Week 8 - Computer-controlled Machining

Considered one of the funnest weeks of the course, it is of course the "Make something big" week. We will learn how to use the magnificent machine known as the shopbot to produce a spectacular something or rather. This could be a big piece of furniture or something smaller we have already made from something like the laser cutting week blown up to a bigger scale.

Assignment:

MAKE SOMETHING BIG!


 

25th March 2015 - How big is big?

So the first question that popped into my head over this assignment was just how big is big, an house?! I settled on deciding that a nice piece of furniture would suffice as big after looking at last year's work. But now my next challenge, what piece of furniture do I make?

In short, two ideas popped into my head. The first one a hammock designed from a piece of wood using a very large living hinge design to give it the flexibility. The second, a kick-ass arm chair with which I can look like an absolute boss! It turns out the latter will be more functional, more practical, and easier to make (not to mention safer!). I refuse to use glue, nails, or screws on this chair, far too Victorian! Time to do some sketching...


26th March 2015 - My big thing is coming together! (In my head, that is)

As promised, here are my scribblings below of what my chair will kind of look like:





As random and rough as they look they really help me organise what the design needs to be like in my head before I start designing on the computer. Also, I needed to do a fair amount of research regarding certain measurements of the chair such as seat height, length, width etc. To translate this to something useful in Rhino I used a separate layer to create a guide for where to put the seat, back and so on. I then used the original layer to draw all the cut lines of my chair. Becuase it was quite a complex design I scaled everything down and made a prototype on the laser cutter with cardboard. Much care is needed when doing this to ensure all the slots are the right length ,width and depth.



Below is my mini prototype:





Having this model was really useful, it showed me some pitfalls which can only be seen by doing this. The main one is the need for more support for the back, both vertical and horizontal. Also, it tips back rather easily so a larger base will help cure that. Apart from that not a bad frist prototype even if I do say so myself! (The prototype also makes an adequate phone stand! Always useful)



My next objectives are to alter the real life design to include these changes and cut some test joints to ensure I have the best fit possible.

I went on to design some test joints for wood which was 18.5mm thick (on average). I made two sets of slots which had slots ranging from 18.3mm to 18.7mm in 0.1mm incraments. This gave me an opportunity to breifly use the shop bot which will make things easier later. My design was pretty simple but I still ran into a few issues:





As good as Rhino is at exporting Illustrator files, sometimes the software using them can tell it wasn't made by Illustrator itself. This was the case for the Part Works software which creates the tool path for the Shopbot. It refused to open it so I had to borrow a computer with Illustrator installed on it for a few moments and re-save an identical file. Then it opened! Next step was to select the vectors and tell Pathworks what you want the shopbot to do with them. I got it to set up a drilling job to make the markers for my different test joints then a cutting job to cut out the pieces. I also added tabs to prevent the pieces from flying off and potentially causing damage. These are easily added onto the Partworks software.

I made myself some unessesary work by making my markers less than 5 mm in diameter. However, it gave me a good opportuniy to learn how to change the cutting tool mid job. This is done my using two wreches and twisting them in opposite directions replacing the tool and collet and doing it all back up again.

I found that for this 18.5mm thick material that joints with a thickness of 18.3mm give a tight, sturdy fit. Now I know this, I can make the final changes to my design and get cutting!

Click here to download the design files.


29th March 2015 - Final Designs Complete!

After toiling away at Rhino for some hours and looking up all the essential measurements needed for my chair, I have my final design! The frist picture shows the individual parts that I deisgned and the second shows how I have fitted them into 4 separate jobs in the shopbot bed. The orange rectangles represent the area of the bed of the machine.





You may think that means I'm ready to process the job but no! This machine isnt as simple as the Laser Cutter or Roland Modela. There are more variables to consider with this machine:

There are a huge range of cutting tools to choose from. One has to consider the benefits of the different choices.

It is essential that the user secures their material sufficiently with screws and be absolutely certain that the tool won't hit the securing screws during the job. Depending on the complexity of the job, a separate drilling job may be required to mark out where the screws go. A very time consuming process but also a worth while one.

In addition to the 4 different jobs, I decided to also do 4 corresponding drilling jobs. I would therefore be exporting 8 DXF files. To keep them organised I created 4 folders labelled A, B, C and D. Then in each folder I would have two DXF files; one called "Drill" and the other "Cut". That way there would be nearly zero chance of me screwing up which job to do!

One issue I had was that I wouldn't know the exact thickness of the material I will be using until tomorrow morning as I still need to purchase it. That meant I wouldnt be able to add any joints until I go to the shop in the morning and see what is avaliable. To make it easier to design and add the joints I made a T-bone joint template which would easily give me a range of 10 joints varying by 0.1mm each. Take a look below:



Time for some sleep followed by an early start as the wood shop opens at 8:00 AM!


30th March 2015 - Shopbot day!

So today was the much anticipated shopbot day! This was gonna be a demanding day; both physically and mentally. First job was to go and find what wood was avaliable at the shop and then add the correct T-bone joints to my design. I was going for as cheap as possible and came across some nice boards that are used for flooring. They weren't the best quality but it came in the thickness I wanted (18mm but was in fact 18.1mm when measured with vernier calipers). I had learned from my test joints that I should make the joints 0.2mm smaller than the thickness of the material to ensure a good, tight fit. I would rather have a joint a tad too small than a tad too big!

I hot footed it to the nearest coffee shop and added T-bone joints of 17.8. Theoretically I should have added 17.9 but I made the assumption that some parts of the wood are bound to be 18mm. Following my "tad too small over tad too big" philosophy, I went with the 17.8.





So I went back to the shop, purchased the wood and brought it to the lab with the help of Shirley, another Fabacademy student.

The next step was to prepare my first file. This was the drilling job (or "drilling job A" rather!).I used a 3mm tool and the whole process took about 20 mins including fitting the screws.

I had a small battle with the program over which side of the vectors it should cut. This was due to the program interpreting the vectors as separate lines. This was fixed by using a function which joins selected vectors together. I could then create the tool path on the correct side of the vector. For the cutting I picked a 5mm tool. It is also essential to add tabs so that the part doesn't move during the job.

When the job was complete I unscrewed the material, gave it a good shake, and all the parts popped out. Then to save time later I would start the next job and smooth down the parts from the previous job during that time.

The first job made an awful lot of noise! To reduce this in the following jobs we reduced the cut depth from 3.2mm to 2.5mm. This would add time to the job as it has to make more passes, but it also meant that the noise was reduced and there was less chance of the tool breaking.

To smooth the parts, I filed down the edges and the tabs followed by a bit of sanding with some sand paper. Last thing I would want is a splinter in my backside!

Below is a much needed time lapse of this work flow. I particularly like this one because you can see me doing the finishing process while the machine runs the job.



After the first job I was able to put two of the parts together so that we had an early indication of the volume of the chair. Got me pretty excited!:



After 3 more jobs for the machine, I had all the parts. I began assembling it during the other jobs but now I could finish it off! The joints were tight... in fact... very tight! It took a lot of effort to hammer together (I looked like a psycopath while doing it!).

Sadly there were two defects/issues with the prototype. The first one was that one job went off of the edge of the bed a few millimeters. Also, the cross sections didn't sit flush with the back of the chair as they should. Sadly not much can be done about the first issue for now but I will return to the lab tomorrow to sort out the second issue. Apart from that the chair/throne looks pretty awesome!





Click here to download the design file.


31st March 2015 - Fixing the miscalculation!

After a bit of sawing, a bit of filing down and a second pair of hands the chair was finally finished. The next step will be making cushions to fit. Here it is along side its little brother!