How a Foundry Can Diversify Into Lost Foam Casting at Negligible Cost
Lost foam casting offers tremendous benefits but has underperformed on commercialization. Part of that is due to the perception of the process requiring large upfront capital and tooling expenses that make it a process only appropriate for large-volume production.
But you don’t have to invest in capital equipment or even tooling to diversify into lost foam casting. For the modest cost of the necessary raw materials, any foundry producing iron, steel, aluminum, or brass castings can quickly produce prototypes and short-run lost foam castings. This lost foam method could be used as an alternative to additive manufacturing or for producing tooling for prototypes, replacement parts, or other small volumes.
Lost Foam Advantages
The benefits of lost foam casting have been well documented. The near net shape process is known to permit complex shapes, and Department of Energy (DOE) research has reported that compared to traditional casting methods, lost foam provides 25%−30% energy savings, 46% savings in labor productivity, 7% less materials used, and production cost reductions of 20%−25%. Less solid waste as well as less particulate air emissions and greenhouse gases are created compared to traditional processes. An additional benefit is that with proper gating design, casting yields over 70% are common, and yields over 80% are feasible.
Based on the authors’ measurements, tolerances of +/- 0.003 in.-per-in. are typical (0.076 mm/mm) and +/-0.002 in.-per-in. (0.05 mm/mm) are feasible in some cases. For the machined foam approach, the tolerance is dependent on the machining accuracy and typically 0.002 in. (0.05 mm) plus the machining tolerance. Recent DOE funded research concerning thin-walled ductile iron showed that even at 0.040 in. (1 mm), the tolerance for lost foam was +/-0.0015 in. (0.039 mm). This key benefit means that lost foam castings can either be used with zero or minimal metal machining. With proper component redesign this can result in enormous cost savings that offset the slightly higher cost of the process compared to traditional green sand or nobake casting.
As shown in Figures 1−3, lost foam provides net or near net shapes that can have complex geometries such as interior channels, blind holes, and true position. Components can have zero or alternating draft (Figure 4). The authors have recently even developed a process for as-cast threads as shown in Figure 5. And lost foam can eliminate the need for tooling; foam can be machined and cast (Figure 6).
Steps to Trying Lost Foam
The lost foam process is shown in Figure 7. The key to having lost foam with minimal capital investment is to eliminate the need for expensive tooling and automated lines. This is done by machining the foam from foam stock and creating a fluidized bed by manually compacting the flasks. The remaining steps are essentially the same as standard lost foam casting.
Step 1: OBTAIN FOAM
The first step is to obtain foam blocks for machining. Lost foam typically uses expanded polystyrene (EPS) foam. The ideal density is 1-1.5 lbs. per cu.ft. (0.016 g/cm3- 0.024 g/cm3).
If carbon control is important such as in iron casting, using Clearcast is desirable—it is a co-polymer made of expanded polystyrene (EPS) and polymethyl methacrylate (PMMA). This can also be obtained in block form for machining.
One essential safety factor: Ensure that any foam used does not contain flame retardant. Foams containing flame retardant tend to explode during lost foam casting because they are designed to not burn. Foundries should verify with their foam provider that the foam does not contain flame retardants, and they should also conduct a flame test themselves.
If the machined foam won’t be used immediately, foundries are advised to age it for at least three weeks prior to use. Newly blown foam is not dimensionally stable due to the moisture slowly evaporating. Depending on climate, after three weeks and possibly sooner, the foam will stop changing dimensionally.
Step 2: PATTERN MACHINING
A CNC machine is necessary for machining the foam pattern. A wide range of machine tools are feasible. It may be necessary to conduct several trials as the feeds, speeds, depth of cut, etc. are going to depend on the machine, the foam, and the cutting tools used.
The key is that the foam can be cut, and tiny chips are formed. If the feed rate is too fast, the depth of cut too deep, or the speed to slow, the foam will tend to tear. This is especially true for blown foam because the foam beads will pull out from the material rather than cut. As a general rule, higher speeds and lower in-feeds and depth of cut are necessary.
When going from the part’s CAD design to CAM and machining, it is important to account for the metal shrinkage that will occur due to the coefficient of thermal expansion for the given metal used. However, due to the lack of mold wall movement, only metal shrinkage is necessary for this adjustment factor. For alloys such as aluminum with its high shrinkage upon solidification, risers can be used or the gating design itself can serve as the riser. For iron alloys, risers are not typically used. For aluminum and steels, the risers are approximately one-third the sand riser size.
Step 3: ADHESIVE
One of the reasons lost foam can permit highly complex shapes is the foam can be joined prior to casting. This means several parts can be designed to be joined together rather than machined or blown as one finished shape. There can be zero draft, alternating drafts, complex channels, or blind holes. One can even cast interlocking components such as chain—so long as they have separate ingates.
The joining of foams can be done using supplies as basic as school glue or tape, but most lost foam foundries use Foam-Lok 70-12-11, a specialty adhesive designed for lost foam. This material is heated and then rapidly solidifies on cooling. Adhesive can be applied by a brush or by dipping.
Step 4: GATING
Typically, lost foam uses a consumable ceramic down-sprue funnel. This is glued to a gating system that feeds the part with one or more smaller in-gates. Often, these are notched to facilitate the breakoff process in finishing. There is no gating ratio and there should not be a choke point in the gating system.
A gating pattern is used to blow the foam gating system for all castings, but the gating system also can be cut out of stock foam.
Gating design is a complex subject and highly dependent on the exact casting design requirements. A thorough coverage of this is beyond the scope of this article. Suffice it to say, the gating rules for normal sand casting do not apply to lost foam.
Aluminum is typically top fed—meaning the castings are underneath the gating system. Iron, steel, and brass are typically bottom fed; the castings are above the gating system.
Minimal glue should be used, as it will produce excess gas. Excess glue also will be cast into the final part, so drips on the foam will result in drips on the casting that need to be ground off. Ensure the stability of the gating system particularly for thin parts.
Finally, the most important thing to keep in mind for gating is to plan how the sand will flow around the part. If lost foam is done properly, it will obtain a near net shape casting with high tolerance control. If it is not done properly, the result will be a mass of metal mixed with sand with little resemblance to the desired part. A key aspect of this is the compaction process.
The gating design directly impacts how the sand can and will flow. While green sand and nobake casting are nearly always designed vertically or symmetrically, you want the opposite in lost foam. Parts should always be somewhat tilted. Otherwise, the sand will not flow into holes, channels, and other features. In addition, sand will only go uphill approximately 0.25 in. (6.35 mm), so gating designs should not require the sand to flow upward.
Step 5: COATING
A variety of coating manufacturers and coatings designed for lost foam are available. The key to coating is having a consistent process. If a foundry is not using a ready-to-use coating, it will need to invest in a viscometer to ensure a consistent coating is achieved. The coating must be mixed thoroughly and then used quickly, as it can settle within a few minutes, which will result in an inconsistent coating and poor casting results. While it is essential to mix the coating, overmixing can cause bubbles in the coating. Those will dry and result as holes in the coating and will cause surface defects and burnt-on sand.
The foams can be dipped into the coating, or the coating can be poured over the foam.
Step 6: DRYING
Moisture in the coatings can cause burn-on sand or even steam explosions. Drying depends on the climate and conditions of the plant. Some places will use fans to circulate air. Others will have special drying rooms with heated and dehumidified air.
Step 7: COATING INSPECTION
Cracked coating will cause burnt-on sand and can cause a complete collapse of the mold. The component and gating system should be inspected for any cracks and touched-up, if necessary, with coating and allowed to redry. Zircon coating also can be used to fill in the cracks. It is not typically used for the overall coating due to expense.
It is dangerous to re-dip or repaint the entire foam assembly. If the coating is too thick, the vaporized foam gas will not be able to escape through the coating walls into the sand as designed. If the gas cannot go through the walls, then the gas may go up the down sprue. If that occurs, the gas can blow the molten metal up out the down sprue and into the operator’s face.
Step 8: COMPACTION
While very expensive, lost foam compaction lines are available and similar to an automated green sand molding investment. Foundries can also opt to make molds by hand. All that is necessary for lost foam casting is to fluidize the sand so it vibrates and flows like water. This causes the sand to pack tightly around the foam. The sand backs up the coating so the system doesn’t collapse when the foam is vaporized in advance of the molten metal front, but it allows the vaporized gas to dissipate.
In manual compaction, a container is needed to serve as a flask, such as a clean 55-gallon steel drum. Depending on the size of the gating system, the drums are often cut to remove one third to half of the height to make pouring easier.
Using ceramic beads for compaction reduces silica dust. However, standard lost foam sand (olivine) can also be used. One can also use regular silica sand, but this adversely impacts the dimensional control. The key is that the sand is unbonded, dry, cooled, and sifted to remove any contaminants. Sand or beads that are used with low melting point metals such as aluminum tend to have EPS residue build up on them over time and then glom together into chunks that need to be removed.
After the drum is filled with 2 in. (50 mm) of beads/sand, it is compacted by repeatedly hitting the drum with a rubber mallet. The sand will become hard to the touch.
The foam assembly is placed onto the starting sand, and sand is added while hitting the drum with the rubber mallet so the sand flows around the part. The sand should not be directly poured onto the part to prevent the coating from wearing off. Also, for thin sections, sand should be filled on both sides so the weight of the sand doesn’t break through the foam.
Once the part is buried, the flask continues to be filled with sand, which is also compacted.
At least 10 in. (25.4 cm) of sand on top of the part is necessary; otherwise the entire mold can float and cause erratic dimensions or form an undefined mass of metal.
Step 9: CASTING
The melting process for lost foam is the same as in other foundry processes. The key is to ensure sufficient melt temperature; a higher-than-normal superheat is required to vaporize the foam compared to pouring into an empty mold cavity. Typically, an additional 50-100F (30-55 C) is needed depending on the precise alloy. Too much superheat also can cause problems. For example, in iron alloys, the metal can boil, which is extremely dangerous; and in aluminum, oxide formation increases at higher melting temperatures.
Lost foam depends on getting the vaporized gas out of the system without the mold wall collapsing. Unlike automated lines, this method of casting does not have a vacuum system to help remove the gas. The molten metal removes the gas by pushing it out of the system using gravity. Pour as fast as possible to make sure there is sufficient weight of metal to offset any gas pressure.
Pouring too slowly can result in mold wall collapse or explosions. The down sprue should always be full during pouring.
Flames may occur around the flask during pouring. That is the styrene gas catching on fire and is normal.
Step 10: FINISHING
The casting should be allowed to solidify at least 30 minutes for aluminum and one hour for iron or steel. At that point, the sand and metal will still be hot, but the sand can be removed and the parts shaken out. Waiting until the parts are fully cooled is also an option. For some metals, this will cause the parts to self-anneal since the lost foam mold is highly insulative.
Assuming the above process steps were done properly, finishing is the same as in sand casting. Typically, the blast time is only one third of the time for a sand casting.
Unfortunately, if an issue occurs in the process, lost foam castings generally are not reworkable. The resulting casting is an undefined mass that only somewhat resembles the starting point and will be full of burnt-on sand as shown in Figure 8.
Ready, Set, Go
For foundries with a need for rapid prototypes or customers desiring small volumes, the method of machined foam and manual compaction for lost foam is an opportunity to diversify with minimal capital expense. The melting, casting, and finishing processes are essentially the same as in their current processes except for needing a higher pour temperature. The molding process for lost foam, while different, is something that can be done on a small scale to first conduct trials. This could even be done with metal that would otherwise be pigged out.
Typically, the machined foam approach works well from a cost perspective for volumes under 100 castings. For over 1,000 castings (total use) it usually makes sense to manufacture foam mold tooling at some point.
An existing sand foundry could conduct trials with the simple process first; once they have had success with prototypes and larger batch sizes are ordered, it may be worthwhile to consider the larger investments in tooling, foam blowing equipment, and an automated compaction line.