Laser jumping while cutting or engraving

We are using Lightburn with a Boxford BGL1390 CO2 laser in a Hackspace with multiple users. For a while now we have had issues where the laser will suddenly jump (usually to the right) by a random distance before continuing. This is only affecting a small number of our users, not all files and in random ways, (array of identical circular designs where several have parts offset but the rest cut perfectly is one example, another is a set of identical designs where alternate columns were rotated 180 degrees with the inverted columns having the interior patterns offset by 1mm). In all cases there seems to be no errors in the design on screen. I’ve included a video showing the latest problem which shows that after the error occurs the laser has jumped very precisely to the centre line of the adjacent pattern before continuing.
Lightburn is being kept updated with the problem not affecting any single version number.

Check if this Lightburn doc helps.

I can see some benefits to this process for a single user who is performing cutting but we have 100+ users performing cut, scan engrave and vector engrave processes. Of these we have a small number, perhaps 3-4 regularly getting errors like the one above and a few more who get an occasional glitch. The above user did manage to partially solve the problem by deleting the larger oval and just engrave the smaller after two failed attempts. The only difference in the engrave portion is size(slight resize of design) and that second set is rotated 180 degrees.

This sounds very much like an intermittent connection garbling or discarding chunks of the data stream.

Is the laser connected by USB or Ethernet?

If USB, does the cable connect to a single PC or to every user’s laptop who brings a design to the machine?

Simply replacing the USB (or Ethernet) cable with a new, known-good cable may solve the problem. Spending enough to get ferrite cores on each end may also buy a better cable in between, which IMO is generally what’s needed.

Short jumper cables inside the cabinet connect the controller to the USB / Ethernet jacks on the side and those side jacks can also wear out. If the new cable doesn’t solve the problem, bypass the jumper cable by connecting directly to the controller’s jack.

If the laser uses an Ethernet connection, check the router to be absolutely certain the laser’s fixed IP address is not in the router’s DHCP range. If the router assigns the laser’s IP address to another device, heartache & confusion will ensue when the two data streams collide.

Is the screen you’re looking at the controller’s display, or in LightBurn?

If the errors are appearing on the controller’s display, this is probably a communications issue, as @ednisley has mentioned. In addition to his suggestions, try clearing files off the controller. Ruida controllers have been known to do some strange things if there are too many files on the machine.

It can also help to Send the file to the laser and start it from the controller, or holding shift while pressing Send to transfer the entire project to the laser, and immediately start it after all data has been transferred. When you use Start, the job begins as soon as enough data has been sent to the laser, while more is still streaming. Occasionally, that causes transfer issues.

If the errors don’t appear on the controller’s display or in LightBurn, this is likely an intermittent mechanical issue. See here for more advice on addressing mechanical issues:

No errors show on controller display and all users send the file before starting the run from the control panel. We regularly clear the files from memory as with so many users it can fill quickly at times. The most recent error as shown in the photo was run a second time from the file already on the controller and failed in the same way, deleting the large ovals and sending again gave a perfect run. This does potentially suggest a comms problem though why it affects this user 1 in 3 times(his words) is a mystery. I would expect comms or mechanical issues to affect more people rather than a consistent few.

It’s a communication problem damaging the data in transit.

The Ruida controller apparently has minimal-to-no error checking, so the only way to determine the transfer worked is to run the job and watch it fail.

Tell us about the cable situation: which interface, how many plugging / unplugging cycles, stuff like that.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.