Week 07 - Computer-Controlled Machining

Group Assignment:

• Complete your lab’s safety training
• Test runout, alignment, fixturing, speeds, feeds, materials and toolpaths for your machine
• Document your work to the group work page and reflect on your individual page what you learned

Please refer to the group page to read about our group assignment.

Individual Assignment:

• Make (design+mill+assemble) something big

Learnings from this week

This week I learned mainly three important lessons.

1. It is important to check the alignment and calibration of the milling machine. Correction of the axes can be easily done and will greatly improve the fitment of parts.

2. I got a feeling on flexures and how certain flexure pattern dimensions will create a more or less flexible flexure.

3. Plywood is an anisotropic material and therefore the milling direction and orientation of the parts will create a different finish, depending on if the cut latitudinal or longitudinal oriented to the grain of the top ply.

Design

For this week I hade some ambitious ideas making a balance stool or a hanging chair, but eventually I ended up designing a more simple, but still quite complex lounge chair with flexure elements and without any glue or fasteners.

First I searched for a picture of a design similar to what I had in mind.

Then I added a side view of this picture to my Fusion file and made a rough sketch.

Because the design contains bent sheets but with large bending radii, I made a surface design and used the Thicken and Convert to Sheetmetal functions to get a flat pattern.

The other parts are just simple flat parts.

After a while I had finished the basic design and joints.

Then I unbent the main part and added the flexure slots and dogbones. This was a lot of manual drawing work, to make everything fit an evenly distributed.

Tip

I used this dogbone addin for Fusion, that makes life much easier.

Here the chair is basically ready for CAM.

Preparation

Because I wanted to try the nesting software Deepnest, I exported all faces as *.dxf.

However, Deepnest didn't accept the main part for import and kept freezing or crashing. All other parts could be important, but there wasn't any point in runnig the nesting without the main part included.

After playing around with the file trying to simplify it and find any abnormal objects in it (using LibreCAD and Inkscape) I decided to not waste more time on this.

Instead my instructor suggested using the nesting function in V-Carve, which I did.

I adjusted the parts manually to make sure the direction of the grain is in the same direction for similar or mirrored parts.

I started adding tooltpaths and when I added the contour, I manually placed tabs to hold the pieces in the stock. I tried to locate the tabs in areas where it would be easy to grind them away later.

Here is the complete layout with all toolpaths. It is hard to see, but I added a drill operation, right after engraving the logo. This would allow me to fix the plate with more screws in the right location.

Fabrication

Once the spindle was warmed up and the plywood sheed was fixed on the bed, I ran the X/Y zero command using the end switches. From there I moved the spindle to the X/Y location I intended to use as the origin and wrote down the coordinates, before zeroing X and Y. In this way I can find and set the origin again, even though the power supply would get shut off or something else would go wrong.

Here are the tools I used.

• 1/4'' endmill, upcut, two flutes
• 1/2'' 60° V-bit, two flutes

settings

I used the speed/feed settings we determined by testing during the group assignment:

- Cutting depth 6 mm
- Speed 2000 mm/min
- Spindle 14000 rpm

The zeroing of the z-axis was done with the aluminium plate.

Then I moved the spindle to different locations on the sheet and verified, that there is not major difference in the height of the bed.

Once everything was ready, I started the first milling operation. It was engraving the text and marking for screws to firmly fix down the sheet.

With the sheet propery fixed down, I started the pocket milling.

Note

Yes, I know that there should be a dust collector shoe, but unfortunately it was broken, so I had to vaccum the chips by hand.

The last operation was cutting the contours.

The milling time all in all was about 2 hours.

Finishing

I broke the tabs using a chisel and hammer and removed all the access material. Here are the raw parts together with some tools I used to round the edges and sand the parts.

Of course that took about as much time as the milling itself.

Assembly

After cleaning the area it was time for assembling the parts. It turned out, that the fit was nice and tight. Using a rubber mallet and some clamps helped to assemble the chair without breaking anything.

... and here it is:

Will it be strong enough?

Yes, it is actually quite sturdy and not wobbly at all. I'm very pleased with the result.

As always, the design files are in the repository. This time the Fusion file was 40+MB, so it isn't included. You can watch the model using this link.

After sanding some small details again I charred the plywood and varnished the chair.

Update 16.3.2025

Because there was some space left on the sheet and the chair turned out to be really comfortable, I decided to make a rest for the feet.

Here is the updated design in Fusion.

And here is the updated v-carve sheet with the improved nesting to make everything fit.

The result looks good and is even more comfortable.

Please feel free to get the files from my repository and make your own chair.

Note

The current dxf and v-carve files are saved for 11,7 mm sheet thickness and 0,05 mm joint clearance.

An idea for the future is to add a joint test piece to the sheet and set up a parametric milling operation in Fusion, which could be adopted according to the results from the test.

Update 14.4.2025

Some time ago Marius Araújo got in contact with me and asked if he could reporduce my design and make a small version of it, by lasercutting from 3 mm MDF.

The result looked very nice, so I went ahead and scaled the design for our 4 mm MDF and lasercut another miniature chair.

Here is the result:

It required two sheets 600 mm x 300 mm and the total cutting duration was 2 h 45 min.