Week 3, CNC Cutting

Objectives


Group assignment: Do your lab’s safety training; Characterize your lasercutter’s focus, power, speed, rate, kerf, joint clearance and types.; Document your work to the group work page and reflect on your individual page what you learned.
Individual assignments: Design, lasercut, and document a parametric construction kit, accounting for the lasercutter kerf, which can be assembled in multiple ways.; Cut something on the vinyl cutter.

Group

Safety Training

David and I reviewed the policies and procedures for our lab and for the laser cutter specifically. We were breifed on the safety systems on the machine, how to turn it on, how to send a job, the importance of the exhaust fan, the location of the fire blanket near the machine, and shown the location of the nearest fire extinguisher.

Laser Characterization

We created a small part to characterize the laser kerf. I drew a small comb in SolidWorks that used a parametric design to take the material thickness as an input and then decrement the thickness of the comb gaps by a set distance.

Then we ran a couple of the parts on a piece of wood and checked the fit. It turned out that the gap that was 0.4mm smaller than the nominal thickness produced a nice snug fit that was still easy to assemble. This is what I used to take forward into my kit project.

Then we did a sweep of laser settings for both rastering and vector cutting. The Epilog software has a nice feature that lets you create a sweep of settings in one laser job.

Setting up a sweep of different speed and power settings for the laser

We started with the focus. We set it up for a sweep of power of 1 to 100% and speed of 1-100%. As we ran this we realized that these settings were too slow and powerful and we were burning the wood.

The first power and speed sweep with too much burning.

We adjusted the settings and did it again. This time we limited the sweep to 25-100% speed and 20-100% power. Each row has the same power and each column has the same speed. This range worked alot better and we were able to see the depth and darkness of the raster of each setting. Since it is advantageous to run at as high of speed as possible, the last column used 100% speed, and the 8th one down had 75% power which looked good for general purpose engraving.

The second power and speed sweep that showed a good range of etching depths and hues.

Then we did a sweep for cutting. We used a range of 5-100% speed and 1-100% power. We knew many of these settings would not cut through, but wanted to see what they looked like anyway. This proved to be a good test as about a third of the settings made through cuts. Columns are the same speed and rows are the same power.

What is more illuminating is to look at the back of the piece. Here we can see which ones got through effectively, which were close and which did not even show through. I flipped the image so that it lines up with image above. The third column is 26.1% speed which is the fastest it can go and still cut 5mm wood. The 7th row is 67% power, so it is possible to cut with 67-100% power.

Results of the cut sweep from the bottom of the piece.

Vinyl Cut Lens Hood Sticker

For the vinyl project I decided to make a decorative sticker for the lens hood of my camera. This allowed me to experiment with the workflow from SolidWorks to our Roland vinyl cutter.

Cutter Setup

The first step was to characterize the vinyl cutter to dial in the settings. The lab at UNCC has a Roland Versastudion GS2-24 machine that has a 24" maximum sheet size.

The vinyl cutter at the UNCC Super Fab Lab

Roll Installation

I used a roll of white vinyl. I laid the roll onto the rollers and took the leading edge under the machine and fed it under the machine and around the back. Then, with the clutch lever disengaged, I fed the material under the rollers and down to the cut groove. Then I slid the left roller as far to the left as I could but still be under the white line. Then I slid the right side roller to the far right of the machine but inside of the white hash mark. The rollers need to be in these areas or the machine will not let you cut. Then I pulled the clutch lever up to drop the rollers onto the sheet to grip it.

Left side tension wheel with the right side far away on the right

Test Cut

Then it was time to see what a good setting for the cutter is. I started by setting the cutting force to 90gf and the Pen Force slider at zero. Then I moved the carriage to the left side of the cutting area and pressed and held the 'Test' button. The machine woudld then run a circle with a cross inside of it as a test cut.

The first cut worked but the knife dragged a little bit, so I wend to -1 pen force and ran again. It still went through well so I went to -2. It was still cutting well, so I went into the menu and adjusted the cut force to 70 and zero pen force. This seemed to be a little bit light so I adjusted it back up to 80gf. This seemed to be a good setting to cut the material without tearing so I used that for the rest of the cuts.

Test cuts showing the different pen force settings for the cutter

Process

Creating the Flat pattern

I started by creating a flattened surface in SolidWorks. I had a surface model of the hood already. So I used the sheet metal tools to create a simple lofted "sheet metal" part that matched the surface. From there I was able to unfold the flat pattern and extract the perimetter.

Adding Text

I then brought a dxf file of the flat pattern into Illustrator. I added text and aligned it with a guide curve to make it centered on the art.

Cutting

I exported the .ai file and brough the geometry into Roland Cut Studio. After many attempts, I realized I had to downsave the file to Illustrator version 8 to get the geometry to load, but it was no problem after that. Then I fed a roll of white vinyl into the machine, setup the default parameters with 0 adjustment on the cutter and tried it. It ran fine and I was able to test the fit on the lens hood, which was quite good.

I then moved on to cut in a reflective piece of vinyl. I loaded the same program and ran the cut profile. I then weeded the letters and put some transfer tape over the artwork. I took the tranfer tape and vinyl pattern and carefully wrapped the lens hood to achieve the final product.

Files

Laser Cut Bokeh Filter Kit

For the parametric instruction kit I made a bokeh filter kit for my DSLR camera. Bokeh are the blurry out of focus areas in a photo that is shot with a narrow depth of field and help to make the subject stand out. Typically, point lights will render as circular or polygonal as they have to pass through the shape of the aperture. However, by adding a different shaped filter in front of the lens, new bokeh shapes can be realized.

Iterations

My camera currently has a fixed focal length 50mm lens that I like to use for portraits and general shooting. It also has a wide aperture setting that is perfect for creating good bokeh. So I decided to build off of this lens. I ended up doing 4 major iterations.

First Iteration

I started by adding a bulkhead to the back of the lens with a simple filter at the front with upper and lower cross braces. I added slits to the rear bulkhead in an attempt to help it conform perfectly to the lens. For the filter I chose a heart shape for testing.

The slits were sketched into the part in a pattern that I thought would work. I did not use any generative software to make them.

There were a number of problems with the first design: Stretch was too fragile Bokeh hole was too big and was not showing the shape Thickness was wrong in some places

Second Iteration

I updated my CAD file to have fewere holes in the stretching section. I also lowered the diameter of the heart shape.

This change was better but still did not work perfectly. The filter hole was a good size and the heart shape started to show in the LEDs I was taking pictures of. However, I was still breaking the rear bulkhead. After measuring it I found that the diameter of the lens was wrong in my model.

Third Iteration

For this iteration I updated the diameter of the rear bulkhead and I made the interior point of the heart in the filter more pronounced. I also updated the geometery of the cross braces to be taller.

Despite the update to the rear bulkhead, it was still breaking and I decided to abandon this idea going forward. I also realized that even though the lens was fixed focal length, there was still about 8mm of telescoping that the lens did to focus the image. So, I realized that the filter piece needed to be able to move axially with the lens.

Fourth Iteration

I completely overhauled the design. I started by going to Home Depot and getting some new wood. The new wood was flatter that what I had been using and was 5mm thick instead of 4mm thick. Then I changed all of the CAD. I decided to keep the rear bulkhead, but to add another bulkhead to the front of the lens just in front of rotating focus ring. This would provide aligned slots that would locate the filter and allow it to slide.

I ran the first test and the parts fit together pretty well, though I did have to fine tune the diameter of the rear bulkhead to get a snug fit. The filter slid well in the 3 slots and the focus motor was able to drive the lens without any noticiable resistance.

I felt like I wanted the filter to stay on the front of the lens, so I attempted to add some wooden spring elements to help. However, after 3 interations they were too fragile and I decided to abandon the idea and leave the solid legs.

I made a few more small changes to the design. I added some cutouts to the rear bulkhead to allow it to stretch onto the lens. I also update the filter with longer legs and added the Fab Lab logo to see if it would work.

The updated filter worked really well and I took a few sample images to show the effect.

CAD Files

As per the assignment, the CAD file was built parametrically in SolidWorks. Global variables were setup for the thickness of the material and for the kerf allowance. In the sketches of the different parts, equations were used to define the gaps so that the pieces would fit together properly. This allowed the CAD to be rebuilt correctly when I updated the material thickness partway through the build.

CAD file for the bokeh filter

Screen shot of one of my layout sketches showing the global variables and the equation driven dimensions.

In order to get the individual pieces of the multi-bodied part into 2D shapes, I saved each body into its own seperate part. Then I added them to a SolidWorks drawing file chosing the correct view required to cut the shape properly. All of the shapes were pulled togher into a single cut. Then I added different color numbers to identify the iteration. In the laser setup I had these run really fast so that they made a mark but did not cut.

Layout of the 2D parts for the laser cutter.

Here is the dxf of the cut file

Cut settings were 20% speed and 100% power for the cuts and 100% speed and 30% power for the number tracing.

Parameterization

The design is parametric as I defined global variables for the design and then used equations in my CAD to define key areas that would be affected by the variables.

The variables I used were material thickness, the press fit amount, and the slip fit amount. For this model I used a thickness of 5mm, a press fit of .7mm and a slip fit of .1mm. So all of the places where the wood parts came together would use equations like thickness - press fit to yeild a press fit when laser cut. If I were to change the global variables the CAD would update accordingly.

I did not parameterize every dimension so it is likely the model will break if the thickess gets too far away from 5mm.