A Knowledgeable Friend Recommended TK Solver;
Veteran Engineer Has Used It Ever Since

Throughout his 45–year career, consulting engineer Dennis Barbeau has spent a great deal of time “thinking outside the box” and on the front lines of technology. His early work was on the design of military aircraft engines and included considerable use of mathematical modeling to investigate and optimize performance. “Military aircraft development has always been led by the engine performance and pushing the envelope has been important”, says Barbeau. In the 60’s and 70’s Barbeau employed mainframe computers to do mathematical modeling. The spectrum of modern commercial software tools that we now take for granted was not available at that time and Barbeau had to develop all of his software from scratch, primarily using Assembler for coding.
In early 2000 Barbeau convinced one of his clients, a small aerospace company, that mathematical modeling could generate major benefits for them. They were spending a great deal of money to develop their database through system tests, with limited results. Initially Barbeau used Excel but found it far too primitive to meet his requirements. He then talked to a friend, who suggested that he use TK Solver. Thanks to his early work in mathematical modeling, Barbeau immediately appreciated the power and versatility of TK Solver and has used it ever since.

Dennis Barbeau recognized early in his work on aircraft engine designs that, properly used, mathematical modeling could serve as “virtual prototyping” to gain significant advantages: projects got done faster and you saved a lot of money. Modeling made it possible to explore many designs on a computer in a short time frame. Instead of the extensive rounds of design / build / test / redesign iterations common to the industry at that time, one could vastly reduce trial and error and far more rapidly build the designs that offered the best potential. The objectives of testing were changed to validation of the mathematical models instead of validation of hardware. TK Solver provided the basis for this process with Barbeau’s customer and resulted in a very rapid and inexpensive assessment of the system issues and solutions.

  Dennis Barbeau
  Home: Mesa, Arizona
  TK User Since: 2001
  Uses TK For: Aero-thermodynamic analysis of gas-turbine and diesel power systems
With his long experience with aggressive, performance-driven, time-driven product requirements using cutting-edge technology, Barbeau can be thought of as a “power user” of TK Solver. His TK models are complex and sophisticated, yet they take very little time to set up. He can continue to refine them as he gains deeper insights into the problem. Barbeau now can get from thousands of possible scenarios to the best few in the shortest possible time—with the help of TK Solver.
Barbeau, Director and Principal of InnSol, Inc., Mesa, Ariz., is a consultant in diesel and gas turbine power systems. He uses TK Solver to assess and optimize complete systems, modeling subsystems and components using aerospace, thermal and mechanical technologies.
“The features of TK Solver allow you to model complex cycles with multi-spool turbo-machinery, heat exchangers, and other component options with relative ease,” Barbeau says. “A system of describing these components with multi-variable non-dimensional characteristics is required. The ability to accomplish this with table look-up and solve routines was one of the things that drew me to TK Solver.”
TK also works for Barbeau because of its layered, object-based architecture. “Modeling complex cycles results in a number of interdependent, implicitly defined variables—multiple iterations with both internal and external dependencies,” Barbeau says. “There can be a dozen in various layers of procedures, two or three layers on top of the basic Rule Sheet.”
His early career entailed “a lot of number-crunching,” Barbeau says. “I began life as a test engineer in a company that was developing the largest gas turbine in the world at that time”—the PS-13, for the Avro Canada CF-105 Arrow all-weather interceptor aircraft.
“I realized I wasn’t going to gain any real understanding of the engine without seeing it from the analytical point of view,” he says, “so I went back to graduate school and then transferred to an analytical group in thermodynamics.
“Then,” Barbeau continues, “I discovered the computer.” He learned Assembler and later picked up Fortran, but quickly reverted to Assembler because of its significant advantages in speed and storage utilization. He later quit coding as the company data processing group developed and concentrated on engineering and the associated computer program design processes.
“What was interesting was that, through this learning experience in using Assembler and trying to cram program size and power into mainframes, I developed and formalized programming techniques that became standards a decade later,” he says. The logic of processes that could substantially improve the efficiency of developing new software became quite apparent.
Barbeau subsequently advanced through department head to vice president of engineering; part of this he attributes to his early love affair and success with the computer: “I was basically an unstructured person; programming provided a structured environment that developed the ‘other half’ of the techniques required for effective problem solving.” Then he diversified into managing organizational turnarounds in engineering, manufacturing, and quality control, as well as new product launches.
Barbeau spent a large part of his career at Teledyne CAE (formerly Continental Aviation and Engineering), where his projects included design and development of gas turbine engines for very high-altitude drones, trainer and light strike aircraft and other specialized applications. “I was fortunate in having the opportunity to concentrate in advanced design concepts,” he says. “System evaluation and optimization through computer modeling were always a fundamental part of that.”
He bailed out of corporations five years ago, and began his consulting firm —“too little patience for the bureaucratic wheel spinning”. His primary focus has been systems engineering for advanced applications that includes “spotty” bits of programming for system assessment. Clients include power system R&D companies such as Ricardo and Concepts NREC, as well as smaller systems developers and manufacturers. He worked with Ricardo on some advanced diesel engine application concepts, taking aim at some of the traditional turbine engine markets, he says.
Barbeau learned of TK through his work with Dr. Darrell Pepper. Dr. Pepper is a professor of Mechanical Engineering at the University of Nevada, Las Vegas, and was Interim Dean of the UNLV Howard R. Hughes College of Engineering and a 2004 ASME Congressional Fellow. “I had a client with a serious system problem and required a programming language that could model it. I called Darrell and explained the issues and he recommended TK Solver.” Actually TK was one of three programming environments that Dr. Pepper identified as having potential, but he noted that TK had particular advantages for Barbeau’s application.
“He was absolutely right: the other two did not provide the structure and ease of component modeling required for multi-component aero-thermodynamic system modeling,” Barbeau says.
That makes the business case, as well as the technology case, for TK Solver. And there’s another TK advantage for Barbeau, the one-time Assembler programmer: “I like to know how something is working ‘under the covers’,” he says. “ ‘Black boxes’ can be dangerous. I like TK because the model is wide open and you can track the system component interactions.”
Barbeau uses the link between TK and Excel, which is the other attractive feature that governed his selection. “I rarely use a model to solve one point. Usually I’m exploring data trends to assess the impact of various variables and design choices. This works well with Excel because of its graphical presentation capabilities, using it as the ‘front end’.”
Barbeau praises UTS tech support. He says he and UTS employee Bill Hardin have successfully worked together on several problems. This support was extremely important in the beginning since it was his first experience coding with a language higher than Assembler; Barbeau adds, it is a different world.