For this month’s project, there’s no new finished piece to show off, nor any fancy engraving technique to experiment with. Instead, we’re going to replace a very small—but very important—broken part that was causing all sorts of engraving problems.

A couple of months ago I noticed that the laser’s engraving precision decreased dramatically. Some of my dialog boxes were slightly misshapen, and an ongoing project had some obvious vector engraving issues that were only popping up intermittently. It seemed like certain pieces that were processed on the rightmost edge of the laser bed would come out with lines out of place, or serious banding in the raster engraving, or both. After some engraving tests, I figured out that the affected area was only the last eight or so inches horizontally. I had several ongoing projects and wasn’t about to send products with engraving errors to my clients so I just stopped using that side of the machine. I had to eat some cost efficiency, but there was simply too much going on to figure out what was wrong in the middle of a busier season.

Still, in the month that followed I discovered by accident that my X-axis Arm was more “flexible” than usual. By that I mean that I was able to slide it up and down the left Y-axis Rail without it moving on the opposite rail. In normal operation, the X-axis Arm stays perpendicular to the Y-axis Rails, so being able to manually rotate it in this way was a hint that led me to the source of the problem.

At the bottom of the right Y-axis belt is a drive gear connecting the belt to the Y-axis Motor. When I was inspecting that area, I discovered that the clamp that held that drive gear tightly to the Y Shaft was broken. It’s a positively tiny piece, smaller than my thumb, but because that clamp gave up the ghost, nothing could be engraved accurately in the right third of the laser bed.

I could have made a service call. After all, this was easily the most complicated repair work I’ve done on any of the laser engravers I’ve worked with. But I was confident that such a small piece would be an easy replacement. It’s also much more economical doing your own repair, provided you’ve got detailed instructions and know how to follow them. Aleks from Scientific Spectrum Midwest provided the replacement part, and I followed the VLS Service Manual (PDF) available through Universal Laser Systems.

The relevant section of the service manual, Y Axis Drive Gears (pg. 58) has only ten steps. Surely I can handle that! Unfortunately, the second step was a doozy: Remove the X-Axis Arm. That process has its own section in the manual, and it has nine steps of its own! It was a little daunting because I knew that removing the X-Axis Arm meant that I’d have plenty of recalibration work to do when it was reinstalled. But there was no other way to remove the belt that was in the way of removing the drive gear.

Thankfully the service manual is detailed and thorough. In fact, it might be too thorough. As part of the process I was to completely disassemble the Y-Axis Motor from the Y Shafts, uncouple it from the chassis and completely remove it from the machine. At the time I was sure there was a reason for this, but during the reassembly phase I came to the realization that the motor could have stayed in place. Maybe the idea was that the motor’s presence would prevent the movement of the Y Shaft in order to install the new drive gear, but there was enough flexibility once the Y-Couplers were removed that it wasn’t necessary. Removing and reinstalling the motor was only a minor inconvenience, though, and didn’t result in any additional calibration.

For all the trouble I thought the X-Axis Arm would give me, it turned out to be a fairly simple removal. Four screws to remove beneath two protective covers and the arm’s ready to pull off of the belt clamps. It’s designed with a spring-loaded arm holding the bearings on one side, so once those screws are removed you just need to give it a little tug to the left and the whole arm lifts up for removal. (Make sure to disconnect air assist tubes if you’ve got ’em!)

Once the X-Axis Arm was removed, the belt clamps that held it to the Y-Axis belts had to be removed too. There are two belt clamps (one for each Y-Axis belt) with three screws each. These are pretty small screws and washers, and I made the mistake of loosening all three before removing any of them. When I finished loosening the third screw, the action of pulling the hex driver out of the screw jostled the belt and knocked both other screws off—into the honeycomb downdraft table I had left installed for no really good reason.

It felt a little like a disaster: that honeycomb table has hundreds of tiny pieces of plastic and wood that have fallen into it over the last several months, and fishing the screws out of that was surely going to be a nightmare. It turns out that wasn’t quite the case: once I pulled out the table and removed the honeycomb, the screws that went flying were still resting on top of the pile of debris. Lucky! In hindsight, if the downdraft table wasn’t installed, those screws might have fallen into a corner of the machine itself, and may have been harder to find. Either way, solve the problem at the beginning by removing all lose screws the moment you loosen them. Don’t let them be jostled away!

Replacing the Y-Axis Drive Gear and its clamp was easy once all of the other parts were out of the way. The biggest issue during reassembly was making sure I re-clamped the belts with exactly the same tension as before. The instructions call for a spring scale to measure tension when you reassemble, but I did not have one on-hand, so I marked which belt teeth that the belt clamps were holding during disassembly. I put them right back where they were. There’s a bigger lesson here about making sure you have all of the required tools before you start a repair project, but my workaround seems to have done the trick for now: all of the engraving tests run after everything was reassembled turned out great!

I made an exact replica of the piece that turned out poorly back when I first noticed the engraving issues. This piece was engraved in the “problem area” of the bed, and used the same settings as the previous piece. The resulting engraving, both raster and vector, were free of visible quality issues. No banding in the raster fills, precise position for all of the vector engraving. You almost can’t even tell there’s a vector line surrounding the AGENT NAME in the picture above, which is how it should be!

It remains to be seen whether my lack of a spring scale will come back to bite my tail in the future. Bad tension on the belts can cause issues that aren’t immediately visible in the engraving quality, but I’m confident enough in how the calibration was done originally that I’m not worrying about it… for now. In the meantime, I’m back to using the whole laser bed to make projects!

PROJECT GALLERY