Patterns for Investment Casting

New Options for Short Run Investment Casting

Tom Mueller, Express Pattern

Over the last several years, the use of direct patterns (patterns created without the use of tooling or machining, usually by a rapid prototyping technique) has become increasingly accepted in the investment casting industry. Consider:

  • More than 90% of non-art investment foundries use direct patterns in their business.
  • The 2010 Wohlers Report, an annual report on the rapid prototyping industry, estimated that 50,000 production castings were created last year from direct patterns with estimated value of $20 million.

In addition, the range of applications using direct patterns has increased tremendously:

  • Direct patterns have long been used to create prototype castings, allowing the foundry customer to verify design adequacy before tooling is ordered.
  • A significant percentage of the direct patterns built today are used forbridge production. Foundries use direct patterns to create limited volumesof production castings while pattern tooling is being built.
  • Direct patterns are increasingly being used in low volume applications. In those cases tooling is never built. Instead direct patterns are used for the entire production run. In most cases, this is new business for the foundry
  • Direct patterns are increasingly being used in process development to fine tune the casting process before tooling is available. The allow the foundry to determine optimum gate locations and orientation on the sprue, determine actual shrink values, and program robotic dipping to obtain an even coat of slurry.

Some foundries have even made short run and fast turn castings from direct patterns a focus of their business.

Issues with Direct Patterns

Will we continue to see the use of direct patterns grow? Probably, but not at the rate we have seen over the last ten years. Direct patterns have two fundamental problems; they are difficult to process, and they are too expensive.

Processing Difficulty

With the exception of wax patterns made by Thermojet or ProJet systems, the process of casting a direct pattern is significantly different than that used for wax patterns. For example, consider the variations necessary for QuickCast patterns:

  • Vents must be added to each pattern, both to allow steam to enter the pattern during autoclaving and to allow airflow through the mold during burnout.
  • The pattern must be punctured at the vent prior to autoclaving to allow steam to enter the pattern and soften the internal structure
  • The patterns may have to be burned out at a different temperature than that at which the furnace is normally run to avoid crystobalite conversion infused silica shells.
  • Additional oxygen may be required in the burnout furnace to support combustion
  • The shell must be cooled down after burnout to allow rinsing and to patch vents.

One foundry estimates that these changes add $500 per assembly to the cost of casting QuickCast patterns.

Even ThermoJet and ProJet patterns require some minor revisions to the conventional process. Any revision to the normal process, however, increases processing cost and increases the probability that a processing error will be made resulting in scrap.

There is no question that if the processing difficulties of direct patterns could be eliminated, they would be much more widely used.

 

Cost

Although costs have come down significantly in the last few years, direct patterns remain expensive compared to the incremental cost of molding a wax pattern.

The high cost of patterns limits the number of applications for which direct patterns make sense.

There are two elements of cost. The first is the cost of the pattern itself. Several elements contribute to the cost of the pattern including the cost of the equipment used to create the patterns, the cost of material, and the cost of labor necessary to finish the patterns as required for the casting process.

The second element of cost is the incremental cost of the process variations necessary to cast the direct pattern detailed above.

To understand the effect of pattern cost, consider Figure 1.

 

figure-1
Figure 1. Pattern cost versus number of castings

 

The cost of molded wax patterns is represented by line 1. The initial cost of tooling defines the intercept with the y axis. The slope of the line is the incremental cost of molding a pattern which is generally very low. As a result, the cost rises relatively slowly as the number of patterns increase.
The cost of direct patterns is shown on line 2. There is no tooling cost so the cost starts at zero and the slope of the line is the effective incremental pattern cost which includes the cost of the direct pattern plus the additional processing cost. That cost is much higher than the cost of molding a wax pattern so the total cost rises much more quickly than the cost of molded wax patterns. The point at which those two lines cross is the break even volume, Va. If the number of patterns required is less than the break even volume, it will be less expensive to use direct patterns than molded wax patterns. If the number of patterns is greater than the break even volume, it will be less expensive to use molded waxpatterns.
Assume for a moment that the additional cost of processing direct patterns could be eliminated. In other words, they could be cast them using the same process used for wax patterns. The effective incremental pattern cost would then be justthe cost of the direct pattern. The cost curve would move to line 3 and the break even point moves to a higher number of patterns, Vb.

 

Now assume that the cost of the pattern itself can be reduced. The cost curve now moves to line 4 and the break even point moves to an even higher numberof patterns, Vc.
What would such changes mean to the foundry? :

  • The foundry could provide higher value to its customers through providing lower cost casting for volumes less than Vc. The lower cost may allow investment casting to compete more effectively in applications where the cost of tooling made it more expensive than other methods of creating metal components.
  • By eliminating the time required to make tooling, foundries can deliver short runs of castings sooner. Customers likely will pay more for castings delivered sooner, increasing the profit of foundries.

 

Clearly, there is a significant value to the industry if progress can be made toward resolving the major issues.

 

New Options

In the last two years, two new options have emerged which move us closer tothat ideal. They include:

  • Voxeljet– Voxeljet is a German technology. Like the SLS Castform process, it is powder based but instead of using a laser to fuse the powderparticles together, the Voxeljet process uses inkjet technology to spray a binder to fuse powder particles together creating the solid. In early testing, Voxeljet patterns have been cast using the same process as molded wax patterns and yielded good castings. An additional advantage is that the Voxeljet material actually shrinks as it heats up, greatly reducing the chances of cracked shells in the autoclave and raising the possibility of fewer coats to build the shell.

    There are currently limitations on surface finish and accuracy that can be achieved with Voxeljet patterns, but they should exceed requirements for most non-aerospace castings.

  • FOPAT – FOPAT patterns are foam patterns created by injecting a two part resin into a mold. The two components of the resin are mixed during injection and react to form a self-skinning foam. It is a relatively cold, low pressure process and consequently can use low cost tooling such as SLA, rubber, epoxy, or spray metal tooling. The patterns are light and durable compared to wax patterns. Like most direct patterns, they will not melt out of the shell in the autoclave, but can be burned out in an extended preheat cycle at temperatures above 1600F.

    One advantage of the FOPAT process is that the patterns can be molded around ceramic cores, something that is not possible with layer-wise direct pattern methods. Surface finish and accuracy are largely dependent on the quality of the tool used. The FOPAT process is also fast compared to direct pattern methods. Once the tool is available, about 50 patterns per shift can be molded per cavity. Up to six tools can be run simultaneously so it is possible to mold 300 patterns per shift.

 

The incremental cost of the foam pattern is more than the cost of a molded wax pattern, but the tooling cost is lower. Figure 2 shows a typical FOPAT cost curve added to the chart of Figure 1.

 

figure-2

Figure 2. Foam pattern cost added to the chart

 

Effect on Pattern Cost

It is apparent that there are now two significant break-even points, Vd and Ve. The first is at the point where the cost of foam patterns equals the cost of direct patterns. The second is where the cost of foam patterns equals the cost of molded wax patterns. Those two break-even points are the boundaries of three volume ranges as shown in Figure 3.

 

figure-3

Figure 3. Low cost pattern ranges

 

For volumes less than Vd, the lowest pattern cost will be provided by new direct pattern technologies with both lower pattern cost and lower processing cost.Note that the accuracy and surface finish requirements of some applications may prevent them from using such new technologies.
For volumes greater than Vd but less than Ve, the lowest pattern cost will be provided by foam patterns.
For volumes greater than Ve, the lowest pattern cost will be provided by molded wax patterns.
The introduction of these technologies has affected pattern cost in two significant ways:

  1. Total pattern cost will be lowered for all volumes less than Ve.
  2. For volumes greater than Va and less than Ve, the lowest pattern cost will no longer provided by molded wax patterns

 

Of course, the exact values of Va, Vb, Vc, Vd, and Ve will vary with the size and complexity of the casting. In addition, these technologies may not be appropriate for all applications, especially those requiring tight tolerances and very good surface finishes. For other applications, however, the potential of lower cost patterns provide opportunities to better compete in low volume applications, provide higher value to customers, provide shorter turn around times, and to earn higher profits
As time goes on, we will no doubt continue to see new technologies developed that will further affect cost. It will be important for foundries to understand how the introduction of new pattern technologies can affect their costs.