SRC Forum - Message Replies


Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum (src_forum@alionscience.com )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Back to message list Topics List About this forum
Original Message:

Posted by: ANIL KUMAR (anilkumar_sng@yahoo.com )
Organization:CENTRE FOR RELIABILITY
Date posted: Thu Mar 7 6:49:23 US/Eastern 2002
Subject: MECHANICAL RELIABILITY
Message:
What is the life characteristic curve and failure distribution for mechanical parts. Electronic parts follow bath-tub characteristic and exponential distribution.What about mechanical components?


Reply:

Subject: MECHANICAL RELIABILITY
Reply Posted by: Johnny Campbell (jcampbell2@bellhelicopter.textron.com )
Date Posted: Wed Mar 13 12:35:26 US/Eastern 2002
Message:
For starters I would question electrical components having a bathtub curve failure trend. Electrical components tend to have a high infant mortality then leveling off to a fairly flat failure distribution. Mechanical components can fall in up to 6 failure patterns with simple items having a low infant mortality and increasing in failures starting at the "wearout age". Most complicated systems and even most bearings follow a more random failure pattern. From a reliability centered maintenance perspective I would refer you to RCM II by John Moubray section 12. I don't have any strictly R&M reference that spells it out as clearly.


Reply:

Subject: Mechanical Failure Patterns
Reply Posted by: Patrick Hetherington (phetherington@alionscience.com )
Organization: Reliability Analysis Center
Date Posted: Wed Mar 13 13:42:28 US/Eastern 2002
Message:
The bathtub curve only has value in explaining three separate and distinct failure phenomena’s i.e. infant mortality, random failures and wearout. Studies have shown that less than 5% of all components actually follow the bathtub curve. Usually one or two of patterns are dominant. Mechanical components generally will follow a Weibull or Lognormal time dependent failure distribution. The parameters for these distributions are dependent on the component type and operating stresses. Systems, whether mechanical, electronic, or electro-mechanical that have many parts (and thus many individual failure patterns) will exhibit a near random failure pattern which becomes very random if failed components are replaced with new components.


Back to message list Topics List About this forum
Reply to this message