RAPID PROTOTYPING COMES TO THE FABRICATING INDUSTRY

by
Jim Mishek
Vista Technologies LLC

Introduction
Plastics, plastics, plastics, this is all that is thought of when applications for rapid prototyping are discussed. Unfortunately, there is a tremendous niche, the metals fabricating area, that is being over looked.

Form, Fit, and Function are the three key words defining the need for prototyping. With the advent of the various rapid prototyping machines, particularly stereolithography (SLA), the need for Fit and Form has been fulfilled. The problem is Function. In the proper application, we have succeeded in building entire functioning models using the SLA material. The projector shown in figure 1 is a perfect example of an ultra successful project using rapid prototyping.

Figure 1

The projector was first prototyped in a conceptual model, just for form. After the marketing people had a chance to critique the design, it was further refined by combining the form ideas with a harder look at the engineering to see if the components would actually fit with the design. After another review with marketing, to further nail down the form, the engineering for fit became serious with all components being installed and tested for function in the prototype model.

When it was decided that all was right, the company then made six totally functioning prototype models using SLA, had them painted, added the decals, then showed them at a major trade show. This avenue gave them a chance to test market a major new product without huge production costs in less than six months. As a point of interest, the product shown in the marketing brochure is a SLA prototype.

The slowest part of trying to assemble the prototype were the few auxiliary pieces that they wanted made out of sheet metal. The sheet metal prototypes delayed the process at each step. Even though this progressive company had extensively used the rapid prototyping (RP) technologies for all plastic components, the old mentality of "if it is sheet metal, it has to be prototyped in sheet metal" still existed.

So, there is still a need to find ways to fabricate parts quickly when metal samples are required. Using the SLA rapid prototyping combined with other new technologies, Vista Technologies is developing techniques which will aid the fabricating community in meeting their Fit, Form, Function, and Speed requirements for prototyping.

Background
The modern age of rapid prototyping began just ten years ago, when Chuck Hull of 3D Systems first made commercially available stereolithograghy. Since then the field has expanded with the introduction of other rapid prototyping technologies: Laminated Object Manufacturing (LOM), Fused Deposition Modeling (FDM), and Selected Laser Sintering (SLS). Every one of these techniques had the plastics industry as their primary market. LOM has also found a major niche in the foundry industry.

Perhaps, this lack of attention to the fabricating industry is because the major unique feature of these technologies is the ability to make three dimensional models. The sheetmetal industry has always been classified as a two dimensional industry. Even the design software for sheetmetal fabrication is two dimensional. Forgotten is the fact that the finished sheetmetal product is almost always three dimensional due to bending and forming. Suddenly, sheetmetal components are ideal candidates for the solid modeling technologies.

Another major reason is that most of the people doing the rapid prototyping development have plastics backgrounds, therefore their initial push was into the areas they were most comfortable with.

The third reason is that the sheetmetal industry was not crying for help. there were so many changes going on with the development of faster lasers, turret punch presses, and tooling abilities, that the shop owners had a hard time keeping up.

Therefore, it is understandable why even though the rapid prototyping technology emerged ten years ago, it was not until just recently that anybody even considered the sheetmetal industry as a possible candidate for its use. Boris Fritz of Northrup Grumman did some of the first work by experimenting with hydroforming. The thought being that the die can be made quickly using SLA, but the material may be too brittle to withstand the forming pressures.

Much to everyone's surprise, Mr. Fritz reported the results that the SLA material was able to withstand the wear and the pressure to successfully hydroform sheet metal. At last report he had tested the material up to pressures of 950 tons.

The problem is, how many other people have hydroforming presses for plates? This technology will be more practical if you are able to make tools that will withstand the force of straight punching, either in a conventional die set or in a turret punch press.

Our experimentation has been twofold: first to show that rapid prototyping is a viable alternative for sheet metal parts, secondly experiment with forming tools to see if they are able to withstand the pressure and shock of conventional punching.

Experimentation
The first experiment was to see if rapid prototyping could aid the fabricating industry for fit and form. The part that was chosen is a stamped part, see figure 2. This part is a motor housing for a small electric motor.

Figure 2

In production, this part is produced by a progressive die, which double hits the metal to produce the draw form, then stamps the holes. The progressive die is approximately $30,000. For this magnitude of money you do not want to be making changes in the tool. This could cost up to half the price of the tool in rework. This is a catch 22. This style of tooling is too expensive to prototype, but without a prototype, the production run is a crap shoot and there is a good chance that reworks will be required.

There are two options for prototyping the part, the first option is to machine the part from a solid. According to the tool maker who was asked to quote a prototype for this motor housing, the cost would be approximately $5000, would take a week minimum, and would be difficult due to the thin walls.

The second option would be to make a model using a rapid prototyping technology. In this case, we made a SLA prototype in 24 hours for a cost of $500. With this prototype, they were able to prove fit and form, they were even able to mount the housing on the motor and test it for air flow.

COST AND TIME COMPARISON FOR MOTOR HOUSING Method of Manufacturing Time to Manufacture Cost to Manufacture Permanent tooling $30,000 12 weeks Mill a prototype $5,000 5 days SLA prototype $500 1 day

In this case, rapid prototyping was proven to be the way to go. Each case should be reviewed individually, but in the vast majority, it will be found that the new RP technologies make sense.

The next two experiments moved from rapid prototyping technology into rapid manufacturing. The first trial was to test a Keltool sample in a forming application. Keltool is a 3D Systems process. In this process, a master is produced (usually using SLA) then a negative mold is made of this. A slurry of powdered A6 tool steel and a urethane binder is then poured into the mold and cured. Upon curing, the mold is pulled and the resulting piece is heated to burn out the binder material. The part is then packed in copper and a copper impregnation occurs. This process takes two weeks.

The resulting part is at a hardness of 30 Rc. The part can then be machined and if desired, heat treated to a maximum hardness of 50 Rc. In this forming application we felt that the harder the part the better - our parts were heat treated to 50 Rc.

It was decided that a fair test was to put a simple emboss on the tool. In reality, there would have been more practical ways to make a simple tool such as this, but the purpose of the test was to see how the material held up under punching conditions. The tool inserts were fitted into a Wilson Tool Series 90 punch holder. The tool was then inserted into a 30 ton Finn Power turret punch press and ran. Since the operator had doubts as to how it would run, he thought he would break it early, so, he chose to run it in 18 gauge (1.1 mm) stainless steel. Much to his surprise, the tool did not break, and in fact, kept running. He ran 1400 forms in the stainless steel. Figure 3 shows the tools and the embossed stainless steel.

Figure 3

At the end of the run, measurements of the punch and the die were taken and there was no measurable difference from when these tools were first put into service.

The last test was run using the SLA material as the punch and the die in a forming application. The Keltool has proven itself longer running punching applications and the SLA material has proven itself under a low impact hydroforming application, now the question is - can the SLA material hold up in a low volume, high impact forming application?

The simple emboss was again chosen as the test shape, see figure 4. The SLA forming tools were punched using a 30 ton Amada Vipros turret punch press. The high shape was chosen because it gave us options to change the tool during the run, if necessary due to tool break down, to continue testing. As it turned out, this was a good idea. Contrary to our initial apprehensions, SLA material held up great, but the forming material could not stretch this far and continually tore, see figures 5.

The first run was done with the full size tool. The punch pad form had a height of .492" (12.5 mm) with .040" (1 mm) radii added to the punch and die pads. In the first two runs there were three pieces of .06" (1.5 mm) aluminum run, followed by five pieces of .06"

Figure 4

.492" (12.5 mm) form height	.375" (9.5 mm) form height
Figure 5

(1.5 mm) cold rolled mild steel. The first punch pad had slight abrasions on the tip from the torn metal stripping from the punch.

The punch pad shape was modified for the second run to try and reduce the metal tearing. The second tip was .375" (9.5 mm) high with a .031" (.8 mm) radius added to the punch tip. This tip suffered no wear from the forming despite the tearing of the metal again.

The final run was made with the emboss height at only .250" (6 mm) and a .031" (.8 mm) radius added to the punch tip. The run began with three pieces of aluminum, all of which tore. The next five pieces of cold rolled mild steel, followed by two pieces of galvanized mild steel, all ran good forms with no measurable tool wear.

We were then able to produce good forms with no damage to the SLA tooling when running the aluminum, mild steel, and the galvanized steel. However, when we ran the stainless steel the tool was able to produce the form, but the after the first hit, the tool did start to degradate. From the second to the sixth hit there was a bulging of the punch and the die pads such that the punch pad finished with a radius of .06" (1.5 mm) and the die pad radius had grown to .125" (3 mm).

The stainless steel forms are more domed by the sixth hit, but still within industry specifications for fabrications of +/- .005" (.12 mm). Figure 6 shows the results from this last run.

Figure 7

Conclusions
The rapid prototyping technology belongs in the fabricating industry. All three functions of prototyping, Form, Fit, and Function, can be fulfilled. It was proven that just because the part is made of metal does not mean the prototype must be. SLA has been proven to be a cost effective alternative.

Figure 8 Rapid manufacturing is the extremely exciting part of where this can lead. The metal sample in figure 8 is an example of how great the possibilities are. This part symbolizes a prototyped, stamped part that was turned around in three days. A person can now make a sample form using a SLA die set, and then, with the aid of a laser or a mill, can put the openings required into the part to simulate the stamped part. A three day turnaround for an inexpensive prototype of a stamping - not bad.

Now, with the Keltool inserts, die makers have an alternative for expensive elaborate forming designs. This could be extremely valuable.

The next application we are going to test will be SLA inserts for tube hydroforming tooling.

The marriage of the fabricating and rapid prototyping technologies is very exciting. Do not have tunnel vision because of the materials being used - open up and experiment. The results may surprise you!

Acknowledgements
A warm thank you goes out to the R&D department at Wilson Tool International for their engineering and testing help in running the sheet metal tests. Thank you to McMillan Electric for allowing us to case study their motor housing and to Dufresne Manufacturing for help in fabrication. A final thank you to 3D Systems for donating the Keltool parts for testing.