut for four or five seconds. When I resume the print it continues from where I hit the pause button. As such it mis-aligns the entire print!
Is there an easy way to resolve this issue?
I am running a standarg GRBL machine off a Mac OS Catalina laptop with LB 1.6.00
Hi Marco,
From your description, it seems the commands are being correctly accepted.
But something in the mechanics is failing on restart
Not sure what machine you have but try running this on a bit of scrap and send results (if not diode change values to something acceptable - but fast - for just engraving)
Dot Test with Angle.lbrn (46.3 KB)
I cannot tell from your image where the stop and start points are. What you describe is what you get if the machine is bumped or shaken while it is paused.
When the machine is paused or stopped, the stepper motors power down. Then they cannot hold a firm position. You can manually move the head, which you should not do, and the laser is now in an unknown location.
If your controller will accept it, try setting $1-255 in the Console window. This will tell the motors to stay energized.
Hi Mikey, I am NOT bumping the stepper motors or the laser cutter at all, the machine is locked down to a heavy mount and the entire print-bed is in an enclosure. I have marked where the offsets are occurring within my image for you here.
Hi! I have run the test. The lines cut out a bit because I was running it as fast as possible but otherwise the alignment and calibration of the motors seems correct. I have re-attached my original images and have indicated where the offsets happen. Once again this ONLY happens when I pause and resume my machine during a long print.
You have any pets in the house?
I fing it really odd the offset is in one axis in the upper part, and the other axis in the lower part. That seems to indicate the possibility that the trolley and gantry cables might be snagging on something. Other than that, I am a loss at this time.
Have you tried setting $1=255 yet?
Hi Mikey,
My machine is in a locked studio room and there is no possibility of pets or anything else bumping it.
These are also TWO cuts on the same material, and in all three examples I have highlighted the image gets offset in the x+ dimension, I have just rotated the material between cuts.
I can also WATCH this offset happen when I hit pause on light burn and can see the laser continue to cut for 5-10 seconds before powering off.
I have not tried the $1=255 trick yet, but this is because I am extremely sure that bumping is not the issue here.
No cats, laser in bunker, images rotated, got it.
$1=255, in the Console or machine settings.
If you can see it happen (shift) when you pause the machine, I am betting the cabling to the laser head is springing it back.
I will absolutely try the console tip you suggested.
The reason why I don’t think that the “shift” is happening from some sort of sprint effect is because the laser head keeps cutting, even switching on and off according to the print after I have hit the pause button. This is why I believe it to be an issue with the memory buffer on the printer. The head does not seem to “spring back” after I hit pause.
Technical:
Lightburn streams the commands to the controller. The controller has a small buffer to catch the commands and store them for processing. When Lightburn sends the stop command, it actually quits sending motion commands, but sends M05 to turn off the laser. That M05 reaches the controller only after the stored motion commands are executed. Only an Emergency Stop button, which kills the power, will give you an immediate stop.
Once the buffered commands are executed, the stepper motors stay powered up for the time delay in the $1 parameter. That parameter is in milliseconds, so $1=254 means power will stay in the steppers for about 1/4 second maximum. After that, holding power is removed and the motors are free to rotate with little effort.
GRBL is written so that $1=255 means Do Not Timeout. So the holding power that keeps the shaft from turning, except under extreme force, is not removed.
Any “jerk back” I was describing, which could be in any direction, would happen after the laser has fully stopped and the beam has shut off.
Be aware the stepper power timeout was created to prevent stepper motor overheating. However, I have not heard of this being a problem. Manufacturers do not push their stepper motor amps to the limit. They want them to last thru the warranty period.
I hope this fixes the disconnect between what you are seeing and what I am thinking.
Hi Mikey, I think this is where the issue is happening. Would there be some way where an error could happen so that the cutter is re-sending the information that is stored in the buffer? The fact that all my offsets are of almost equal distance and always in the same direction as the cutting head is moving makes me to consider this as probable cause.
Not sure what you mean here. Once the buffer is empty, the last issued command will be in effect. That command will be processed until completion. The laser will then stop. As the commands are executed, the contents of the buffer becomes less and less. It is not like it is reading from a list. It is like a marbles in, marbles out kind of thing.
I think you are misusing the term offset. Offset refers to a position correction (X-Y) to the path, not an extension of it.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
