Collaborative Product Design
Takes Flight

In today’s product design landscape, two forms of the design process are typical. The first process, a sort of “build it and they will come” method, rests entirely on supplier-side engineering teams designing a product and hoping it sells.

The second form of design common in today’s engineering environment is custom engineering. In the custom engineering process, customers turn over specifications to a supplier and let the supplier design from there.

But an evolution of the product design process from the supplier-side paradigm is underway. The emerging process, called collaborative product design, lets teams of supplier and customer engineers work together to drive product innovation, shorten life cycles, reduce costs, and limit the many design changes common in a single-sided design process.

For Flowserve and NASA, the collaborative process that led to a new valve design combined the resources of each team’s engineers and testing facilities. The end-result was a better product that satisfied the needs of each participant. Flowserve earned the rights to a new valve technology, saving time and effort in the process, and delivered a valve design that will save NASA money and improve its rocket-testing procedures. Through effective collaboration, Flowserve and NASA produced an innovative product that might not have been as easily achieved using conventional design methods.

Collaboration is not a simple process. It requires engineers willing to innovate inside an unusually cooperative customer/supplier partnership that features a high level of trust by participants for both partners. Certain industries, such as information technology, were early adopters of collaborative product design processes. Computer Sciences Corporation, a leading global information technology services company, is a case in point. Computer Science’s director for manufacturing solutions, Michael Bauer, believes there are defined ways in which companies can make the collaborative product design process effective and rewarding for all participants. Bauer says a first step is the elimination of duplication in process, people, and technology.

To do this, organizations in a collaborative engineering process need to designate one owner per process, based on best fit. This requires managers in a collaborative project to match the right person to the right process by identifying and agreeing on key skill sets. By streamlining a project in this way, managers give individual process owners responsibility and control while reducing waste and cost.

Critical also in the management of a collaborative product design process is setting agreements on functions like administration and finance and clearly spelling out the ownership of intellectual property rights. Being proactive in delineating responsibility for these sometimes thorny issues is the hallmark of a well-managed collaborative design project.

Other ways to improve product design collaboration include: integrating process teams to fuse silos of product design, strategic sourcing, buying, and program management; encouraging shared governance to drive radical changes in performance, cycle time reduction, and costs; maintaining a balanced scorecard to reward new skills and roles for process owners; and promoting changes to the status quo, upstream and downstream. Ultimately, the success of collaborative product design rests on how well supplier engineers can build and manage the process to completely tailor the project to meet the customer’s needs. “Everyone on the supplier side needs not to lose the focus on the needs of the customer during every phase of the product design process,” says Bauer. “Our goal should be not just to deliver a product, but to make our customer successful. It just can’t be business as usual.”

Bauer says companies taking on the collaborative product design process must, at the bottom line, be able to remove the barriers that commonly hamper internal and customer-supplier communication by insisting on upsetting the status quo. Continually adapting to each partner’s procedures and needs builds flexibility that can drive the process to a design solution that both sides desire.

Karlin Wilkes, a Flowserve engineer who worked on the NASA project, agrees. “Fundamentally, collaborative product design is a new way of doing business,” he says. “It allows Flowserve to take advantage of and deepen long-standing customer relationships, and employ complementary engineering and design strengths to build the best possible solutions for our customers. A natural byproduct of a successful collaborative effort is a trust-based relationship that builds continued success for all partners.”

At the Stennis Space Center near Bay Saint Louis, Miss. (, NASA engineers had a problem. Their process of testing various rocket propulsion systems was running into an expensive snag, hanging on high-pressure control valves that needed to be switched out and replaced for each test.

Stennis uses these valves, in both ambient and cryogenic configurations, for the flow control of oxygen, hydrogen, and other propulsion propellants used in its rocket engine and component test facilities. Stennis is famous as the main propulsion testing center for NASA, and it is where the Space Shuttle’s main engines are tested and certified.

Propulsion tests can include measurements of temperature, pressure, flow regulation, engine design validation, and material and component performance. Engineers at Stennis have historically employed inline, top-entry valves in these high-pressure performance tests because the top-entry design was perceived to limit external leakage.

But inline, top-entry valves have their drawbacks. They are expensive and have long face-to-face lengths of up to six feet. NASA also needed the flexibility to change the trim size on its valves to meet the differing flow control requirements of individual tests. Since the trim size cannot be changed on the type of inline, forged valves that NASA normally uses for these applications, NASA engineers were forced to stock multiple valves for a single test sequence — for example, testing an engine or component using several propellant flow rates.

When Failure Is Not an Option
To find a solution, NASA engineers turned to Flowserve Valtek Control Products, a provider of flow control solutions for the aerospace industry. The contact between NASA and Flowserve was originated through the Stennis Propulsion Test Directorate, which directs and manages the operational and engineering functions for rocket engine testing at Stennis.

Considering the drawbacks of inline, top-entry valves, NASA engineers thought a solution might be to try an inline, split-body valve, which would allow them to change trim sizes without taking the valve out of service to meet testing requirements. If Flowserve could build a split-body valve for NASA, it would provide tremendous flexibility in testing and save on valve inventory expense.

NASA engineers were cautiously optimistic. In their experience, split-body valves, particularly those used in cryogenic applications, were prone to external and seat leakage. Additionally, in NASA’s high-pressure engine testing applications, engineers who wanted to use a split-body design were typically limited to an offset body as compared to an inline design. This is because inline split-body valves typically have to use castings instead of forged body assemblies. NASA prefers inline valves because they are inherently easier to work with, as their body geometry enables simpler installation than that of offset valves.

Valves for Rocket Science
To see if their idea would fly, field and design engineers from NASA and Flowserve worked to create a valve that would simplify installation and maintenance, allow for trim size changes, and minimize inventory requirements, thus reducing cost. Combining NASA’s experience with high-pressure aerospace applications and Flowserve’s application, design, and testing experience in the liquid natural gas, upstream oil and gas, and aerospace markets, the team generated a set of technical solutions that met NASA”s requirements for high-pressure, cryogenic, rocket propulsion tests.

Because NASA’s requirements included both ambient and cryogenic applications, Flowserve began by integrating its most advanced control valve technology into the valve”s seals, stem packing, and mechanical design for the required temperature ranges. In the first stage of the process, the body of the valve was divided into an upper and a lower section, with its seat ring sandwiched in between. To maintain the stem packing at an acceptable sealing temperature for cryogenic service, the design team added heat-absorbing fins to the upper-body section of the valve.

With the valve body made of stainless steel, project team engineers designed the seat ring in a nickel-based alloy with a coefficient of thermal expansion less than that of the body material. This enabled the body surrounding the seat ring to contract more than the seat ring when the valve interior was cryogenically cooled, preventing external leakage at the body-seat joint. Engineers also machined the seat ring to have small, raised-face sealing surfaces on both sides of the seal groove, which concentrate the body bolt-load over a small area and work to prevent external leakage.

Preparing for Launch
The new valve’s body bolt-circle design is also different from those in conventional high-pressure control valves, with half of the bolts clamping the split-body together from the top and half from the bottom. This design allows a short, clean flow path, minimizing frictional flow losses and shortening the face-to-face length of the valve. While a conventional class 4500 valve with a nominal Cv of 120 has a face-to-face dimension greater than 40 inches, the NASA-Flowserve split-body valve is only 25.5 inches long.

“This was an entirely collaborative and iterative process, with contributors from NASA and Flowserve providing their expertise and contributions to the success of this project,” says Michael Yentzen, an engineer on the project who is now a legislative affairs specialist at NASA headquarters in Washington, D.C.

NASA trusted Flowserve as a cooperative partner to help it design the solution it needed. Flowserve solved every issue as it arose, getting input and feedback from NASA’s engineers during each step of the design process. All told, Flowserve and NASA worked together on the project for about a year.

Pre-launch tests included cryogenic testing at Flowserve labs, then at Stennis, and then together in final validation tests. At Stennis, the valve was bubble-tight at 11,250 PSIG, even at cryogenic temperatures.

Mission Accomplished
Following the design process, NASA ordered additional Flowserve split-body valves in varying sizes for other applications in the same system. The new applications for the additional split-body valves include transfer-line isolation and tank venting at Stennis. NASA has slated the new split-body design as a flow control valve in an RP-1 (refined petroleum) fuel run-line in a facility currently under construction at Stennis.

“We’re very pleased with the product we developed,” says Yentzen. “This valve design will save NASA a significant amount of money and will make rocket test operations at Stennis simpler and more efficient.”

NASA engineers wrote up the results of the design process in a NASA Tech Brief and are looking forward to working with Flowserve again. Flowserve engineers believe the new valve will have multiple applications across the aerospace and industrial gas industries, as well as in markets such as upstream oil and gas and steel.

About the Author
Karlin Wilkes, who worked on the Stennis Space Center valve design project, is an engineer with more than 14 years of experience in field testing and troubleshooting, application engineering, and design and testing for cryogenic liquids, such as LNG, LOX, and hydrogen. Wilkes began his career as field testing engineer with General Electric Company in 1990. He joined Flowserve Corp. in 1993 as an application engineer for the Valtek control valve line and was subsequently promoted to engineering manager in 1996. Mr. Wilkes can be reached at or 801 489-2645.

For More Information: