SRC Forum - Message Replies
Forum: Reliability & Maintainability Questions and Answers
Topic: Reliability & Maintainability Questions and Answers
Topic Posted by: Reliability & Maintainability Forum
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998
Posted by: Matt Wiltshire
Date posted: Thu May 27 11:51:59 US/Eastern 1999
Subject: MIL-STD-810E Salt Fog Test
How many hours would the MIL-STD-810E Salt Fog Test (48 hours, 5% salt solution, 95 degrees F) be equivalent to in a worst case land environment such as the coast of Florida in the summer and a bast case in Nebraska (if there is a better one let me know)? If numbers can be directly applied to the 48 hour test, can it be applied linearly to an extended test (much > 48 hours)? Can you recommend any studies that address this?
Subject: Test Acceleration
Reply Posted by: Dave White
Organization: Spectral Solutions
Date Posted: Tue Jun 1 14:29:49 US/Eastern 1999
That is the $64,000 question! The trouble is, there just isn't enough information to answer it. Things like product geometry, activation energy and material selection play too big a part, so without these any estimate of test acceleration comes out as little better than a guess. Your best bet is to compare test results to actual field results for your product (if these are available). If it is a new product, and field data doesn't exist yet, you may get some insight from "similar" products that do have both field and lab data available. In this case "similar" doesn't have to be functionally similar, but similar materials and geometries.
I once had a manager who just wouldn't accept this. In the end he told his boss that one day of testing equalled one year of field use (he didn't say where in the field, as you point out - all fields are not the same!). I don't know where he pulled this number from, and I would not like to try and defend it. If you do find any referance material that addresses this I'd like to know about it too. Good luck, and please post any info you find. Thanks.