Equipment Reliability Institute
ERI News - your reliability newsletter
May 2005 - volume 19

Wayne TustinHello, readers -

In this space, in too many prior issues, we've told you about my new text on "Random Vibration & Shock Testing". Some readers may have doubted that the book would ever be published. Below you will see photographic evidence that the book is here!!! At left, how it looks unwrapped. At right, with each of the corners protected inside shrink-wrapping around a hard cover that further protects the 444 4-color printed pages. The current description is at our website. In late April, we received a few copies by air to use at ERI's course at Aberdeen Proving Ground, in Maryland. We are now shipping to our back orders and we are accepting new ones. Please visit our purchase form page or give us a call at (805) 564-1260.

Down below we have some useful Steve Brenner remarks re fixtures, some Joel Newberger ideas re "natural cooling" of electronics, another vibration myth explained by John Starr, mysterious "deciBels" explained by yours truly and the 8th installment of Bob Renz' very practical "test lab musings".

Best wishes,
Wayne Tustin
*******************************

A bad fixture can ruin a good vibration and shock test

Did your DUT (device under test - be it wiring card, box, rack or system) recently fail an important vibration (or shock) test? Is it possible that you inadvertently overtested the DUT? Might a fixture resonance have been the real reason (rather than a DUT weakness) for failing that test?

Or let's change those questions from past tense to future tense. How can we give our DUT every chance to pass an upcoming important vibration (or shock) test? How can we be sure that we don't inadvertently overtest the DUT? How can we avoid (or moderate or shift) some fixture resonance that might cause an overtest?

We must obtain or develop some fixture design, fabrication and usage expertise. We can, of course, obtain some help by "going outside" to a fixture specialist firm. Even then, we need help in fixture usage. Where, for one example of perhaps a dozen needed decisions, shall we place the control accelerometer? See Steve's list, which follows.

  1. From what material should we fabricate the fixture?
    What fabrication method?
  2. What is the maximum allowable fixture weight or mass?
  3. Where should we put the control accelerometer?
  4. How many control accelerometers will we need?
  5. What control strategy will we use?
  6. What resonant frequencies must the fixture avoid?
  7. How much damping do we need?
  8. Should we use damping foam?
  9. What are our form factor limitations?
  10. Must one fixture serve in all three axes? Or, if necessary, can we use additional fixtures?
  11. What specimen mounting hardware is needed? How will the specimen attach to the fixture?
  12. Where must the combined fixture + speciment center of gravity be?
  13. Do we need inserts? What type?
  14. Will we need an adapter plate to secure the fixture to the shaker
    armature?
  15. Against what standards should we experimentally evaluate the fixture?

A big step in developing the needed expertise is specialized training. Steve Brenner can assist at your facility via (1.) consulting on a specific fixture or test or via (2.) formal classroom training at your facility to avoid future fixture and test problems. Alternately, you can send individuals to Steve's vibration and shock test fixture design course at Las Vegas February 20-22, 2006.
*******************************

Naturally Cooled Electronic Equipment
by Joel Newberger

“Natural cooling” is Nature’s way of cooling electronic equipment.
Don't overlook that the natural cooling mode includes two thermal transfer mechanisms . . .(1.) “free” (or “natural”) convection and (2.)

During natural cooling, radiation is generally the more dominant thermal transfer mode when the electronics package’ external surface is painted (or anodized) any color, and as long as it is a flat, dull paint. The paint color is an important radiant design factor when equipment is illuminated by a high temperature heat source like the sun.

At sea level, naturally cooled equipment - when external surfaces are painted - can transfer as much as 70 percent of packaged heat to the surrounding environment by thermal radiation, and can transfer the remaining heat load by natural convection. This 70/30 percent thermal transfer mode split does not require large temperature differentials between packaged surface and the surrounding environment. This type of mode transfer split occurs when temperature differentials are as low as a few degrees. When external surfaces of packaged equipment are unpainted, the above transfer mode split is reversed; i.e., a minimum of 70 percent is transferred by convection. Therefore, an electronics package with unpainted external surfaces transfers heat to the surrounding environment at a greater temperature differential (between surface and ambient) than when external surfaces are painted.

When electronic equipment, located outdoors, is naturally cooled, Nature helps equipment cooling to a greater extent than when located indoors. Outdoor equipment cooling is enhanced since bulk average air velocity (i.e., wind speed) is higher than occurs indoors. The higher outdoor velocity removes surface heat more effectively then the “still” indoor air ambient. As a result, the 70/30 transfer mode split does not apply for naturally cooled outdoor equipment. Cooling of outdoor equipment is enhanced by the radiation mode if the external surfaces of packaged equipment “see the sky.” The “sky” temperature is not the outdoor ambient temperature, but behaves as an atmospheric thermal radiation heat sink, and is sometimes referred to as the “sky vault” temperature. The sky vault temperature can be as much as 20o to 40oC below the prevailing ambient temperature. Of course, solar illumination can add an additional thermal load to outdoor operating equipment.

Solar impingement on equipment surface is negligible with heavy cloud cover. In addition, cloud cover acts as a radiation shield and limits heat transfer between package external surfaces and the sky vault. In this manner, cloud cover reduces equipment thermal load while simultaneously limiting equipment surface radiant cooling to sky vault. As a result, Nature keeps the heat flux input/output “ledger” balanced.

Naturally cooled airborne packaged electronic equipment, when located in an unpressurized compartment, experiences both thermal enhancement and degradation as aircraft operational altitude increases. The convective/radiative, transfer-mode-percent split varies with altitude. At sea level, natural convection transfer to surrounding environment is maximum. As altitude increases, rarefied air reduces convective thermal transfer while radiative transfer increases, since aircraft skin (i.e., radiation heat sink) temperatures are reduced. Therefore, as aircraft altitude increases, the convective/radiative transfer changes from a dominant convection mode to one where radiation is dominant. Above 50,000 feet altitude, the external surfaces of a naturally cooled electronics package are radiation cooled with almost negligible convective heat loss to surrounding environment.

Whether your electronic equipment is naturally cooled on the ground, at sea, or in the air, heed an old TV commercial: “Don’t fool with Mother Nature.” Understand her!

Joel is President of Thermalogics, Inc., and a principal in SNA Engineering. SNA specializes in mechanical design/packaging and in thermal/structural analysis of electronic equipment used in both commercial/industrial and military applications.

For a listing of available short courses (listing of 15+ short course descriptions) and more information about Joel, please visit his page at ERI. Joel invites readers to "mix and match" these courses and thus to identify electronics cooling training that can in 2005 meet at their facilities.

(back to the top)
*******************************

Understanding Vibration Myths: Myth #2
by John Starr

Electronics Vibration Myth #2: Discovering a product's weakness through HALT or Step Stress Testing can cause an Environmental Stress Screen to be artificially lowered in intensity due to the weakest link.

Tests to failure, as in Highly Accelerated Life Tests (HALT) or Step Stress fragility testing are often wrongly criticized for producing failures. This criticism is not uncommon in fragility testing, ruggedization or Environmental Stress Screen (ESS) development.

Fragility Testing
Fragility testing to define a product's weaknesses can be performed through Step Stress Testing or through the HALT process (stopping short of the product redesign cycle). Fragility tests are often criticized when failures are found at equivalent environmental exposure levels above those expected in the field. Criticism occurs without considering the strength of the test unit relative to the production population. Electronic products are exceptionally vulnerable to strength variations. These variations occur due to lack of control of design details critical to life capability. These details include dimensions and material properties of many parts. As an example, solder shape can significantly affect life.

Service life failures occur due to high load to strength ratios. There are distributions of exposure and distributions of part strength. Failures can occur (1.) because the part is subjected to loading extremes or they can occur (2.) as part of the population on the low end of the strength distribution. Establishing a failure point in the distribution of failures through testing can help define design margins to minimize the potential for service life failures.

Ruggedization
Ruggedization testing includes Step Stress or HALT methods that include redesign cycles; these develop a product that has an adequate margin on service life conditions, allowing for expected product variations. One very important benefit of ruggedization is that it increases vibration capability. The redesign is not limited to changing the weak link, it can be a change that reduces the stress at the weakness.

Increased vibration capability usually allows the product to be screened at higher levels. This leads to more effective vibration screening of all parts of the product.

Stress Screens
Stress screens can be developed through step stress fragility testing. The product's weakest link determines maximum intensity of the screen vibration, avoiding excessive life use for the weakest part. However, the fragility test does not create the weakness, merely discovers the weakness. Quantitative evaluation of a product's weakest link does not create the weakness, but defines the weakness. Determining a product's weakness is essential to the screen process.

Lack of understanding of the product can result in screen vibration being ineffective or overly damaging.

Lack of understanding of damage experienced during a vibration screen also contributes to Vibration Myth #3:

Electronics Vibration Myth #3: For environmental stress screening of electronics, thermal screens are more effective than vibration screens.

J. Starr is a registered professional engineer in the state of Pennsylvania with over 35 years of continuous and varied experience in structural capabilities in Nuclear, Chemical, and Defense industries. He works as a Consulting Engineer at CirVibe Inc. To contact John send an e-mail to jstarr@equipment-reliability.com.

(back to the top)
*******************************

The DeciBel
by Wayne Tustin

The following “deciBel” discussion is taken from Section 1.26 on decibel scaling, from Wayne Tustin’s new text “A minimal mathematical Introduction to the Fundamentals of Random Vibration & Shock Testing, HALT, ESS & HASS, also Measurements, Analysis & Calibration”.

How can we show 0 to 1000g and 0 to 0.001g on the same graph?

The magnitudes we encounter in vibration, shock and sound measurements vary over a huge dynamic range. Logarithmic scaling “compresses” large values that would drive an ordinary meter off scale (or plotter off the paper). And it “expands” small values that would not lift an ordinary meter’s pointer (nor a plotter pen) off zero.

With logarithmic scaling of amplitudes we can evaluate those tiny contributions. We could see those small contributions by greatly increasing the “gain” or amplification. But then each of the large contributions would drive the pen off the paper.

DeciBel scaling implements log scaling of magnitude. The original unit, the Bel, was simply the base 10 logarithm of a sound power ratio. That unit was later subdivided by 10 to get dB.

The original “classic” relationship is number of dB =

Since power is proportional to voltage squared, we can say that

number of dB =

or

number of dB =

The denominator Po or Eo is our standard or reference value, against which we are comparing some new or unknown value in the numerator P or E. Always state what that reference value is, e.g. 20 mPa (micropascals) for acoustics.

You will use the “power” equation for sound power measurements, for power spectral density random vibration measurements, etc. Far more commonly, you will use the “voltage” equation for voltage, current, pressure, force, displacement, velocity, acceleration, etc.

Did the statement that power is proportional to voltage2 agree with your previous knowledge? Maybe it is worth exploring. Let’s (in our imaginations) double the voltage in a simple circuit, and observe that power quadruples.

Power 1 = 1 watt, while Power 2 = 4 watts

Insert power quadrupling into the earlier x10 “power” equation or insert voltage doubling into the earlier x20 “voltage” equation. Either way, you should get +6 dB.

Tables similar to Table 1-1 are commonly found in Electrical Engineering and other handbooks. Notice the red line pertaining to +6 dB in our “voltage doubling” imaginary experiment. Note that a halving of voltage or quartering of power can be described as -6 dB.

Table 1-1 Decibel relationships

Volts, amps etc. ratio

Power ratio

No. of dB

Ü- +Þ

Power ratio

Volts, amps etc. ratio

1.00

1.00

0.00

1.00

1.00

0.89

0.79

1.00

1.26

1.12

0.79

0.63

2.00

1.59

1.26

071

0.50

3.00

2.00

1.41

0.50

0.25

6.00

3.98

1.78

0.35

0.13

9.00

7.94

2.82

0.62

0.10

10.00

10.00

3.16

0.18

0.03

15.00

31.62

5.62

0.10

0.01

20.00

100

10.00

0.03

0.00

30.00

1000.00

31.60

Wayne Tustin, ERI's president, can be reached by e-mail or phone (805) 564-1260. Read more about Wayne at ERI's website.

(back to the top)
*******************************

Test Lab Musings (part 8)
by Robert L. Renz

Even though your fixture is a very solid chunk of magnesium or aluminum, be careful to gently set it place. Don't slide it into position. Before you emplace your fixture, take a moment to wipe off the head expander, or shaker table and the bottom of the fixture. A particle or a lost piece of fixture hardware can wind up pressed into your fixture.

Before you set up for a test, take a flashlight and look into the threaded inserts on your slip table. Remove any small nuts and bolts or other objects that you find – or the next time you thread a bolt into the insert, it might "bottom out" on this old hardware and damage your slip table.

If one of the threaded inserts in your shaker table suddenly seems like it's too short, remove that bolt and take a look with a flashlight. You’ll probably find some lost hardware in the bottom of the hole. If you’re lucky, you can remove it with tweezers – if you’re less lucky, you’ll need a drill. If you were attaching to a slip table, delay the test while you remove the slip table and examine its bottom surface. Your machine shop may need to remove a bump.

When you throw out a damaged accelerometer cable, cut off one end – otherwise, some helpful soul might “rescue” it from the trash and put it back in your stock.

Robert L. Renz of General Dynamics - Advanced Information Systems at Bloomington, Minnesota.

(back to the top)

Climatics
Message Board


Have you questions about humidity (condensing or non-condensing)? Temperature? Altitude? Salt fog? Fungus? About measuring any of the "climatic" environments? Or about conducting "climatic" environmental tests?

Direct your questions (or observations) to ERI's Climatics message board. Sometimes the Climatics board moderator, Steve Brenner, himself, will
assist you. Steve is ERI's
climatics test specialist.
 
Don't hire - train (I)

Instead of hiring measurement engineers for our data acquisition and other needs........ why don't we train some of those we already have working here? People we already know. People who already know us, our products, our customers, our problems. We can save a bundle over all the visible and invisible costs of hiring new people (who don't always work out).

Plan 1: Let's have Chuck Wright come here to teach a few days.

Plan 2: Let's send individuals to Chuck's October 17-19, 2005, Las Vegas, Nevada.

If we train 8 or more, plan 1 is less expensive and it's focused on our needs. If we only train 2 or 3, plan 2 is less expensive, and our people will learn outside class from their classmates.
 
Could we host an ERI course?


Have you noticed that most ERI “open” or “public” short courses are held at military/naval/commercial sites, rather than at hotels?  Why? 
Well, partly to save rent, but mostly to facilitate visits to working areas.  Two examples: We like to visit vibration and shock labs to see (and demonstrate) shakers and shock test machines. We like to visit electronic development areas to see cooling and EMC activities. 
We recognize that hosting an ERI short course involves effort and we try to compensate by offering two no-charge participations.  Might this idea interest your organization?

 
Don't hire - train (II)


Have you noticed that electronics assemblies are getting smaller? And hotter? "Natural" cooling is seldom enough, nowadays. Joel Newberger, teaches ERI short courses about cooling of electronics.

Plan 1: Let's have Joel Newberger come here to teach a few days. Based on the 19 Newberger "mini courses", we can select two or three that we need. Joel can combine them into say three days here.

Plan 2: Let's send some
individuals to Joel’s October 24-26, 2005, course at Las Vegas, Nevada.

If we train 8 or more, plan
1 is less expensive. Further, it will be focused on our needs. If we only train 2 or 3, plan 2 is less expensive, and our people will learn both in and outside class from their
classmates.
 
Test lab becomes school


In the critiques of my 3-day "open" courses (that emphasize, among other subjects, vibration and shock testing) one theme appears often.  It goes something like this "I valued most the lab tour and shaker/shock test machine demonstrations."

Often the critique's author is a designer who has been sending his hardware to a test lab.  For some reason, he has never accompanied his hardware.  Often he's never seen a shaker, or a fixture, or accelerometers. 

I suspect that in future he will personally accompany his hardware to the test lab.  As a result of that visit, he will learn more about his hardware than he has in the past.  Design improvements will follow.  Future designs will avoid some past difficulties.
 
Vibration and Shock courses coming up


Wayne Tustin will teach short courses in vibration testing, shock testing, measurement, analysis, calibration, HALT, ESS and HASS at the following locations:

June 6-9, 2005
Pfinztal (Berghausen), Germany

July 12-14, 2005
Bothell (Seattle), Washington

August 24-26, 2005
Santa Barbara, California

September 27-29, 2005
Mississauga, Canada

October 4-6, 2005
Longmont, Colorado

December 6-8, 2005
Detroit, Michigan

Perhaps you’d like to have customized training presented at your facility, for your designers and test specialists. Contact us.
 
New "Cooling"
Message Board


Have you questions about
the "natural" or "artificial" cooling of your electronics packages?
Direct your questions (or
observations) to ERI's
Electronics Cooling message board.
Sometimes the
board moderator, Joel Newberger, himself, will
assist you. Joel is ERI's
electronics cooling specialist.
 
Announcements


CEEES Conference
European environmentalists will meet at Nurnberg, Germany May 11 and 12, 2005 for the Third European Nurnberg Conference on Environmental Testing and Engineering. This will gather CEEES members of the Confederation of European Environmental Engineering Societies, the European umbrella organization of national technical societies in France, Germany, United Kingdom, Italy, Sweden, Switzerland, Finland, Belgium, The Netherlands and the Czech Republic. The Conference is organized and hosted by the German CEEES member society GUS. The Conference title is "Methods and Benefits of Environmental Testing & Engineering."
Wayne Tustin will represent ERI, presenting "Shock Response Spectrum SRS - Understanding Mechanical Shock Analysis in the Frequency Domain."

New Book released
The "New Management of Engineering" book from Patrick O'Connor is now available. Visit his website for more details.
 
America's Riviera®


Will you be visiting Southern California during 2005? Might a visit to ERI at Santa Barbara - the America’s Riviera®
(between mountains and seashore, about 1/4 of the way from Los Angeles to San Francisco) provide a suitable business purpose to qualify all or part of your trip as business expense? Consider participating August 24-26 in Wayne's basic vibration and shock course.
 
"Manager speak"


Every test engineer has felt frustrated when he couldn't "sell" his managers on some technical matter.  Unless managers have engineering backgrounds, managers don't and probably will never understand engineers. 

So engineers (if interested in convincing managers) need to learn to talk to their managers in their language. 

ERI is looking for an ex-engineer to teach short courses in "manager speak"
 
Show this to your boss


Have you noticed, boss, that there's a lot of turnover among your engineers and senior technicians?  Have you suspected, boss, that more of your engineers and senior technicians are looking elsewhere?

One of the reasons, boss, is that your engineers and senior technicians perceive that they're not advancing technically.  They are stagnating.  They aren't getting sufficient training.  You're not bringing in short courses.  You're not sending them away for short courses. 

Don't look at training as an expense, boss.  Consider it an investment .
 
Contact information


ERI - Equipment Reliability Institute
1520 Santa Rosa Ave.
Santa Barbara - CA - 93109
Tel: (805) 564-1260
Our fax number:
(805) 966-7875

Wayne Tustin tustin@equipment-
reliability.com

Webmaster webmaster@equipment
- reliability.com

Websites
http://www.equipment-
reliability.com

http://www.vibrationand
shock.com

Copyright © 2000-2005 Equipment Reliability Institute. All rights reserved.

 
Free Newsletter


Subscribe
If you would like to subscribe to ERI News, go to either website, fill in the form "Free Newsletter" and hit the Submit button. Subscribe now!

Recommend
If you enjoy reading ERI News and want to recommend it to a friend, just hit "forward" on the menu of your e-mail program or tell your friend to subscribe at our website.

Previous issues
Missed the previous issues? It is not a problem. Just visit our newsletter archives section and find all ERI's News issues.

Unsubscribe
If you do not want to receive ERI's quarterly newsletter, please reply to the newsletter message with "remove" as subject or click here.

 
Visit www.equipment-reliability.com
Visit www.vibrationandshock.com