🔩Week 7: CNC Machining
Project Heros
Giant hydrocarbon molecule kit:
Group Assignment
Our group assigment was one of my first times operating a CNC machine of this size and it was a lot easier than I thought it would be. I don't know why but in my head I had envisioned this process to be far more complicated (the CAMing was a little involved but the operating the machine was really easy). I was amazed at how great the alignment was on the machine, that 100 mm square part was milled down to a scarily accurate square. I also realised how dangerous these machines can be and how many risks they can prevent with stock potentially flying off the bed. If I ever get one in the future, I will opt for an enclosure as this really helped minimise the dangers (and dust). As soon as my parts were finished cutting, I also realised how incredible of a tool this is - massive parts being cut from a CAD file with ease and I think I will return to make some furniture and other projects I now have the ability to make.
Designing the Kit
This week I am scaling up the molecule kit that I made in laser cutting week and attempting to cut it out of ply wood on our CNC machine. I haven't had much prior experience with CNC machines so I am excited to give this a go. For the sake of simplicty, I will only be attempting to make a molecule of ethane as it will fit nicely on 2 sheets.
We will be trying to make half of this:
Modelling this week wasn't very interesting as I would just be scaling up the model I made in Onshape last week. Or so I thought. Our machine requires Fusion 360 to be used for the CAM process and so our CAD model is needs to be in Fusion. I could of exported a .step file from Onshape and imported it into Fusion, but I instead decided to remodel it from scratch as it will be easier to edit if needed. Exporting and importing the .step file removes the edit history of the model, meaing I cant go back and adjust a prior sketch.
The process of doing this was identical to what I did in week 3, so I wont be covering it here. There are differences though:
- I scaled up all the dimensions by a factor of 6.5
- I also extruded the sketches 13mm thick. We are using 12mm plywood, but the extra millimeter will give us some breathing room.
- And the tolearnces on the slots were set to 0.2mm. Our ply was fairly consistantly 12mm so a 12.2mm slot width was chosen for a nice push fit.
- I extruded the depth of the larger centre slots an extra 3mm. This is because the 6mm bit will not create a sharp corner on the inside and will prevent the pieces from slotting together.
With this, we had our model now in Fusion 360, and apropriately sized and toleranced for our CNC and machined version
CAM in Fusion 360
This was my first time using the CAM abiltiies of fusion, and thankfully I only needed to make a single contour so this process wasn't too difficult. To start with I duplicated the parts to the amount that I needed per sheet. Then I converted all of these to components.
From here we were ready to enter the Manufacturing sectiong of Fusion. First things first, we ensured that the tools and post processing libraries were installed for the bits and machine we were using.
Next I created a sketch the size of the footprint of the stock (minus 20mm on the edges to give some breadthing room), and I created a new arrange to arrange all the parts to fit nicely on the stock. I ensured that I gave some spacing on the edge of the stock, and between the parts.
I then created a new milling operation setup, and set the origin of my stock to the model origin point. The previous sketch for the arrange was drawn from this world origin point which allowed us to easily align our pieces to our stock like this.
I then defined the dimensions of the 12mm ply stock we were using, and we dont need any offsets because everything is aligned to the world origin point.
Now that we have our stock set up and our parts placed where we will cut them, we can assign the cuts. This is really easy for my project as I have no pockets and just need a single contour to cut all my parts out. I went ahead and added a new 2D contour cut, selected the milling bit from the library I imported, and set the feeds and speeds to the reccomended numbers for the bit on ply.
Then I assigned the bottom edges of the parts as geometries I wish to be cut, as well as adding some smll tabs to hold everything ain place.
And then I assigned the passes for it to take. There is quite a lot of perimeter to cut with this and I need 2x of these AND there was limited time in the lab so I opted to do deep roughing passes only and clean up any issues later. The bit we are using is good upto 13 mm of ply at these speeds so 5 mm passes should be more than fine.
And with that we have generated our required toolpaths.
Now we can export this by right clicking on NC Programs and creating a new one.
Then we can select the process we just made and hit post to save our numerical command file.
Cnc Machining
Now that we have our NC file we can cut it out on our Fab Lab's WorkBee CNC. First I placed the stock on the workbed and fastened it down with the clamps.
Then we zeroed in our piece using CNCjs which is our control software for the machine. Our machine has a piece of zeroing hardware consisting of a wire that clamps onto the bit and a metal block that is placed on the edge of piece at the origin. The bit is aligned to roughly the centre of the metal block and by hitting home the machine will slowly lower the bit till it contacts the metal and finds the z home, then it repeats the process for the x and y axis'. From here we would then move the bit in about 15 mm along the x and y axis and reset the home position to this. This just gave us some extra breathing room to ensure our piece is fully in the stock.
Then we import the files from fusion into CNCjs, turn on the vacuum head on the machine, remove the homing parts and secure them on the head, close the cnc housing and run the cut.
And it started cutting away!
My first sheet came out a little strange, 2/3 of the parts were properly cut, but 1/3 of them still had some material left. I don't know what the issue here exactl was, maybe the z-axis slipped in the final run. I chose not to recut it at this point as I was running out of time. I then ran another lot through but this time I dropped the z-axis by 1 mm to help fix any issues.
Not recutting it turned out to be a huge mistake as processing the not-fully-cut parts took several hours. It was such a pain that I forgot to get photos of this but it was just sanding down the edges, and if that wasn't enough - cutting down the edges, and if that wasn't enough- hacksawing down the edges, and if that wasn't enough - hammering the edges to snap off large chunks.
After about 10 hours of clean up, the parts were smoothed down and everything joined together with a light tap of a mallet.
Add on several layers of the cheapest acrylic paint I could find.
Re-assemble it, and you have the finished molecule of ethane.
All in all, this was a great experience. The molecule was huge and turned out to be 135 cm in the longest axis. The testing at the beginning to find a nice fit worked out really well, but If I were to do this again I would give it a bit more as once it was painted the fit became even tighter. I think the wood has maybe swelled a little bit from the paint, but we will see if It shrinks again when it evapourates. This was one of my first times using a large CNC machine like this and I think I will be back to make some larger scale projects that would benefit from this machine.