This article describes a failure analysis precipitated by system outages caused by failures in a snubber circuit. The subsequent failure investigations include electrical stress analysis and investigation of capacitor component failures. The effects of various electrical stress and environmental factors and their impact on various capacitor failure modes is discussed. Several clues lead to capacitor corona as a possible culprit, with physical part inspections leading to the final source of the failure.
As an application engineer I once had a customer call me on a Friday afternoon needing help that week-end to get a gas-turbine generator up and running. The customer would not start the generator unless the accelerometer monitoring the vibration and it related electronics were working and verified.
Throughout the week-end I talked with the engineer multiple times to explain how to calibrate a charge-mode accelerometer, what parts he would need, and what kind of tolerance on the results he could expect.
By Sunday afternoon the generator was running and the engineer was very, very happy. Without my help he would not have been able to get that generator back on line for an additional two or three days.
Yes, engineers sometimes don't mind having their week-end interrupted! The 'best' pay is often that of a very satisfied person whom you've helped through a sticky problem. Money is a necessity, but the sure knowledge of having done a good job to help someone else has no second.
AEi Systems was established as Analytical Engineering in 1995 to support increasing demand for Worst Case Analysis in the Space market. Services provided by AEi Systems include Circuit and... Read More
Technical Paper
0
1
Download to Print
Add to bookmarks
1 comment
write a commentRDentonSr Posted Dec 23, 2010
As an application engineer I once had a customer call me on a Friday afternoon needing help that week-end to get a gas-turbine generator up and running. The customer would not start the generator unless the accelerometer monitoring the vibration and it related electronics were working and verified. Throughout the week-end I talked with the engineer multiple times to explain how to calibrate a charge-mode accelerometer, what parts he would need, and what kind of tolerance on the results he could expect. By Sunday afternoon the generator was running and the engineer was very, very happy. Without my help he would not have been able to get that generator back on line for an additional two or three days. Yes, engineers sometimes don't mind having their week-end interrupted! The 'best' pay is often that of a very satisfied person whom you've helped through a sticky problem. Money is a necessity, but the sure knowledge of having done a good job to help someone else has no second.
reply