Michael Edwards: Fab Academy 2019

Week 10: moulding and Casting

2019.03.20

Individual Assignments:

  1. Design a 3D mould around the stock and tooling that you'll be using, machine it, and use it to cast parts. Incomplete

Group Assignments:

  1. Review the safety data sheets for each of your moulding and casting materials, then make and compare test casts with each of them. Incomplete
  1. Mould A: Cable Holder
  2. Mould B: M Tile
  3. Vibration Table

Nomenclature

What do we call things?

Rightly or wrongly, when learning new fields I like to know the words that are used. The closest I have come to casting in the past is jelly moulds (before I became vegetarian).
The part that is produced can also be called the casting.

A bivalve mould is a two part mould with each mould casting half the item.
An articulated mould is a multi-part mould.
A mould might be coated with release agent before casting, so that the part can be easily removed.
We will be using the MDX-50 to mill hard wax (the blue stuff we used as a sacrificial layer for milling the boards). This wax mould will then be used to create a silicon mould that we use to cast the actual part. Therefore we are actually creating at least two moulds. The first mould is a mould for the mould. The part will come out of the second mould.

A bivalve mould is a two part mould with each mould casting half the item.
An articulated mould is a multi-part mould.
A mould might be coated with release agent before casting, so that the part can be easily removed.
We will be using the MDX-50 to mill hard wax (the blue stuff we used as a sacrificial layer for milling the boards). This wax mould will then be used to create a silicon mould that we use to cast the actual part. Therefore we are actually creating at least two moulds. The first mould is a mould for the mould. The part will come out of the second mould.

Mould A: Concrete Cable Holder

Fusion 360 Modeling

To start this week I wanted to make something useful. I started off drawing a cable holder for my desk (especially useful for a programmer cable...). The idea is that I'm going to draw the cable holder as a body, then draw a box around the outside of the model to create the mirror (the mould) then one more time to make the mould of the mould... I have no idea if it's going to work, but Fab Academy is teaching me to go for it and see what happens.

This is the initial sketch. I made the square 60mm x 60mm, the construction lines to divide the x axis of the shape into thirds are 20mm apart. Then the construction lines for the arcs are 6mm either side of the thirds dividers. The arc bends to 2.25mm at its widest point, so the gap for the cable will be 7.5mm.

Next I extruded the whole square down and everything but the cable channels up. I added fillets to the internal corners of the cable channels:

Cast Object View

I flipped the design just to see what the final cast object will look like in use. The top of the cable holder will be self-levelled by the pour of the concrete or resin. For this first mould I'm trying not to make it too difficult for myself.

With the design flipped back upside down, I drew this box around the whole model. I thought originally I would cut the original out of the box, but I think I am going to have to extrude the cable channel dividers.

It's very hard to tell, but I have added a 1 degree draft to the sides. This shape would create the casting I need, but not the top of the cable holder, so I needed to create walls to encase the top.

The walls are too thin, but the negative space in this model would produce the cable holder, so I'm on the right track.

Mould View

For the next step I thickened the walls of the mould. Now I have the mould as a model. I am going to draw a 3D box around it and cut the mould shape out of it and... fingers crossed... it will work.

Final mould tool

After drawing the box I used a Modify > Combine > Cut operation with the box as the target and the previous body as the tool. This created a new body which is almost right, but needs a base.

I span the design upside down and extruded the walls 3mm.

I drew a 3mm rectangle on one of the sides of the newly extruded walls, then extruded it across the base with a join operation.

The design flipped back right side up, I extruded the square around the channels down back to the base.

Then I pulled the shapes between the dividers up.

I then filleted the corners of the channels and the outside edges.

Milling the mould

With the MDX-50

The

Back to the top

Mould B: Food Safe Small Wax Cut E Tile

Mixing and filling the mould

The casting material is called Polycraft Food Grade Silicone Rubber from MB Fibreglass in Belfast. The kit consists of 1KG of white(ish) base and 100g of blue catalyst. The material cures blue. The mix ratio is 10:1 and the reported properties are:

Back to the top

Vibration Table

Meanwhile...

A vibration table helps to make nice and smooth castings by removing all the air bubbles and filling all crevices. For some casting materials (concrete for example) vibration causes liquification and creates much smoother castings.

The first photo below is the parts that I have cut out of 3mm MDF for the table. I have designed some tabs that will connect to the frame via springs and cut out a wedge shape to off-balance the motor, creating the vibration. I also cut a slot to allow the wedge to rotate through the board. The whole thing came out of one sheet of A3 MDF which was quite pleasing. The spring holders at each end will connect to a fairly sturdy plywood frame made of 18mm ply and the table will be supported on springs below too.

Next I got the parts I need laid out, the cable tie will connect the motor to the table, the springs and screws are for the frame and to allow the table some movement.

Just about to measure and cut the 18mm ply. That saw turned out to be woefully underpowered and I used a standard hand saw.

I connected the springs to the tabs on the laser cut MDF to check the plan would work. It seems fine! Once I connected the springs I stretched the assembly out slightly and used it to measure the long frame base bars.

Here's the parts for the base frame laid out ready to connect. I have some angle brackets to attach the crossbeams.

I assembled the table top with the springs and cable-tied the motor to the centre of the table.

The frame is quite agricultural... Not my finest joinery, but it is going to be absolutely fit for purpose.

I cut two more bars and set them perpendicular to the first set. Once both the bars are screwed on the frame feels much more secure and doesn't twist.

I then screwed the table and spring assembly on top of the frame.

The next thing I did was wire up the motor to a switch and a couple of batteries. I took the motor out of a broken toy, so I don't know anything about it, but I tested it with 3v from two AA 1.5v batteries and it works fine. The switch could do with being mounted, but with spiral development in mind, this working prototype is good enough for the first iteration.

The motor could probably do with being strapped down a bit better, the single cable tie allows a lot of movement, but it seems to work. I have added the strip of balsa on this video to negate the movement of the moulds caused by the vibration. In the video you can see the wax blocks shuffle along the strip. I need to add another one where the wires come out and along the back edge to stop the moulds falling off the table.

I had some time while while my third mould was milling, so I added some springs underneath the table bed to support the hanging table. This will allow me to put some more weight on the table and relives some of the stress on the hanging springs.

Back to the top

Chocolate Casting

With moulds from the £100 CNC

Back to the top

Obstacles

Problem 1

Back to the top

References

Online

"If you are cast in a different mould to the majority, it is no merit of yours: Nature did it." - Charlotte Bronte

Polymerisation: Addition
Cured Colour: Blue
Cure Time (25C): 16 Hours
Shore A Hardness: 59
Mixed Viscousity: 900-1300
Tensile Strength (psi): 650
Tear Strength (Die B): 90
Linear Shrinkage: NIL