7. Make Something Big

Goals

Group assignment:

[x] do your lab’s safety training
[x] test runout, alignment, fixturing, speeds, feeds, materials, and toolpaths for your machine Individual assignment
[x] Make (design+mill+assemble) something big (~meter-scale)
[ ] Extra credit: don’t use fasteners or glue
[x] Extra credit: include curved surfaces

Tools Used

Concepts - My go-to sketching app on the ipad
Fusion 360 - 3D CAD/CAM software from autodesk
ShopBot PRSAlpha 96 48 - CNC router primary for woods, plastics, and metal

TLDR; Nice images

Group Project

For our group project test we created a test piece that would look at the effect of how more and less aggressive strategies affect the quality of the cut.

I have the results of the testing is below but the key takeaways are:

On a gantry style machine reducing tool pressure for finishing passes greatly improves the accuracy of the cut.
To improve edge quality in wood, finishing should always be done with a conventional cut direction .

CMT Orange included this very helpful guide in one of there brochures. Link

Link to full documentation here

Results of testing

9.5mm Ply Test	1 Pass	2 Pass	3 Pass
Chipload	0.2	0.28	0.28
Feed Rate	4800 mm/min	6750 mm/min	6750 mm/min
RPM	12000	12000	12000
Depth Of Cut	10	6	6
Direction of Cut	Climb	Climb	Climb + Conventional Finish
Nominal Ext X and Y	100	100	100
External X	100.75	100.55	100.2
External Y	100.88	100.55	100.2
Nominal Int X an Y	50	50	50
Internal X	49.65	49.7	49.9
Internal Y	49.3	49.3	49.9
Surface Finish	Poor	Poor	Good

Summary

The feeds, speeds and strategies that were used in the group project translated directly into my individual project. I actually added an additional “spring pass” to the design to improve the overall accuracy of the cuts.

Individual Project

1. A Furniture Designers First Chair

For this week I decided to design a chair. My background is in furniture design but to date I have never designed a chair so I thought that this week would be a good opportunity.

1.0 Design

For my inspiration I took 3 main sources, the clasic Alvar Alto Stool, The high back Charles Rennie Mackintosh Chair, and the Tuam Chair. The later was a subject of a project of a friend of mine from ATU Connemara. Our Irish Chair - ATU Connemara project I developed the form of the design through sketching in Concepts. This was the result of the ideation. I took care to ensure that I was not creating features that could not be manufactured with 3 axis CNC milling. The design is based around 24mm birch plywood.

1.1 CAD

In Fusion I opened a blank file and created a new component. I created the seat pan using a extruded circle and then created a sketch on the surface for the holes for the legs. I used a circular pattern to create the holes for the other legs. I used the offset face command to widen the hole for the back leg. This is driven by the material thickness parameter. A new component is created and the inside surface of the hole is used to sketch in the leg. Which is then extruded as a new body. The same process is used for the large back leg. Which is dimensioned And extruded. At this stage I made a liner pattern to see the full shape so I can asses the overall design.

1.1.1 3D Printing Break

I like to make scale models as I go so even though I am not finished my model once I have something close I exported a STL file and brought it into Orca slicer.

This was printed at 1:10 scale.

Now back to the design…

1.1 Part 2 Return of CAD

I need to add dogbones to inside radius of all the joints to compensate for the radius left by the 6mm endmill. I used an addon is called Dogbone can be found here I am going to use dowels to join the two halves of the legs together so to do this I created a sketch of points on the surface of the leg and then made a hole feature at each point. Then the whole leg can be mirrored to create the other half. I fixed up some the transitions on the back leg using blend curves I created a new componant for the middle section of the back leg, as this one will have dowels on both sides I will use a through hole feature. This is actually very easy as all I need to do is project the face of the outside leg adn extrude it to create the new inside leg. Both the inside and outside of the back leg are now mirrored to create a 4 layer wide leg. I made a new component called “sheet” and I drew the size of the plywood I will be using. I then set it as the envelope in fusions Arrange feature. And arranged all my parts on the sheet. Note that this is a starting point but you can manually move parts around to make things more efficient.

1.2 CAM

In Fusions Manufacturing workspace I created a new Setup and setup my “WCS” and stock size. For my Work Coordinate System (WCS) I want the x axis to be aligned with the long side of the sheet and my z zero to be aligned with the top of the sheet. My first toolpath is a set of drill for all my 8mm dowel holes. I make use of select same diameter and select same depth to select all the dowel holes and then use the drill tool to create the toolpath. Next was the pocket toolpath for the seat pan Next was the pocket toolpath for the seat pan. I split this into 3 separate toolpaths.

Rough - 7mm depth of cut, 50% stepover, feedrate 6720mm/min, RPM of 12000, 0.5mm stock to leave, Climb Cutting
Finish - Full depth of cut, 50% stepover, feedrate 6720mm/min, RPM of 12000, No stock to leave, Conventional Cutting
Spring pass - As above. I used the 2D Contour to cut out the pieces, a key setting for this toolpath is setting tabs and there placement. One of my things that I would change if I was to cut this part again is the number of tabs I used. I use the same Rough, Finish and Spring pass settings as I did for the seat pan. Just to note of the key things that has helped me in making furniture on the CNC is roughing in Climb direction and finishing in Conventional direction. Im actually not sure as to why it works well but I have found that it creates a smoother finish that requires less sanding. I ran a final Simulation to check for any issues with the toolpaths.

1.3 Control

After creating the toolpaths the next stage in the process is to post process them into a format that the CNC can understand. I used Shopbots OpenSBP post that is included in Fusion. I just make sure to uncheck the Minimize Feed Rate button. This is what the final post processed toolpaths look like. I created additional test drilling toolpaths to confirm the depths. A note on naming conventions: Use consistent naming conventions for your toolpaths to ensure clarity and ease of use. In production environments I have found that adding the oder of operations in the name is very useful, then adding a descriptive name such as `Pocket Roughing” and then tooling information to the end. This looks something like 08-Pocket Spring-6mm comp.sbp

1.4 CNC

I confirmed the thickness of the material and updated the design to 23.8mm And turned on the dust extraction. I removed the tool in the machine and inserted a 2 flute 8mm dowel drill. After using the z zeroing procedure is confirmed the z height was accurate byt setting the z height to 18mm And using a scrap piece of wood to check. I ran the program in the shopbot console. It is very important to realize that the PC cannot turn the spindle on, so you need to manually turn it on before running the program Turn on the spindle manually before running the program. Confirmed that the blind hole was accurate. The through hole was actually not deep enough to reach the bottom of the material. So the program was updated and I used a caliper to confirm that it was getting all the way through. Ran all the Drill toolpaths. And swaped of the tool for a 6mm compression endmill Used the side of the material to ensure I had minimized the sickout of the tool, this reduces vibrations and improves accuracy. Shopbot includes a aluminum plate that can be used to zero the z-axis using a macro called C2 I once again confirmed the accuracy of the through hole using a piece of scap wood. Another helpful feature from shopbot is these messages to remind you about the tool change. Between roughing and finishing toolpaths, I used a caliper to confirm that my dimensions are as expected. We continue with the rest of the programs. Until all the parts are cut out. However!!

After removing the first part I noticed that the bottom edge was terrible quality So I inspected the tool and noticed that the the bottom edge was damaged. After inspecting my cuts I found something that looked like debris near one of the tabs. Using some tweezers I was able to remove the a piece of a twist drill that was lodged the the bed. I replaced the damaged tool,re-zeroed the machine and re-ran the spring pass

1.5 Assembly

I used a multi-tool to cut the tabs. I personally like to cut the tabs in the middle when removing the pieces from the machine. I then used a sharp chisel to flush up all the tabs. Sandpaper and a piece of scrap as a sanding block helped clean up the edges that were affected by the broken tool. First dry fit was successful! Though I noted that the joints had a very tight fit, this will need to be sanded down later. Glue and dowels were used to secure the pieces together. And clamps holding the pieces together. You can never have too many clamps. In this case I had to use our big sash clamps as I ran out of smaller clamps. All the parts displayed. Before I assembled the pieces together, I tried to do as much sanding and cleaning up of glue as possible. Glue was applied to the joints. And the seat pan And more clamps for another 24 hours! A shot of the piece next to its 3D printed model!

1.6 The Sit test!

The clamps are off and the seat is able to sustain my ample weight! Im very happy with the spring in the chair thanks to the long flexure of the back leg. Hero shot of the chair. And the classic small vs far away shot with all the development scale models.

In Summary

I am very happy with how the project turned out! I have always wanted to make a chair from scratch and this was a wonderful excuse. Thanks again to Jeremy at ATU Connemara for much of the inspiration behind the design. Our Irish Chair - ATU Connemara project

The things that are useful to take from this project are:

Making scale model is very useful for CNC projects. I would normally use a laser cutter to make the scale model but in this case I had the 3D printer available.
Having as much design work done before you get to CAD and CAM is really helpful, the hour I spent sketching made all the modeling flow so much faster.

The workflow that I teach here at the Enterprise Fablab is based on the following steps:

Design: Sketching and planning
CAD: Taking your design and adding specificity and dimensions in CAD software.
CAM: Creating toolpaths based on your CAD model.
Control: Posting those toolpaths to machine friendly code.
Machine: You in the mode of a operator setting up your machine and your material

Things I would do differently next time

I decided not to add and offset to my joints as I only had one shot at making this in the time frame o I thought sanding them would be lower risk, But I think if I was doing this again I would just have added even a small offset to the joints to make them more even.
The whole chair is a bit of a mystery as to how stable and sturdy it will be. If I had more time I would have made a complete chair in a cheaper material and tested it more thoroughly.
Time allowing I would have just worked slower and more carefully. There are a few parts of the chair that are a bit rough around the edges and I did not have time to finish it in oil. (Though I will do that as ill be keeping this chair for myself)
I would have liked to bring more ergonomic considerations into the design.

See below link to to files created this week:

Week 7 Files