Imagine it's just another day at the plant, and you pass by that same old turbine deck on the way to your office next to the unit. The floor of your office is elevated within the structure, and you can tell which mode of operation the plant is in just by feeling the familiar strength and speed of the vibration passing through your chair as you sit in it. In today's petrochemical environment, all kinds of process equipment is elevated within structures for a variety of reasons. Some typical reasons are shortage of real estate, economics or process requirements. Sometimes this can lead to interesting structural dynamics issues, so the real challenge is predicting how the equipment will function in the structure and implementing measures upfront to prevent unwanted vibration.
This challenge is typical when vendors design process equipment such as turbines, compressors, fin-fan coolers, blowers and shakers without considering all aspects of support design, but it is not necessary. What is important is that the designer of the structure has some idea as to how the equipment functions in various modes of operation. There are similar challenges in running high-pressure piping from reciprocating compressors and pumps. The pulsation can cause high-force vibration and, in some cases, resonant vibration due to acoustics. Another good example is a pressure-relief valve. The transient pressure-momentum conditions can lead to high "kick down" forces, causing structural failure. A plan of attack may be as follows:
- Lay out the equipment on a plot plan.
- Try to locate any equipment with potential vibration issues as close to the ground elevation as possible. If that is not possible, put it as close to primary members as possible.
- Meet with the process and mechanical folks and review potential vibration problems. Also, tour any plant facilities with similar equipment and learn from the past.
- Perform preliminary sizing of structural members and lay out the structure.
- Define any known or calculable vibration-forcing functions, or "exciters."
- Develop a simple structural dynamics model of the system.
- Assess the results.
- Sometimes the vibration may be so high, isolators or "inertia blocks" might have to be installed. In any case, it is easier to "tweak" a model than to modify it on the fly during the startup of a new plant. In exotic cases, a variable-speed driver to detune the system might be a solution.
- Meet again with process, production and mechanical teams, and see if things make sense.
But there is also another scenario, which is getting back to your office in the structure. How does one address the vibration problems of an existing system? To address this, we need some knowledge of the process, the mechanical equipment and the structural design. There may not be a quick fix. It always amazes me how people are willing to give the "college try" in search of a quick fix. Sometimes you're lucky, but more often not. Nevertheless, a structural problem can be fixed quickly and cost-effectively for most applications. The following is a plan of attack that, with experience, can work:
- Define the driving forces.
- Develop a structural dynamics model, preferably using the finite element method.
- Perform a field study of the problem.
- Normalize the finite element model.
- Revise model as required to solve the problem.
These are some ideas on structural design, with the sole purpose of getting you thinking. Don't let structural dynamics problems bite you when you least expect.
For more information, visit www.knighthawk.com or call (281) 282-9200.
How to fix structural problems:
- Define driving forces.
- Develop a structural dynamics model.
- Perform a field study of the problem.
- Normalize the model.
- Revise model as required.