By Leslie Langnau, Managing Editor

The use of mechatronics principles should make new product/device design faster, easier, and deliver fabulous and inexpensive products. But many engineering groups grapple with this design approach. Why do some groups work while others struggle?

We’ve heard about the promise of mechatronics for many years. Off-the-record, we hear comments about the “problems with mechatronics.” Some engineering groups get it and apply it with great success. Others don’t even want to hear the term. But there is no denying that whatever you call it, this approach to design is necessary in today’s world of multifunction, multitasking equipment and systems.

You don’t have to refer to this approach as mechatronics. Said Kevin C. Craig, Ph.D., Professor of Mechanical Engineering, Marquette University, “I define mechatronics as multidisciplinary engineering system design.” This definition is much more descriptive.

A number of engineers and managers are looking into why this approach appears to either not deliver on its promises or why it only works for some. Their research so far indicates that there are three main problems: Education, corporate structure, and the lack of truly collaborative design tools.

Education should break down the walls, but …
Years ago, the wall between manufacturing and engineering had to come down before industry realized measureable improvements in productivity. A similar situation faces those who wish to implement mechatronics, only this time the walls that must come down are those between engineering disciplines.

Education has played a role in building those walls, partly in response to demands of last century’s corporations and labor unions who segregated engineering manpower into separate functions; mechanical, electrical, and others. Today, inertia maintains the status quo with many universities and colleges continuing to segregate engineering disciplines. Even the professors don’t collaborate with each other! The result is mono-functional engineers (a new term that you may hear more of soon).

This singular focus has created engineers who speak a different engineering language from each other. Noted John Pritchard, global product manager, Kinetix Motion Control, Rockwell Automation, “At a recent workshop with 50 engineers pulled from all areas of a company, the language discrepancies were clear. We were discussing how to take a mechatronic approach to robot design. In the conversation, the mechanical engineers spoke about their struggles with reverse dynamics. The control guys said their biggest challenge was Cartesian to joint transforms. This conversation went on for ten minutes before they realized they were talking about the same thing, just using different words. The control guys were thinking about math while the mechanical guys were thinking about links, angles, and so on. For this group, the solution was to speak mathematics.”

A few educators are aware of this issue and are initiating a profound change, which we will go into shortly.

Another educationally based problem involves awareness; the decisions any engineer makes can affect other engineers’ choices for a design. “Lack of such awareness trips up many projects,” agreed Pritchard. “The choice of material is a fairly common decision that causes problems. For example, in the design of a reciprocating mechanism controlled by a servo system, a mechanical engineer may choose steel over aluminum. The steel may be more readily available, less expensive, standard practice, and so on. The control engineer, however, is now confronted by several constraints because of this choice. The servo motor must have three times the peak torque to accelerate at the same rate it would have needed had the mechanical engineer gone with aluminum. In addition, the design will need a bigger motor, bigger drive and circuit breaker, heavier wiring, bigger amp supply, bigger everything.

“The mechanical engineer may have no idea how the design of one part impacts the overall machine. A 10¢ per part saving may really result in up to $10,000 additional cost in order for the control engineer to deal with the larger inertia. And there are many choices like this; couplings, compliance, gearbos backlash, and so on,” continued Pritchard. “And the control engineers and the electrical engineers do the same thing; trapezoidal acceleration, for example, can excite resonances which can frustrate the mechanical engineers. Another example is the common practice of putting acceleration at 100% rather than a lower percentage, which can impact wear.”

“And control systems is one of the more important disciplines for mechanical and electrical engineers to have some knowledge of,” added Razvan Panaitescu, manager of Engineering for Mechatronics, Siemens. “It stands in between mechanical and electrical. You don’t need to know electronics deeply, just enough to model.”

A few professors have witnessed this lack of awareness and are developing programs that will not only solve it, but that will create shifts in the traditional engineering labor pool.

A change is coming
Ken Ryan, Director of the Center for Applied Mechatronics at Alexandria Technical College in Minn., spoke about what educational institutions can do to resolve these issues. He sees the engineering role shifting into two main categories: the specialist engineer (which is probably most of you) and the cross functional engineer.

The Specialist or mono-functional engineer is the traditional Mechanical Engineer (ME), Electrical Engineer (EE), Controls Engineer (CE) and so on. These individuals are experts in their chosen field. “Industry will always need these individuals,” said Ryan, “but not in the numbers that they have hired previously. I see a day when a company’s engineering labor force will consist of about 20% of these specialists.”

The Cross-Functional engineer is essentially the mechatronics engineer. This individual has more of a breadth of training, learning much about multiple engineering disciplines but typically not to the depth of the specialist engineer. These are the people corporations need to make mechatronics programs successful. Noted Ryan, “I think these people will make up about 40% of the engineering labor pool in a typical corporation.”

The cross-functional engineer can be further divided into two categories:

The Technologist: This individual is meant to be the functional extension of the traditional engineer; they implement the designs of the specialist. She/he is a member of a mechatronics team and will often function as a liaison among the specialists. This individual’s role is coordinative and integrative, both vertically and horizontally.

The Technician: This individual does what an engineer tells him/her to do. They are responsible for installation, service, and maintenance of mechatronically designed equipment. The remaining 40% of a corporation’s engineering pool will likely consist of these skills.

Mechatronics requires that either you master more than one or two engineering disciplines, or you develop a group of generalists to support the specialists. The cross-functional engineer will never replace the specialist engineer because they do not have a comparable depth of knowledge.

At Alexandria Technical College, the program is very successful. The college is in the middle of a huge packaging machinery area. By developing a cross-functional engineering program, graduate students find placement in all kinds of industry including transportation, mining, marine, automation, and other areas. “Once we took ourselves out of the packaging box,” said Ryan, “then we started finding lots of people interested in our students because these fields are all trans-functional fields.”

Corporate structure needs to nurture collaboration, not impede it
Global locations and engineers grouped by discipline do more to create miss-communications than solve it. “The biggest problem is interaction among disciplines,” noted Panaitescu. “Many corporations still physically group engineering disciplines so that engineers either work only with other engineers of their discipline, or they work in isolation.” The most successful companies have an open culture and nurture it.

Then there is the issue of cooperation, which can be sidetracked by corporate structure. “Engineers are naturally competitive,” said Panaitescu.

“But companies with more successful mechatronic design programs leverage the competitiveness between project-focused cross-functional engineering groups rather than having individual engineers competing against each other,” noted Pritchard. “The strategy of ‘which group will produce the best machine’ works well.”

Successful users of mechatronics also use a common design process that everyone sticks to. “One goal of a common design process is to ensure engineers check with each other throughout, ensuring that one decision does not impede future decisions from other engineers involved in the design,” said Panaitescu. “Corporations do not need to mandate that engineers attend communication classes; that is not the issue.”

Part of this common process involves the creation of a requirements document. It lays out in the beginning, what the design must do. Noted Panaitescu, “it is not often used because its not very interesting paperwork. But it can help speed product development.”

“The first step is to sit with the customer and decide what the device must do,” continued Panaitescu. “It will not significantly differ among projects. But if you define soundly, thoroughly, then everyone thrives. Naturally, the requirements will include performance, precision, timing, vibration and so on. But the requirements should also include how a system performs and how it will be designed; did you optimize that machine, reduce its carbon footprint? How much material did you put into the machine? These factors should be part of the mechatronics concept. The requirements change as we change. If you have such a process that incorporates physical mechatronics concepts with requirements concepts, then you have everyone in the team looking at the same goal, a common perspective.”

Proctor & Gamble, for example, has resolved many of these issues. Said Craig, “P&G has developed internal programs that have broken down the silos, embraced mechatronics, developed integrated design, and offer in-house courses that look at the mechanical, electrical, and controls. It’s doable.”

The need for truly interoperable software tools
The biggest issue with the various CAD and other product-development tools is that they do not offer the required level of interoperability that lets a controls engineer interact with the design of an electrical engineer.

“At first glimpse,” said Craig Therrien, product manager, Dassault Systèmes SolidWorks Corp., “it might appear that a simple movie of a machine in operation is all that is necessary for a collaborative mechatronics approach.

However, although a 3D-based mechanical CAD animation of intended machine function is a huge improvement over 2D drawings – and can help pinpoint potential collisions – it does not convey important engineering information that electronics and controls engineers need to select, size, and program the appropriate system. Nor can an animation alone help engineers factor the effect of their decisions into the mechanical design.”

Something more than moving pictures is needed to take advantage of mechatronics. Programs should provide control engineers access to mechanical engineering information, such as mass, material properties, moments of inertia, and force/torque requirements, to choose the most suitable electronic control mechanism. Mechanical engineers need to combine the loads created by specific electronic controls with the output of dynamics analyses to validate a system’s structural integrity. Controls programmers need to be sure the system functions as intended without any mechanical or electronics systems issues. In short, everyone involved needs an integrated mechatronics design environment that moves mechanical and controls design information in both directions. This helps the team to make important decisions and design modifications during the design cycle rather than as a result of costly prototyping.

Two soon-to-be-released examples of such a mechatronics environment are the integration between SolidWorks® Motion kinematics and dynamics analysis software and controls automation packages LabVIEW® from National Instruments and Motion Analyzer® from Rockwell Automation.

“With these integrated tools,” continued Therrien, “the mechanical engineer can model a machine in SolidWorks 3D CAD software and conduct kinematics and dynamics analyses in SolidWorks Motion software. Then, electronic systems engineers and control programmers can access the entire motion simulation from either LabVIEW or Motion Analyzer, including pertinent engineering data such as force, torque, and friction requirements, to design and program the control system. Finally, the mechanical engineer can access detailed controls information, such as the type of device or the size of the motor, to conduct additional stress and vibration analyses.

Noted Marc Monaghan, engineering systems manager at Hartness International, a manufacturer of packaging systems, “We are constantly looking for ways to reuse our design data, and the merging of mechatronic control simulation with mechanical design is an excellent approach. This integration extends the benefits of kinematic simulation into the arena of control programming, allowing the initial concepts of control logic to be designed and tested simultaneously with the mechanical function that it needs to control.

“Project timelines are more aggressive than ever, giving us much less time to develop designs with iterations of physical prototyping,” Monaghan added. “The integration of 3D modeling, analysis, and control development allows us to identify potential issues and opportunities for innovation long before the first part is produced. It is another step towards getting more problems solved during the design phase of a project, when cost savings and efficiency improvements deliver the most benefit.”

Engineers at NCR Ltd., a leading manufacturer of ATM machines, also desire and require better product design tools. According to Dr. John White, chief engineer at NCR, “We use mechatronics to optimize performance. An interoperable program, such as the SolidWorks and LabVIEW connection, gives our R&D teams the ability to develop a digital prototype in advance of a physical build. LabVIEW controls the motion trajectories while SolidWorks is used to calculate the driving forces, power requirements, and stresses. Connecting the control software to the mechanical assembly provides our engineers with the data needed for full design analysis and optimization. For us, it’s all about reliability through optimization.”

Dassault Systèmes SolidWorks Corp.
www.solidworks.com

National Instruments
www.ni.com

Rockwell Automation
www.rockwellautomation.com