RE: More info on freeze cast process (FCP), ATTN: Andy Scott

Date: Thu Apr 08 1999 - 05:05:43 EEST

Glenn & Andy,

I wonder if following will make for a repeatable process, without air

1. place the mold and water in a vacuum casting machine (Kugelgen, MCP,
2. degass the water in a vacuum chamber
3. after degassing, but before the water freezes, pour water into mold
4. contiune applying vacuum until frozen

Any thoughts?

Dan Davis
PROTON Rapid Prototyping & Manufacturing Center
Hicom Industrial Estate, Batu Tiga, PO Box 7100
Shah Alam, 40918 Selangor MALAYSIA
+60 3 515-2380 phone/fax

> -----Original Message----- > From: Monica & Glenn Whiteside [] > Sent: Thursday, April 08, 1999 8:57 AM > To: Rapid Prototype Mailing List > Subject: More info on freeze cast process (FCP), ATTN: Andy Scott > > Andy: > > In reply to your question I have quoted some more information from the > article that might answer > some questions that people may have: > > "Fine-tuning the process:" > > "Not surprisingly, many problems were encountered during the > development > of the freeze cast process. > Mold design was particularly challenging, because water expands as it > freezes. Fortunately, ice exhibits the same linear expansion > repeatability, > if frozen at the same temperature. However, the expansion is > multidirectional (I take this to mean that water is isotropic - GW). > Therefore, the elastomer mold must be designed to provide enough extra > volume to absorb multidirectional forces. Moreover, the thickness and > consistency of the elastomer must be correlated with the different > modules > for the different shapes of the castings". > "After conducting hundreds of tests and measurements, problems > involving > the morphology of castings and the induced stresses were resolved. A > system > was developed in which several basic geometries were designed, to > which > certain semi-constants could be applied in the design of a specific > mold. A > major breakthrough came when directional solidification was viewed in > terms > of expansion rather than shrinkage". > "Another major problem was how to prevent the cracking caused by > stresses induced in the ice pattern by constricted freezing. This was > studied and attacked from three different directions: mold design, the > freezing medium's velocity and temperature, and air content in the > water/ice > system. It was discovered that > extremely low temperatures trigger cracking, particularly when the > patterns > are stripped at higher temperatures. This problem was adressed by > "conditioning" the ice, just as wax is conditioned prior to being > injected. > Conditioning involves controlling the air content of the ice and > regulating > additives (amount and chemistry) in the water". > "Warmer mold temperature was found to be crucial for fast, clean > stripping of rubber molds. Parting sprays that do not freeze, > pre-chilling > the water solution, and higher temperatures are used for > fast-stripping of > the molds". > "Air bubbles trapped on the surface of the ice were a recurring > and > persistent problem, in spite of molds designed to permit air bubbles > to > escape. Unfortunately, these bubbles would be reproduce on the > surface of > the casting. This problem was minimized by controlling turbulence and > counter pressure within the mold, through proper mold design and > venting. A > vacuum was also applied in a sealed chamber, prior to and during the > mold-filling procedure. A combination of these methods, along with > special > additives for the water, finally eliminated the air bubbles". > > Fascinating process!! > Hope this additional information helps. > > Best Regards, > > Glenn Whiteside > > > > >In a message dated 99-04-05 19:39:55 EDT, > writes: > > > ><< Then water > > (with "special" additives to minimize air bubbles) is poured into > the > rubber > > mold and frozen. >> > >Water expands when it freezes. Is it isotropic? Then you need to > compensate > >for the metal shrinkage as it cools. Sounds to me, it might take a > few > >iterations to get the numbers dialed in. > >Andy Scott > >Lockheed Martin Aerospace > > > >For more information about the rp-ml, see > > > > > For more information about the rp-ml, see

For more information about the rp-ml, see

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:51:25 EEST