Week 3

03 — Computer Controlled Cutting

For the group assignment we were tasked with finding the kerf of the laser for three different materials. In the table below, the measured material thickness are given for the different materials.

Cardboard	4.70mm	Plywood	4.07mm	Acrylic	2.39mm

Figure 1: Measuring the different material thicknesses

The kerf is the full “width” of material removed by the laser beam during cutting. It depends on factors such as the laser type, material, material thickness, power, speed, and focal settings.

Based on previous students work found in the fab academy repository:

https://academany.fabcloud.io/fabacademy/2024/labs/barcelona/students/group-assignment/assignments/week03/

We decided to design a comb to investigate the kerf: the idea is that:

when the laser makes a cut, material is consumed by the laser itself, so when joining two pieces of the same material of the same thickness together it will be loose. Thus a positive offset needs to be added to the joint profile to compensate for material lost during the laser cutting process. We chose a comb design based on the approach Dani’s team documented for their group in 2024 (link to Dani’s project). We incremented our value by +0.05 per slot this later proved to be too conservative.

Andrew explaining how he is implenting the comb design in Autodesk Fusion

Parametric Design Workflow in Fusion

1. In Autodesk Fusion we created several variables we knew we would use across the design.

   1. matthick — our measured material thickness
   2. kerf adjustment— half the value of the kerf as it is applied per side as a face offset command
   3. acrylic — our measured acrylic thickness (for refference when changing matthick)
   4. plywood— our measured plywood thickness (for refference when changing matthick)

   5. cardboard — our measured cardboard thickness (for refference when changing matthick)

      

2. With our variables configured we drew a rectangular sketch profile in a new component [60 x 120 mm] that was then extruded to the value matthick .

   

3. We then place a center rectangle object with it midpoint centered in the middle of the horizontal base of the rectangle. This object was then dimensioned to mathick for width and 60mm for height [30*2]. Shape was then negatively extruded to -matthick.

   

4. We use an array set to features + symetrical and a value of [11] to pattern the cutaways from the center.

   

5. Positive offsets were manually added to each interal edge (except for the first slot) using the offset face command.

   

6. Some labels were added and extruded from a sketch to supply some useful notes on the test.

7. To generate the DXFs another sketch was created with a linked projection updating the matthick variable and naming and exporting the DXFs we needed for each material.

   

Laser Cutting

Positioning the laser and focal point

Figure LC1 below shows the key concepts a laser in a laser cutter:

• the light is concentrated using a convex lens at a focal point.
• the focal point corresponds to the location of maximum power
• this is the optimal location to cut the medium
• A poor focus can lead to a poor cut quality because the maximum energy is situated either before or after the medium

Figure LC1: Laser beam concepts

There are three main laser cutters at IAAC, the Rayjet 400, the Rayjet 500 and the Speedy 100R. Each machine has a different laser and therefore a different laser. They each have a calibration piece, where the user either adjusts the height of the platform using the controls for the smaller laserjets R400 and Speedy 100R or by manual adjusting the tube for the R500 as seen in the Figures LC2 and LC3 below:

Figure LC2: Speedy 100R focal point calibration

Figure LC3: Raserjet 500 (left) and R400 (right) focal point calibration

Workflow and Learnings

Once the machine were basic operation was appropriated it was time to cut some material for the first time and attempt to determine the kerf for one of the machines. The practical workflow is as follows:

1. Generate design DXFs from 3D files
2. Place on IAAC cloud Server
3. On machine terminal download an open DXF in Rhino

4. Process document in Rhino

   Check for duplicate lines. We made this mistake a couple of times so we started using the two reccomended commands in Rhino below. These command insure connected lines and lines that are not layered on top of eachother. These are both things to check to ensure good quality profiles are cut at a uniform kerf.

   • SelDup : select duplicates and delete duplicates
   • select everything and use command join
   • explode to break up a block

   Recolour layers we used Black for Rasters as the first pass, then red for cuts. Rasters moves pixel by pixel in x,y where as the cuts follow the defined vector.

   

   Figure LC4: Josep instructing Andrew on how to use the Rhino7 and Raserjet Software

5. We then click print from Rhino → select workspace window → click print again then go into rayjet control software that will automatically open.

6. Several things need to be configured in Rayjet including moving the design into the cut space, configuring the laser cut settings (power and speed) for different layers you are using (in our case black and red)

   Ideal power and speed ranges for different materials, thickness are given for each machine on a wall panel in the lab. Sometimes the power and speed settings will have to optimised for the specific job which is often due to the cleanliness of the lens or the varying precision of the dimensions given for the material to be cut. The Rayjet 400 laser test settings for the material we are testing are:

   Cutting material:	Cardboard 4mm	Plywood 4 mm	Acrylic 3mm
   Power	20	45	40
   Speed	2.0	1.8	1.8

   Marking material:	Cardboard 4mm	Plywood 4 mm	Acrylic 3mm
   Power	12	7.5	12
   Speed	2.5	12.5	1.8

   Mistakes made: It ended up being good practice to delete superfilous jobs in the rayet software because we loaded and old file at one point. Similarly Job reset needs to be used often when you are cutting the same project with similiar setting. A full screen mode also appears when double clicking objects and the only way to get out is to press escape twice.

7. Press the link machine button

8. Add your material to machine
9. Focal point calibration : adjust z offset on machine using the height offset acrylic piece for that specific machine. Adjust machine x, y offset using the controls on the machine.
10. Return to control software and adjust cut item to the live updated crosshairs in the software that correspond to the machine tool head.

Laser Cutting

The software successfully launched the cutting of the comb measurement piece.

Figure LC5: First comb print with the R400, calibration and cut above, video of the laser cutter in action below

Laser Enmarcha.mp4

R400
	thickness (mm)
Cardboard	4.7	4.65	4.6	4.55	4.5	4.45	4.40	4.35
	too loose	too loose	too loose	too loose	too loose	too loose	too loose	loose
Plywood	4.07	4.02	3.97	3.92	3.87	3.82	3.77	3.72
	too loose	too loose	loose	loose	loose	loose	ok	ok
Acrylic	2.39	2.34	2.29	2.24	2.19	2.14	2.09	2.04
	Rethink	research range

After testing the first two materials on the laserjet R400 we realised we were too conservative with our stepped offset value of .05 per slot and decided to update our Fusion model to +.1mm offset per slot.

Figure LC6: a little too snug for comfort

---

Figure LC7: back to the fusion to adjust the kerf

Kerf results

After another iteration, ideal kerf measurements were noted for the three separate materials

Figure LC8: Kerf measurements

The Rayjet 400 laser cutter power and speed settings are given in the table below. These setting have been slightly optimised compared the laser test settings given previously as slightly more power and less speed was necessary to cut through the plywood more cleanly and for the corners on the cardboard.

Cutting material:	Cardboard 4mm	Plywood 4 mm	Acrylic 3mm
Power	20	50	40
Speed	1.6	1.5	1.8

The final kerf findings for the Rayjet 400 laser are:

	Cardboard	Plywood	Acrylic
thickness (mm)	4,7	4.07	2.39
kerf (mm)	1.1	0.5	0.05

Project Files

03_laser_kerf_test v8.f3d.zip