Application of Generative RP Technologies with Rubber-Like Materials for Pneumatic Seals

Seals are pressure-retaining technical elements. Widely used in industrial and other applications, they are important for the elementary function of fluidic devices. Sometimes seals have complex geometries and are often made of elastomers or elastomer-like materials. Several new rapid prototyping materials have appeared in recent years that demonstrate elastomer-like behavior. This article provides a survey of the related methods (stereolithography, laser sintering and 3D printing) and the available materials.

Against the backdrop of ever-shortening deadlines and new competitive situations, it is becoming increasingly important to procure all components and prototypes required for testing product concepts, approving new products and even for starting up production and manufacturing small batches within very short periods of time. This issue applies in particular to complex shaped parts made with molds such as die castings, “metal injection molded,” or MIM parts, and elastomer parts. Although they have only been available since the end of the 1980s,1, 2 generative technologies have become helpful3, 4 in recent years in the field of industrial product development, along with machining options provided by modern CNC processes such as simultaneous five-axis machining and HSC.

For the purpose of offering a vision for the future, these examinations are intended to show which contribution generative manufacturing technologies have longevity in the area of sealing components.

Seals are frequently an inconspicuous component within the overall product. Although their importance is sometimes overlooked, they have immense technical and economic significance.

What are “Generative” Technologies with Regard to Elastomer Parts?

How are parts produced from elastomer materials using generative technologies? At the beginning of 2006, there were three process concepts and three material suppliers available on the market. The processes include laser sintering (a granular polymer material is melted and “baked” with the preliminary layer), stereolithography and 3D printing (the latter is achieved by polymerization using UV laser or lamps in accordance with coating information for photo-monomers).

Laser sintering material is a powder based on polyester (TPE-E). The pressure density is achieved by using appropriate fluid infiltration material. It allows for the adjustment of hardness values within a range of 60 to 75 ShA and elongation at break values from 120 percent to 130 percent depending upon laser power.5, 6

Opposite to this, a thermoplastic elastomer based upon polyurethane resin (TPE-U) is used in stereolithography, which demonstrates typical values of 70 ShA where hardness is concerned, and 75 percent with regard to elongation at break.6 Comparative figures for 3D printing uses a thermoplastic elastomer based upon a polyurethane resin (TPEU), which amount to (depending upon coloring) 60 ShA (gray) to 75 ShA (black) for hardness, and roughly 50 percent for elongation at break (see Figure 1, page 12).6

Although the hardness values, and to some extent, the values for elongation at break, are within the ranges demonstrated by common elastomer materials, the tensile strength values—which serve as a rough orientation for mechanical load capacity—deviate greatly. Values for RP materials are approximately 10 times less. In the examinations described below, only materials with hardness values ranging from 70 to 75 ShA were used.

Geometric Reproduction Based on a Case Study Involving a “Double Nipple”

Comprehensive international benchmarking was conducted on the basis of a double nipple required for a customer-specific project with the technologies listed in Figure 1 in cooperation with research institutes, system suppliers and service providers (even two suppliers were involved in the case of the stereolithography process). At least 10 parts were produced per process and per supplier. Figure 2 depicts the basic geometry and the essential characteristics of the part.

Figure 3 shows the macroscopic appearance of typical representative parts fabricated with each of the processes. At first glance, the outside dimensions and essential characteristics have been very well reproduced, and depending upon the used process, the surface ranges from glossy and smooth to somewhat granular.

However, if the functional dimension (sealing diameter) is measured with greater accuracy at the sealing area using a profile comparator, differences become apparent (see Figure 4). Most of the parts are within the specified tolerances, but based upon the example of stereolithography, it is plain to see that although the same process, material and initial geometry were used, in reality, data interpretation and process control influenced reproduction of the CAD model.

The need for the improvement of the process capability can also be seen during a microscopic examination. Even under transmitted light microscopy (see Figure 5, page 14), the divergent shape of the sealing area (sealing semi-circle) is apparent in comparison with a series of manufactured parts produced with a mold which is identical in the functional area. An exact reproduction of the sealing radius is only conditionally possible using generative methods. With regard to reproducibility, especially in the case of minimal absolute size and small radii (typically < 1 mm), all of the processes demonstrate room for optimization with respect to geometric reproduction as well as surface quality (relatively rough or torn instead of smooth). Improvement may be possible in this area with more finely focusable lasers and machines (and associated mechanisms), which are micro-technically more suitable, as well as fine-grained powder or base materials which have been improved during photo-polymerization.

Two peculiarities are conspicuous as well:

  1. A step effect, which results from the used production principle of all generative processes, is detected at inclinations and radii as the layers are built up. For this reason, the selection of an appropriate layer thickness, which is feasible for the respective process, plays a decisive role in the quality of the finished part. Just how divergently this fact may reveal itself in the final results is demonstrated by a direct comparison of two parts produced by means of stereolithography.
  2. The need for knowledge of the respective application is demonstrated by the 3D printing example. Micro-channels, which promote leakage, result intrinsically due to the build-up of layers parallel to the longitudinal axis (all other parts were built up perpendicular to the longitudinal axis).
Despite all (moderate) criticism with regard to the physical reproducibility of the initial virtual model, it must still be observed that 10 generatively produced parts were finished in less than two to three hours (fabrication time) with each of the used processes.

Usability as a Sealing Element

Leakage measurements were performed in addition to the material examinations in order to determine whether or not the parts can be used as pneumatic sealing elements. The nipples were inserted into a fixture made of PMMA, which simulates the actual geometry of the mating part for this purpose (see Figure 6, page 14), and were measured with a precision leakage measuring instrument. The photos depict the fixture with the mounted sealing nipple in the unpressurised state.

Leakage measurement was performed on at least seven samples in new condition for each process using a differential pressure within a range of 2 to 7 bar. Figure 7, summarizes the individual leakage results in a graphic representation.

Although the mean values ascertained in this way (5 to 10 liters per hour) differ greatly from those obtained for series parts produced with a mold (typically < 1 liter per hour), fundamental usability is established for at least two of the four processes with regard to the short periods of time within which a prototype can be made available, if concessions can be made where leakage is concerned (e.g. 5 l/h). The large spread demonstrated by the leakage measurement values within the individual processes is still unsatisfactory (up to a factor of 10). On the one hand, this is surely caused to a great extent by the manufacturing tolerances of the functional dimension (compare Figure 5), and on the other hand, it’s also caused by the varying reproduction of the micro-topography in the sealing area (see Figure 6).

However, significant improvement might well be expected in both cases (absolute value and leakage value spread) as a result of the optimization potential already discussed above as part of the microscopic examination.

When the double nipples became fabricated due to 3D printing, the measured leakage values confirmed the unfavorable outcome shown in Figure 8 with the resulting channels, which open fully as of a differential pressure of 3 to 4 bar so that no sealing at all is possible.

Summary and Future Prospects

In summary, the examinations elucidated in this article demonstrate that suitable sealing elements for static pneumatic pressure loading can, in principle, be produced in very short periods of time using generative technologies. However, process improvements will be necessary before these elements can be used to any great extent, and further examinations with regard to material and process performance will be required as well.

Looking to the future, the following may be deduced:

  • Generative manufacturing methods, for elastomer parts as well, can and will represent a time-saving opportunity in the field of new product development for fluidic components.
  • Direct, tool-free production of small lot sizes of geometrically complex sealing elements is conceivable in the mid-term or the long-term. 

Klaus Müller-Lohmeier is the manager of technology for Festo AG & Co. KG (Esslingen, Germany), where he is responsible for materials sciences and application, failure analysis, environment questions, advanced prototyping and new production techniques. He has his MSC in Aerospace Engineering and about 25 papers and presentations on materials behavior and rapid prototyping and manufacturing. 

Notes

  1. Hull, Charles W. Apparatus for Production of Three-Dimensional Objects by Stereolithography. U.S. Patent 4,575,330 of March, 11, 1986.
  2. Grimm, Todd. User’s Guide to Rapid Prototyping. United States of America. Society of Manufacturing Engineers, 2004.
  3. Müller-Lohmeier, Klaus/ Edelmann, Oliver. “Direct Prototyping of Parts in Light Metal Alloys and Steel by Laser Melting.” Rapid Prototyping & Manufacturing 2005 Conference, Dearborn (MI). 11, May 2005.
  4. Müller-Lohmeier, Klaus. “Hands-On Experience with Pneumatic Seals Made by Generative RP Techniques.” 14th International Sealing Conference, Stuttgart (Germany), 10 October 2006.
  5. Levy, Gideon. Sintaflex “A New Elastomer Powder Material for The SLS Process.” Rapid Prototyping & Manufacturing 2005 Conference, Dearborn (MI). 11 May 2005.
  6. Company information: www.3dsystems.comwww.dsmsomos.com. www.2objet.com.

Zones


IMTS 2012
Register today for The MFG Meeting, March 8 - 11, 2012, Orlando, Florida.
3D Printing – The New Frontier for Manufacturing
I had the privilege of touring one of the prominent companies in this rapidly growing field of 3D printing,


Read more


Featured Zones: Hardware | Management | Materials | Processes | Product Development | Software | View More Zones...

Zones | Suppliers | Products | Articles | Calendar | Industry Links | Contact Us

© 2012 AMT-The Association For Manufacturing Technology

All Rights Reserved | About Us