By Greg McMillan and Stan Weiner
Greg McMillan and Stan Weiner bring their wits and more than 66 years of process control experience to bear on your questions, comments, and problems. Write to them at email@example.com.
Stan: Process control has been very good to us. More than the money earned, there was a sense of pioneering advancements in the understanding and implementation of what works best.
Greg: More than any other field, process control is open to innovation because there is no script or guide. Undergraduate programs in process control are rare or non-existent. Industrial short courses and books provide pieces of puzzles that are as diverse as the products in the process industry. When I worked on the initial development of the ISA Certification of Automation Professionals (CAP) program, I was surprised that there was no set of books—including my own—that would develop a new engineer into a proficient practitioner. The books from industry were not written for teaching. Numerical examples and test problems were"missing in action."University textbooks generally focus on the math needed for graduate degrees and research. Some textbooks have provided an overview of instrumentation, valves and control systems, but often the representation is dated and without guidance as to what a process control engineer really needs to know or do on the job.
Stan: Our courses in chemical engineering and physics did little to help us to select, specify, configure, install, checkout, start up and maintain automation systems. In control theory classes, we had perfect measurements and valves, negligible dead time and state space controllers. When these courses did show instrumentation systems, the figures often had pneumatic signals, DP flow meters downstream of control valves, and actuators without positioners—all disasters.
Greg: Quantum mechanics, the theory of relativity and astrophysics didn't help us much in figuring out why a loop was messed up, although the Heisenberg Uncertainty principle (http://plato.stanford.edu/entries/qt-uncertainty/) explained why our management couldn't get a fix on both where we were and what we were doing. We hoped that there was conservation of spin, so that executive decisions in the wrong direction might be resulting in decisions in the right direction somewhere in the universe.
Stan: Talking about being out of touch, why are operator graphics so functionally primitive? How come we don't have 3-D plots of batch, column, fluidized bed, reformer and sheet profiles? Why don't we have dashboards of equipment and process performance? Why no live maintenance records? How come sensor locations and the signals for feedback, feed-forward, cascade and split range control are left to the imagination?
Greg: I know we don't want to clutter the screens, but wouldn't better integration and visualization of data and control system functionality improve the operability and maintainability of plants? Why not an engineering view on the operator console that is an evergreen combo process flow diagram (PFD) and simplified process and instrument diagram (P&ID)? Why not have a drill down to stream info both static and dynamic? Coriolis meters could provide live density, temperature and inferential composition data besides mass flow. Most PFD and P&ID drawings are out of date as soon as the plant starts up.
Stan: This lack of a comprehensive picture of our profession and plant performance is a problem for everyone, especially the new engineer. To get an idea of how a new engineer comes up to speed, Greg interviewed Sarah Tremblay and Ted Stillwell, who have three months and 30+ years of experience, respectively.
Greg: Sarah, what exposure to process control did you had have prior to this job?
Sarah: Process control was only mentioned in passing in some of my courses in mechanical engineering. We had a lab, but the scenarios were too scripted—no room to explore. I think I would have benefited from a perspective of the role of process control and the types of applications and jobs in the process industry.
Greg: Ted, are there any courses offered at your company?
Ted: We have "lunch and learn" presentations and demos and half-day seminars sponsored by suppliers.
Greg: When I started, I was sent to an eight-week intensive course and lab on instrumentation and control. I was then sent to E&I construction where I was responsible for the installation, checkout, and start-up of automation systems for a half-dozen production units. In between construction assignments I was given the chance to be the lead engineer on a project under the tutelage of Stan. Do you have developmental program for new engineers?
Ted: Companies today don't have the money or time to invest in training programs. Also, projects are quicker and dollar-driven, so you end up working on several at the same time instead of just one big job. Luxuries of the 1970s and 1980s have gone away.
Sarah: An intensive multi-week course would not benefit me much right now. You need enough experience and time on the job to put what you would learn in a course in the proper context.
Ted: Sarah has been given pieces of a number of projects. I provide guidance and help as needed. Sarah has had a chance to see control systems assembled at a panel fabricator. She is learning by doing and asking a lot of questions. It takes a while for it all to come together—what was done and why, and how it all fits together. Part of this is working with systems in the field at the completion of projects.
Stan: It looks like we're relying more than ever on the time-tested method of learning on the job and pairing up new and experienced engineers.
Greg: Maybe some interactive labs, such as the one I'm teaching for the ISA New Orleans section on March 2-3 can provide basic answers, but this assumes the person has had enough experience and guidance to appreciate the questions. Since expertise is disappearing at an alarming rate, companies should hire new engineers right now while there is still time to transfer knowledge. ( See the Control Talk series, "Going, Going, Gone," Aug '09, www.controlglobal.com/articles/2009/ProcessIndustryExperts0908.html; Sept '09, www.controlglobal.com/articles/2009/ProcessControlExpertise0909.html; and Oct '09, www.controlglobal.com/articles/2009/ProcessControlExperience0910.html). Packaging expertise for retirement and holding back on hiring more than just mortgages the future. It may lead to foreclosure. The future is now.
Stan: The"problem solving flow sheet"in the cartoon by Ted may be more applicable than ever, except you will probably be stuck in the"Do Loop.”
Greg: For more comic relief we finish with the conclusion of our list started last month from the Sanders family to help you embrace your inner geek.
You Know You're a Geek If...
12. Your work clothes are not nice enough to be dinner clothes
11. You speak a second language—and that language is Vulcan
10. Your most critical"heartbeat"is between your server and client
9. You mail punch cards via FedEx as your version of "electronic mail."
8. You know what "DOS"or "UNIX"stands for. (DOS is disk operating system, and UNIX is uniplexed information and computing system [originally UNICS.])
7. You find the naming of UNIX humorous. (Uniplexed doesn't mean anything.)
6. You spend extensive time debating the difference between explorer.exe vs. cmd.exe
5. You can name every MS operating system since DOS and when it was released
4. You think this list would work better as a Top 8 or Top 16. (I think binary jokes are still funny, too.)
3. You come dressed as a DCS to a costume party
2. You can recite complete lyrics to the "Elements"song. (www.youtube.com/watch?v=6b2Uy1TDAl4)
1. You read this list and laughed at any of the items.