AutoTestCon 2005
Sep 26-29, Orlando, FL

Introducing
The Testability Director Version 3.2

How to Select the Right Vibration Testing Service?

 By: Wayne Tustin, Equipment Reliability Institute and Rick Smith, Wyle Laboratories, Inc.

Whose Fault is the New Orleans' Tragedy?  By: Louis Y. Ungar, Editor

Before and beyond anything we can write we want to convey our heartfelt sympathies to the too many victims of Hurricane Katrina all along the Southeast Gulf states of the United States.  Certainly no human being can be blamed for natural disasters such as this one and we don’t intend to blame anyone for the deaths of so many (the true extent still unknown) and the great suffering our people are still enduring as rescue is hampered by the flooding, especially in New Orleans.

Our piercing question in the title is a result of New Orleanean history, dating back to 1713 when the French colonist Jean Baptiste Le Moyne de Bienville ignored his engineers’ warnings about the hazards of flooding in the settlement he planned.  Unconcerned by dangers, he mapped the settlement in a pinch of swampland between the mouth of the Mississippi River, a massive Lake Pontchartrain to the north and the Gulf of Mexico to the south.  The city of New Orleans grew to be one of the most beautiful cities in the United States undisturbed by its water boarders, despite its altitude being below sea level.  Somehow, until this week, the threat of flooding did not materialize.  Even early Tuesday morning it appeared that Katrina’s winds left the water boarders from flooding.  When the massive rainfall followed the Hurricane, however, the levees holding back the lake and river were overwhelmed and when the city’s pumping stations could no longer keep water out, Bienville’s miscalculations could no longer be ignored.  Several levees have 100 to 200 foot (33 m to 67m) gaps that the US Army Corpse of Engineers estimates will take several days to shore up using a massive number of helicopters and will take months to repair.  In the meantime some 80% of New Orleans is under standing water ranging from 4 to 30 feet  (1.3 m to 10 m) deep.  It is too early to estimate the number of dead, and too late to second-guess the wisdom of a businessman who built a wonderful city that mocked the warning of his engineers for nearly 300 years, only to be devastated in less than 300 minutes.

Certainly none of us engineers were around to be blamed for this, nor can we blame city planners who built the best possible flood prevention mechanism they could.  Perhaps this tragedy has a message for us engineers that we may be taking our jobs less seriously then we should.  How many of us would have said to Bienville, “I will not build this city because as an engineer I know the disaster that could befall it and I cannot contribute to a disaster?”  How many of us test engineers stop a product from being shipped to customers when we know it is not tested sufficiently?  Perhaps we can rationalize it by saying we only make (cell phones, or computers, or radios, or PDAs, or oscilloscopes, or memory boards, etc.) – nobody’s life is depending on it.  But we know that it could, and perhaps one day it will, and we will know whose fault it will be.  Let’s keep to our mission as test professionals.  Let’s remember New Orleans (as we are sure we will rebuild her) and make all faults ours – and let’s work to identify and eliminate them before anyone else gets hurt.  We cannot correct the mistakes of our colleagues in the 18^th century, but we can stop from making similar mistakes in the 21^st.

Selecting a laboratory to perform your vibration test can be daunting..   Let’s consider a hypothetical case where a “newcomer” needs a vibration test performed.  (For a temperature, altitude or humidity test, his questions would be somewhat different.) Here, his “To Whom It May Concern” inquiry to several candidate labs might resemble this:

I’m seeking your advice. My company makes flight (or shipboard or land vehicle) hardware that I’ll call “widgets”.  Our Air Force (or Navy or Army) contract states that widgets will be vibration tested to a standard.  My company doesn’t want to invest capital in nor devote space to shaker, power amplifier and controls, and doesn‘t want to pay for operator training.  Rather, my company wants to employ a commercial testing lab.  We have already taken ordinary business precautions such as checking with the Better Business Bureau.  Candidate labs are ISO (International Standards Organization) certified and several belong to A2LA (American Association for Laboratory Accreditation).  Several are somewhat active in the IEST (Institute for Environmental Sciences and Technology). 

I’m assigned to evaluate candidate labs for these tests.  What should I look for during a visit to each lab?   What questions should I ask?

Thanks for your assistance. 

Sincerely, Joe Labseeker

The authors suggest the following questions.  They urge Joe to visit each lab to (1) discuss his test, (2) discuss his questions (and any the lab may suggest), and (3) tour the facility.  Before each visit, he should get information about each facility and its operations.

1. Does the lab have a website Joe can visit for general information?  Hopefully the website will describe the lab’s history and pertinent experience.

2. Does the candidate lab seem willing and able to help Joe understand his test requirements?  Many of the words and phrases in standards (such as PSD in g²/Hz) confuse readers. Some terms seem open to interpretation.

3. Is the lab able to communicate directly with Joe’s customer at Joe’s request?  What if something unforeseen needs to be clarified before the test starts?

4. Have test personnel documented any formal training in vibration testing, measurement and analysis? 

5. Have test personnel shown Joe, in their personal (or the lab’s) library, texts that answer Joe’s many questions?  Joe wants his company library to obtain those texts.

6. Is the lab able to prepare a Test Plan or a Test Procedure that Joe and Joe’s customer can review and approve well before the test?

7. Does the lab have a sufficient number of shakers that the lab can match Joe’s probable schedule??  These must be sufficiently large (adequate force, adequate table area, adequate frequency range) to do Joe’s job.

8. Joe’s customer may call for widget vibration tests at temperature extremes. Can the eligible lab combine environments?

9. Can the potential lab perform other types of environmental tests, if required?

10. Can the aspirant lab help Joe organize complex test programs?

11. Does the hopeful lab have the resources to design, manufacture and experimentally evaluate before use (against what standards?) the fixture(s) that attach widgets to the shaker? Or will an outside service be needed?    

12. Can the possible lab safely machine and weld magnesium (which has been recommended to Joe) and install threaded inserts to firmly attach our widgets?

13. Will the competing lab provide and attach accelerometers and measurement instrumentation systems, including the display and recording systems, not only to measure and record vibration input to Joe’s DUT (device under test) but also DUT vibratory responses?  

14. Does the contesting lab have a significant inventory of accelerometer channels?

15. Will the lab assist in selecting accelerometer locations?

16. Beyond the specified test, Joe’s structural analysts and design engineers want to correlate and validate their mathematical widget models, and so may ask for additional accelerometer channels.  Can the lab accommodate this?

17. Can the candidate lab provide instrumentation to monitor widgets during test?

18. Some widgets require special power supplies, at 400 Hz, for example.  Others require 28 volts DC.  Does the eligible lab provide various power sources?

19. Some widgets require hot air or other gas flow.  Others require hydraulic oil, gasoline and/or diesel fuel. Some require pressurization.  Some require static structural loading.  Which of these can the aspirant lab provide?

20. In addition to ASCII and EXCEL formats, in what other formats can the hopeful lab provide data?

21. If DUT resonances result in test failure, will lab personnel help Joe (and Joe’s  designers) learn from those failures?

22. Can the potential lab respond on short notice, in case of sudden need for testing?

23. Shortly after Joe’s visit, Joe hopes to receive from several possible labs several written proposals for the anticipated vibration testing, with fixture design, manufacture and evaluation shown separately.

24. Are Joe’s colleagues, and other witnesses that may be required, welcome to observe tests?

25. Joe has learned that there are government agencies established that require their representatives to witness certain tests.  Does the lab have familiarity and close contact with government witnessing agencies?  Does the lab have a standard procedure to timely notify the government witness (to avoid test delay)?

26. Is the suggested lab authorized to handle sensitive information?  Do lab personnel have government security clearances, and does the candidate lab have a Facility Security Officer?

27. Joe will need a report to show the results of each test. What kinds of reports does the recommended lab offer? 

28. In addition to digital photographs of each setup, what else does a typical test report include? 

29. How quickly can Joe get reports (after each test is completed)?

30. Can the competing lab submit reports electronically, as well as on paper?

31. Can the contesting lab provide preliminary test data for review as tests progress?

32. The final report should list all equipment used.  The measurement equipment used must have been calibrated with records traceable to NIST (the National Institute of Standards Technology).

33. Finally, what is each lab’s hourly rate?  Or have the labs other ways to provide estimates for their services?  To what services do those rates apply?  For what services will Joe’s firm not be charged?   

Bert Horner - The Test Connection, Inc.

The Staff.

Knowing who is testing your product is as important as knowing what equipment is used. It is not sufficient to have an operator who runs the boards through the ATE tests. They also need to have a higher level of technical competency to verify the faults on the boards. Consider, for example, that if the fixture or probe does not make contact to a point on the board the ATE will likely (and mistakenly) call out the board as faulty.  If your test partner does not have the right person analyzing the failures, this might result in the “sin of sins,” namely, failing good products.

Knowing What Is Being Offered.

Test engineering and testing services can sometimes be dismissed with the incorrect notion that everyone is equal and comments like “a test program is a test program” or “the system does all the work”. Such clichés and generalities have no scientific bases. There are distinct differences in what is being offered by a test service, both in documentation and in quality of service. The key here is to find out what is being supplied from each potential partner and evaluate it to see what that buys your company. 

Getting It There.

Sometimes distance can seem be a concern with one’s partner, but remember, if boards are shipped overnight, whether it’s 50 or 250 miles, it’s still overnight. Communication, whether it’s via emails or phone conversations, is key to successfully meeting time schedules.  The cost factor always comes into play and this is where it can either make or break dealing with the test partner.    As in many business operations, however, the volume users always get the best pricing and this holds true with the shipping companies too. 

Selecting one’s test partner is not rocket science, but there are some key items that need to be looked at during the evaluation. Getting to know what the test development and test service partner is about as a company will give you a better feel of how they will be able to support your production team.

• If your friend forwarded this newsletter to you, please register as a member and receive The BestTest Newsletter -- absolutely free!
• If you wish to update your news preferences or cancel the subscription, please unsubscribe.
• If you have any questions, please email experts@BestTest.com