Computer-Controlled Cutting

The laser cutter at OpenDot is a GCC SpiritGLS CO2 laser cutter. It has a work area of 960x610mm and can comfortably cut acrylic and plywood of up to about 12mm strengths.

The Lasercutter is connected via USB to a dedicated CAM PC installed here at the fablab. The CAM PC runs Rhino 7 on Windows, which is configured to open a template file with the laser cutters work area on a locked layer. I had never used Rhino before, but with Antonio’s instructions it wasn’t too hard to get started. The workflow for cutting a file is something like this:

• use the Import command to import a DXF file from a USB key
• clean up the file contents
  • position everything as wanted in the work area
  • merge segmented paths by selecting everything and invoking the Join command
  • make sure there are no duplicate curves
  • move curves to different layers depending on the process to apply (cut vs engrave)
  • make sure all shape’s Print Color is set to By Layer
• if there are any shapes to fill in, use the Hatch command to do that
• open and set up the Print dialog
  • open the properties window
  • set the preferences for each layer by changing the ‘pen’ settings
    • generally, the workflow is to leave Power at 100% and adjust only the speed
    • for specific cases it can be necessary to adjust the Power and PPI settings as well
  • on the advanced page set the position mode to relative
    • this way the laser can be positioned manually before cutting, and the laser cutter will relative to this positon. This makes it easy to accurately position the laser on material that is already partially used.
  • in the main print window, the origin of the print area can then be adjusted using the window selector. This can be useful to align file contents with the laser.
• turn on the laser cutter
  • wait for it to finish its initialization (bounds check)
  • auto-level by attaching the level probe, moving the laser over the material and pressing the auto-level key
• after double-checking everything in Rhino, pressing print sends the document over to the printer
• after positioning the cut head and turning on the air pump, the hood can be closed and the cut can be started using the laser cutters start/stop button.

To characterize the kerf of the laser and experiment with different settings, we used a ‘kerf test generator’ Sol wrote using SVG.js. You can try the generator directly in the browser on his personal documentation.

We measured the plywood that we were going to cut with calipers, which came to a thickness of 2.9mm. We then used the generator to generate a test pattern with corresponding cutouts and a kerf scale from 0.25 to 0.35mm. We measured a kerf of 0.3mm and found that under-adjusting slots for a kerf of 0.25mm gave a good tight fit for press-fit joints.

We then ran a series of tests in order to characterize the speed and power settings in relation to each other. We used the lab-provided guideline of 6% speed and 100% power at 500 PPI as a starting point, and made a matrix of test cuts adjusting relative to this initial value. The columns have a relative speeds of 80%, 90%, 100%, 110%, and 120% (left to right) wrt. the 6% guideline, whereas the columns have a power setting of 100%, 90% and 80% (top to bottom). Towards the low-power and high-speed corner of the matrix the laser started to struggle with cutting through the whole material, so that the final piece in the 3rd column was already not able to be removed from the surrounding material easily anymore.

We cut the Matrix row by row with the same setup in Rhino: At the beginning we placed the five cuts, one per column, on five layers and gave them five different colors. Then we labelled them on a separate layer for engraving and set up the different pen settings:

To print a column it was now enough to quickly adjust the power of all layers, manually move the cut head down one column and do another series of cuts.

On the back of the tiles (pictured here in the same arrangement as the fronts above), it is clearly visible how excessive power and lack of speed cause charring on the edges of the cuts.

Based on these results we concluded that cutting at 6% speed and 80% power gave the best results out of the settings we tested, but noted that 120% speed and 100% power provided similar results while using the laser’s maximum output power, using the time a bit more efficiently.