jerome arul

< back to main

this week i'm gonna explore the geometry of the tetrahedral skeleton further

looking back at CAD from wk4 and wk6 the skeletal (voronoi?) has an internal angle of 109.471 degrees between members, not 120 degrees like i previously thought.
three members are always 120 degrees to the other normal member.

i have a bunch of sch 40, 1" pvc pipe
for which the parts will tenon into.
• ID: 1"
• OD: 1.315"
• master constrained by 3x6x1.5" machinable wax stock
• master-mold (wax) wall thkness: min 1/4"
• mold (urethane) wall thkness: min 0.26"

halves connected by 5/16th bolts

109.471 degrees internal angle

assembly of diamond cubic connector

half component with removable core

quarter component with sprue and registration

half of half the component
for a two-sided mold
• add sprue(s)
• add registration bumps
• no taper added..
• 3/8" ⌀ removable core
• 0.26" tool clearance between part and wall
• 0.26" t-bone for mating face
• export as .step

1/8" flat, bottom and internal corners

1/4" ball, finishing pass

thurs
mastercam ain't easy (thnks jen!)

• check surfaces = avoid surface
• containment lines = tool boundary sketch

• parallel (1/4" flat) for roughing
• contour (1/4" flat) to clean wall
• scallop (1/8" flat) for bottom, internal corners
• scallop (1/4" ball) for finish surfaces

• simulate with "verify"
• "G1 icon" is post
• check post-processor
• tool library (offsets) is checked by jen

feeds and speeds:
• SFM for wax = 270
• S4000 F76 (for 1/4")

3 tools, exported separately

cnc controller - zero XY under "setup"

intelitek is locked out. get permission.
home machine
• open gcode file
• double check tool #
• insert tool into spindle
• jog to XY (eyeball)
• zero XY under "setup"
• Z0 is spoilboard.

• square stock to bed (eyeball)
• hot glue to spoilboard

rough mill (1/4" flat)

vacuum between toolchange

finishing pass (1/4" ball)

master model

pour the mold

fri noon
1st half

using reoflex 30, urethane
pam for mold release, brush in corners
1:1 ratio
pour A or B into the other more viscous liquid
stir well, try not to fold air into it
• chip brushes
• gloves
• plastic cups
• tongue depressors
• fan+hepa for ventilation
• n95

should de-gas

basement shop

sat noon
2nd half

potlife: 30mins
cure: 16hrs

sun 10AM
demold 2nd half, looks good

go to shop to make core
• find 3/8th round stock in scrap bin
• cut close to length
• face
• break edges

facing 3/8" core

core

dry run, make paper cone

turn on hepa fan, don't spill

2 pours on sunday
2 pours on monday

smooth-cast 310 - hard white plastic
1:1 ratio

potlife: 7mins
cure: 4hrs

air was getting trapped at the top of the mold
at two different locations.
i cut some relief into the urethane.

mold needs surgery

2x tetra / diamond cubic connectors

the kit

connected to 1" pvc

need better hepa fan
i should build this one.

• 2 spills - need to calculate mold volume
• restock: gloves, stirrers

mold started failing after 3rd cast
• got sticky. i didn't demold soon enough?
• is it because i used pam and not real mold release?