Inconsistent Engraving Quality Issues with Boss LS-3655

brandong · January 18, 2024, 10:15pm

Running out of ideas with this one and I hope you all could offer some advice. My colleague and I have been testing, adjusting, and testing more for months now and still aren’t able to produce a consistent quality. Trouble is, we don’t even know where to start looking…

We’re using a Boss LS-3655 with 105W tube.

Our issue has been inconsistent quality across a sheet of legend plates on each batch. Some would come out excellent like this:

But others would be out of focus to varying degrees like this:

The frustrating part is the error is not consistent. We’re cutting a 12x24" sheet each time, so 96 plates per job. Some will be perfect on the sheet, others not. I estimate it’s a 75/25 split between good and bad respectively.

After looking at the fuzziness closely, we thought it might be a tracking issue (for lack of a better term) as if the laser was not firing at the correct location when the head would move from left to right or right to left. So we set lightburn to not use bi-directional fill and we got a single full sheet of perfect text.

But the next day, using the same settings, we’re back to our longstanding issue. We are the only ones using the machine so nothing was changed overnight.

Boss has not been of much help because after spending hours on support calls with them and aligning, shimming, realigning, cleaning, and you guessed it, more aligning…they pronounced our fuzzy results as “looks good to us”. And compared to the disastrous results we were getting at first setup it was an improvement. However our old laser, an Epilog Helix, has been producing excellent legend plates for 10 years or more without issue.

And this is only one of a few issues we’re having, none of which we can consistently replicate such as:

Some legend plates in a sheet not cutting out completely
Artifacts in the engraving (random dots, dashes, and lines appearing in places there shouldn’t be)
Having to put a pipe character ( | ) for engraving on the far left in the scrap area otherwise the far left column text will be skewed

Here are our lightburn settings for engraving:

I’ve also attached our file:
22LP1-4530C-311_QTR.lbrn2 (939.4 KB)

Do we have our lightburn settings wrong? Are we having a software error? Or maybe a controller error? Mechanical error?

Thanks

gilaraujo · January 18, 2024, 10:23pm

bed location, are the 25% bad in a specific zone of your machine bed or they are also random all over?

Colin · January 18, 2024, 10:27pm

Im inclined to agree - do you have gauge blocks or even just a ruler to check that the focus distance does not change across the area of the bed?

Personally, I’ve experienced this particularly in the center of my 1300x900 bed machine when using vacuum holddown for thin materials. I installed a 1.5" x 1" aluminum bar to prevent bowing.

brandong · January 18, 2024, 10:27pm

Seemingly random unfortunately. We tried positioning the sheet at the top right of the bed and at the bottom left and got bad in both. At first we thought it was limited to the center of the sheet, as if the plastic was warped, but the burn this afternoon had the bad clustered around the top left corner of the sheet.

gilaraujo · January 18, 2024, 10:29pm

this is what puzzled me
because initially it defines that problem to be mechanical, so belts tension on the x

but if then it started apearing again without bi directional it shows something a bit more severe

Still inclined to think it is to do with the axis - a deep clean of the rail and a deep inspection of the belt seems in order

Maybe even test the motor by hand when unplugged to see if it feels “sandy” gritty

brandong · January 18, 2024, 10:30pm

We used some shim stock to check focus at five points across the bed, each corner and the center; it’s pretty flat. But I’ll confirm that with my colleague in the morning when he’s back. On his exact procedure for checking that.

Colin · January 18, 2024, 10:32pm

That’s a good step. Try doing it with the material on the bed. Cast acrylic can vary much, and when you add a lamination of another material I would suspect the same for any of the two material films like you’re using here - make sure that sheet is flat and consistent in thickness. Let us know how that goes!

brandong · January 23, 2024, 3:45pm

Thanks for the suggestions everyone.

We checked the belt and stepper motor; they seem to be fine. However when we were cleaning the rail we could feel bumps along the track where screws have been countersunk and then covered with a plastic cap. If you push the laser head by hand, you can feel a small resistance as it goes over each cap.

So we’ve sanded down each cap the best we could then cleaned and re-greased the rail. You can still feel it if you push the laser head by hand, but if you just rest your hand on it while letting the stepper motor pull it along, you don’t feel anything.

Our next run of plates turned out better but there’s still some inconsistency between them on the same sheet. However, this may just be the best that can be managed without creating some form of clamping system for the plastic or modifying the deck with a vacuum mount as suggested by Colin. Because he’s correct, there is some slight variation in the plastic.

But we’re trying to make this machine as easy to use for our team as the Epilog was. So, being able to just set the plastic in and run a job is our goal. But this may be something we’ll just have to live with even though the Epilog seemed to handle it just fine.

Here’s the latest results. The bottom is our ideal but the top is very close: