💧Week 12: Casting and Moulding
Project Heros
Resin cast encaps sitting on the end of the 2020 extrusion of my final project.
Group Assignment
This week's group assignment was to compare materials safety data sheets, run some test casts and compare a milled vs printed negative mould. One of the things I learn't this week was that UV resin can come in a non-toxic form, I have always dealt with 3d printing resin which has always been toxic. The other major thing I got out of it was printing my own moulds. I have cast a few dice for dungeons and dragons before (quite poorly), but that was using pre-made moulds. I had never even thought of the feasibility of using 3d prints to make moulds and now that I have, I think I would be inclined to use them in projects where it calls for it. It still had all the print defects and lines in the final resin part, but I think there may be avenues to explore to remove them if needed.
3D Modellign the Negative
We had to make a silicon mould and to do this we had to cast the silicon into a wax-block negative and to do this we needed a model. This was the most difficult part of this entire week and I spent about an hour trying to wrap my head around whether this model needs to be a negative or positive. The easiest way that I understood it is that we would model the part we wanted, then we would make a box around it that we would use to fill up resin with.
We will be making some end caps for my 2020 extrusion for my robot. It is a really boring use, but I wanted something for my project, and I wanted to test some testing of manufacturing resin parts, how brittle they are, what are the tolerances of the end product, how much can it flex etc.
To start with I modelled the end cap that I was going to use. It is a simple half box and I gave the hole in the centre an extra 0.25 mm of clearance over the 2020 extrusion as I felt this would be a safe amount.
Then I added this tiny little fillet on the inner edge. It is purely aesthetic, but I wanted to see how detailed this process could produce with a milled block. The fillet was 0.5 mm and I would be impressed if this process captured this feature.
I then duplicated it as I wanted my block to cast 2 at a time. This would turn out to be a terrible mistake as it added an extra 2-3 hours of milling time to the process. If I had instead made it a single part per cast, it would only take an extra 15 minutes to pour and cure another UV resin endcap.
I then modelled the box around it, and envisioned the empty space it formed being filled with silicon to create the mould. And with this, we had our final 3d model.
Milling the Block
To mill this we used the lab's SRM-20 and to generate machine code for it, we used MODELA Player 4 which was very intuitive to use. To start with we imported the model by hitting import.
To start with we will mill a little off the top surface of the block to give us a nice and square volumeto to work with. To do so we set the origin to the top of the block, set it to a surfacing pass, chose the 3mm Square bit we were using, took 0.5 mm of depth off the top using scanlines, and set apropriate speeds. All of these settings look like this (its a high resolution image you may need to open it):
And we can see the result of that top pass.
Now that we have a square block, we need to run a roughing pass over it with the 3mm bit. This will remove most of the material we need, and we will come over it again with a finishing pass after. To do so add another process on the one we have created and make it a countoured roughing surface with the 3mm square bit we are using. These settings look like this:
Then we will add a final finishing pass, make it a contoured finishing cut with the 1.5 mm round ball bit. These settings look like this:
And thats it! Suprising straight forward and we can see a final simulation of the cut in the simulate tab:
Now we can hit cut and these files will load into Vpanel and mill away. We can add all the operations in the order we made them in. We then went ahead and loaded the 3 mm square bit into the chuck, let the first 2 passes run, then Vpanel prompted us for a tool change and we loaded in the 1.5 mm ball bit. This process is highlighted in depth in my electronics production week.
And we just let that mill away. It was about 3 hours into the milling process that I realised that making a mould for 2 of them was a mistake, it would of been far quicker to make a mould for just one of them.
But after a good 5-6 hours of milling, the wax block was finally cut. I was extremely impressed by how smooth it came out, it was possibly one of the smoothest things I have ever manufactured and even a high quality resin printer would struggle to produce this.
Making the Mould
To make the mould we are using 2 part silicon. I started by pouring out a bit more than I estimated I would need, mixing it 50-50 by weight.
I then went ahead and applied some vaseline with a paint brush - this will act as our release agent to prevent the silicon from sticking to the wax block.
After letting the silicon sit for a few minutes to remove any air bubbles, I poured it into the block until it just reached the top. I then tapped it on the bench for a few minutes to try and remove some of the air bubbles that formed after mixing it.
It was the end of the day, so I let it set over night and when I came back, it popped right out of the block and was suuuuper smooth. The silicon had captured the subtle detail on every surface including that 0.5 mm fillet on the inner edge.
Casting the Resin
With the mould now ready, we could pour in the resin. Using tiny mixing cups, I poured in some of our UV curing resin (a little bit more than I estimated I would need), and mixed in some purple dye.
I poured it into the mould, then popped it into our UV resin curing station. The UV curing process produces heat so I kept putting them back in untill they werent producing heat anymore and found that I needed about 10-15 minutes for it to fully cure.
And my first batch was ready! these came out... pretty bad. When I had mixed in the dye I had introduced a whole heap of air bubbles into it and these had not only risen to the top to create a bubbly surface, but some had collected at the bottom of them ould to create these indentations. But for my purposes these are still functional parts.
The second batch I mixed in some green UV fluorescant dye and accidentally squeezed the cap off dumping half the bottle into it. Below is a freeze frame of that accident. I drained as much as I could but I decided to role with it to see how too much dye would effect the curing.
And the UV fluorsecant dye made the curing process really cool, you could see shadows being cast from the UV light and you could also see the intensity change as the plate rotated. I found that these were extremely malleable after 15 minutes and I ended up curing them for a good 40 minutes or so and they stiffened up.
And this batch turned out a lot better because I let the mixed UV resin sit for a few minutes to try and elimated any bubbles that I created. There were still some on the top and a quick hit with a blowtorch or a vacuum chamber would of fixed these. But it came out nice and the excessive amount of dye left it slightly flexible, you could deform it a few mm by squeezing it hard. AND THAT 0.5 MM FILET HAD PERFECTLY BEEN CAPTURED, I can't get a good photo of it but its there in the final part and adds a smooth feeling to that edge.
And these fit really well on the 2020 extrusion. Going for an extra 0.25 mm of clearance was a good fit for this.
All in all I this was quite a fun and rewarding week in Fab Academy. I didn't have much experience in making moulds before this and I think its a valuable tool that I have in my maker toolbox now. I also see the possiblilties for this method, embedding fibres to produce strong parts, embedding electronics, dyes, pigments, casting ceramics for extremely heat resistant parts - it is a production method that has a lot of room for creativity.
So much so that I tried to cast some chocolate. I printed off a 2 part mould of an Among Us space man and tried casting directly into that as an experiment. I thought that the chocolate would shrink as it cooled and some oil as release agent would allow it to pop right out, but it turned out horribly. I think with some silicon moulds it would work and it is a project that I am exploring because chocolate robots sounds fun.
