RSS Feeds
Back to Basics
FEA still key for verifying physical tests
Paul E. Teague -- Design News, September 12, 2004
|
| Test sample: Here's what Herrera, Stafford engineers studied to determine failure cause. The dome of the cover was unrecoverable from the accident site. |
The wrench did it.That was the conclusion forensic engineers at Herrera, Stafford, and Associates (HS&A) reached after a series of tests in their lab on the reasons for failure of a compressor head cover at a Texas oil well. The cover failed after the compressor was restarted following repairs, releasing natural gas that exploded, injuring workers and causing millions in damage. The tests showed that the person holding the wrench was the culprit. He used too much force. Lab tests were verified using finite element analysis (FEA).
Workers had used a pneumatic wrench to tighten the bolts on the head cover instead of the recommended torque wrench. The lab tests with a similar wrench revealed that they got the recommended torque (80 ft lbs) with only 20 psi in the air supply. There were several other errors during the repair operation too, including not lubricating the bolt/nut assembly, not tightening all eight bolts equally, and not following the recommended torquing sequence.
But ensuring that the cause of the failure was the force used to tighten the bolts and not design or manufacturing flaws was critical, given the liability potential. So, HS&A engineers went back to the classic use of finite element analysis software, and used it to verify their physical tests. They used Algor software for the job.
First, they created structured-mesh models so they could specify the exact location of nodes for the loads and constraints, says Anselmo Najera, HS&A design and analysis engineer. To simulate friction between the cover and its aluminum seal, he modeled a thin layer of elements and gave it weak material properties so it would easily deform.
For all models, he completely restrained the aluminum seal and applied the material properties based on information from the manufacturer. Among the loading conditions necessary to consider: the force of the bolts holding the cover in place and the pressure within the compressor. Najera considered a variety of bolt force loads ranging from 1,024 to 15,000 lbs. He also considered a scenario where the bolt forces were not the same for all bolts. He applied the maximum pressure of the compressor, 600 psi, to the inside surface of the dome for most of the models. In all, he analyzed 18 variations of the model. For each linear static stress analysis, he looked at the maximum principal stress and compared it to the yield stress of the material. He also looked for areas of stress concentrations.
"We found that neither the groove nor the asymmetry of the dome was significant enough to cause failure," he says. "The most significant factor was the force used to tighten the bolts."
Other projects where Herrera, Stafford engineers have used FEA to solve design problems were a study to determine why one company's light bulb filaments kept breaking, and others involving automobile-accident reconstruction.
While FEA is increasingly used today to optimize product design, the kind of application HS&A used it for is the technology's roots. "The traditional use of FEA was to verify physical test results," says Suchit Jain, an executive with SolidWorks, which develops and markets COSMOS FEA software. "Physical tests don't tell why things fail. You get many more data points with FEA."
|
| In spec, out of spec: FEA was used to study 18 variations of the compressor head cover. Bolt force on the cover (above) was the same for all bolts and within recommendations. The bolts were unevenly torqued (below) and forces exceeded recommendations. |
|
