Lectures on Molding and Casting and Machining a Mold
On Thursday, Mickael gave us a very good lecture on molding and casting. An overview can be found here.
I decided to use the Roland Modela MDX-20 to machine my mold. Emma gave us a very good lecture on 3D milling using a Roland Modela.
Some notes
Metal plate as base
Double-sided tape to fix block of wax
Remove bit
Open Fab Modules and choose STL as source and RML as process
Load STL file
Select mm, 1/8, wax, rough
make .png
STL units (mm/unit)
1.0
PNG resolution (px/mm)
25
Side
z
make .path
type
3D rough
diameter (mm)
3.175
offsets (-1 to fill)
-1
overlap (0-1)
0.25
error (px)
1.5
top intensity (0-1)
1
top z (mm)
e.g. 0 or -1 in order to trim surface
bot intensity (0-1)
0
bot z (mm)
top z - object’s height
cut depth (mm)
1
Insert bit and calibrate Z
set and note XY origin
Send it and start machining mold
Wait until job done
Select mm, 1/8, wax, finish
make .path
type
3D finish
tool diameter (mm)
e.g. 1.5875 (1/16 inch)
overlap (0-1)
0.9
error (px)
1
top intensity (0-1)
1
top z (mm)
same as previous
bot intensity (0-1)
0
bot z (mm)
same as previous
clearance (mm)
0
clearance diameter (mm)
same as tool diameter
Send it and start finishing
Collect all filings in order to melt and reuse
Designing a Mold For a Chocolate Cat
My idea was to have a chocolate cat for easter as a present. Well, I made some research to study physiognomy and anatomy of cat and drew some sketches.
In Rhinoceros, I designed a 3D model. The cat’s tail was a real nerve-jangling challenge! It wasn't easy to build a specific organic shape using a 3D application and create a form of it which a milling machine is able to machine. It took me several days to build the model and design and adjust the 3D mold. My first nice design showed a prettily curved tail. Unfortunately, I wasn't able to design a machinable mold for it.
Finally, I revised my design and created a straight tail. Now, it was curved in one direction only. The top of the tail was round. So, I had to design a 3 part mold which in turn didn't work with the ears. Another 4 part mold wasn't easily machinable. I had to redesign the tail with a straight machinable top and keep it simple. I added two vents and a filling hole on the bottom.
I had to do some final adjustments. The positive molds’ edges had to be flattened because the end mill wasn’t long enough and the whole machine head would come down. I measured end mill and machine head and drew them. In Rhinoceros, it was easy to simulate now if the head would have enough space to move. Besides, the cat’s neck had to be filled with a kind of ruff. So, the end mill could properly work. The wax block (Kerr AccuCarve 34888 Block 1 Lb) wasn't big enough for the chocolate cat and flattened edges. The cat couldn't be smaller because it would become too detailed to be machinable. So, I spared top and bottom edges of the positive mold. I would use a frame later in order to edge the molds.
In the end, I was very happy with my design and excited to machine my mold. I came to know better Rhinoceros and understood the milling machine’s limits.
Please download an original Rhinoceros design file here.
Machining the Positive Mold
I opened the Fab Modules and loaded my STL file. My model got some details and had to be milled with a 1/16 inch end mill at least. The Fab Modules calculated a very long milling time—approx. 14 hours. I was told to set the PNG resolution to 25 px/mm and use a 1/8 inch end mill for the rough cut in order to reduce the milling time. I sent the RML file and started the milling process.
The rendered PNG showed a black mark and I wondered what it was. This shouldn't be there. I quickly checked the STL file and discovered a hole in the mesh. I stopped the milling process and analyzed the milling path. I couldn't see any problematic path and restarted the job. The rough job should be done well though the PNG file showed this tiny mark. The result was good, indeed. I repaired the 3D file and loaded a a new STL file. I set up the finishing job which was done after a whole day.
Photos of back mold production
Rough job settings
Diameter
3,175 mm
Offsets
-1
Overlap
0,25
Error pixels
1,5
Finishing job settings
Tool diameter
1,5875 mm
Overlap
0,9
Error pixels
1
Clearance
0 mm
Clearance diameter
1,5875 mm
Unfortunately, the final result wasn't satisfying. Loading a repaired file messed the alignment. Somehow, the pin holes weren’t circular anymore. And some parts of the chocolate cat’s body weren’t finished smoothly all over. I wanted to machine a new mold but I was told to go on and start the 2nd part of the mold. Some other students had to machine their molds as well and time was running. I could review my first result and probably optimize my design.
Photos of front mold production
Machining the front mold worked out perfectly. I did a rough job using a 1/8 inch end mill and a finishing job using a 1/16 inch end mill. I got a precise and smooth mold.
Casting the Negative Mold
-
Casting the negative molds worked well. In order to edge the molds, I laser cut and fixed two acrylic glass pieces with double sided tape as well as single sided tape to be sure that no liquid could flow out. The silicone rubber compound suitable for food related applications (Smooth-Sil 940) had to cure 24 hours at room temperature, 2 hours at 80 °C and 1 hour at 100 °C in an oven.
Photos of production
Unfortunately, the thin base of the chocolate cat’s back wax mold broke loosening the negative mold. I cut out the non-circular pin holes using a cutter.
Casting a Chocolate Cat
In order to tightly align both molds, I used a fix. I melted the chocolate and poured it into the closed mold.
Photos of production
I wanted a hollow chocolate cat and let all surplus chocolate pour out. Unfortunately, the figure broke opening the mold. The chocolate layer was too thin. I decided to cast a solid chocolate cat which in turn perfectly worked.