AN PUBLICATION | AUGUST 2010

ISO 281: 2007

Part II of our
great debate.
Follow the money
When science and
politics clash
Presidents Report
In praise of STLE
corporate members
Rate-ef-chane
alarms
A better metric fer extendin lube life.
Check out our digital TLTsponsored by Polaris Laboratoriesat www.stle.org
Afton Chemical and Polartech. More choices, more convenience.
Now, in one convenient place, you have a much bigger selection of additive technologies. Aftons family of lubricant and fuel additives
has been strengthened with the addition of Polartechs metalworking uid additives. Together, well bring superior additive solutions to
the metalworking industry by adding Aftons strong focus on security of supply to Polartechs people, expertise and exibility.
With additional metalworking additive products, Afton and Polartech are a great t. And we promise
future breakthroughs from our shared Passion for Solutions.
2010 Afton Chemical Corporation is a wholly-owned subsidiary of NewMarket Corporation (NYSE:NEU). www.aftonchemical.com
Afton Chemical and Polartech are the registered trademarks of Afton Chemical Corporation.
PLATukL5
20 MINUTES WITH...
16 Ricardo J. Hein
Based in Atlanta, this Argentine-
born STLE member is the head of
Conexo, which provides tribology
innovations in domestic and
foreign markets.
By Thomas T. Astrene
20 Student Poster Abstracts
Extended abstracts written by
winners of the Student Poster
Competition held at STLEs 2010
Annual Meeting & Exhibition.
Tribological Improvement of
Carbon/Carbon Composites by
Inltration of ZnO/Al
2
O
3
/ZrO
2

Solid Lubricant Coatings
By Hamidreza Mohseni and
Thomas W. Scharf (Advisor)
LUBRICATION FUNDAMENTALS
22 Follow the money
What happens when national
energy policy clashes with the
reality of engineering science?
Check your wallet for the answer.
By Dr. Robert M. Gresham
BEST PRACTICES
26 Strategic oil analysis:
Time-dependent alarms for
extended lubricant lifecycles
(Fourth of a ve-part series on
oil analysis)
Following these strategies
allows you to calculate rate of
wear generation, a more
insightful metric than total wear.
By Mike Johnson
COMMENTARY
30 In search of a fatigue limit:
A critique of ISO Standard
281:2007
In Part II of our debate, a leading
gure in the bearings community
presents his case that the new
standard should be withdrawn.
By Erwin V. Zaretsky
PEER-REVIEWED PAPER (EDITORS CHOICE)
42 Probabilistic Stress-Life
(P-S-N) Study on Bearing
Steel Using Alternating
Torsion Life Test
By Shigeo Shimizu,
Kazuo Tsuchiya and
Katsuji Tosha
30 1b
TLT / AUGUST 2010 / VOLUME 66 / NO. 8
L U N T L N T 5

' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 1

b
LPAkTHLNT5
LUNTLNT5
LULuHN5
10 Tech Beat
Size does matter for nanoscale
friction; Extracting soybean oil
with a switchable-hydrophilicity
solvent; Presence of nanobub-
bles on superhydrophobic
surfaces.
By Dr. Neil Canter
54 Newsmakers
This months newsmakers
include Quaker Chemical, Palmer
Holland, ConocoPhillips and
more.
56 Sounding Board
Readers discuss which profes-
sional and personal organiza-
tions they belong to and why.
61 New Products
Diaphragm valves for reactive
uids; Bioscope catalyst
perfusing stage incubator;
Fuel ethanol additive; Automatic
open cup ash point tester; MQL
systems for CNC operations.
65 Advertisers Index
66 Resources
Keep up to date with the latest
technical literature available in
print and online.
67 Tribology Transactions
Review
Take a moment to review the
offerings from the most recent
issue of STLEs peer-reviewed
journal.
4 Presidents Report
The value of corporate
membership
6 From the Editor
Remembering Dons octopi
8 Headquarters Report
Business book delivers more than
happiness
68 Selling Points
No really means not now
70 Shop Floor
Enforcing the leash law
72 Cutting Edge
Sliding on cells
5b
PUBLISHER/EDITOR-IN-CHIEF
Thomas T. Astrene
MANAGING EDITOR
Karl M. Phipps
CONTRIBUTING EDITORS
Dr. Neil Canter
Dr. Robert M. Gresham
Mike Johnson
Jean Van Rensselar
CIRCULATION COORDINATORS
Myrna Scott
Judy Enblom
DESIGN/PRODUCTION MANAGER
Joe Ruck
ADVERTISING SALES
Tracy Nicholas VanEe
Phone: (630) 922-3459
Fax: (630) 904-4563
[email protected]
Dr. Pranesh B. Aswath
University of Texas
at Arlington
Arlington, Texas
Robert F. Baker
King Industries
Norwalk, Conn.
Dr. John K. Duchowski
Hydac Filtertechnik
GmbH, Werk 8
Dr. Hong Gao
ConocoPhillips
Ponca City, Okla.
Dr. Seong H. Kim
Penn State University
University Park, Pa.
Dr. Chih Lin
Baker Hughes
The Woodlands, Texas
Dr. Ashlie Martini
Purdue University
West Lafayette, Ind.
Myron McKenzie
American Roller
Bearing Co.
Greer, S.C.
Dr. Jun Qu
Oak Ridge National
Laboratory
Oak Ridge, Tenn.
Dr. Brajendra K. Sharma
NCAUR/ARS/USDA
Peoria, Ill.
Allison Toms
GasTOPS Inc.
Pensacola, Fla.
Dr. Min Zou
University of Arkansas
Fayetteville, Ark.
Dr. Qian Zou
Oakland University
Rochester, Mich.
TRIBOLOGY AND LUBRICATION TECHNOLOGY (USPS 865740)
Vol. 66, Number 8, (ISSN-1545-858), is published monthly
by the Society of Tribologists and Lubrication Engineers,
840 Busse Hwy, Park Ridge, IL 60068-2376. Periodicals
Postage is Paid at Park Ridge, IL and at additional
mailing ofces. POSTMASTER: Send address changes to
Tribology and Lubrication Technology, 840 Busse Hwy,
Park Ridge, IL 60068-2376.
EDITOR
Evan Zabawski
The Fluid Life Corp.
Edmonton, Alberta, Canada
TECHNICAL EDITORS
b
2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
www.lubrizol.com
2010 All rights reserved.
Fresh as the Day it was Rolled
Salt may go great with Sushi, but its tough on metal sheets, coils, and components. For assurance
that your customers metals will arrive as fresh as the day they were rolled, Lubrizols new family
of ALOX

rust preventive additives can provide exceptional salt spray protection. Globally registered
ALOX thin lm additive packages are available through the Lubrizol global manufacturing network,
ensuring consistent supply.
With you every step of the way.
hen I was asked to speak at the corporate member break-
fast at STLEs 2010 Annual Meeting in Las Vegas, I thought
of the usual speech to give. I planned on discussing the many ben-
ets corporate members re-
ceive, including free registra-
tion to annual meetings, free
education courses and discounts
on exhibit space and the Com-
mercial Marketing Forum.
For most companies involved
in STLE, the decision to join as a
corporate member becomes a
no-brainer when analyzing the
cost vs. the benets. STLE actu-
ally loses money when a com-
pany utilizes all of its corporate
member benets. There is cer-
tainly a lot of value in the pack-
age.
While all of this is interest-
ing and relevant, I knew this
was not the message I wanted
to concentrate on in my speech.
After giving it some thought and
looking back at the history of
our society, I decided to talk about the role corporate members
have played in STLEs ongoing success.
The formation of our organization back in 1944 as the American
Society of Lubrication Engineers (ASLE) would not have been pos-
sible without the support of corporations from the steel industry,
lubricant suppliers, lubrication-device manufacturers, additive
suppliers, oil companies and the aluminum industry. If not for the
support of these titans of industry 66 years ago, our society would
not have ourished. The focus in 1944, as it is today, was on creat-
ing a society that meets the needs of industry.
Some of the principles established back then include:
1. Encouraging the active cooperation of industry and other multi-
disciplinary professions in advancing the knowledge and applica-
tion of the art of lubrication.
2. Supporting and stimulating the study and development of the
theory and practice of lubrication techniques in chemistry, phys-
ics, engineering and other related sciences.
3. Accumulating and disseminating to lubrication professionals
and industry authoritative and scientic information available
through research, testing, eld experience, etc.
Education and the dissemina-
tion of knowledge were the
principles upon which our so-
ciety was founded. STLEs
founding fathers recognized
that an educated workforce
benets employers. Corpora-
tions recognized and em-
braced this concept and sup-
ported ASLE.
The generous support of
our corporate members over
the years has contributed to
STLEs success, and today our
society is recognized as the
global leader in tribology and
lubrication science. Corpo-
rate member support comes
not only in the form of dues
but, more important, through
their commitment at annual
or local section meetings either through attending or by teaching
a course or presenting a paper.
Today, as back when STLE was created, corporations still play
an important role in the success of our society. They also realize
the importance of the society and the benets they reap through
STLE education and training programs that allow for a more edu-
cated workforce that produces better products and services.
STLE grew through the years with the support of corporations.
Even if you are an individual member and your employer pays your
dues, consider this a form of corporate support. It means your
company recognizes the benets and value of STLE membership.
STLE has thrived for more than 66 years due to the support
of a very special group of companies.
PRESIDENTS REPORT
Peter Drechsler
Peter Drechsler is a senior tribology specialist
with The Timken Co. in Canton, Ohio. You can
reach him at [email protected].
1he ve|ae of corporete membersh|p
W
The generous support of our corporate members over the
years has contributed to STLEs success, and today our society
is recognized as a global leader.
4 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
LuII for PreaenIuIiona
51L 2011 AnnuuI HeeIing 6 xbibiIion
STLE-sponsored meetings are a respected venue for presentation of the
latest technical information and for professional development and business
networking opportunities. The Annual Meeting showcases more than 350
technical presentations, application-based case studies and best practice
reports, and discussion panels on technical or market trends. Education
courses support professional development needs and prepare people for one
of STLEs three certication programs. Exhibits and commercial presentations
spotlight the latest products and services of interest to lubrication profes-
sionals. Typical attendance is 1,100 -1,400 people.
AbaIrucI 5ubmiaaion
If you are interested in presenting at the 2011 meeting, submit a 100-150-
word abstract of the presentation through the STLE website at www.stle.org.
Site opens July 1. Abstracts due Oct. 1, 2010. Notication of acceptance will
be sent in November 2010. You do not need to prepare a full manuscript to be
included on the Annual Meeting program, but a 2-3-page extended abstract is
requested for the Proceedings CD. However, you are invited and encouraged
to submit a manuscript for review and possible publication in the STLE peer-
reviewed journal Tribology Transactions.
For more information, please contact:
HerIe hedIund
630-323-7212
mbedIundQaIIe.org
- Ae|cspate
- te|amlts/tcmpcslles
- tcallas
- tcaalllca hcallc|la
- lalae b |lvel|ala
- lavl|camealally l|leaaly llulas
- llula lllm ea|las
- bea|s aaa bea| luc|ltallca
- b|ease
- luc|ltallca luaaamealals
- helalwc||la llulas
- Naacl|lcclcy
- Ncale||cus helals
- lcwe| beae|allca
- l|atlltal luc|ltallca Applltallcas
- kcllla llemeal ea|las
- !alely, Neall| b keulalc|y Allal|s
- !eals
- !clla luc|ltaals
- !u|late lalaee|la
- !yal|ellt b Nya|aullt luc|ltaals
- !|lccleslla
- wea|
2011 PreaenIuIiona ure being aougbI in Ibe foIIowing ureua:
H
uy
15-19, 2011
-
A
I
Iu
n
I
u
h
i
I
I
o
n
-
A
I
I
u
n
I
u
,
6
eorgiu
{u
5
A
]
hen I rst met Don Riddell at a seminar in 1998, his pre-
sentation on viscosity grades converted me to 0W-30, an
oil he promised would run better year-round and give better fuel
economyit does. Don had well over 10,000 other students, or
converts, through the latter
part of his 41-year career
with Imperial Oil, a Canadian
petroleum company largely
owned by ExxonMobil. Though
he passed away on June 19,
he left an indelible mark on
all members of his audience
whether at one of his semi-
nars or at a local STLE meet-
ing, where it was practically
an annual tradition to have
him present a topic.
As soon as Don would be-
gin speaking in his energetic,
anecdotal style, you had two
guarantees: (1.) You were going to learn something new, and (2.)
You were about to be entertained. Most of our local members who
are in a position to give presentations will borrow an anecdote or
two of Dons, and I would like to share the (publishable) one that is
my favorite.
Whether explaining the merits of 0W-30 or simply discussing
multigrade oils, Don would be compelled to talk about the viscosity
improver additive. Rather than wax technical about polymethacry-
lates, he would just tell you to picture an octopus swimming in the
ocean.
When that octopus is swimming in cold water, he would be like
this, Don would say as he would enthusiastically pull his arms
around his body and mimic shivering. But, he would continue, he
would go like this when he swims into a nice, hot current, and then
he would comically op out his arms, cock his head and loll out his
tongue as though he was suffering from the heat.
Paying little attention to the weird stares from the audience
members who were wondering where he was going with this, Don
would just smile and carry on with a look that told you to trust him.
Now, imagine there are millions of tiny octopusesoctopiwhat-
ever, oating around in the oil. The oil wants to thin as it is heated,
but if all these octopuses/octopi are inging their arms, er, tenta-
cles out, then the oil isnt going to thin out as much because they
are taking up more space and banging into each other.
At this point, a rst-time viewer would have little idea that
they were witnessing a carefully crafted analogy that Don had
been using for many, many years, because his energy level fooled
you into thinking he was genuinely excited about this apparently
new topic, and his delivery always contained quasi-errors.
Now if we take a
0-weight oil and cram it full
of octopisee, I got it right
this time, giving a twin-
kling wink as he talked, we
can make that oil behave
less like a 0-weight at the
higher temperature and
more like a 30-weight.
Sometimes he would then
interrupt himself and say,
Did you know that the W
stands for winter, not
weight? Crazy, huh? with
an incredulous tone as if he
himself just learned this.
Concluding with, Thats how we make multigrade oils. We stuff
them full of long-chain polymers that act just like octopi.
But what happens to the tentacles if we shove the octopi
through tight clearances? Don would ask as he paced around with
his arms sticking out. (If you were lucky enough to be sitting at the
right angle, his demonstration would play out like a magic trick.)
Making a chopping motion with one hand, against the opposite
shoulder, he would sh for somebody to say they would be cut. As
soon as he got his required response, he would whip his one arm
behind him, making it disappear. Whats another word for cut? he
would ask and, seldom waiting for a reply, answer his own ques-
tion with Shear.
And in that instant you realize that youre now learning about
the difference between temporary and permanent shear and why
multigrade oils are the most prone to this damagethey are full of
Dons octopi.
We miss Don, and his memory will be kept alive in the form of
sharing his anecdotes, even if we cant do an octopus impression
half as well as he did.
Evan Zabawski, CLS, is manager of training and
education services for The Fluid Life Corp. in
Edmonton, Alberta, Canada. You can reach him at
[email protected].
FROM THE EDITOR
Evan Zabawski
W
Paying tribute to a mainstay of the Canadian lubrication world.
kemember|a 0oa's octop|
6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
0oa's coav|ac|a case for 0R30 mu|t|rade o||: 'P|cture aa octopus
sw|mm|a |a the oceaa.'
IN 1960, DETROIT LAUNCHED BOATS, NASA LAUNCHED MONKEYS, AND
WE LAUNCHED A COMPANY.
ANALYSTS, INC. CELEBRATES 50 YEARS OF LEADERSHIP IN OIL ANALYSIS.
Much has changed since 1960. The land yachts and space monkeys are long gone. Analysts, on the other hand, has not only survived,
weve thrived. We have grown from a single laboratory into a global leader with seven labs across the US, Japan and Mexico. Along the way,
weve pioneered some of the industrys most ground-breaking patents and services in testing, sampling accessories and outstanding
customer relations. From the start, our commitment to providing innovative services has never wavered. Where do we expect to be on our
100th anniversary? The same place we are now: the leading laboratory in quality and innovation serving Industrial clients around the globe.
To learn more, visit us at www.analystsinc.com
I T S B E T T E R T O K N O W
LOS ANGELES, CA | CHICAGO, IL | LOUISVILLE, KY | ATLANTA, GA | HOUSTON, TX | MONTERREY, MEXICO | TOKYO, JAPAN
elivering HappinessA Path to Prots, Passion and Purpose
is a best-selling business book that made it to my summer
reading list. The author, Tony Hsieh (pronounced Shay), is CEO of
Zappos.com, the online shoe and apparel retailer.
When I was presented with a copy as an advance assignment
for a professional retreat, I had no idea what Zappos was, let alone
why the company and its CEO are
so well knownat least to other
people! But Hsiehs personal nar-
rative and the interwoven story
of Zappos unconventional path to
nancial success inspired me to
comment on this exceptional book
and recommend it as a worthwhile
read.
The Zappos story is built
around the companys relentless
pursuit of creating what they call
WOW experiences for custom-
ers, vendors and employees. For
example, free shipping upgrades
and a 365-day, no questions asked
return policy.
Zappos commitment to serving customers sometimes pushes
beyond the exceptional to the humorous or bizarre. As a test,
Hsieh once dared a business associate to make a late night call to
the Zappos 24-hour customer service line and ask if the operator
could help locate a place to order late night pizza. Another caller
once asked a Zappos employee to answer his questions without
using any pronouns. The customer service reps were successful in
both cases!
According to Hsieh, it is company culture that sparks this level
of employee commitment and leads to a powerful and valuable
brand. Zappos culture has grown out of a set of 10 core values
that dene how the company operates internally and externally.
The allegiance to these values is so strong that the company of-
fers new hires $2,000 to quit in their rst week as a way to test
their commitment to these long-term goals vs. short-term nan-
cial gain.
The WOW factor also extends to investors. The company, which
was founded in 1999 and grew to $1 billion in gross merchandise
sales within a decade, was sold to Amazon.com in 2009 for $1.2
billion.
Although Im not one of the lucky 100 investors who got a cut
of this $1.2 billion payday, the consolation prize is the wisdom to
be gained from reecting on Zappos core values. My favorite is
Core Value Number Five, which is to Pursue Growth and Learning.
For both personal and professional reasons, its a message that
has broad appeal and immediate
impact.
While a good place to start
on this goal might be by getting
a copy of the book, I hope you
will consider STLE as another op-
tion for professional growth. The
options vary from something as
simple as reading copies of TLT to
the challenge of sitting for one of
STLEs four professional certica-
tion exams. As Hsieh points out,
You have to want to challenge
and stretch yourself in order for it
(personal and professional growth)
to happen.
STLEs HQ staff is adopting this
philosophy. Each employee has been given the assignment to read
at least one business book this year, with STLE covering the cost of
the book. The catch is that they have to share what they learned
from the book with our extended staff of 12 people at one of our
weekly staff meetingssort of a book report for adults. The ex-
pected result is that each of us will be exposed to a dozen business
books, and our capabilities and culture will be better for having
participated in the experiment.
Reading Delivering Happiness convinced me that this sort of
professional development exercise is a great way to counteract
the negative impact of the current recession. Take a look at the
book for yourself and at STLEs suite of professional development
products and opportunities. I am certain youll nd it to be an anti-
dote to the daily grind and a way to refresh your
sense of personal and professional enthusiasm.
HEADQUARTERS REPORT
Edward P. Salek, CAE / Executive Director
You can reach Certied Association Executive
Ed Salek at [email protected].
D
If advice from an online shoe CEO ts, why not take it?
8as|aess book de||vers
more thea hepp|aess
8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
You have to want to challenge and stretch yourself in
order for personal and professional growth to happen.
exxonmobilsynthetics.com 800-892-4449 (within the U.S.) +1 281-368-5803 (outside the U.S.)
ExxonMobil, the ExxonMobil logo, SpectraSyn, SpectraSyn Plus, SpectraSyn Ultra, and SpectraSyn Elite are trademarks of Exxon Mobil Corporation and or its afliates. exxonmobilchemical.com
Extreme
Extreme Conditions Call for Extreme Performance
Extreme conditions demand more from lubes. They require synthetic lubricant basestocks with greater exibility, improved stability
and endurance. Our high-performing product portfolio combines with proven basestock experience and formulation expertise
delivering opportunities for breakthrough formulations to meet your toughest challenges.
SpectraSyn PAO
SpectraSyn Plus PAO
SpectraSyn Ultra PAO
And introducing...
SpectraSyn Elite Metallocene PAO
esearchers continue working to gain a better understanding of friction at the
nanoscale because it has an impact on how materials are prepared for such
applications as nanoscale data storage systems, nanocomposites and nanoelectro-
mechanical devices. One thing that has been learned is that atomic-scale interac-
tions are very important in determining how friction behaves at the nanoscale.
In a previous TLT article, work was discussed that correlates friction at the
nanoscale with friction at the macroscale by taking into account that surfaces at
the nanoscale are not smooth but, in fact, are rough with topography similar to a
mountain range.
1
Atoms and molecules are present at the tops of peaks and can
rub against each other in a fashion similar to frictional behavior at the macroscale.
In looking at materials on the nanoscale, two-dimensional interactions become
very important. Graphene is one such material that is being closely examined be-
cause it consists of individual atomic layers of carbon and is a precursor for graph-
ite. A previous TLT article highlighted work done to show that graphene is the
strongest material ever examined.
2
One objective in working with materials at the atomic level is to determine the
frictional effects seen with small numbers of atomic layers. Such work has not been
carried out until now.
A1HILALLY 1hIN 5h15
STLE-member Robert Carpick, professor in the department of mechanical en-
gineering and applied mechanics at the University of Pennsylvania in Philadel-
phia, and James Hone, professor in the department of mechanical engineering at
Columbia University in New York, headed up a group of researchers that evalu-
ated how friction occurs among atomically thin materials. The materials evalu-
ated are hexagonal boron nitride, graphene, molybdenum disulde and niobium
diselenide.
Carpick says, Our objective was to evaluate a range of layered materials that
have different lattice constants, elastic constants and electronic properties in order
to see if we could nd some commonality for frictional behavior that is indepen-
dent of their different structures and electronic properties.
Graphene and molybdenum disulde are certainly familiar materials widely
used as solid lubricants. Hexagonal boron nitride is also utilized as a solid lubri-
cant in applications such as a coating for cutting tools. The nal material, niobium
diselenide, is not widely known. Carpick says, Niobium diselenide is an exotic
material of interest for its low temperature, superconductivity. As far as we know,
there are no known tribological applications.
The researchers used an atomic force microscope (AFM) in a technique known
as friction force microscopy to measure the frictional forces encountered with
New research reveals that atomic-scale interactions
play a major role in determining how friction behaves.
KEY CONCEPTS
Frictional effects among
atomically thin sheets of
materials with different
structures and electronic
properties were studied.
An atomic force micro-
scope determined that
friction decreases as the
thickness of atomic layers
increases from one to
around four and then
levels off.
A single sheet of atoms
produces more friction
because it more readily
bends to conform to an
AFM tip, leading to more
contact area and higher
friction.
TECH BEAT
Dr. Neil Canter / Contributing Editor
R
T
E
C
H

B
E
A
T
S|ze does metter
for aeaosce|e fr|ct|oa
1 0 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
atomically thin sheets of these materials. The sheets of these
four materials were prepared by exfoliating them from a bulk
source onto a silicon oxide substrate. The AFM also was
used in a topographic mode to determine the thickness of
the atomic sheets.
The researchers found that
the friction generated for all
four materials increases with
decreasing atomic thickness.
This result may not seem in-
tuitive, but Carpick provides
an explanation.
When a material gets
thin, it also becomes very ex-
ible, he says. A macroscale
example is the fact that bend-
ing back a single sheet of pa-
per is far easier than bending
a piece of wood. As the AFM
tip approaches a single sheet
of atoms, an attractive Van
der Waals force occurs be-
tween the tip and the single
sheet, causing the sheet to bend as it conforms to the tip.
This means more contact area and, thus, higher friction.
The researchers also observed that the friction force ex-
hibited a stick-slip motion, slipping once for every unit cell
of the lattice for graphene and molybdenum disulde. The
force required to slip was seen to build up for the rst 1-2
nanometers of sliding, consistent with the idea of deforming
the puckered sheet as sliding begins.
Carpick continues, In contrast, thicker sheets of atoms
do not conform as readily because they are stiffer, leading
to a lower level of friction. The bending effect of the single
sheet around the tip is also known as a puckering.
Figure 1 shows the puckering effect as a single atomic
layer of a graphene sheet adheres to an AFM tip. Graphene
atoms moving out of the plane to conform to the tip are
shown in blue and red.
The researchers estimate that a monolayer sheet causes
20% greater friction than bilayers of atoms and 2-3 times
higher friction than bulk sheets of atoms. This phenomenon
was seen regardless of changes in scanning speeds, applied
forces, AFM tips made from different materials and humidity
changes. In the latter case, a reduction in humidity from 30%
to 5% did lead to overall lower friction, but the variation of
friction as a function of thickness remained constant.
Carpick indicates that friction does not continue to de-
crease as the thickness of atomic layers increases. He says,
Friction decreases as the number of atomic sheets increases
to around four but then levels off. The friction seen for four
atomic layers is comparable to friction seen for 50 layers or
when the material is present in a bulk state.
The researchers then
placed the materials on a
surface that they strongly ad-
here to in order to assess the
frictional effect. The surface
used was muscovite mica,
which forces the materials to
be atomically at when fac-
ing the tip of the AFM.
The result is that no in-
crease in friction is seen with
a decrease in atomic thick-
ness for any of the materi-
als. Carpick says, When
the AFM tip encounters the
atomic layers of the materi-
als, they do not come off the
substrate and adhere in the
same fashion as when they
are on the silicon oxide surface.
A second follow-up experiment determined how the
materials perform if suspended over 300-nanometer holes
placed in the silicon oxide substrate. Friction did increase as
the thickness declined in a similar fashion to when the mate-
rials were placed on the silicon oxide substrate.
These results clearly suggest that the thickness depen-
dence of friction at the nanoscale applies to all thin materials
that are either attached loosely to a substrate or suspended
over a substrate. Carpick is hoping that this research will
aid the design and performance of nanoscale mechanical de-
vices.
Further information can be found in a recent article
3
and
by contacting Carpick at curpickQaeua.upenn.edu.
RFRNL5
1. Canter, N. (2009), Understanding Friction Laws at the
Nanoscale and their Relation to the Macroscale, TLT, 65
(7), pp. 1011.
2. Canter, N. (2009), Graphene: The Strongest Material ever
Examined, TLT, 65 (2), pp. 2829.
3. Lee, C., Qunyang, L., Kalb, W., Liu, X., Berger, H., Carpick,
R. and Hone, J., Frictional Characteristics of Atomically
Thin Sheets, Science, 328 (5974), pp. 7680.
As the AFM tip approaches a single sheet of atoms, an attractive
Van der Waals force occurs between the tip and the single sheet,
causing the sheet to bend as it conforms to the tip.
Figure 1 | Adhesion of a sliding AFM tip on a single layer of
graphene atoms leads to higher friction than on thicker layers
due to a puckering effect. Atoms depicted in blue and red ad-
here to the tip and move out of the plane. (Courtesy of the Uni-
versity of Pennsylvania)
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 1 1
oybean oil is currently obtained from soybeans through a process involving
extraction of the oil from akes. Hexane is used as a solvent in this process
because it is very compatible with soybean oil and easily removed by distillation.
Philip Jessop, professor of chemistry and Canada Research Chair in Green
Chemistry at Queens University in Kingston, Ontario, Canada, indicates that this
is not an environmentally friendly process. He says, Hexane is a volatile solvent
that can cause a signicant amount of smog formation. In Canada, it is estimated
that 4,400 metric tons per year of hexane are emitted with a third sourced from
oilseed processing. One other factor is that the nal distillation process is very
energy-intensive.
In a previous TLT article, the concept of a switchable solvent was introduced.
1
Jessop has been looking at this approach to use solvents in a more cost-effective
and environmentally friendly standpoint.
His original switchable solvents could be converted reversibly from a nonpolar
to a polar state. Jessop has been successful in doing this using carbon dioxide as a
reagent. Carbon dioxide is utilized because it is very cost-effective, readily avail-
able and easily removed. Jessop has now reported a new kind of switchable solvent,
which he calls a switchable-hydrophilicity solvent (SHS), meaning it can switch
from a hydrophobic state to a hydrophilic state.
The switching process is shown in Figure 2. A hydrophobic organic solvent is
typically insoluble in water, as shown on the left. Reaction with carbon dioxide in
water (carbonated water) switches the solvent to a hydrophilic state that is soluble,
leading to the formation of a homogeneous mixture on the right. Eventually, the
SHS can be isolated in its original hydrophobic state through removal of the carbon
dioxide.
Using a SHS to process soybean oil without the use of hexane and distillation
has potential. The solvent in its hydrophobic state could be used to extract soybean
oil and then switched to a hydrophilic state to enable the oil to be isolated. A suit-
able solvent that can reversibly react with carbon dioxide has now been found.

Carbonation of water produces a weak acid (carbonic acid) which means that the
desirable solvent should be basic. Jessop and his coworkers have known that ami-
A carbon dioxide is used in a new process that
is more environmentally friendly and cost efcient.
S
KEY CONCEPTS
The current method for
extraction of soybean oil
requires the use of a
volatile solvent (hexane)
and an energy-intensive
distillation process.
A switchable-hydrophilicity
solvent has been found to
extract soybean oil without
the need for hexane and
distillation.
Green Center Canada
wants to use this concept
to extract residual motor
oil from used plastic
bottles to facilitate
recycling and reduce
waste disposal costs.
T
E
C
H

B
E
A
T
xtrect|a soybeea o|| W|th e
sW|tcheb|ehydroph|||c|ty so|veat
We found that the difference in the wavelengths of maximum
absorption for the solvent in the absence and presence of water is
far greater than any other solvent we have ever tested and
shows a change in true solvent polarity.
1 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
dines, amines and guanidines can react with carbon dioxide
and water to become hydrophilic. Jessop says, We looked at
a number of candidates by varying the number of alkyl groups
attached to these nitrogen-based functionalities. Guanidines
were eliminated as possible solvents because they are too ba-
sic and do not switch back to
a hydrophobic state.
The parameter used to
determine the best possible
choice is the octanol-water
partition coefcient. Jessop
says, We added a tiny amount
of the solvent to a beaker
with one-part octanol and
one-part water. The percent-
age of solvent in octanol vs.
water is measured. A higher
value means that the solvent
is more hydrophobic.
The researchers found that
a solvent with a logarithmic
octanol-water partition co-
efcient above seven is too
hydrophobic and will not be
miscible with carbonated wa-
ter. One specic amidine sol-
vent that worked well exhibited a value just above six.
This amidine solvent also happens to change polarity
when exposed to carbon dioxide. Polarity measurements
were obtained by determining the wavelength of maximum
absorption of Nile Red dye in the absence and presence of
carbon dioxide. Jessop says, We found that the difference
in the wavelengths of maximum absorption for the solvent
in the absence and presence of water is far greater than any
other solvent we have ever tested and shows a change in true
solvent polarity.
The reason for this major change is that the solvent is very
immiscible in water in the absence of carbon dioxide. This
effect changes dramatically when carbon dioxide is added.
This amidine solvent was then used to extract soybean
oil from the soybean akes. An extraction experiment was
conducted to compare the ability of the amidine to extract
soybean oil vs. hexane. After stirring soybean akes in both
solvents overnight and then ltering, an equivalent amount
of soybean oil was detected in both solvents.
Carbonated water was then introduced to remove the
amidine solvent from the water. The efcacy of this tech-
nique was evaluated by using 1H nuclear magnetic resonance
(NMR) spectroscopy of deuterium water. After only one wa-
ter wash, 96% of the amidine solvent had been removed.
Jessop anticipates that additional washings will remove the
remaining solvent.
The solvent can then be isolated from the water through
removal of carbon dioxide by heating the mixture at a tem-
perature of 80 C for one hour. Jessop indicates that further
work needs to be done to determine the amount of residual
solvent left in the akes and
how to remove it. This fac-
tor is important because the
akes are used in other ap-
plications.
One area of concern for
Jessop is solvent durability.
He says, We are uncertain
that amidine solvents can
be reused in this process
because they can hydrolyze
over time. We have recently
looked at amines which are
chemically more robust. Sev-
eral promising amine candi-
dates have been identied
that will perform better than
the amidine solvent we ini-
tially evaluated.
Jessop also indicates that
further work will be done to
determine how much energy can be saved by not distilling
hexane.
RLYLLIN6 H1R IL
Jessop works with Green Center Canada, which is looking
to commercialize sustainable technologies developed in the
academic community. He says, One new project that Green
Center Canada is working on is to evaluate the ability of
switchable solvents to extract residual motor oil from used
plastic bottles. If an approach can be developed, then recy-
cling of the plastic bottles will become feasible.
Such a process will signicantly reduce waste disposal
costs as plastic bottles with residual motor oil now have to
be land-lled in Canada.
Additional information on the use of switchable solvents
to extract soybean oil can be found in a recent article
2
and by
contacting Jessop at jeaaopQcbem.queenau.cu.
RFRNL5
1. Canter, N. (2006), Analyzing Switchable Solvents, TLT,
62 (2), pp. 1516.
2. Jessop, P., Phan. L., Carrier, A., Robinson, S., Durr, C. and
Harjani, J. (2010), A Solvent having Switchable Hydrophi-
licity, Green Chemistry, 12 (5), pp. 809814.
Several promising amine candidates have been identied that will perform
better than the amidine solvent we initially evaluated.
Figure 2 | A switchable-hydrophilicity solvent is used in the
hydrophobic state to extract soybean oil and then is converted
into a hydrophilic solvent through the introduction of carbon-
ated water. Isolation of soybean oil is then achieved followed
by switching the solvent back to its hydrophobic state through
removal of carbon dioxide. (Courtesy of Queens University)
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 1 3
mproved wettability of lubricants on surfaces continues to be an objective for lu-
bricant suppliers. They are striving to improve the ability of lubricants to adsorb
on surfaces, which leads to better friction reduction and wear loss.
One surface type that has been looked at actively is superhydrophobic. In a
previous TLT article, a superhydrophobic surface is dened as one in which water
exhibits a contact angle of approximately 150 degrees.
1

One of the key features of a superhydrophobic surface is that of water repel-
lency. Water remains a major problem in lubrication systems and any means that
can be done to keep it away from surfaces helps extend the life of the system.
Dr. Antonio Checco, associate scientist at the Brookhaven National Laboratory
in Upton, N.Y., says, One of the perceived reasons for the extreme water repellency
of superhydrophobic surfaces is the presence of small gas pockets near the surfaces.
These bubbles further reduce the amount of surface area that can be in contact with
water, thereby increasing the macroscopic contact angle.
The presence of surface texture encourages the formation of these bubbles.
Checco says, Texturing of surfaces with topographical features such as cavities
promotes the formation and stabilization of bubbles. From a thermodynamic stand-
point, coalescing of small bubbles into larger ones is encouraged. This particular
effect is more likely to occur on at as opposed to textured surfaces because in the
latter case the bubbles are tightly conned within the surface textures.
On superhydrophobic surfaces that exhibit textures on the nanometer scale,
nanobubbles are expected to form. Firm evidence of the presence of nanobubbles
would be helpful to better determine how these nanotextured surfaces repel wa-
ter. However, nanobubbles on superhydrophobic surfaces have not been directly
probed until now.
k-RAY 5LA11RIN6
Checco and his coworkers devised an approach to conrm the existence of nano-
bubbles of gas on superhydrophobic surfaces. The researchers developed a textured
surface and then immersed it in water. Nanobubbles were detected through the use
of small angle X-ray scattering (SAXS).
The textured surface was prepared through a block copolymer, self-assembly-
based fabrication technique. Checco says, We started with a surface of at, silicon
wafers similar to the material used in the electronics industry. Subsequently, a thin
lm of a block copolymer consisting of polystyrene and poly(- methyl methacry-
Freseace of aeaobabb|es oa
saperhydrophob|c sarfeces
I
KEY CONCEPTS
Textured, superhydro-
phobic surfaces, which
encourage the formation
and stabilization of
bubbles, were prepared
and immersed in water.
X-ray scattering analysis
shows direct evidence for
the presence of nano-
bubbles on these sur-
faces.
The results show that
water only penetrates
about 5-10 nanometers
into 24-nanometer-
diameter cavities, about
15-30 layers of water
molecules.
T
E
C
H

B
E
A
T
Using small angle X-ray scattering, researchers have devised a
way to detect nanobubbles on superhydrophobic surfaces.
Texturing of surfaces with topographical fea-
tures such as cavities promotes the formation
and stabilization of bubbles.
1 4 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
late) was deposited
on the silicon sur-
face.
The block co-
polymer was an-
nealed at 180 C
in a vacuum oven
causing a micro-
phase separation to
occur. This heating
leads to the forma-
tion of poly(- meth-
yl methacrylate)
domains within a
polystyrene ma-
trix. Checco adds, The result is the creation of poly(-methyl
methacrylate) cylinders that are perpendicular to the silicon
substrate.
Treatment with ultraviolet light leads to the degradation
of the poly(-methyl methacrylate) cylinder blocks and the
crosslinking of the polystyrene. Checco says, Ultraviolet
light breaks bonds between the polystyrene and the poly(-
methyl methacrylate). With the degradation of the poly(-
methyl methacrylate), cylindrical cavities are formed that
are 24 nanometers in diameter, 40 nanometers apart and ar-
ranged in a hexagonal lattice.
These cavities are transferred onto the silicon surface
through an etching process. The right image in Figure 3 is a
scanning electron micrograph of the nanocavities.
To render the surface superhydrophobic, the silicon sur-
face was passivated with a 2.5-nanometer-thick monolayer
of octadecyltricholorsilane. Ultrapure water was then placed
between the textured surface and a thin Mylar lm. The
depth of the water layer was approximately 100 microns.
SAXS was used to detect the presence of the nanobubbles.
Checco says, We used SAXS because the size of the cavi-
ties in the textured surface is small enough to scatter x-rays.
SAXS was used to compare the difference between scattered
x-ray intensity in air and when water is placed on top of the
surface.
In analyzing the results, the researchers noted a reduction
in scattered x-ray intensity when water is used. Checco says,
Water is denser than air and should generate less scattered
x-ray intensity if it penetrates into the cavities. The results
show that the scattered x-ray intensity is higher than ex-
pected if water penetrates completely into the cavities. This
strongly indicates
that water penetra-
tion is only partial
and nanobubbles
are present in the
remainder of the
cavities.
Specically, the
researchers found
that water only
penetrates about 5
to 10 nanometers
into the cavities,
which corresponds
to approximately
15 to 30 layers of water molecules.
The left image in Figure 3 is a cartoon illustrating the
nanobubbles shape as inferred from x-ray measurements. In
the center image, the optical prole of a water drop placed on
the nanotextured surface is shown.
The scattered x-ray intensity measurements were taken at
various time intervals to evaluate the stability of the nano-
bubbles. Checco says, We found that the values remained
constant, which means that the nanobubbles are stable.
Conrmation of the existence of nanobubbles on supe-
rhydrophobic surfaces means that they clearly contribute
to water repellency. One potential application for utilizing
nanobubbles is in reducing the friction encountered by mi-
cro- and nanouids as they ow into small channels.
Further information can be obtained in a recent article
2

and by contacting Checco at cbeccoQbnI.gov.
RFRNL5
1. Canter, N. (2008), Controlling Surface Wettability, TLT,
64 (5), pp. 1213.
2. Checco, A., Hofmann, T., DiMasi, E., Black, C. and Ocko,
B. (2010), Morphology of Air Nanobubbles Trapped at Hy-
drophobic Nanopatterned Surfaces, Nano Letters, 10 (4),
pp. 13541358.
Neil Canter heads his own consulting company,
Chemical Solutions, in Willow Grove, Pa.
Ideas for Tech Beat items can be submitted to
him at neiIcunIerQcomcuaI.neI.
Ultraviolet light breaks bonds between the polystyrene and the
poly(-methyl methacrylate). With the degradation of the poly(-methyl
methacrylate), cylindrical cavities are formed that are 24 nanometers in
diameter, 40 nanometers apart and arranged in a hexagonal lattice.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 1 5
Figure 3 | The image on the left is a cartoon showing the shape of a nanobubble
inferred from x-ray measurements. The center image is an optical prole of a water
drop placed on a nanotextured surface and the image on the right is a scanning
electron micrograph of the nanocavities (Courtesy of Brookhaven National Laboratory)
k|cerdo J. he|a
Based in Atlanta, this Argentine-born STLE member
is the head of Conexo, which provides tribology innovations
in domestic and foreign markets.
By Thomas T. Astrene/Publisher
Ricardo Hein:
The quick le
- lcuaaea tcaexc, a luc|ltaals-
solution company, in 2004 in
Atlanta, Ga. The company has
three business divisions.
- Nls pa|lae| la tcaexc ls alsc |ls
wife, Alejandra Liernur Hein.
- A aallve cl ueacs Al|es,
Argentina, he has lived and
worked on three continents.
- la|aea a masle|'s cl stleate
degree in mechanical and marine
engineering from UdeMM
University in Buenos Aires.
Earned a masters in business
administration in strategic
management from DePaul
University in Chicago.
- hc|e l|aa ? yea|s cl expe|leate
in the lubrication eld for
industrial products. Twelve
years in international sales and
lubricant engineering for Castrol,
Shell and Fuchs.
- l|cvlaea luc|ltallca l|alala lc
internal and external customers in
75 countries.
- llueal la lalls|, !paals|, be|maa
and Portuguese.
MinuLes wiLh.
1L1: PIeuae expIuin Lonexo'a Ibree muin buaineaa diviaiona.
hein: Our goal in creating three divisions was to give focus to each busi-
ness activity. The rst division handles the North American market for
our industrial maintenance products and Rewitec Nanocoating.
The second division is dedicated to market testing instruments. Our
primary focus is on the Optimol SRV tribometer.
The third division markets these innovations in Latin America and
is co-managed by team members in 10 countries in the region. Each of
these divisions responds to each businesss unique challenges using the
knowledge and skill sets of our team members.
1L1: wbuI ia Ibe roIe of Ibe 5Rv in IriboIogy reaeurcb?
hein: To stay competitive our industry needs tools that bring the com-
plexity of eld conditions to the lab to decrease time and costs involved
in tribology research. Our SRV machine does just that. It simulates any
eld application with dynamic variables, different components, materials
and shapes under any condition.
1 6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
One of our customers, who researches solu-
tions for fretting wear, was able to use the SRV to
reproduce the conditions for fretting to appear in a
four-hour test with high load. The nding evolved
into a recent SAE norm for fretting wear testing.
Other customers use the engine-simulation
module for research in lubricants for modern en-
gines using fuels that change interaction with tri-
bological surfaces. The test results in the lab nd
direct correlation in the eld. The design of the
SRV is successful because it precisely congures to
specic testing needs.
1L1: wbuI ure Ibe aoIuIiona IbuI Lonexo
offera for geura?
hein: We market a gear conditioner, Rewitec
Nanocoating, which reverts wear and micropitting
and increases load-carrying capacity. The compo-
nents of the nanocoating intercalates in the metal
surface at the atomic level using the energy from
friction and load. It forms an interspersion of dif-
ferent ions that diffuse into the crystal structure of
the surface layers.
These layers sustain the rolling/sliding components even
where the lubricating lm breaks. We are working on wind
turbine gears where we increase the load-carrying area by 18
times on gear surfaces that are affected with micropitting.
In the aggregates and cement industries, we can stop vi-
bration stress peaks and smoothen the surface roughness by
3-5 times in large Falk and Symetro gears at full operation.
The benets for our customers range from doubling the life-
time of mechanical equipment to increasing productivity to
improving operational efciency.
1L1: 1bia ia u cbuIIenging Iime for Ibe IubricunIa
induaIry. wbuI udvice do you buve for IubricunI
murkeIera?
hein: We are facing a technological change in our industry,
with challenges that range from fuel economy to carbon
emissions to drain intervals. For transportation and mobile
equipment, higher efciencies, longevity issues and environ-
mental concerns challenge our industrial markets.
We must accept the current conditions as the new norm
without expecting the old market back. These times call
for stronger investment in research and in higher quality
standards for our products in order to answer the demand
for new and more evolved equipment where our previous
products no longer t. Innovation also must meet the new
standards of quality required by our customers. Lubricant
marketers should also consider accommodating the needs
of the international markets; the payoff is worthwhile. One
should see these challenges as a fresh start and restrategize
from here.
1L1: how did you end up in u cureer in IriboIogy/
IubricuIion engineering?
hein: After I graduated as a marine engineer in Argentina, I
was hired by Shell for the lubricants group in 1984 where I
was trained in this eld. Later on, working for Fuchs, I had
the opportunity to learn about metalworking and industrial
uids. Then, working for Optimol/Castrol some years later, I
gained a deep understanding of specialized and performance
products.
1L1: wbuI bud Ibe moaI inuence on you, peraon-
uIIy or profeaaionuIIy?
hein: Throughout these 25 years, I have worked with many
seasoned lubrication engineers and tribology experts who
shared their knowledge with me. Moving from Argentina to
Germany for six years before coming to the United States
15 years ago exposed me to remarkably different technologi-
We must accept the current conditions as the new norm
without expecting the old market back.
Ricardo (center) with Conexo customers at an industry trade show in
Buenos Aires, Argentina.
We are facing a technological
change in our industry, with
challenges that range from
fuel economy to carbon emissions
to drain intervals.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 1 7
cal mindsets. During those
years, I visited more than 80
countries and was exposed to
innumerable industries and
equipment to which I applied
my knowledge of lubricants.
1L1: If you couId buve
uny job or profeaaion in
Ibe worId, wbuI wouId
you cbooae?
hein: I would choose what I
do today. While I was com-
pleting my MBA about 10
years ago, I decided to build
my own company. The idea to
start Conexo and dedicate my
efforts to it did not resolve itself until I found new technolo-
gies and innovations in tribology. I am passionate about this
industry and understand it thoroughly. I feel very condent
working independently in it.
1L1: wbuI ure Ibe
greuIeaI cbuIIengea in
your job on u duy-Io-
duy buaia?
hein: I have to confront
skepticism; that requires re-
searching and documenting
cases to use as references.
In addition, when I export
to different countries on the
continent, I come across dif-
ferent cultures and languag-
es. This requires an open
mind and taking the time to
make sure that communica-
tion is clear and accurate.
Finally, being an entre-
preneur requires double the enthusiasm for growing the busi-
ness, especially after the current global nancial crisis.
You can reach Ricardo Hein at rbeinQconexoIeaI.com
Lubricant marketers should consider accommodating the needs
of the international markets; the payoff is worthwhile.
Ricardo and Alejandra Heinpartners in life as well as the
international tribology company Conexo.
1 8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
YOU ALSO GET SEA-LANDS WINNING TECHNICAL SUPPORT AND DEPENDABLE SERVICE.
821 Westpoint Parkway Westlake, OH 44145
440-871-7887
www.sealandchem.com E-mail: [email protected]
When you want a certain additive, chances are weve got it right
in stock. We maintain a huge inventory of quality products from
the best names in the business and we get them to you FAST!
Alkylpolyglucosides
Antioxidants
Biocides
Chlorinated Paraffins
Corrosion Inhibitors
Emulsifiers
Esters
Ether Amines
Fatty Acids
Fatty Alcohols
Lard Oils
Lubricant Additive Components
Lubricant Additives Packages
Methyl Esters
Neodecanoic Acid
Petrolatum
Polyalkylene Gylcols
Polyalphaolefins
Rust Preventatives
Sulfonates
Sulfurized Compounds
Synthetic Lubricants
Vegetable Oils
Waxes
White Oils
CHE MI CAL CO.
IN1RuL1IN
Carbon/Carbon composite (CCC), a well-recognized struc-
tural material, is known for its high temperature strength
and low relative density.
1
Since most of its applications in-
volve an oxidizing environment, the improvement in high
temperature oxidation behavior needs to be addressed. Fur-
thermore, fretting wear of uncoated CCC limits its lifetime
in applications, such as bushings in jet engines. A number of
investigators have reported enhancement in oxidation and
wear resistant of CCC in the presence of protective coating
layers.
2,3
But application of a surface and subsurface coating
system that can preserve its oxidation resistance along with
maintaining lubricity at high temperature remains an issue.
Therefore, the main motivation of this research was to in-
ltrate the porous CCC with atomic layer deposited (ALD)
lubricious oxides, which have been shown to provide good
tribological properties,
4
and determine the solid lubrication
mechanisms responsible for the improvements in wear resis-
tance with electron microscopy.
kPRIHN1AL
CCC contains ~21% total porosity and ~75% open poros-
ity. ALD was used as the deposition technique with metal-
organic precursors of diethyl zinc (DEZ), trimethyl alumi-
num (TMA), tetrakis (dimethylamido) zirconium (IV) and
DI H
2
O for deposition of ZnO, Al
2
O
3
, and ZrO
2
, respectively.
Pure ZrO
2
and one trilayer of ZnO/Al
2
O
3
/ZrO
2
of total thick-
ness ~220 nm were inltrated into the CCC monoliths. To
simulate the fretting wear of uncoated and coated CCC, a
high frequency reciprocating rig (HFRR) was operated at
room temperature with a normal load of 1 N using a 440
C stainless steel (SS) counterface under a stroke length of 1
mm and reciprocating frequency of 20 Hz for 130 min.
1r|bo|o|ce| lmprovemeat of 0erboaf0erboa 0ompos|tes by
lah|tret|oa of Ia0fk|
2
0
3
fIr0
2
So||d Labr|ceat 0oet|as
STUDENT POSTER ABSTRACT
Hamidreza Mohseni and Thomas W. Scharf (Advisor)
Department of Materials Science and Engineering & Institute for Science and Engineering Simulation (ISES)
University of North Texas, Denton, Texas
Hamidreza Mohseni received
his masters of science and bach-
elors of science degrees from
the University of New South
Wales (2006) and Sharif Uni-
versity of Technology (2004),
respectively. Currently, he is a
doctoral candidate working un-
der the guidance of professor
Thomas W. Scharf in the de-
partment of materials science
and engineering at the Univer-
sity of North Texas. His research interests include investigating
the tribological behavior of solid lubricant thin lms deposited
by atomic layer deposition and determining tribological mecha-
nisms using FIB-SEM, FIB-HRTEM/HRSTEM and XRD charac-
terization techniques. You can reach him at bumidrezumobae-
niQmy.unI.edu.
2 0 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Editors Note: The 2010 STLE Annual Meeting & Exhibition
in Las Vegas has given us another batch of winners from
this years Student Poster Competition, as we will be
publishing the winning abstracts written by student
members of STLE in this semiregular feature in TLT. I feel
it is important to foster as much growth and interest as
possible in these upcoming tribologists. The world has
many challenging technical issues, particularly in the en-
vironmental and energy elds that must be solved if we
are to continue evolving as a people. Young people who
are just entering the world of tribology will be the ones
tackling these challenging problems. They are the future
of this planet and the future of our professional society.
Please feel free to contact the authorsperhaps you can
offer some guidance, a mutual research or work opportu-
nity or even employment within your organization. This
is your chance to review tomorrows ideas and talent be-
fore they become todays products and competitors.
Evan Zabawski, CLS
Editor
ILLu51RA1Iv R5uL15
The ALD coatings exhibited excellent conformality and pore-
lling down to ~100 microns into the CCC.
Discernible visual differences in the wear track width
and amount of transfer lm on the counterface in Figure 1
indicates improved wear resistance with the ALD coatings.
A ~65% improvement in the wear factor (down to 1.5x10
-
6
mm
3
/Nm) was achieved with the ALD inltration of the
1 trilayer of ZnO/Al
2
O
3
/ZrO
2
compared to uncoated CCC.
Cross-sectional transmission electron microscopy (XTEM)
studies of the worn surfaces were performed to elucidate the
solid lubrication mechanisms responsible for these improve-
ments. Figure 2 shows XTEM images of the 1 trilayer (a)
before and (b) after HFRR testing. Wear is only observed
in the top ZnO layer, in which there was a high density of
sliding (shear)-induced stacking faults inside the wear tracks
shown in (c) in comparison to the unworn trilayer. The cor-
responding FFT-pattern in (d) was indexed to the (0002)
basal plane. Activating subsurface basal stacking faults will
promote intralm shear/slip and hence improve wear. The
slip of partial dislocations likely resulted from a dislocation
glide process along the basal planes. Also, a high resolution
scanning TEM (HRSTEM) image and corresponding energy
dispersive x-ray spectroscopy (EDS) elemental line scan are
shown in Figure 2 (e) and (f), respectively. The image and
line scan show the existence of a mechanically mixed layer
(MML) that contained no iron from the counterface. Intra-
lm shear with this friction induced subsurface (mechani-
cally mixed layer) aids in shear accommodation (prevents
brittle fracture).
5uHHARY
ALD ZnO/Al
2
O
3
/ZrO
2
nanolaminates are good candidates for
providing low friction and wear resistant surfaces and inter-
faces in moving mechanical assembles such as CCC bush-
ings that experience fretting wear.
ALkNwL6HN15
The authors acknowledge the support of UNTs Center for
Advanced Research & Technology (CART).
RFRNL5
1. Sheehan J.E., Buesking K.W. and Sullivan B.J. (1994),
Carbon-Carbon Composites, Annu. Rev. Mater. Sci., 24,
pp. 19-44.
2. Huang, J.F., Li, H.J., Zeng, X.R., Deng, F., Xiong, X.B. and
Li, K.Z. (2007), Oxidation Resistant Yttrium Silicates Coat-
ing for Carbon/Carbon Composites Prepared by a Novel In-
Situ Formation Method, Ceramic Inter., 33, pp. 887-890.
3. Park, S.J., Seo, M.K. and Lee, J.R. (2002), Effect of Oxi-
dation Inhibitor on the Low Energy Tribological Behavior of
Carbon-Carbon composites, Carbon, 40, pp. 835-843.
4. Doll, G.L., Mensah, B.A., Mohseni, H. and Scharf, T.W.
(2010), Chemical Vapor Deposition and Atomic Layer De-
position of Coatings for Mechanical Applications, J. Ther-
mal Spray Tech., 19, pp. 510-516.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 2 1














Figure 2 | HRTEM analyses: (a) 1 trilayer before HFRR test, (b) 1
trilayer after HFRR test, (c) high density of basal stacking faults
inside the wear track along the reciprocating sliding direction, (d)
FFT pattern shows diffraction spot of (0002)-basal planes of ZnO,
(e) HRSTEM drift corrected line scan prole of MML and (f)
corresponding elemental line prole. Pt is a protective layer for
FIB-sectioning.
Figure 1 | Optical images of worn surfaces (Top: transfer lms on
440 C SS counterfaces and Bottom: wear tracks).
CCC-Uncoated CCC-ZnO/Al
2
O
3
/ZrO
2
CCC- ZrO
2

rom time to time I and others have written about the techni-
cal problems besetting the wind turbine industry. In the short
term, wind turbines have been a boost to our industry in the
way of funded research into how to manage these problems. Our
friends at ASME have a new technical committee, and STLE will
form either a new industry council or technical committee focus-
ing on the technology. In addition, the Society of Maintenance
and Reliability Professionals is interested in the condition moni-
toring problems associated with wind turbines. So the technical
community is clearly ramping up, thanks to your tax dollars.
Now Ive been accused of being a troglodyte, curmudgeon,
grumpy old man or, at the very least, benignly cynical. All are
likely true, especially when one thinks of politicians as they
intermingle with the economics of science and technology. So
What happens when national energy policy clashes with the reality
of engineering science? Check your wallet for the answer.
LUBRICATION FUNDAMENTALS
Dr. Robert M. Gresham / Contributing Editor
F
Fo||oW the
moaey
kY LNLP15
- wlaa la|ms la l|e U.K. a|e |l|ly cve|ameal
subsidized. As a result, the government pays
a so-called, renewable obligation.
- !|e cla |ale lc| eletl|ltlly a|cuaa l|e
country is on the order of 10 cents a kilowatt
hour.
- Ualess we l|lcclclsls taa pull a |accll cul
of a hat pretty soon, new wind turbines will
incur high maintenance costs.
2 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
when someone suggests politics or
politicians being involved, I imme-
diately resort to the clich, Follow
the Money!
Lets start with the United King-
doms experience, since Europe
seems to be a little ahead of the
U.S. in wind energy. Many of their
wind farms are located in the Cam-
brian Mountains, a pretty windy
place and remote as well. It seems
the debate in the U.K. over the ef-
cacy of wind farms, as in the U.S.,
is not without a certain hyperbole
on both sides, and the claims seem
to border on the ridiculous. For
example, a spokesman from Re-
newables U.K., says, The U.K. is
the windiest country in Europe, so
much so that we could power the
country several times over using
this free fuel.
He further claims that existing wind turbines have the
capacity to prevent 3.7 million tonnes of carbon dioxide per
annum. However, on closer inspection, to do this the wind
turbines would have to generate electricity at 100% capacity,
100% of the time. Of course, the wind does not blow 100%
of the time. Further, commercial wind turbines need about
6-10 mph wind to operate and automatically stop at 55 mph
to protect the equipment. When wind is in this outlet range,
electricity is not constant. According to U.K. government g-
ures, wind turbines operate at only 27% of capacity.
That wouldnt be so bad if we could store electricity, but
of course we cant. Thus, you have to either install many
more turbines or have standby alternative generating capaci-
tylike a fossil fuel or nuclear power plant. Germany, which
has the largest number of wind turbines in Europe, is build-
ing ve new coal power stations, which it does not otherwise
need, purely to cover the power uctuations from its wind
farms.
When you follow the money, what you nd is that in
the U.K. wind farms are highly government subsidized. The
government pays a so-called, renewable obligation (RO).
The cheapest way to collect the RO is to build a wind farm.
The way it works out, it costs about 2 million to build a
wind turbine. The RO is about 138,000 annually, and with
sales of the generated electricity added an industrious wind
farmer can clear about 300,000. Thats about a 6-7 year pay-
back, not bad. However, in essentially double that time, you
can still get your money back without selling a single spark
of power. Thats a pretty good low-risk deal, unless youre a
taxpayer.
Recently, after 10 years of negative lobbying by the pre-
sumed liberals of New England, the U.S. government ap-
proved the Cape Wind project in Massachusetts. The idea
for this project, youll recall, was to install a huge wind farm
in Nantucket Sound, playground to a large number of sea-
oriented folks, some of whom live in Hyannis Port. Their
opposition was primarily based on aesthetics, or lack thereof,
and some notions about negative environmental impact on
the ecosystem, hazards to sailors, shipping and the like.
For the pro-wind people, this project makes a lot of sense,
as the farm is going to be in a windy place with a relatively
shallow ocean bed and a relatively high density of people
(which reduces transmission distance and cost) and who,
theoretically, should be more culturally predisposed to such
a solution. However, the debate quickly turned into a case of
its OK in your backyard but not mine. Alas, but follow the
money.
According to a May Wall Street Journal article, construc-
tion of the Cape Wind project is massively subsidized by the
federal government. This, of course, means that not only will
New Englanders taxes rise but so will mine and yours. But,
hey, were all good guys.
But thats not all.
The going rate for electricity around the country is on the
order of 10 cents a kilowatt hour. Apparently, Cape Wind
folks have asked the state for a 15-year contract at ~20 cents
a kilowatt hour with a 3.5% escalation clause. In other words,
not only will New Englanders taxes rise, but the cost of elec-
tricity, at least from this source, will essentially double. Add
to that is the problem that has STLE all abuzz these days
premature failure of wind turbine gearboxes and some key
bearing systems. Unless we tribological types can pull a rab-
bit out of a hat pretty soon, these brand new wind turbines
are likely to incur high maintenance costs as well.
So, as the folks of Nantucket, Hyannis Port and Marthas
Vineyard are sitting on their verandas and yachts, sipping
Chablis, eating brie and watching their new windmills going
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 2 3
Many people believe wind turbines will despoil the aesthetics of areas like Nantucket Sound.
round and round, they can contemplate the fact that they
paid extra to build them, are required to buy the product
(electricity) and must pay about double the normal price for
the privilege. Further, they can only hope that we tribolo-
gists, design engineers, fabricators and operators can x the
technical problems soon or else their costs will climb dis-
proportionately higher yet. When you follow the money, its
kind of a lousy deal.
But in keeping with my curmudgeonliness, lets look at a
current energy-related crisis and see what happens when we
follow the money again. Of course, Im referring the BP deep-
water oil spilla tragedy by any and all standards. Lets say
we, as a nation, take the approach we
did with nuclear power. We also had
serious problems at Three Mile Island,
followed not long after by Chernobyl.
These two catastrophic events made
us very adverse to risk in the nuclear
power industry in the U.S. So rather
than deal with the problem, we quit
building nuclear power plants and be-
came all the more dependant on fossil
fuels as an energy source for electrical
power, further diverting fossil fuels
from other applications.
France on the other hand contin-
ued with nuclear power development
and, with no major negative events,
now derive about 85% of its electric-
ity from nuclear power. To that extent,
France is that much less dependant on
foreign oil.
What if, after the BP oil spill crisis
is over or at least mitigated, we con-
tinue to ban most offshore oil drilling,
both deep water and shallow? What if
we raise taxes on oil suppliers to raise
so-called trust funds to mitigate
possible future risk or spills? What if
we excessively tighten the regulatory
environment to near punitive levels?
What if several of these companies
crumble and go into bankruptcy or
just leave the country?
We will be even more dependant
on foreign oil, and prices will signif-
icantly rise. And it is not just gaso-
line prices that will rise but also the
myriad of other products that are de-
rived ultimately from crude oil such
as plastics, synthetic bers for clothes,
furnishings and a wide variety of addi-
tives that go into foods, paints, fertiliz-
ers, detergents, cosmetics, packaging,
drugsthe list is almost endless.
If this happens our friends in Nantucket Sound might feel
a little smug. Their 20 cents-plus-per-kilowatt hour electric-
ity might suddenly seem like a bargain instead of a lousy
deal.
If you feel like you need a crystal ball to sort all this out,
just remember one rulefollow the money.
You can reach Bob Gresham,
STLEs director of professional development,
at rgreabumQaIIe.org.
2 4 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
NCeec: ccmmiImenI Ic fncing re|icL|e cnc cffcrccL|e prccucI:
hc: Leen unwcvering. Mcre cnc mcre Lu:ine::e: cre fncing cuI
whcI cIher: hcve kncwn c|| c|cng- NCeec ccre: cLcuI Ihe uc|iIy
cf prccucI iI :e||:, cnc Lcck: IhcI up wiIh Ihe Iechnicc| experIi:e Ic
Le cn cI-ycur-:ice cc||cLcrcIcr.
Lecrn mcre cLcuI hcw e:Ier: cf SeLccic /cic
cnc cc:Icr ci| cerivcIive: ccn he|p ycur
ccmpcny. Vi:iI nceec.ccm cr e-mci| Ncreen
Cherry cI [email protected].
www.nceec.ccm
v0'r0 cr)0
t0 )[0 +r
t0 mt q++rJq[)0
q0J r0))q[)0
cqtr prJct
8eg|ster today at www.po|ar|s|abs.com/tra|n|ng 877-808-3750 ext. 24
Stop reacting to
individual reports...
...and start managing a more
effective program today!
Oil Analysis & Cooling System Maintenance
Atlanta August 17-18 Oil Analysis
Salt Lake City September 20 Cooling System
September 21-22 Oil Analysis
Edmonton October 26 Cooling System
October 27-28 Oil Analysis
TRAINING
Coolant Analysis Webinars
Cooling System Preventive
Maintenance for Todays Engines September 8
Coolant Formulations & Management September 15
Testing & Data Interpretation September 22
Maintaining Equipment in the Field September 29
Fuel Analysis Webinars
Why Test Diesel Fuel? October 6 & November 3
How Clean is Your Fuel? October 7 & November 4
OMA Level I
Certication
il analysis is a powerful tool in the machine condition monitoring
toolboxif used properly. Much like other technologies, it performs
best within a well-developed plan. When accomplished, a well-devised
plan can provide an effective long-term view into the health of any ma-
chine with lubricated components.
TLT has provided STLE members with information about test methods,
alarm methods and about the best alarm t for the noted test methods to
construct an effective oil analysis approach. The November 2009 TLT pro-
vides an overview that would be a worthwhile preview to this article. This
article can be found on the STLE Web site (www.aIIe.org).
The 2009 article indicates that there are three common alarm types for
grading the underlying problems for sumps and lubricated components.
BEST PRACTICES
Mike Johnson
Fourth of a ve-part series on oil analysis (January, April, July, August, December)
O
Following these strategies
allows you to calculate
rate of wear generation,
a more insightful metric
than total wear.
Strete|c o|| eae|ys|s: 1|medepeadeat
e|erms for exteaded |abr|ceat ||fecyc|es
kY LNLP15
- !lme-aepeaaeal ala|ms |eveal l|e
actual rate of wear per unit of
time.
- !|e uall cl llme tcula ce |eplatea
with other incremental units,
including production values, miles
or operating cycles.
- As llme lale|vals pe| cll t|aae
increase, the rate of top-up
volume should be factored to
reect dilution effects.
2 6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
These are statistical (alarms used to identify machine wear
problems), absolute (aka aging alarms, used to identify lubri-
cant health and degradation) and percentage-based alarms
(used for lubricant health and contamination monitoring).
The focus of Part IV of this ve-part series addresses rate-
of-change (ROC) and volume-compensation alarms. These
are common process alarms and could be effectively used to
track machine conditions operating under a variety of consid-
erations. When coupled with top-up volume normalization,
ROC alarms may help the engineer make decisions about
the lubricants long-term surface protection characteristics,
something difcult to track with routine lubricant-analysis
processes. Accordingly, this approach helps with evaluations
between high-performance and commodity grade products.
Aside from wear debris analysis, this technique also could
be used to measure contamination control effectiveness for
hydraulic and circulation systems and improve sump life-
cycle management.
1IH-PNN1 ALARH5
There are a variety of circumstances under which a ROC or
time-dependent alarm is useful. For instance, if the site had
a machine that was sputtering through nal cycles/hours/
units of production and it was necessary to stop the machine
before it failed in service, one might use ROC alarms to track
the increase in the selected indicator of failure over short
blocks of time.
For example, suppose a machine has already produced
wear debris indicating an aggressive wear pattern from the
routine oil analysis cycle. Perhaps a degraded bearing or gear
surface condition is also evident in vibration data. The oil is
changed and an inspection conducted to obtain corroborat-
ing evidence of a failure symptom. Following conclusion
that a repair is pending, the owner wants to squeeze as much
time from the machine as possible but wishes to
do so safely to avoid collateral damage. An ROC
alarm could provide the owner with a wear de-
bris value per unit of time that tells more about
the ongoing rate of wear development than an
interval reading.
A ROC alarm could be used to measure the
rate of oxidative degradation of a large sump vol-
ume as well. For instance, as oil ages its rate of
degradation often increases. If the oil is hot, wet
and/or contaminated with iron or copper wear
debris, the rate of decay could accelerate.
Coupled with oil top-up volumes, this meth-
od could be used to gauge engine wear for large
industrial engines driving ships, trucks and
earth-moving vehicles. Particularly, as it pertains
to wear debris for engines, the small wear particle
size in engine oil analysis is below typical OEM lter element
capture size, meaning wear data should be well represented.
wAR RA15 u5IN6 RL ALARH LIHI15
A ROC alarm is applied to a system where the amount of
change must be considered relative to the amount of time
through which the change occurs.
Consider the data set in Figure 1. This represents the
amount of wear that has occurred in a compressor sump
during a 24-month
period, beginning
with an oil change
(Month 1) and ac-
counting for rou-
tine top-ups that
have occurred dur-
ing the analysis pe-
riod.
Figure 2 shows
the iron val-
ues plotted out
over the period
of samples. The
plot would seem-
ingly suggest that
the wear problem
is becoming sig-
nicant. The black
trend line adds cre-
dence to the con-
cern. However, the
timeline for wear
generation should
be considered.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 2 7
Figure 1 | Compressor sump 24-month
iron wear and top-up volume record.
Figure 2 | A 24-month wear debris plot and trend.
The chief concern is the ongoing rate of wear generation.
Total wear is important but can be misleading.
The chief concern is the ongoing rate of wear generation.
Total wear is important but can be misleading. The iron value
shown in Month 24 reects two years of accumulation, less
leakage, and does not reect the extent to which the rate of
wear is increasing or not. Figure 3 reects the same data set
adjusted to reect wear produced per month.
After accounting for growth over time, it appears that the
rate of wear generation uctuates between 5 and 30 ppm per
month but is steady. The plateau toward the end of the se-
quence may be explained by a variety of conditions, includ-
ing weather, operational inconsistencies or operating loads.
The formula for normalization for any type of data set,
both process and maintenance measurements, is:
(
P
0
P
-1
)
Pf =
(
T
0
T
-1
)
where:
P
f
= Factored data point, in this instance iron
P
0
= Current data point
P
-1
= Previous data point
T
0
= Current time period
T
-1
= Previous time period
The values for P, time period, also could be units of pro-
duction, tons, miles, years or any other parameter. The ac-
tual units must be the same, but the nature of the units can
be any parameter dened by the user.
Also keep in mind that the values for V
-1
should reect
an actual condition. In this instance, the iron reading for V
0
,
which is Month 1, follows an oil change. The wear debris
from the previous period is not provided, so the value for V
-1
is zero. If the parameter measured (AN/BN, RPVOT hours,
viscosity, etc.) has a denable starting point, use that value.
Wear rate for period zero is not measurable.
wAR RA15 6 k1N RAIN IN1RvAL5
Extension of drain intervals is an expected outcome of im-
provements in lubrication practices. Mineral oil lifecycles
should be extendable by a factor of three to ve if the lubri-
cant is maintained in a cool, clean (no atmospheric contami-
nants and/or wear debris) and dry state. Heat, contaminants
and moisture all contribute markedly to oxidation and short-
ened lifecycles.
Following the example, assume this
compressor has a 55-gallon sump and
experiences nominal leakage across
separator and seals. Rather than change
oil at the traditional one-year intervals,
the owner is operating on a condition-
based change plan. As time passes, to-
tal wear in the sump increases, as re-
ected in the concentration of wear per
unit of oil (ppm). When oil is added to
the sump, the existing concentration of
wear is diluted, being distributed into
a larger unit of oil volume. If a signi-
cant amount of oil is added, it can ap-
pear that the rate of wear is less severe
than it truly is. To avoid misinterpre-
tation, the top-up volumes should be
factored into the wear rate.
A simple formula to account for added volumes is:
(
V
0
+ V
1
+ V
2
+ V
n
...
)
Vn
p
=
(
V
0
)
where:
Vn
p
= Normalized volume per period
V
0
= Initial volume
V
1
= Period 1
V
2
= Period 2
V
n
= Period N
In this example, wear debris is factored by the period
(month) to determine the rate of wear. This value can be
further factored by the cumulative top-up for that period to
provide a monthly wear rate accounting for the total amount
of oil added to date.
Treating each monthly data point with the two factors
(P
0
* P
f
* Vn
p
) enables the reliability engineer to see the ongo-
ing change in light of both time and top-up volume. When
2 8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Figure 3 | The 24-month wear trend factored to reect debris generated each month.
There are three common alarm types for grading the underlying problems
for sumps and lubricated components: statistical, absolute and percentage.
applied to the data set from Figure 1, the current plot and
trend line, as shown in Figure 4, provides a clearer picture of
the machines health. Comparing Figures 2, 3 and 4, it is ap-
parent that the wear rate is still tolerable for the machine, but
after factoring for top-ups, the rate is twice that perceived
from the ROC values. In both instances, the impression
of dramatic growth in wear over the 24-month period is
lessened.
PRAL1ILAL APPLILA1IN FR NRHALIZ A1A
Whether a compressor sump, gearbox, engine or other ma-
chine type that produces wear debris, factoring data and
then applying rate alarms allows the owner to make more
informed decisions. Showing the allowable wear rate of <10
ppm per month for the rst alarm level overlaid on the graph
(see Figure 5), the reliability engineer can see that the rate of
wear is higher than desired and warrants action but is not a
cause for dire concern. Options to consider for additional
inspections and analysis could include:
Iirsi, vcrily ilai ilc samplcs arc
representative of current condi-
tions.
Lvaluaic wiil wcar dclris analysis
(ferrography) to identify the wear
mode. For cutting wear, improved
system cleanliness; for scufng
wear, improved lubricant quality or
increased viscosity.
Inspcci sysicm lccd lincs io assurc
lubricant delivery.
Vcrily ilai ilc sump icmpcraiurc
is within the suggested operating
range.
Vcrily ilai maclinc pcrlormancc
is within the expected prole (no
misalignment, looseness or load
balance issues).
5uHHARY
Machine owners can expect lubricant
lifecycles to become extended as they
pursue precision lubrication activities.
Total wear metals values may give the
appearance of a problem where none
exists. Total wear debris measurement
is most common but does not reect
the state of ongoing machine change
and machine health,
To avoid mischaracterizing data, re-
liability engineers may wish to factor
key data points (lubricant health, ma-
chine wear debris, contaminant load)
to allow for changes over time. Time normalization allows
the user to grade the data points for change per unit of time.
Sump volume normalization allows the user to grade the
data points for change, both for time and for any changes in
total sump volume during the whole time increment.
Statistical alarms are particularly helpful if data sets
from identical makes and models of a machine type can be
grouped to create an expected typical prole for that make
and model. Once done, alarm sets can be constructed at 1, 2,
3 and 6 sigma values to provide truly meaningful reporting
health interpretation.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 2 9
Mike Johnson, CLS, CMRP, MLTII, MLA1, is
the principal consultant for Advanced Machine
Reliability Resources, in Franklin, Tenn.
You can reach him at mike.jobnaonQ
preciaionIubricuIion.com.
Figure 4 | Data set normalized by time and top-up volume.
Figure 5 | Initial alarm overlaid to contrast actual against allowable wear results.




















kY LNLP15
- Ier nearly Ihree decades many experIs n Ihe rellng
hearng cemmunIy have advecaIed Ihe neIen ef a
faIgue lmI.
- Newever, Ihs arIcle's auIher heleves daIa dees neI
supperI Ihe cencepI ef a faIgue sIandard, whch s a
crucal parI ef Ihe IS0 ?81: ?001 sIandard.
- Ihe auIher heleves Ihe IS0 ?81: ?001 s a msIake
and sheuld he wIhdrawn.
la seerch of e fet|ae ||m|t:
k cr|t|ae of lS0 Steaderd
28I:2007
COMMENTARY
Erwin V. Zaretsky, P.E.
J 0 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
In Part II of our debate, a leading gure in the
bearings community presents his case that the new
standard should be withdrawn.
SO Standard 281:2007 is based on a mistake. It is a disser-
vice to the technical and industrial communities around
the world. A fatigue limit should not have been includ-
ed in the standard based on available data. I recommend
that ISO withdraw Standard 281:2007.
whA1'5 1h PRLH?
ISO Standard 281: 2007 incorporates the concept of a fatigue
limit within its bearing life calculation method. Further-
more, this standard assumes that the bearing is made from
AISI 52100 bearing steel, is lubricated with mineral oil and
has values for raceway and ball nishes that are not stated.
It is based on work rst published by Stathis Ioannides
and the late Tedric A. Harris at the SKF Engineering and Re-
search Centre in Nieuwegein, The Netherlands, in 1985, as
well as later work published by Ioannides together with G.
Bergling, and A. Gabelli, also of SKF, in 1999. The fatigue
limit used in ISO Standard 281:2007 is a Von Mises stress of
900 MPa (130,500 psi), which corresponds to a maximum
Hertz contact stress of 1500 MPa (217,500 psi). To many
bearing companies and some engineers, it is the preferred
method to predict ball and roller bearing lives under load
and speed. However, what if a fatigue limit does not exist for
high hardness alloy steels such as AISI 52100? Is a fatigue
limit for through hardened bearing steels a reality or a prod-
uct of imagination? Where is the data?
whA1 I5 A FA1I6u LIHI1?
Before we begin our search for this fatigue limit, we should
understand what it is that we are looking for so that we can
recognize it.
To begin, it is generally accepted that the life of a compo-
nent in terms of repeated stress cycles is inversely related to
a critical shearing stress, , to a power where
L ~ 1/
n
[1]
I
Abcut the authcr
Erv Zaretsky is an engineering consultant to industry
and government, noted author and lecturer, adjunct
professor of engineering at Case Western Reserve
University and distinguished research associate at the
NASA Glenn Research Center in Cleveland, where he
recently retired. He has more than 50 years of experi-
ence in mechanical engineering related to tribology,
rotating machinery, fatigue and probabilistic life pre-
diction. He has written over 200 technical papers and
two books and has lectured widely throughout North
America, Europe, Asia and the Middle East. He is a
Fellow of both ASME and STLE and a member of two
ANSI/ABMA Committees on Rolling Bearing Stan-
dards. In 1998 he was appointed to the Senior Scien-
tic and Professional Corp., the highest rank achiev-
able by a federal engineer or scientist. He has received
four I-R 100 Awards established by the editors of R&D
Magazine. In 1999 STLE presented Zaretsky with the
Wilbur Deutsch Memorial Award, which honors the
most outstanding paper written on the practical as-
pects of lubrication. He is also the recipient of nu-
merous NASA awards for his contributions to the
Space Program, among which are the NASA Medal for
Exceptional Engineering Achievement, NESC Direc-
tors Award and the Astronauts Silver Snoopy Award.
You can reach him at evzureIakyQgmuiI.com
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 3 1
The standard is a disservice
to the technical and industrial
communities around the world.
The critical shearing stress, , can be the orthogonal shear,
maximum shear or octahedral shear stress. Some investiga-
tors have also used the Von Mises or effective stress as the
decisive stress. This relation when plotted on log-log graph
paper will plot as a straight line, as illustrated in Figure 1(a)
and referred to as Case 1. The slope or tangent of the line is
the value of the stress life exponent, n.
The classic Whler curve illustrated in Figure 1(b) in-
troduces the concept of a fatigue limit where the stress-life
relation is that of Equation [1] until the stress reaches the
value of the fatigue limit,
u
, at approximately 10
6
to 10
7
stress cycles, where the life is considered innite. That is,
no fatigue failures would be expected to occur. This will be
referred to as Case 2.
In practice, fatigue data for material types that have fa-
tigue limits do not manifest a linear line on an log-log S-N
plot but a curved line, as illustrated in Figure 1(c). This will
be referred to as Case 3. The apparent relation between life,
shearing stress and fatigue limit,
u
, is
L ~ 1/( )
n
[2]
u
From Equation [2], for an applied stress and the pres-
ence of a fatigue limit, the resultant life will be longer than
that from Equation [1] with a fatigue limit. ISO Standard
281:2007 with a fatigue limit takes the form of Case 3, Figure
1(c).
whA1 Au1 1hR 51ANAR5?
The American Standards Institute (ANSI)/American Bearing
Manufacturers Association Standards 9 and 11 are used for
the load ratings and life prediction of ball and roller bearings,
respectively. These standards with various updates through
the years were adopted by the ABMA in 1953. ABMA change
its name from the Anti-Friction Bearing Manufactures Asso-
ciation (AFBMA) in 1993.
ANSI/ABMA Standards 9 and 11 are based on the Lun-
dberg-Palmgren theory published in 1947. However, this
theory of Gustaf Lundberg and Arvid Palmgren dates back to
1924. At that time, Palmgren who had been working at SKF
in Sweden since 1917 published a paper in Germany outlin-
ing his approach to bearing life prediction. He presented an
empirical formula based on the concept of an L
10
life, or the
time that 90% of a bearing population would equal or exceed
without rolling-element fatigue failure. This 1924 paper by
Palmgren is the rst time in the literature that a probabilistic
approach to life prediction of a machine element was formu-
lated.
A fatigue limit should not have been included in the standard
based on available data.
3 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
ln L, stress cycles
Fatigue limit
Fatigue limit

u
10
4
10
6
10
7
10
5
10
8
10
9
10
10
10
11
0.1
(a)
0.5
1.0
n

ln L, stress cycles
0
4
10
6
10
7
10
5
10
8
10
9
10
10
10
11
0.1
(b)
0.5
1.0
n

ln L, stress cycles
0
4
10
6
10
7
10
5
10
8
10
9
10
10
10
11
0.1
(c)
0.5
1.0
n

1/n
1/n
1/n
L ~ 1/
n
L ~ 1/
n
L ~ 1/
n
L ~
L ~ 1/(
u
)
n
ISO 281: 2007
Figure 1 | Stress-life (S-N) curves illustrating concept of fatigue
limit. (a) Case 1, no fatigue limit. (b) Case 2, classic Whler curve
where life is innite below fatigue limit u. (c) Case 3, life
approaches innity as applied stress approaches fatigue limit
u, as asymptote.
What if a fatigue limit does not
exist for high hardness alloy steels
such as AISI 52100?
ln L, stress cycles
Fatigue limit
Fatigue limit

u
10
4
10
6
10
7
10
5
10
8
10
9
10
10
10
11
0.1
(a)
0.5
1.0
n

ln L, stress cycles
10
4
10
6
10
7
10
5
10
8
10
9
10
10
10
11
0.1
(b)
0.5
1.0
n

ln L, stress cycles
10
4
10
6
10
7
10
5
10
8
10
9
10
10
10
11
0.1
(c)
0.5
1.0
n

1/n
1/n
1/n
L ~ 1/
n
L ~ 1/
n
L ~ 1/
n
L ~
L ~ 1/(
u
)
n
ISO 281: 2007
Assuming that a ball or roller bearing was properly de-
signed, manufactured, installed, lubricated and maintained,
rolling-element fatigue would limit the useable life of the
bearing. In the life equations that Palmgren presented, he in-
corporated a fatigue limit or load below which no failure will
occur as well as a time or location parameter before which
time no failure should occur. This fatigue limit is represented
by Equation [2], Case 3 of Figure 1(c).
Over the next 12 years, Palmgren evolved his bearing life
prediction formulae. In 1936 Palmgren published the fol-
lowing:
For a few decades after the manufacture of ball bearings
had taken up on modern lines, it was generally considered
that ball bearings, like other machine units, were subject to
a fatigue limit, i.e., that there was a limit to their carrying
capacity beyond which fatigue speedily set in but below
which the bearings could continue to function for innity.
Systematic examination of the results of tests made in
the SKF laboratories before 1918, however, showed that no
fatigue limit existed within the range covered by the compar-
atively heavy loads employed for test purposes. It was found
that so far as the scope of the investigation was concerned,
the employment of a lighter load invariably had the effect
of increasing the number of revolutions a bearing could ex-
ecute before fatigue set in.
It was certainly still assumed that a fatigue limit coex-
isted with a certain low specic load, but tests with light
loads nally showed that the fatigue limit for innite life, if
such exists, is reached under a load lighter than all of those
employed and that in practice the life is accordingly always
a function of load.
In other words, Palmgren in 1936 concluded that for
bearing steels, and more specically for AISI 52100, no fa-
tigue limit existed as a practical matter. His conclusion is
represented by the stress-life relation of Equation [1], Case
1 of Figure 1(a).
hw I LIF FAL1R5 HR6?
The life equations were further empirically rened by
Palmgren until 1947 when he and Lundberg published their
life theory based on the Weibull distribution function and
life theory. The foundation of the Lundberg-Palmgren theo-
ry and the ANSI/ABMA and later the ISO (International Or-
ganization for Standardization) Standards are based on three
variables and three exponents. These are as follows:
| |
e
h
e e
c
z
V
L
1
1 1
~
(

t
[3]
where L = life; = critical shear stress; V = stressed volume; z
=depth to critical shear stress; c =shear stress-life exponent; e
= Weibull slope or modulus; and h = exponent relating depth
to the maximum shear stress to life. This formula becomes

p
eq
D
P
C
L
|
|
.
|

\
|
=
10
[4]
where L
10
= life at a 90% probability of survival or a 10%
probability of failure; C
D
= dynamic load capacity or the
theoretical load on the bearing that will result in an L
10
life
of one million inner-race revolutions; P
eq
radial load on the
bearing and p = load-life exponent.
Equation [4] is identical to that proposed by Palmgren
in 1924. From the 1947 Lundberg-Palmgren theory, the
load-life exponent, p, equals 3 for ball bearings and 4 for
roller bearings. However, Lundberg and Palmgren in 1952
proposed p = 10/3 for roller bearings. These exponents were
chosen to reect the pre 1940 SKF bearing data.
In the late 1950s and early 1960s, the steel industry intro-
duced vacuum-processed bearing steels. Vacuum processing
resulted in bearing steels that were relatively free of hardened
inclusions and resulted in signicantly longer bearing lives
than were achieved with air-melted (AM) steels. These vacu-
um-processed steels, such as AISI 52100, exhibited load-life
exponents p of 4 for ball bearings and 5 for roller bearings,
instead of 3 and 10/3, respectively used by the ANSI/ABMA
and ISO Standards. This coupled with improved manufac-
turing techniques and lubrication technology resulted in
bearing lives signicantly longer than those benchmarked
for the Lundberg-Palmgren equations.
In order to account for these technological achievements,
I was invited by Dr. Robert L. Adamczak, chairman of the
ASME Lubrication (now Tribology) Division to chair a Life
Factors Committee for the ASME Lubrication Division in
1969 to codify these improvements. The committee was
comprised of prominent engineers and scientists from in-
3 4 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Vacuum processing resulted in bearing steels that were relatively free of
hardened inclusions and resulted in signicantly longer bearing lives than
were achieved with air-melted steels.
dustrial users, the bearing companies and government. The
results of the committees work were published in September
1971 by ASME in Life Adjustment Factors for Ball and Roller
BearingsAn Engineering Design Guide.
There were ve life factor categories. These categories
were: Material, Processing, Lubrication, Speed and Misalign-
ment. Nineteen sets of supporting bearing and specimen
rolling element fatigue data also were published in the guide.
From these data, the ASME committee concluded that, for
a specic material and heat treatment, the life factors were
independent of each other. The material and processing
life factors were benchmarked to air-melt (AM) AISI 52100
steel.
Six years after publication of the ASME design guide, the
concept of life factors was adopted by ANSI, ABMA and ISO
in their bearing life standards. Instead of the ve life fac-
tor categories of the ASME, the standards used three. These
were: a
1
, Reliability; a
2
, Material and Processing Properties;
and a3, Operating Conditions. The ASME categories became
subcategories under Categories a
2
, Material and Processing
Properties; and a
3
, Operating Conditions. The individual
bearing company and/or user could choose the life factors
they think were appropriate for their application. Equation
[5] can now be written as:

p
eq
D
P
C
a a a L

3 2 1 10
[5]
In the early 1970s after the ASME publication, Eric Bam-
berger, Hans Signer and I, under NASA sponsorship, were
the rst to design and test bearings that we had manufac-
tured from double-vacuum melted (VIM VAR) AISI M-50 at
elevated temperatures. This was the rst time double-vac-
uum melted AISI M-50 was used. We tested at speeds to 3
million DN (DN is a speed parameter, the bore of the bearing
in millimeters multiplied by its speed in RPM) and tempera-
tures of 218 C (425 F). These new test data represented ap-
proximately 150,000 bearing test hours. The bearing lives
were over 100 times that predicted by the unfactored Lund-
berg-Palmgren equations. This bearing technology has been
incorporated into most, if not all, commercial and military
aircraft ying today.
Since the ASME publication, much additional bearing
fatigue data have been generated. Improvements have been
made in bearing manufacture, heat treatment and raceway
surface nish. Better quality control techniques have been
instituted by the bearing companies. There was also the in-
troduction of hybrid bearings. Bearing rework became an
acceptable procedure. As a result of these changes, it be-
came apparent that the 1971 ASME life factors needed to be
updated.
In 1985 I was invited by STLE to form a Life Factors
Committee to update the life factors that were published by
the ASME in 1971. A thorough search of the literature and
inquiries to researchers in the eld at that time revealed no
data to support a fatigue limit for through-hardened bear-
ing steels. STLE published Life Factors for Rolling Bearings
in 1992.

hw I A FA1I6u LIHI1 R-HR6?
In 1982 H.K. Lorosch of FAG Bearing Co. (now part of INA-
Schaefer KG) published results of fatigue tests on three
groups of vacuum-degassed 7205B-size AISI 52100 inner
races at maximum Hertz stresses of 2.6, 2.8 and 3.5 GPa
(370, 406 and 500 ksi), respectively. These were very highly
loaded bearings. From these tests, Lorosch concluded that,
Under low loads and with elastohydrodynamic lubrication,
there is no material fatigue, thus indicating that under such
conditions bearing life is practically unlimited.
O. Zwirlein and H. Schlicht, also of FAG Bearing Co.,
in a companion paper published concurrently in 1982 with
that of Lorosch and using the same 7205B-size bearing inner
races, reported large amounts of compressive residual stress
due to the transformation of retained austenite into martin-
site. Bearing research performed at the General Motors Re-
search Center in Warren, Mich., in the 1950s and early 1960s
showed that these compressive residual stresses can signi-
cantly increase bearing life.
My colleagues and I performed an analysis in 1965 at the
NASA Lewis Research Center in Cleveland. We superimposed
compressive residual stresses on the principal stresses in a
ball or roller race contact. This reduced the critical shearing
stress in the bearing raceway, thus increasing bearing life. As
an example, a 10% reduction in the shearing stress due to the
imposition of a compressive residual stress can result in a life
increase of approximately 160%.
Lorosch, Zwirlein and Schlicht failed to account for the
signicantly large presence of these induced compressive re-
sidual stresses in their bearing raceways. Instead they as-
sumed that the large increases in life that they reported were
due to a fatigue limit. Zwirlein and Schlicht concluded that,
contact pressures (maximum Hertz stresses) less than 2.6
GPa (370 ksi) do not lead to the formation of pitting within
a foreseeable period. This corresponds to true endurance.
However, their observation is not supported by rolling-ele-
ment fatigue data in the open literature for maximum Hertz
(contact) stress levels under 2.6 GPa (370 ksi). If Lorosch,
Zwirlein and Schlicht were correct, no bearing in rotating
machinery applications would fail by classical rolling-ele-
ment fatigue since maximum Hertz stress levels in the range
of 1.2 GPa (175 ksi) 10 1.9 GPa (275 ksi) are typical.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 3 5
It became apparent that the
1971 ASME life factors needed
to be updated.
The concept of a fatigue limit made its way into the FAG
bearing catalog. In the 1995 FAG Catalogue WL 41520 ED,
every page that has bearing dimensional information and rat-
ings has a statement in the upper right-hand corner, Rolling
bearings can be fail-safe if C
o
/P
o
is equal to or greater then 8,
see page 41. On page 41 in the section on Adjusted Rating
Life the following is stated:
Bearings are fail-safe if there is utmost cleanliness in the
lubricating gap, full separation of the surfaces by the lubri-
cating lm and load corresponding to C
o
/P
o
greater than or
equal to 8 where, C
o
, is the static load capacity and, Po, is the
static load applied to the bearing.
Based on the FAG criteria, for a ball bearing the fatigue
limit occurs at a maximum Hertz stress of 2.0 GPa (292 ksi).
For roller bearings, the fatigue limit occurs at a maximum
Hertz stress of 1.4 GPa (205 ksi). It is difcult for me to rec-
oncile that for the same bearing steel there are two separate
fatigue limits, one for ball bearings and the other for roller
bearings, that are so signicantly different. It appears to me
that the FAG fatigue limit is based on the classic Whler
curve, Case 2, illustrated in Figure 1(b).
In 1985, based on the results reported by Lorosch, Zwir-
lein and Schlicht, Stathis Ioannides and Tedric A. Harris at
the SKF Engineering and Research Centre in Nieuwegein,
The Netherlands, applied Palmgrens 1924 concept of a fa-
tigue limit to the 1947 Lundberg-Palmgren equations where
Equation [3] above becomes
| |
e
h
e
e
c
z
V
L
u
1
1 1
~
(

|
|
.
|

\
|
t t
[6]
where,
u
, equal the fatigue limit in shear. Ioannides and Har-
ris do not disclose that Palmgren, also from SKF, discarded
the concept of a fatigue limit in 1936. From Equation [6],
Equation [4] can be rewritten to include a fatigue-limiting
load, P
u
,:
p
u eq
D
P P
C
L
|
|
.
|

\
|

=
10
[7]
where
[8] ( )
u u
f P t =
When P
eq
P
u
, bearing life is innite and no fatigue failure
would be expected. When P
u
= 0, the life is the same as that
3 6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
2.50
2.25
2.00
1.75
1.50
1.25
1 10 100 1000 10,000 100,000
Ioannides-Harris
(1985)
Lundberg-
Palmgren
(1947)
M
a
x
m
u
m

H
e
r
t
z

s
t
r
e
s
s
,

G
P
a
350
325
300
275
250
225
200
M
a
x
m
u
m

H
e
r
t
z

s
t
r
e
s
s
,

k
s
Normalized life
2.50
2.25
2.00
1.75
1.50
1.25
Ioannides Harris
(1985)
Lundberg
Palmgren (1947)
(equal 1)
350
325
300
275
250
225
200
(a)
(b)
V
a
r
a
t
o
n

f
r
o
m

c
a
c
u
a
t
e
d

f
e
,

p
e
r
c
e
n
t
600
800
1000
1200
400
400
200
0
200
6 10 20
Number of bearings failed, r
40 100 200 400 1000


Deep grove ball bearing
g



Figure 2 | Comparison of life models for rolling-element bearings
normalized to ANSI/ABMA standards where life L = 1. Fatigue-
limiting shear stress for
45
assumed for Ioannides-Harris model,
276 MPa (40 ksi). (a) Ball bearings. (b) Cylindrical roller bearings.
Figure 3 | Variation between actual and calculated L
10
bearing
lives with STLE life factors for 51 sets of deep-groove and
angular-contact ball bearings and cylindrical roller bearings from
Harris and McCool compared with Monte Carlo variations and 90%
condence limit.
It is difcult for me to reconcile that for the same bearing steel there are
two separate fatigue limits, one for ball bearings and the other for roller
bearings, that are so signicantly different.
2.50
2.25
2.00
1.75
1.50
1.25
1 10 100 1000 10,000 100,000
Ioannides-Harris
(1985)
Lundberg-
Palmgren
g
(1947)
M
a
x
m
u
m

H
e
r
t
z

s
t
r
e
s
s
,

G
P
a
350
325
300
275
250
225
200
M
a
x
m
u
m

H
e
r
t
z

s
t
r
e
s
s
,

k
s
Normalized life
2.50
2.25
2.00
1.75
1.50
1.25
Ioannides Harris
(1985)
Lundberg
Palmgren (1947)
(equal 1)
350
325
300
275
250
225
200
(a)
(b)
V
a
r
a
t
o
n

f
r
o
m
c
a
c
u
a
t
e
d
f
e
,

p
e
r
c
e
n
t
600
800
1000
1200
400
400
200
0
200
1 2 4 6 10 20
Number of bearings failed, r
40 100 200 400 1000
Deep grove ball bearing
g
for Lundberg and Palmgren. If the fatigue limit in Equa-
tion [6] or the fatigue limiting load in Equation [8] were
to be eliminated, the IoannidesHarris equations are iden-
tical to those of Lundberg and Palmgren. A comparison of
Ioannides-Harris, Lundberg-Palmgren and the ANSI/ABMA
Standards is shown in Figure 2.
The ostensible reason Ioannides and Harris used the fa-
tigue limit was to replace the material and processing life
factors, a
2
, that are used as life modiers in conjunction with
the bearing lives calculated from the Lundberg-Palmgren
equations (Equation [5]). However, I have speculated that
the reason it was done was to allow SKF to compete with the
FAG catalog life ratings. Ioannides and Harris made the same
mistake as Lorosch, Zwirlein and Schlicht; they mistook the
effect of compressive residual stresses in a bearing raceway
for a fatigue limit. The life equations of Ioannides and Harris
were subsequently incorporated into the SKF bearing cata-
log.
In the online June 28, 2006 issue of The eBearing News,
the headline reads, ISO Adopts SKF Bearing Life Calcula-
tions. The article goes on to say, The International Or-
ganization for Standardization (ISO) has adopted SKF Life
Theory in its upcoming update to ISO 281. The new ver-
sion will replace ISO 281:1990 as modied in 2000. ISO
281:1990 itself replaced ISO 281:1977. SKF Life Theory has
already been incorporated in the German standard DIN ISO
281, as of 2004. The new ISO 281 incorporates a new ser-
vice life formula which, for the rst time, takes into account
SKFs work on fatigue limit, lubrication and contamination,
all as dened by fundamental SKF Life Theory, as rened and
updated.
whA1 5 AL1uAL ARIN6 A1A 5hw?
In 1995 Dr. Tedric Harris of Pennsylvania State University
and professor John McCool, under contract with the U.S.
Navy, analyzed 62 rolling element bearing endurance sets.
These data were obtained from four bearing manufacturers,
two helicopter manufacturers, three aircraft engine manu-
facturers and U.S. Government agency-sponsored techni-
cal reports. The data sets comprised deep-groove radial
ball bearings, angular-contact ball bearings and cylindrical
roller bearings for a total of 7,935 bearings. Of these, 5,321
bearings comprised one sample size for a single cylindrical
roller bearing, leaving 2,614 bearings distributed among the
remaining bearing types and sizes. Among the 62 rolling ele-
ment bearing endurance sets, 11 had one or no failure and
could not be used for the analysis. These data are summa-
rized and plotted in Figure 3.
At the NASA Glenn Research Center, Brian Vlcek, Robert
C. Hendricks and I randomly assembled and tested 340 vir-
tual bearing sets totaling 31,400 radially-loaded and thrust-
loaded rolling-element bearings. We determined the L
10
maximum limit and L
10
minimum limit for the number of
bearings failed, r, using a Weibull-based Monte Carlo meth-
od. These limits are shown in Figure 3 together with a 90%
upper condence limit using the method of Leonard John-
son at General Motors. In Figure 3, the Harris-McCool data
sets were superimposed on these plots.
Of these bearing data shown in Figure 3, 39% fall be-
tween the maximum and minimum life variations from
Monte Carlo computer simulated bearing tests performed by
us. Four bearing sets representing 8% of the bearing sets had
lives less than that predicted. Thirty bearing sets, or 59% of
the bearing sets, exceeded the maximum life variation of the
Monte Carlo simulation. Eight of these bearing sets or 16%
exceeded the statistical 90% condence upper limit estab-
lished by Leonard Johnson. However, only one bearing set,
representing 2% of the bearing sets, fall below the lower life
limit. It can be reasonably concluded that 98% of the bearing
sets had acceptable life results using the Lundberg-Palmgren
equations with the STLE life adjustment factors to predict
bearing life.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 3 7
Figure 4 | Effect of CVD AISI 52100 steel and load-life exponent
on variation between actual and calculated L
10
bearing lives with
STLE life factors. (a) Load-life exponent p is 3 for ball bearings
and 10/3 for cylindrical roller bearings (from Fig. 3). (b) Load-life
exponent p is 4 for ball bearings and 5 for cylindrical roller
bearings.
V
a
r
a
t
o
n

f
r
o
m

c
a
c
u
a
t
e
d

f
e
,

p
e
r
c
e
n
t
600
800
400
400
200
0
200
1 6 10 20
Number of bearings failed, r
40 100 200 400 1000
600
800
1000
1200
400
400
200
0
200
4
Number of bearings failed, r
40 100 200 400 1000
Minimum Monte Carlo variation
Maximum Monte Carlo
variation
Deep grove ball bearing
Angular contact ball bearing
Cylindrical roller bearings
p = 4 for ball bearings;
p = 5 for roller bearings
Upper 90%
confidence
limit
(a)
(b)
p = 3 for ball bearings;
p = 10/3 for roller bearings
Minimum Monte Carlo variation
Maximum Monte Carlo variation
V
a
r
a
t
o
n

f
r
o
m

c
a
c
u
a
t
e
d

f
e
,

p
e
r
c
e
n
t
600
800
400
400
200
0
200
1 2 4 6 10 20
Number of bearings failed, r
40 100 200 400 1000
600
800
1000
1200
400
400
200
0
200
1 2 4 6 10 20
Number of bearings failed, r
40 100 200 400 1000
Minimum Monte Carlo variation
Maximum Monte Carlo
variation
Deep grove ball bearing
Angular contact ball bearing
Cylindrical roller bearings
p = 4 for ball bearings;
p = 5 for roller bearings
Upper 90%
confidence
limit
(a)
(b)
p = 3 for ball bearings;
p = 10/3 for roller bearings
Minimum Monte Carlo variation
Maximum Monte Carlo variation
The data of Figure 3 were broken down and replotted by
Vlcek, Hendricks and I in Figure 4(a) for bearings made from
CVD AISI 52100 steel. These data were adjusted for a load-
life exponent p of 4 for ball bearings and 5 for roller bearings
in Equation [5] and are shown in Figure 4(b). The adjusted
life results (Figure 4(b)) correlated best with those of the
Monte Carlo tests.
Based upon these material and processing life factors and
load-life exponents, each bearing data set appears consistent
with the other. These data further suggest that the load-life
exponent, p, for ball bearings in the standards should be 4
instead of 3. And for roller bearings, the load-life exponent,
p, should be 5 instead of 3.33 or 4. However, what is most
important is that for these bearing sets, there was no fatigue
limit manifested by the data. If a fatigue limit were to be
incorporated in this data, the bearing lives would be over
predicted and/or the bearings undersized for its application.
These results were published in 2003 in STLEs peer-reviewed
journal Tribology Transactions.
hw LAN I 5 5uR?
For nearly 30 years there have been passionate advocates in
the rolling bearing community for the existence of a fatigue
limit. On the other hand, there have been those such as my-
self asking, Where are the data to support the existence of
a fatigue limit? There are no data! However, that made no
difference to those who wrote and had adopted ISO Stan-
dard 281:2007. People such as myself were considered ob-
structionist to technical progress. With the adoption of ISO
Standard 281:2007, we were now in the position of having
to prove a negative that a fatigue limit for AISI 52100 steel
does not exist.
Funding to run full-scale bearing tests at low enough
stress levels to prove or disprove the existence of a fatigue
limit would be extremely expensive. And who would pay for
it or conduct the tests? Our colleagues in Japan came up
with the perfect solution.
A leader in this effort is professor Shigeo Shimizu of the
School of Science and Technology at Meiji University in Ka-
wasaki, Kanagawa, Japan. He reasoned that a material when
tested should only react to its state of stress and environ-
ment. In other words, if AISI 52100 steel had a fatigue limit,
a stress below which no failure will occur, it should exhibit
this limit no matter what type of fatigue test was performed.
A statistically signicant number of fatigue tests are required
over a range of shearing stresses to assure with reasonable
engineering and scientic certainty that the steel either did
or did not exhibit a fatigue limit. And, if a fatigue limit did
exist, what is its value?
In 2008 Shimizu and his colleagues, professors K. Tosha,
D. Ueda and H. Shimoda, published a journal paper in Tribol-
ogy Transactions reporting the results of rotating beam fatigue
experiments for through-hardened AISI 52100 steel at very
low shearing stresses as low as 0.48 GPA (69.6 ksi). The test
results produced fatigue lives in excess of 100 million stress
cycles without the manifestation of a fatigue limit.
In order to assure the credibility of his work, additional
research was conducted and published by Shimizu together
with his colleagues, professors K. Tsuchiya and K. Tosha, in
a 2009 issue of Tribology Transactions. They tested six groups
of AISI 52100 bearing steel specimens using four-alternating
torsion fatigue life test rigs to determine whether a fatigue
limit exists or not and to compare the resultant shear stress-
life relation with that used for rolling-element bearing life
prediction.
The number of specimens in each sample size ranged from
19 to 33 specimens for a total of 150 tests. The tests were run
at 0.5, 0.63, 0.76, 0.80, 0.95 and 1.00 GPa (75.5, 91.4, 110.2,
116.0, 137.8 and 145 ksi) maximum shearing stress ampli-
tudes. The stress-life curves of these data, as shown in Figure
3 8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
M
a
x
i
m
u
m

s
h
e
a
r
i
n
g

s
t
r
e
s
s
,

m
a
x
,

G
P
a
1.0
0.3
0.5
10
3
10
4
10
5
10
6
10
7
10
8
10
9
Life L, stress cycles
n = 10.34
n = 10.34
L
10
life
L
50
life
Figure 5 | Probabilistic stress-life (P-S-N) curves for AISI 52100 bearing steel in reverse torsional stress.
From three-parameter Weibull analysis of Shigeo Shimizu, Kazuo Tsuchiya and Katsuji Tosha.
M
a
x
i
m
u
m

s
h
e
a
r
i
n
g

s
t
r
e
s
s
,

m
a
x
,

G
P
a
1.0
0.3
0.5
10
3
10
4
10
5
10
6
10
7
10
8
10
9
Life L, stress cycles
n = 10.34
n = 10.34
L
10
life
L
50
life
5, show an inverse dependence of life on the shearing stress
as in Case 1, Figure 1(a), but do not show an inverse rela-
tion on the difference of the shearing stress minus a fatigue
limiting stress as in Case 3, Figure 1(c). The shear stress-life
exponent, n, for the AISI 52100 steel was 10.34 from the
three-parameter Weibull analysis and was independent of
the Weibull slope, e.
In 2007 Dr. T. Sakai discussed experimental results ob-
tained by the Research Group for Material Strength in Japan.
He presented stress-life rotating bending fatigue data from
six different laboratories in Japan for AISI 52100 bearing
steel. He also presented stress-life fatigue data for axial load-
ing. The resultant lives were in excess of a billion (10
9
) stress
cycles at maximum shearing stresses (
max
) as low as 0.35
GPa (50.8 ksi) without an apparent fatigue limit.
It can be reasonably concluded from the volumes of data
from Meiji University and Dr. Sakai that the fatigue stress-life
relation that incorporates a fatigue limit for AISI 52100 steel
as in ISO Standard 281:2007 is incorrect. These data preclude
the existence of a fatigue limit to maximum shearing stresses
(
max
) as low as 0.35 GPa (50.8 ksi). This is equivalent to a
maximum Hertz stress for ball bearings of 1.09 GPa (158.1
ksi) and 1.17 GPa (169.7 ksi) for roller bearings.
Arvid Palmgren got it right 74 years ago!
FR FuR1hR RAIN6:
1. Bamberger, E.N., Harris, T.A., Kacmarsky, W.M., Moyer,
C.A., Parker, R.J., Sherlock, J.J. and Zaretsky, E.V. (1971),
Life Adjustment Factors for Ball and Roller BearingsAn Engi-
neering Design Guide, ASME, New York.
2. Bamberger, E.N., Zaretsky, E.V. and Signer, H. (1976), En-
durance and Failure Characteristics of Main-Shaft Jet Engine
Bearing at 310
6
DN, Journal of Lubrication Technology, 98
(4), pp. 580585.
3. Harris, T.A. (1995), Final Report Establishment of a New
Rolling Bearing Contact Life Calculation Method, U.S. Na-
val Air Warfare Center, Aircraft Division Trenton, Contact
No. N6833593C0111.
4. Harris, T.A. and McCool, J.J. (1996), On the Accuracy
of Rolling Bearing Fatigue Life Prediction, ASME Journal of
Tribology, 118 (2), pp. 297310.
5. Ioannides, E. and Harris, T.A. (1985), New Fatigue Life
Model for Rolling Bearings, ASME J. Tribol. Trans., 107 (3),
pp. 367378.
6. Ioannides, E., Bergling, G. and Gabelli, A. (1999), An
Analytical Formulation for the Life of Rolling Bearings, Acta
Polytechnica Scandinavica, Mechanical Engineering Series,
137, Finland.
7. ISO 281:2007: Rolling Bearings-Dynamic Load Ratings
and Rating Life, International Organization for Standardiza-
tion, Geneva, 2007.
8. Lundberg, G. and Palmgren, A. (1947), Dynamic Capac-
ity of Rolling Bearings, Acta Polytechnica, Mechanical Engi-
neering Series, 1 (3), Stockholm, Sweden.
9. Lundberg, G. and Palmgren, A. (1952), Dynamic Capac-
ity of Roller Bearings, Acta Polytechnica, Mechanical Engi-
neering Series, 2 (4), Stockholm, Sweden.
10. Lorosch, H.K. (1982), Inuence of Load on the Magni-
tude of the Life Exponent for Rolling Bearings, Rolling Con-
tact Fatigue Testing of Bearing Steels, J.J.C. Hoo, ed., ASTM
STP-771, American Society for Testing and Materials, Phila-
delphia, Pa., pp. 275292.
11. Palmgren, A. (1924), Die Lebansdauer von Kugellagern
(The Service Life of Ball Bearings),Zeitschrift des Vereines
Deutscher Ingenieure, 68 (14), pp. 339341, (NASA TT-F-
13460, 1971).
12. Palmgren, A. (1936), Om Kullager Barformaga Och
Livslangd, (On the Carrying Capacity and Life of Ball Bear-
ings), Teknisk Tidskrift, Mek, h. 2, (in Swedish) and The Ball
Bearing Journal, 3, 1937, pp. 34-44.
13. Parker, R.J., Zaretsky, E.V. and Bamberger, E.N. (1974),
Evaluation of Load-Life Relation With Ball Bearings at 500
F, Journal of Lubrication Technology, 96 (3), pp. 391397.
14. Saki, T., Takeda, M., Shiozawa, K., Ochi, Y., Nakajima,
M., Nakamura, T. and Oguma, Y. (2000), Experimental Re-
conguration of Characteristic S-N Property for High Car-
bon Chromium Bearing Steel in Wide Life Region in Rotating
Bending, J. Soc. Mat. Sci., Japan, 49 (7), pp. 779-785 (in
Japanese).
15. Saki, T. (2007), Review and Prospects for Current Stud-
ies on Very High Cycle Fatigue of Metallic Materials for Ma-
chine Structure Use, Proc. 4th International Conference on
Very High Cycle Fatigue (VHCF-4), TMS (The Minerals,
Metals and Materials Society), pp. 3-12.
16. Shimizu, S., Tsuchiya, K. and Tosha, K. (2009), Proba-
bilistic Stress-Life (P-S-N) Study on Bearing Steel Using Al-
ternating Torsion Life Test, Tribology Transactions, 52 (6),
pp. 807816.
17. The eBearing News, ISO Adopts SKF Bearing Life Calcu-
lations, http://www.ebearing.com/news2006/062801.htm,
June 28, 2006.
18. Tosha, K., Ueda, D., Shimoda, H. and Shimizu, S. (2008),
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 3 9
A Study on P-S-N Curve for Rotat-
ing Bending Fatigue Test for Bear-
ing Steel, Tribology Transactions,
51 (2), pp. 166-172.
19. Vlcek, B.L., Hendricks, R.C. and
Zaretsky, E.V. (2003), Determina-
tion of Rolling-Element Fatigue Life
from Computer Generated Bearing
Tests, Tribology Transactions, 46
(3), pp. 479493.
20. Zaretsky, E.V. (1992), Life Fac-
tors for Rolling Bearings, STLE SP-
34, Society of Tribologists and Lu-
brication Engineers, Park Ridge, Ill.
21. Zaretsky, E.V., Poplawski, J.V.
and Peters, S.M. (1996), Compari-
son of Life Theories for Rolling-El-
ement Bearings, Tribology Transac-
tions, 39 (2), pp. 237248.
22. Zaretsky, E.V. (1997), Tribology
for Aerospace Applications, STLE SP-
37, Society of Tribologists and Lu-
brication Engineers, Park Ridge, Ill.
23. Zaretsky. E.V. (1998), A
Palmgren RevisitedA Basis for
Bearing Life Prediction, Lubrica-
tion Engineering, 54 (2), pp. 1824.
24. Zaretsky, E.V., Poplawski, J.V.
and Miller, C.R., Rolling Bearing
Life PredictionPast, Present, and
Future, in Proc. of the Internation-
al Tribology Conf. Nagasaki, 2000,
1, Japanese Society of Tribologists,
Tokyo, Japan, 2001, pp. 101107.
25. Zaretsky, E.V. (2010), Rolling
Bearing Life Prediction, Theory and
Application, Recent Developments
in Wear, Prevention, Friction and Lu-
brication, G ed. George K. Nikas,
Research Signpost, Kerala, India.
26. Zwirlein, O. and Schlicht, H.
(1982), Rolling Contact Fatigue
Testing of Bearing Steels, J.J.C. Hoo,
ed., ASTM STP-771, American Soci-
ety for Testing and Materials, Phila-
delphia, Pa., pp. 358379.
Adding value to your formulations
$Nll00 l0 $0ll0K 8$8900.
The Bright Stock Replacement
Concerned about the uncertainty of bright stocks?
Now theres a better option.
Soltex BSR-900 is a polybutene-based polymeric thickener
developed specifically as a 150 bright stock alternative. Its
used in a wide range of high-viscosity, industrial lubricants.
F0fl0fM800 800lll8
Excellent thickening at low treat levels
High shear stability
Superior oxidation resistance
Soluble in all base oils, including synthetics,
for clarity of finished products
Greater viscosity index boost to finished lubricants
Lower cost than PAO, PAG &other alternatives
Soltex BSR-900 a more enlightened alternative to 150 bright stock.
Call now for details and technical assistance.
Need another option? Ask about our line of severely hydrotreated naphthenics.
+1 281 587 0900 www.soltexinc.com
See our full line of performance chemicals:
POLY8uTElES SuLF0lATES 0ETER6ElTS 6REASE A00lTlVES Vl lMPR0VERS
SYlThETlCS P0uR P0llT 0EPRESSAlTS RuST 8 C0RR0Sl0l llhl8lT0RS Al0 M0RE
4 0 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Biocide for lubricant applications The FDA has approved
PREVENTOL

CMK PRESERVATIVE as a Food Contact

Substance with intended use at levels up to 1 percent as an
antimicrobial preservative in lubricants that may have incidental
contact with food. Additionally the EPA and NSF have approved
this product for use in industrial lubricants. For customers requiring
these approvals, this is a new choice for a lube preservative that
is effective against bacteria, yeast, and mold fungi.
Biocides for metalworking fluids effective and econ-
omical fluid control. LANXESS Material Protection offers
biocides based on PCMC, OPP, IPBC, BIT and Bronopol.

Corrosion inhibitors for aqueous and non-aqueous metal
working fluids. LANXESS corrosion inhibitors are based on tolyl-
triazole and benzotriazole. They offer thermal stability, resistance to
atmospheric oxidation, low toxicity and non-irritation to skin.
LANXESS Material Protection Flexible solutions you
can always trust
New EPA, FDA and NSF Approved Biocide
For Lubricant Applications
LANXESS Corporat|on Mater|a| Protect|on Products
111 PIDC Park West Dr|ve, P|ttsburgh, PA 15275
Phone. 1 - 800 - LANXESS Fax. 412-809-1068
Ema||. mpp-|nfo@|anxess.com www.mpp.us.|anxess.com
Effect|ve, econom|ca| b|oc|des based on PCMC.
PREVENTOL

CMK 40
PREVENTOL

CMK 30
PREVENTOL

CMK Preservative
PREVENTOL

is a registered trademark of LANXESS Deutschland GmbH.

kY wR5
Alternating Torsion; Bearing Steel; Fatigue Limit; Log-Normal Distribution; P-S-N
Curve; Rolling Bearings; Rolling-Contact Fatigue; Weibull Distribution Function;
Whler Curve
A51RAL1
Probabilistic stress-life (P-S-N) studies using alternating torsion life test were conducted
with the bearing steel JIS SUJ2/AISI 52100 having a hardness range of Rockwell C scale
of 58-62 HRC to determine the existence of a fatigue limit for this through-hardened
steel. One hundred and fty (150) alternating torsion tests were performed at six tor-
sion maximum shear stress amplitudes of 0.5, 0.63, 0.76, 0.80, 0.95, and 1.0 GPa. The
number of specimens that were run at each stress amplitude varied from 19 to 33. There
is no observable fatigue limit for JIS SUJ2/AISI 52100 bearing steel. On log-log plotted
data, life is linear and is inversely proportional to shearing stress to a 10.34 exponent. The
failure mode of the experiment is due to the conjugate tensile and compressive alternating
stresses in accordance with alternating shearing stresses. The three-parameter Weibull life
distribution function with a constant value of the Weibull slope m = 3/2 was found to have
the best conformity followed by the lognormal distribution and the two-parameter Weibull
distribution function. The shear stress-life exponent (A = 10.34) for JIS SUJ2/AISI 52100
bearing steel was independent of the Weibull slope.
IN1RuL1IN
Although the concept of a fatigue limit for rolling-element bearings was rst pro-
posed by Palmgren in 1924
1
, it was apparently abandoned by him again in 1936
(Palmgren
2
) followed by again in association with Lundberg in 1947 (Lundberg and
Palmgren
3
). Palmgrens statement in his 1936 paper about the fatigue limit is de-
picted as follows:
For a few decades after the manufacture of ball bearings had taken up on modern lines,
it was generally considered that ball bearings, like other machine units, were subject to
a fatigue limit, i.e. that there was a limit to their carrying capacity beyond which fatigue
speedily set in, but below which the bearings could continue to function for innity. Sys-
tematic examination of the results of tests made in the SKF laboratories before 1918,
however, showed that no fatigue limit existed within the range covered by the compara-
tively heavy loads employed for test purposes. It was found that so far as the scope of the
investigation was concerned, the employment of a lighter load invariably had the effect of
increasing the number of revolutions a bearing could execute before fatigue set in. It was
certainly still assumed that a fatigue limit coexisted with a certain low specic load, but
tests with light loads nally showed that the fatigue limit for innite life, if such exists,
is reached under a load lighter than all of those employed, and that in practice the life is
accordingly always a function of load.
Frobeb|||st|c StressL|fe (FSh) Stady oa
8eer|a Stee| 0s|a k|teraet|a 1ors|oa L|fe 1est
SHIGEO SHIMIZU, KAZUO TSUCHIYA, and KATSUJI TOSHA
Meiji University School of Science and Technology, Kawasaki, Japan
PEER-REVIEWED
Manuscript accepted June 8, 2009
Review led by Bob Errichello
Copyright STLE
Editors Note: This months
Editors Choice paper authored
by STLE-member Shigeo Shimi-
zu et al. is referenced in Part II
of our two-part series on ISO
281:2007 bearing-life stan-
dard. Erv Zaretsky, an STLE-
member and living legend in
the bearings community, pres-
ents the other side of this
topic in a commentary article
featured in this issue (see page
30). It only seems tting to
print one of Ervs many refer-
ences outlining the potential
failings of the proposed stan-
dard. This EC paper details a
very thorough study, using one
type of steel, showing inter-
esting results regarding fa-
tigue life and interpretation
thereof. For those who found
the ISO 281:2007 discussion
interesting, I highly recom-
mend reading this paper and
some of the other referenced
articles to fully appreciate the
complexity of deriving a new
standard to satisfy all.
Evan Zabawski, CLS
Editor
4 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
In 1985, Ioannides and Harris
4
came out with their theory
by applying Palmgrens 1924
1
concept to fatigue limit in the
Lundberg-Palmgren equations (Lundberg and Palmgren
3
),
which was later used in the ISO standards (ISO
5,6
); Barnsby
7
;
ISO
8
) for rolling bearings leading to the limiting stress based
on the von Mises stress of
vm
= 0.684 GPa corresponding to
the Hertz stress
max
= 1.14 GPa (Barnsby
7
) or
max
= 1.5 GPa
corresponding to
vm
= 0.9 GPa (ISO
8
) for bearing steel. The
result is the inclusion of a life modication factor, a
XYZ
or a
SL

or a
ISO
. These fatigue limits are based on Ioannides-Harris
4
as
well as the sophisticated work of Ioannides, et al. in 1999
9

and are evaluated by using a conventional Whlers fatigue
limit.
Table 1 shows representative fatigue limits expressed by
von Mises stress
vm
, maximum Hertz stress
max
, maximum
orthogonal subsurface shear stress
0
, and alternating torsion
specic lives N
10
and N
50
obtained in this research work with
Whlers fatigue limit as
0
.
In the usual S-N test data, however, for bearing steels heat
treated to hardness above 58 HRC, none of the experts in
the eld including Shimizu
10,11
and Zaretsky, et al
12
had men-
tioned about the tendency of the test data showing a fatigue
limit based on the probabilistic stress-life (P-S-N) study of
the material and/or the probabilistic force-life (P-F-L) study
of the rolling bearing.
According to Tosha, et al.
13
, the P-S-N study conducted for
rotating bending for the bearing steel (JIS SUJ2/AISI 52100)
in a similar manner did not show the fatigue limit. Further,
even in the examples of S-N (stress-life) characteristics at
very high cycle domain up to giga(10
9
) cycles for rotating
bending and axial loading, the bearing steel did not show
any fatigue limit (Sakai
14
). These numbers of giga stress cy-
cles for S-N tests correspond to the numbers of contact stress
cycles for the rolling bearing raceways and/or the rolling ele-
ments, which may be actually assumed above 10
7
-10
8
revolu-
tions bearing lives.
Because of the insertion of an assumed and specically
dened fatigue limit for the bearing steel JIS SUJ2/AISI 52100
into the ISO standards
5,6,8
, it is important to establish with
absolute certainty whether there is a fatigue limit, its charac-
teristic and value, as well as with a stress-life or load-life ex-
ponent together. Because, the fatigue limit affects the stress-
life or load-life exponent strongly, they must be reevaluated
by using applied stresses or loads minus the fatigue limit or
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 4 3
TABL PRESENTATIV F TIGUE L MIT ;
vm
V MISES S RESS,
max
HERTZ S RESS, ND t
0
MAXIMU O THOGONA SUBSU
FA SHEAR STRESS (IN GPA ANNIDE AN HARRI (4); ( ARN BY (7) ; (IS (8) RRE NDIN 0% AND 50% AIL LIFE N
10
AND
N
50
FOR ALTERNATIN T ON LIFE TES

vm

max

0
a
N
b
10
N
b
50

0
c
N
b
10
N
b
50
ASME (7) 0.684 1.14 0.285 2.92E 09 8.14E 09 0.245 1.37E 10 3.86E 10
(4)
d
0.72/0.78 1.2/1.63 0.30 1.73E 09 4.80E 09 0.30 1.73E 09 4.80E 09
I & H (4)
e
0.84/0.98 1.4/1.63 0.35 3.58E 08 9.82E 08 0.35 3.58E 08 9.82E 08
ISO281 (8) 0.9 1.5 0.375 1.77E 08 4.83E 08 0.323 8.25E 08 2.28E 09
Min value 0.6 1.0 0.25 1.11E 1 3.14E10 0. 1 .20E 10 1.48E 11
a

0
0.25
max
for line contact.
b
N
n
a
1
(
0
/2.27)
1034
(
0
/2.27)
99
Eq. [15].
c

0
0.215
max
for circular contact.
d
Bad material
e
Good
material.
N MEN LAT RE
A Stress life exponent, 1/A slope of P-S-N curve
A

Stress life exponent for

a
1
(lnR ln 09)
1m
reliability fa tor
B C
A
S
fatigue strength coefcient
C
S
B
1A
basic dynamic stress rating for material
C
S
Basic dynamic stress rating for
c Exponent related to the stress-life exponent A, where A cm
D Minimum diameter of pecimen
E

Log base coefcient of variation for population

I Number of stress amplitudes
j , 2, . . . , N
S
orde umber
K
n
Percentage point for standardized normal distribution N(0, 1)
m Weibull slope, depends on stress distribution state in specimen
M
t
Torsional moment or torque
N Number of stress cycles to failures
n Probability of failures, %
N
10
10% failure life/90% rating life for material
N
50
Median life
N
90
90% failure life
N
n
n% failure life
N(, ) Normal distributio t ean and standar viation
N(0, 1) Standardized normal distribution
N
S
Number of sample size
R 1 n100 reliabil ty
S Stress amplitude
S
f
Fatigue limit
S

f
Fatigue limit for
Minimum life for population
/ log base coefcient of variation
(ln 0 9)
1m
(L
10
) scale parameter
ln N
50
log base mean or location parameter
ln N
84
log base standard deviation

max
Maximu ertz stress

vm
vo ses stress

0
Maximum orthogonal subsurface shear stress

max
Maximum shearing stress amplitude for torsion
TABLE 1REPRESENTATIVE FATIGUE LIMITS;
vm
VON MISES STRESS,
max
HERTZ STRESS, AND t
0
tt MAXIMUM ORTHOGONAL SUBSUR
FACE SHEAR STRESS (IN GPA; IOANNIDES AND HARRIS (4); (BARNSBY (7)); (ISO(8)) CORRESPONDING TO 10% AND 50% FAILURE LIFE N
10
NN AND
N
50
NN FOR ALTERNATING TORSION LIFE TEST

vm

max

0
, L. C.
a
N
b
NN
10
NN N
b
NN
50
NN
0
, C. C.
c
N
b
NN
10
NN N
b
NN
50
NN
ASME (7) 0.684 1.14 0.285 2.92E 09 8.14E 09 0.245 1.37E 10 3.86E 10
I & H (4)
d
0.72/0.78 1.2/1.63 0.30 1.73E 09 4.80E 09 0.30 1.73E 09 4.80E 09
I & H (4)
e
0.84/0.98 1.4/1.63 0.35 3.58E 08 9.82E 08 0.35 3.58E 08 9.82E 08
ISO281 (8) 0.9 1.5 0.375 1.77E 08 4.83E 08 0.323 8.25E 08 2.28E 09
Min value 0.6 1.0 0.25 1.11E 10 3.14E10 0.215 5.20E 10 1.48E 11
a

0
0.25
max
for line contact.
b
N
n
NN a
1
(
0
/2.27)
1034
(
0
/2.27)
99
Eq. [15].
c

0
0.215
max
for circular contact.
d
Bad material
e
Good
material.
NOMENCLATURE
A Stress life exponent, 1/A // slope of P-S-N curve
A

Stress life exponent for

a
1
(lnR ln 09)
1m
reliability factor
B C
A
CC
S
C fatigue strength coefcient
C
S
CC B
1A
basic dynamic stress rating for material
CCC
S
C Basic dynamic stress rating for
c Exponent related to the stress-life exponent A, where A cm
D Minimum diameter of specimen
E

Log base coefcient of variation for population

I Number of stress amplitudes
j 1, 2, . . . , N
S
NN order number
K
n
K Percentage point for standardized normal distribution N(0, 1)
m Weibull slope, depends on stress distribution state in specimen
M
t
MM Torsional moment or torque
N Number of stress cycles to failures
n Probability of failures, %
N
10
NN 10% failure life/90% rating life for material
N
50
NN Median life
N
90
NN 90% failure life
N
n
NN n% failure life
N(, ) Normal distribution with mean and standard deviation
N(0, 1) Standardized normal distribution
N
S
NN Number of sample size
R 1 n100 reliability
S Stress amplitude
S
f
S Fatigue limit
S

f
S Fatigue limit for
Minimum life for population
/ log base coefcient of variation
(ln 0 9)
1m
(L
10
) scale parameter
ln N
50
NN log base mean or location parameter
ln N
84
NN log base standard deviation

max
Maximum Hertz stress

vm
von Mises stress

0
Maximum orthogonal subsurface shear stress

max
Maximum shearing stress amplitude for torsion
load limit for the P-S-N or P-F-L test data on log-log graph
paper at the same time.
If a fatigue limit does not exist, then the current ISO stan-
dard
8
must be reexamined. Otherwise, it is likely that the
user of the standard will undersize a ball or roller bearing for
a specic application and/or over-predict the bearings life.
If, however, a fatigue limit does exist, the question becomes
what is the correct value or magnitude of this fatigue limit as
well as the load-life exponent. Without both correct values,
there is a probability of either over-or underpredicting the
life of a bearing for a specic application.
It is the objective of the research work reported to deter-
mine if a fatigue limit exists for the most widely used bear-
ing steel JIS SUJ2/AISI 52100. Probabilistic stress-life studies
were conducted with the bearing steel JIS SUJ2/AISI 52100
having a hardness range of Rockwell C-scale of 58-62 HRC.
One hundred and fty alternating torsion tests were per-
formed at six torsion stress amplitudes of 0.5, 0.63, 0.76,
0.80, 0.95, and 1.0 GPa. The number of specimens that were
run at each stress amplitude varied from 19 to 33.
LIF I51RIu1IN FuNL1IN
In the statistical treatment of life distribution, which fairly
obeys the life test data, conventionally used distribution
functions are the lognormal and the Weibull distribution
function. In the alternating torsion fatigue test, if the maxi-
mum stress amplitude at the external circumference of the
test specimen is
max
and the life corresponding to n%(0 n
100%) failure of the test specimens is N
n
, then the prob-
ability density function f (N
n
) for the lognormal distribution
is given by Eq. [1].
f (N
n
)
1

2N
n
exp

1
2

ln N
n

[1]
where = ln N
50
= log base mean (location parameter), N
50
=
exp[] = median, = ln N
84
= = log base standard
deviation, and = / = log base coefcient of variation.
On the other hand, the three-parameter Weibull distri-
bution function with a probability consideration that the
material will not fail, that is to say, with a reliability of R =
1 n/100 (0 R 1), can be expressed as shown by Eq. [2]
or [2a].
ln
1
R

N
n

ln
1
09

1m
(N
10
) 0 [2]
ln
1
R
m ln

N
n

[2a]
In this equation, m = Weibull slope, = scale parameter,
and = the minimum strength for a population of material
showing no failure, which may be expressed as the mini-
mum life of the material under a stress amplitude
max
. For
= 0, this equation will be considered as the two-parameter
Weibull distribution function.
For such statistical distribution functions, data analysis
can be carried out usually by taking the probability of failure
n% on the vertical axis and ln N on the horizontal axis, and
then the life distribution data are plotted on the lognormal
and/or the Weibull probability paper (see Appendices, Figures
A1 and A2).
Now if these results are successively projected onto the
horizontal axis corresponding to ln
max
of the life distribu-
tion data for various stress amplitudes, one can obtain the
P-S-N curve. Therefore, even though the equation formats of
both the life distribution and the P-S-N curve may be differ-
ent from each other, the P-S-N curve should be expressed by
an equation given from the same life distribution function.
R6R55IN uA1IN5 FR P-5-N 51uY
F 1h 151
Considering A as the stress-life exponent, B as the fatigue
strength coefcient of the material, and the newly intro-
duced C
S
as the basic dynamic stress rating, the P-S-N curve
will be expressed as shown by Eq. [3] for a fatigue limit S
f
(S
f

0) and the minimum life ( 0) corresponding to a stress
amplitude
max
.
N
n
B(
max
S
f
)
A

max
S
f
C
S

A
C
S
B
1A
[3]
Now, for the lognormal distribution in Eq. [3], C
S
is conven-
tionally adopted as 50% basic dynamic stress rating with
= 0 and n = 50%. If the expected coefcient of variation for
the population is expressed by E

as shown by Eq. [4], the

standard deviation for an ith stress amplitude could be put
= E

, and the percentage point of the standardized normal

distribution N(0, 1) is given by K
n
, then the P-S-N curve can
be expressed by Eqs. [5] and [6] (Tosha, et al.
13
).
E

1
I
I

i1

i
I number of stress amplitudes [4]
N
50
B(
max
S
f
)
A

max
S
f
C
S

A
S
f
0 [5]
N
n
N
1EKn
50

max
S
f
C
S

A(1EKn)
[6]
According to this, it is understood that a semi-logarithm
regression model (Shimizu
10
) often used for most of the S-N
tests for obtaining a regression equation between the fatigue
lives and the respective stress amplitudes (both as medians)
cannot be employed as a correct numerical equation model.
Subsequently, if n = 10%, that is to say the reliability R =
0.9, C
S
is 90% basic dynamic stress rating from the relations
of Eqs. [2] and [3], and a
1
is the reliability factor, then the
three-parameter Weibull-based P-S-N curve can be expressed
by Eqs. [7] and [8].
N
10

max
S
f
C
S

A
S
f
0 [7]
N
n
a
1

max
S
f
C
S

A
a
1

lnR
ln09

1m
[8]
4 4 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Furthermore, the regression equation for the minimum life
will be given by Eq. [9], considering parameters S'
f
, C'
S
, and
A' as the replacement of S
f
, C' S, and A.

max
S

f
C

[9]
FA1I6u LIHI1
From Eq. [3], conventional lognormal based S-N curves
can be drawn as shown in Figures. 1(a) and 1(b) in the case
where = S
f
= 0, the median life N
50
is inversely related to the
power function of the maximum shearing stress amplitude

max
as shown by Eq. [3a].
N
50

max
C
S

A
or
max
C
S
N
1A
50
[3a]
This relationship when plotted on log-log graph paper will
be plotted as illustrated in Figure 1(a) showing no fatigue
limit. The slope of S-N curve is expressed by 1/A, minus the
inverse of the stress-life exponent A.
The classic Whler curve illustrated in Figure 1(b) in-
troduced the concept of a fatigue limit where the stress-life
relation is that of Eq. [3a] until the stress reaches the value of
the fatigue limit S
f
, at approximately 10
6
to 10
7
stress cycles,
and that the life is considered innite as shown by Eq. [3b].
That is, no fatigue failures would be expected to occur.
N
50

max
C
S

A
or
max
C
S
N
1A
50

max
S
f
N
50

max
S
f

[3b]
In practice, fatigue data for those materials showing fatigue
limits do not manifest into a straight line on a log-log S-N
plot but rather to a curved line shown as an asymptote to a
fatigue limit illustrated in Figure 1(c) (Shimizu
10,15
). The ap-
parent relationship between the life, the shearing stress, and
the fatigue limit is
N
50

max
S
f
C
S

A
or
max
S
f
C
S
N
1A
50
[3c]
From Eq. [3c], under an applied stress on a material that
shows the existence of a fatigue limit, the resultant life will
be longer than that determined from Eq. [3a]. A linear re-
gression is obtained by plotting the data for
max
S
f
. In this
gure, the P-S-N curve for an arbitrary probability of failures
converges to the fatigue limit as the asymptote shown by Eq.
[6].
The three-parameter Weibull-based S-N curves are shown
in Figures 1(d) and 1(e) with a minimum life 0 consid-
ered as a function of the shearing stress as shown by Eq. [9].
A linear regression is obtained by plotting N
10
as shown by
Eq. [3d]. This equation corresponds to the three-parameter
Weibull-based S-N curve with fatigue limit S
f
=0.
N
10

max
C
S

A
or
max
C
S
(N
10
)
1A
[3d]
The last case is shown in Figure 1(e) with minimum life
0 and fatigue limit S
f
0. A linear regression is obtained by
plotting N
10
as shown by Eq. [3e] at rst and then plotting

max
S
f
as shown in Figure 1(e).
N
10

max
S
f
C
S

A
or
max
S
f
C
S
(N
10
)
1A
[3e]
In this gure, the P-S-N curve for an arbitrary reliability con-
verges to the fatigue limit as the asymptote shown by Eq.
[8].
Palmgren
1
once proposed a similar equation with the
condition where < 0 ( =
B
: a life constant related to the
ultimate strength of material) instead of Eq. [3e] in 1924, but
he disregarded
B
as well as S
f
in 1936
2
and 1947
3
twice from
his huge life tests. However, he recognized the existence of a
minimum life in 1936 as follows:
The variation is such that there is no pronounced accumu-
lation of values around a mean value, and the minimum life is
clearly not governed by laws to more than a triing extent.
Consequently, neither the average nor the minimum life pro-
vides a satisfactory basis for a denition.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 4 5
Figure 1 | Fatigue limit concept for lognormal and three-parameter Weibull-based S-N (stress-life) curves with slope 1/A. (a) No fatigue
limit. (b) Classic Whler curve where life is innite below fatigue limit S
f
. (c) Life is an asymptote to fatigue limit S
f
. (d) No fatigue limit
S
f
= 0, the minimum life as a function of
max
. (e) With fatigue limit S
f
>0, most generalized S-N curve.
kPRIHN1AL vIL AN 151 H1h
Figure 2 shows the front view (a) and side view (b), sec-
tion A-A of the experimental setup on the test rig manufac-
tured for the alternating torsion life test. The test rig consists
of a V-pulley speed reduction ratio of 1/2.5 from a 3.7-kW,
4-pole, 3-phase AC motor with an inverter and a lever crank
mechanism capable of applying torque by changing the crank
radius (eccentric cam) to various values. The alternating tor-
sion stress amplitude
max
is set on the test rig by keeping the
torque variation within 0.2% for the torque converter with
maximum torque
max
=500 Nm. The test speed for varying
the stress cycles is set from 350 to 960 rpm for high to low
stress amplitudes.
Table 2 shows the chemical composition of the material
supplied by the steel manufacturer as per the JIS standard
values. According to the steel manufacturer, the retained aus-
tenite for SUJ2, 58-62 HRC may be measured as 7-10%.
An alternating torsion fatigue test specimen is shown in
Figure 3. The test specimens made of SUJ2 hardened to 58-
62 HRC are ground to 8 0.025 mm diameter with surface
roughness R
y
< 1.6 m and a radius R30 on the radial por-
tion. The portion D 0.025 is nished to D = 14.73 mm or
D = 12.69 mm on an NC lathe, respectively.
Figure 4 shows the schematics of the test section of Fig-
ure 3. The torsional moment M
t
of circular shaft with diam-
eter D and the conjugate tensile and compressive alternating
stresses 1 due to maximum surface shearing stresses
max

are shown in Figure 4(a). The maximum shearing stress am-
plitude
max
= 16 M
t
/D
3
is obtained from this gure. Mohrs
circle in pure shear is shown in Figure 4(b).
Figure 5 shows the mechanical properties of the test spec-
imens, which are collected for the distribution of the hard-
ness HRC, the tensile strength
B
, the torsion strength
B
,
and the proportional limit
p
for the sample size N
S
= 20-34.
The resulting distributions followed the normal distributions
with mean and standard deviation such as N(60, 0.65) HRC,

B
= N(2.1, 0.12),
B
= N(1.7, 0.02), and
p
= N(0.82, 0.09)
GPa, respectively.
151 R5uL1 AN I5Lu55IN
FuiIure Hode
Figure 6 shows a typical example of the broken test piece
obtained in this alternating torsion test for the P-S-N study.
All specimens were broken like this photo. A closeup of the
broken surface is also shown in Figure 7. The ridge marks
are observed to be converging to the top of the small white
surface portion, but no sh eye is observed, as was the case in
4 6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
AC Motor 3.7kW-4P
V-pulley i = 1/2.5
Torque Converter
Tmax = 500 Nm
Eccentric Cam
Specimen
(Fig. 2)
Flexible
Coupling
ETP-T25
Table
#7210
CDB
Lever Spindle
Con-rod
ETP-T25
2-Guide Rail
Feed Screw
Surface Plate 1200-900-150
4-Linear Bearing
Lever
ETP-T25
Con-rod
ETP-T25
AC Motor 3.7kW-4P
V-pulley i = 1/2.5
100
A
A
900 600
Eccentric Cam
SECTION A-A
600
(a) Front view (b) Side view
ETP-T25
Test Rig
Center Line
TABLE 2CHEMICAL COMPOSITIONS FOR BEARING STEEL, JIS SUJ2/AISI 52100, IN WT%, O: PPM
C Si Mn P S Cr M Ni Cu O
Product 0.98 0.18 0.37 0.016 0.003 1.42 0.03 0.07 0.12 6
JIS 0.951.1 0.150.35 0.5 0.025 0.025 1.31.6 0.08
Figure 2 | Experimental setup on alternating torsion life test rig
the P-S-N study under rotating bending test (Tosha, et al.
13
).
However, neither the sh eye nor the ridge marks were found
in the case of JIS S55Cx, 60-64 HRC material equivalent to
AISI 1055 steel used for linear bearing rail in alternating tor-
sion fatigue test (Shimizu
11,15
). An example of the SEM photo
for ridge mark (Figure 7(b) top point) is shown in Figure 8.
In the photo, there is no nonmetallic inclusion that can be
observed at the origin of the failure.
From the above discussions, the failure mode is due to
the conjugate tensile and compressive alternating stresses in-
clined at 45 to the test piece centerline in accordance with
the torsion surface shearing stresses as shown in Figure 4.
This failure mode was the same as in S55Cx (Shimizu
10
),
which was also found in the railcar axle of Whlers alternat-
ing torsion fatigue test in 1870 (Whler
16
).
5IuIiaIicuI 1reuImenIa
The life test data of six lots corresponding to the standard
alternating torsion stress amplitudes
max
= 0.5-1.0 GPa are
D
Mt
max
max
(a)
(b)
Mt
max

m
a
x

=
45

m
a
x
max

max
0
max
D

16Mt
max 1

90
max

0.005

D
D
o
u
b
l
e

c
e
n
t
e
r

d
=
2
.
5
0.005
0
.
0
2
5
R
1
A
0.025 8

2
5
h
7
A
R
3
0
100
216
66 50
D=12.69
D=14.73
1.6
50
Clamp bush
ETP-T25
Flexible
coupling

2
5
h
7
3.2
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 4 7
Figure 3 | Test specimen, SUJ2/AISI 52100, 58-62 HRC. Figure 5 | Mechanical properties for SUJ2 test specimen.
Figure 6 | Failure mode, broken SUJ2 torsion test specimen. (a) Test
section. (b) Surface failure origin.
Figure 7 | Closeup of failure origin of SUJ2/AISI 52100 steel torsion
test specimen. (a) Ridge marks showing crack propagation. (b)
Point of fatigue crack initiation.
Figure 8 | SEM photo for ridge mark top point (Figure 7).
Figure 4 | Schematic of test section of Figure 3. (a) Torsional mo-
ment of circular shaft showing conjugate tensile and compressive
alternating stresses due to surface shearing stresses. (b) Mohrs
circle in pure shear.
summarized in Table 3. There were no censored test data.
All specimens were run until failure. The data in Table 3 are
plotted as Benards median rank (see Abernethy
17
) shown by
Eq. [10] on the lognormal or Weibull probability paper.
n
j 03
N
S
04
100 % [10]
At rst, the life test data (
max
, N
j
) from Table 3 are plot-
ted on the lognormal probability paper and the estimations
for logarithmic based mean and standard deviation are
carried out as shown in the life distribution of Figure 9(a)
(cf. Appendix Figure A1 for the linear regression). According
to this, it is understood that for each one the life distribution
on the lognormal probability paper shows a very good tting
and linearity. As a result, the obtained coefcients of varia-
tion varied in the range = / = 0.044-0.070 as arranged in
Table 4. Therefore, the coefcient of variation for the popula-
tion may be estimated by Eq. [4] to be E = 0.05 by taking
the average value of six stress amplitudes.
4 8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
TABLE LTERNATIN TORSIO F TIGUE TEST D TA FO S SUJ AI 210 AT SIX SHEARIN STRESS A PLITUDE ,
max
Life N
j
, stress cycles
Maximum Shearing Stress Amplitude,
max
, GPa (ksi)
Order Number j 0.5 (72.5) 0.63 (91.4) 0.76 (110.2) 0.80 (116) 0.95 (137.8) 1.0 (145)
1 5.42E 0 8.65 05 1.23E 05 6.30E 04 1.20E 04 3.28E 03
.62E 0 9.67 05 1.34E 0 8.40E 04 1.30E 0 58E 03
.63E 0 1.09 06 1.78E 0 1.03E 05 1.60E 04 5.77E 03
.00E 0 1.25 06 1.86E 0 1.13E 05 1.90E 04 7.61E 03
5 1.00E 0 1.35 6 2.30E 0 1.17E 07 2.10E 04 8.20E 03
6 1.25E 0 1.37 6 2.37E 0 1.18E 05 2.50E 04 8.29E 03
7 1.76E 0 1.53 2.51E 0 1.24E 05 2.70E 04 8.89E 03
8 2.23E 0 1.74 2.57E 0 1.47E 05 2.80E 04 9.56E 03
9 2.38E 0 1.79E 2.65E 0 1.47E 05 2.90E 04 1.17E 04
10 2.56E 0 2.20E 2.70E 0 1.49E 05 3.40E 04 1.22E 04
11 2.71E 0 2.21E 3.08E 0 1.88E 05 3.70E 04 1.29E 04
1 74E 0 2.93E 3.26E 0 2.18E 05 3.80E 0 36E 04
13 2.88E 0 3.01E 3.35E 0 2.30E 05 3.80E 04 1.39E 04
14 2.89E 0 3.12E 3.43E 0 2.38E 05 4.00E 04 1.40E 04
1 .27E 0 4.42E 3.63E 0 2.55E 05 4.10E 04 1.42E 04
1 91E 0 5.02E 3.68E 0 3.11E 05 4.30E 0 57E 04
18 4.22E 0 5.25E 4.22E 0 3.44E 05 4.50E 0 76E 04
19 4.45E 0 6.82E 4.98E 0 3.72E 05 4.70E 04 1.84E 04
20 5.95E 0 5.11E 0 4.41E 05 5.10E 0 85E 04
21 5.60E 0 4.70E 05 5.20E 0 10E 04
22 5.64E 0 4.97E 05 5.50E 04 2.45E 04
23 5.78E 0 5.36E 05 5.70E 04 2.47E 04
24 6.08E 0 5.72E 05 6.60E 0 76E 04
25 6.12E 0 7.56E 05 7.00E 04 2.87E 04
26 6.13E 0 3.78E 04
27 6.34E 05 3.83E 04
28 6.49E 05 68E 04
29 6.70E
30 7.00E
31 8.34E
3 8.43E
33 9.82E
Figure 9 | Lognormal-based life distribution (a) and P-S-N curve
(b) for alternating torsion life test results.
TABLE 3ALTERNATING TORSION FATIGUE TEST DATA FOR JIS SUJ2/AISI 52100 AT SIX SHEARING STRESS AMPLITUDES,
max

Life N
j
NN , stress cycles
Maximum Shearing Stress Amplitude,
max
, GPa (ksi)
Order Number j 0.5 (72.5) 0.63 (91.4) 0.76 (110.2) 0.80 (116) 0.95 (137.8) 1.0 (145)
1 5.42E 06 8.65E 05 1.23E 05 6.30E 04 1.20E 04 3.28E 03
2 6.62E 06 9.67E 05 1.34E 05 8.40E 04 1.30E 04 4.58E 03
3 7.63E 06 1.09E 06 1.78E 05 1.03E 05 1.60E 04 5.77E 03
4 1.00E 07 1.25E 06 1.86E 05 1.13E 05 1.90E 04 7.61E 03
5 1.00E 07 1.35E 06 2.30E 05 1.17E 07 2.10E 04 8.20E 03
6 1.25E 07 1.37E 06 2.37E 05 1.18E 05 2.50E 04 8.29E 03
7 1.76E 07 1.53E 06 2.51E 05 1.24E 05 2.70E 04 8.89E 03
8 2.23E 07 1.74E 06 2.57E 05 1.47E 05 2.80E 04 9.56E 03
9 2.38E 07 1.79E 06 2.65E 05 1.47E 05 2.90E 04 1.17E 04
10 2.56E 07 2.20E 06 2.70E 05 1.49E 05 3.40E 04 1.22E 04
11 2.71E 07 2.21E 06 3.08E 05 1.88E 05 3.70E 04 1.29E 04
12 2.74E 07 2.93E 06 3.26E 05 2.18E 05 3.80E 04 1.36E 04
13 2.88E 07 3.01E 06 3.35E 05 2.30E 05 3.80E 04 1.39E 04
14 2.89E 07 3.12E 06 3.43E 05 2.38E 05 4.00E 04 1.40E 04
15 3.27E 07 4.42E 06 3.63E 05 2.55E 05 4.10E 04 1.42E 04
16 3.33E 07 4.74E 06 3.68E 05 2.88E 05 4.20E 04 1.52E 04
17 3.91E 07 5.02E 06 3.68E 05 3.11E 05 4.30E 04 1.57E 04
18 4.22E 07 5.25E 06 4.22E 05 3.44E 05 4.50E 04 1.76E 04
19 4.45E 07 6.82E 06 4.98E 05 3.72E 05 4.70E 04 1.84E 04
20 5.95E 07 5.11E 05 4.41E 05 5.10E 04 1.85E 04
21 5.60E 05 4.70E 05 5.20E 04 2.10E 04
22 5.64E 05 4.97E 05 5.50E 04 2.45E 04
23 5.78E 05 5.36E 05 5.70E 04 2.47E 04
24 6.08E 05 5.72E 05 6.60E 04 2.76E 04
25 6.12E 05 7.56E 05 7.00E 04 2.87E 04
26 6.13E 05 3.78E 04
27 6.34E 05 3.83E 04
28 6.49E 05 4.68E 04
29 6.70E 05
30 7.00E 05
31 8.34E 05
32 8.43E 05
33 9.82E 05
Co-sponsored by:
Society of Tribologists and
Lubrication Engineers
840 Busse Highway
Park Ridge, Illinois 60068
P: 847-825-5536
F: 847-825-1456
www.stle.org
[email protected]
ASME International
Three Park Avenue
New York City, New York 10016
P: 800-843-2763 (U.S./Canada)
95-800-843-2763 (Mexico)
973-882-1167 (outside North America)
www.asme.org
[email protected]
2010 INTERNATIONAL
JOINT TRIBOLOGY CONFERENCE
Oct. 18-20, 2010
Hilton San Francisco Financial District
San Francisco, California (USA)
Golden Gate Bridge
Aquarium by the Bay
Fishermans Wharf
Asian Art Museum
Wine-tasting tours
Chabot Space & Science Center
Charles M. Schulz Museum
Whale Watching in Monterey
Napa Valley Museum
USS Hornet
For more information,
please contact:
Merle Hedland
630-323-7212
[email protected]
Please join hundreds of your colleagues for a unique event
focusing on cutting-edge research in the multidisciplinary
eld of tribology. Technical presenters are from more than
20 nations, including the U.S., U.K., Israel, Taiwan, Canada,
India, Pakistan, Japan, China, Poland, Singapore, South Korea,
Czech Republic, Switzerland, Mexico and the Netherlands.
The conference is soliciting papers in
tribology areas that include:
||at||aa|a
taat+ct Hcc|+a|c:
||+:ta|1|a1a+a|c |aa||c+t|aa
|ac||a lcc|aa|a|c:
|a|acc|c1 \a||+cc:
|1|a1a+a|c |aa||c+t|aa +a1
Fluid Film Bearings
|aa||c+at: +a1 |11|t|c:
H+c||ac taaaacat: l||aa|a
H+act|c \ta|+c l||aa|a
H+aa|+cta||a/Hct+|a|||a l||aa|a
+aat||aa|a
|a|||a ||cacat |c+||a:
\a||+cc |a|acc||a
l||aaa+tc||+|:
W|a1 la|a|ac |ac|t|+|a l||aa|a
SAN FRANCISCO ATTRACTIONS
The IJTC is a unique conference for tribology
thought-leaders looking to:
|c+| :aac 11 tcc|a|c+| |c:cat+t|aa:
cta|| |t| t||aa|a': ta |a|c::|aa+|:
from around the world.
||c|c catt|ac1c |c:c+|c|
||:cac| |c|c |aa||c+t|aa tcc|aa|a |: |c+1|a.
Registration is open. Register at www.stle.org
The P-S-N curve determined from this result and from
Eqs. [5] and [6] is also shown in Figure 9(b). The stress-life
exponent A = 10.36 and the 50% rating stress C
S
= 2.58 GPa
are obtained. It is understood that they are constants and
independent of the probability of failures n%. Furthermore,
from this result it is revealed that when the estimation is car-
ried out with S
f
= 0, such a material may be said to have no
fatigue limit.
Figure 10(a) (cf. Appendix Figure A2(b) for linear re-
gression) shows the result of the test data conforming to the
two-parameter Weibull distribution function as a reference.
In the life distribution Figure 10(a), it is found that in the
lower and also the higher failure regions, the life values are
moving towards the larger deviation side from the theoretical
straight line. A variation in the estimated Weibull slopes is
found in the range of m = 1.7-2.45 with an average value of
E(m) = 2.0. This variation of the m-value is not so big com-
pared with the case of S55Cx material, but we cannot treat it
as a constant m-value. Here, if Eq. [8] is used for = 0, then
for the arbitrary values of n = 10, 50, 90%, the P-S-N curve
in Figure 10(b) will be obtained with A = 10.13, 10.37, 10.56
and C
S
= 2.44, 2.60, 2.71 GPa, by ignoring the variation of
m-value as well as the reliability factor. As a result, it is found
that the stress-life exponent A and the rating stress C
S
can-
not be constants for the life N
n
corresponding to an arbitrary
probability of failure n% in the case of the two-parameter
Weibull regression.
At the same time in the life distribution for S55Cx, 60-64
HRC (Shimizu
11
), the Weibull slope remained as m = 2.8-6.4
and E(m) = 4.3. In this P-S-N study test with S55Cx, the m
values for the two-parameter Weibull regression have larger
values and wider scatter, and the fatigue life is smaller than
this SUJ2 experimental test data. It is also revealed that the
Weibull slope for the two-parameter Weibull distribution
function does not follow the conventional physical meaning
(Weibull
18
and Lundberg-Palmgren (Ioannides and Harris
4
;
(Lundberg and Palmgren
19
), i.e. the Weibull slope depends
on the material and will be larger for the material with larger
fatigue strength.
In general, in accordance with the method in which the
Weibull slope m = 3/2 = a constant (independent of stress
amplitudes and also the strength of materials) and taking the
estimated values of the scale parameter and the minimum
life , the regression result will be as shown in Figure 11(a)
(cf. Appendix Figure A2(a)). The and estimations are car-
ried out by using Eqs. [11] and [12] with m = 3/2 (Shimizu
10
).

N
10
02848N
50
07152
for m
3
2
[11]

ln
1
09

1m
(N
10
) 4483 (N
10
) [12]
According to this, the life distribution in Figure 11(a)
obeys extremely well the three-parameter Weibull distribu-
5 0 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
10
N10 (/2.44)

m
a
x

G
P
a
SUJ2
5862HRC
0.1
3
10
1
0.3
10
50
90
99
n


%
2PW
1.0GPa
0.5
P-S-N
1
Number of stress cycles N
4
10
5
10
6
0.80
0.95
0.76 0.63
10 10
7
(a)
8
0.5GPa
(b)
N90 (/2.71)
N50 (/2.60)
10.13
10.37
10.56
T BL E TIMATED P RAMETER LOG RMA D TRIB -
F TI FIG E() 05
Lognormal Estimations, in 10
3
Stress Cycles

max
GPa N
10
N
50
N
90
0.5 16.85 0.76 0.045 7,859 20,790 54,998
0.63 14.63 0.68 0.046 946 2,258 5,392
0.76 12.89 0.57 0.044 191 396 822
0.8 12.32 0.71 0.058 90.3 224 556
0.95 10.45 0.52 0.050 17.8 34.5 67.2
1.0 9.55 0.67 0.070 5.96 14.0 33.1
TABLE 4ESTIMATED PARAMETERS FOR LOGNORMAL DISTRIBU-
TION FUNCTION IN FIG. 9, E() 0.05
Lognormal Estimations, in 10
3
Stress Cycles

max
GPa N
10
NN N
50
NN N
90
NN
0.5 16.85 0.76 0.045 7,859 20,790 54,998
0.63 14.63 0.68 0.046 946 2,258 5,392
0.76 12.89 0.57 0.044 191 396 822
0.8 12.32 0.71 0.058 90.3 224 556
0.95 10.45 0.52 0.050 17.8 34.5 67.2
1.0 9.55 0.67 0.070 5.96 14.0 33.1
Figure 10 | Two-parameter Weibull-based life distribution (a) and
P-S-N curve (b) for alternating torsion life test results.
Figure 11 | Three-parameterWeibull-based life distribution and
P-S-N curve for alternating torsion life test results. (a) Three-pa-
rameter regression with m = 3/2. (b) P-S-N plot. (c) P-S-(N-) plot.
tion function with m = 3/2. Using Eqs. [7]-[9], in any of the
cases for the failure probabilities in the P-S-(N-) curve in
Figure 11(c), the stress-life exponent remains constant at A
= 10.34 and the basic dynamic stress rating for one cycle
also remains at C
S
= 2.27 GPa. Then the nal P-S-N curve is
shown in Figure 11(b) including the regression.
However, while comparing with S55Cx, 60-64 HRC
(Shimizu
11,15
), it is found that the life distributions for both
materials can hold the same value m = 3/2; therefore, it is
conrmed that the Weibull slope m does not depend on the
fatigue strength of the material. Moreover, in case of S55Cx
with C
S
= 2.33 GPa, though not much different from that
(C
S
= 2.27 GPa) observed for SUJ2, the difference lies in the
stress-life exponent; that is, A = 8.3. However, in this test
with SUJ2 of high fatigue strength as A = 10.34, it can be
said that the stress-life exponent for materials with higher
strength has a higher A value.
Tables 5 and 6 summarize the estimated results for the
two-parameter and three-parameter Weibull life distribution
function corresponding to each parameter and specic lives
N
10
, N
50
,and N
90
in Figures 10 and 11.
For the comparison of both materials, Figures 12 and 13
present the 50%-S-N and 10%-S-N curve, respectively, based
on the lognormal and the three-parameter Weibull distribu-
tion function. From each of these gures, it is made clear
that the difference between the material fatigue strength is
mainly due to the stress-life exponent A for the region of
high cycle fatigue. Further, it is understood that a fatigue
limit could not be observed in any materials and any regres-
sion functions.
From Lundberg and Palmgren
3
, the shear stress-life expo-
nent A is a function of the Weibull slope m where A = c/m.
From Lundberg and Palmgren
3
, m = 10/9 for the ball bearing,
m = 3/2 for the roller bearing, and c = 31/3. These yield to the
values of A = 9.3 for the ball bearing and A = 6.9 for the roller
bearing. Subsequently, Lundberg and Palmgren
19
changed
the value of the Weibull slope for the roller bearings to m
= 9/8. As a result, for roller bearings, A = 9.2. However, as
shown in our article, A is independent of the Weibull slope
m. Therefore, A = c = 31/3 for the ball and the roller bear-
ings. This value is approximately the same as that shown by
us experimentally where A = 10.34 for JIS SUJ2/AISI 52100
material.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 5 1
TABLE 5 TIMATED P RAMETERS FOR TWO-PARAMETER
WEIBULL D STRIBUTIO FUNCTIO I FIG E(m)
Two-Parameter Weibull Estimations,
i
3
Stress Cycles

max
GPa m N
10
N
50
N
90
0.50 1.70 28,800 7,665 23,215 47,039
0.63 1.77 3,090 867 2,512 4,950
0.76 2.1 08 182 430 743
0.80 1.71 308 82.6 24 02
0.95 2.45 43 17.2 37.0 60.4
1.0 1.85 19.3 5.72 15.8 30.3
TABLE TIMATED P RAMETERS FO THREE-PARAMETER
WEIBULL D STRIBUTIO FUNCTION WITH m 3/2 IN FIG
Three-Parameter Weibull Estimations,
i
3
Stress Cycles

max
GPa N
10
N
50
N
90
0.5 ,200 1,930 7,105 20,101 42,384
0.63 2,590 445 1,023 2,474 4,961
0.76 45 9 423 858
0.80 264 31 89.9 238 491
0.95 36 7.8 15.8 36.0 70.6
1.0 17 2 5.79 15.3 31.6
T BLE TIMATED PARAMETER -S-N C RVE BY
LOG RMA D TRIB TION F TI T REE METER
WEIB D TRIB TION F TI N WITH m 3/ C MPARISON
WIT M TERIAL DIFFEREN F GS. 1 ND 13
Lognormal Three-Parameter Weibull
Mate. SUJ2 S55Cx SUJ2 S55Cx SUJ2 S55Cx
E

0.05 0.03 m 3/2

A 10.36 7.38 10.34 8.3 A

9.9 6.6
C
S
2.58 3.5 2.27 2.33 C

S
2.27 3.89
TABLE 5ESTIMATED PARAMETERS FOR TWO-PARAMETER
WEIBULL DISTRIBUTION FUNCTION IN FIG. 10, E(m) 2.0
Two-Parameter Weibull Estimations,
in 10
3
Stress Cycles

max
GPa m N
10
NN N
50
NN N
90
NN
0.50 1.70 28,800 7,665 23,215 47,039
0.63 1.77 3,090 867 2,512 4,950
0.76 2.19 508 182 430 743
0.80 1.71 308 82.6 249 502
0.95 2.45 43 17.2 37.0 60.4
1.0 1.85 19.3 5.72 15.8 30.3
TABLE 7ESTIMATED PARAMETERS FOR P-S-N CURVES BY
LOGNORMAL DISTRIBUTION FUNCTION AND THREE-PARAMETER
WEIBULL DISTRIBUTION FUNCTION WITH m 3/2 IN COMPARISON
WITH MATERIAL DIFFERENCES IN FIGS. 12 AND 13
Lognormal Three-Parameter Weibull
Mate. SUJ2 S55Cx SUJ2 S55Cx SUJ2 S55Cx
E

0.05 0.03 m 3/2

A 10.36 7.38 10.34 8.3 A

9.9 6.6
C
S
CC 2.58 3.5 2.27 2.33 C

CC
S
C 2.27 3.89
TABLE 6ESTIMATED PARAMETERS FOR THREE-PARAMETER
WEIBULL DISTRIBUTION FUNCTION WITH m 3/2 IN FIG. 11
Three-Parameter Weibull Estimations,
in 10
3
Stress Cycles

max
GPa N
10
NN N
50
NN N
90
NN
0.50 23,200 1,930 7,105 20,101 42,384
0.63 2,590 445 1,023 2,474 4,961
0.76 453 68 169 423 858
0.80 264 31 89.9 238 491
0.95 36 7.8 15.8 36.0 70.6
1.0 17 2 5.79 15.3 31.6
Figure 12 | Lognormal based 50%-S-N curve for comparison of
materials
14
.
Figure 13 | Three-parameter Weibull-based 10% S-N curve for
comparison of materials (Shimizu
10
).
For reference, the P-S-N regressions corresponding to
the lognormal and the three-parameter Weibull distribution
function for both materials are shown in the following equa-
tions.
Lognormal-lascd P-S-N curvc rcgrcssion
N
n

max
258

1036(1005Kn)
(JIS SUJ2/AISI 52100) [13]
N
n

max
355

74(1003Kn)
(JIS S55Cx/AISI 1055; Sakai (14))
[14]
Tlrcc-Paramcicr Wcilull lascd P-S-N curvc rcgrcssion
N
n

max
227

1034

lnR
ln09
2
3

max
227

99
(JIS SUJ2/AISI 52100) [15]
N
n

max
233

83

ln R
ln 09
2
3

max
389

66
(JIS S55Cx/AISI 1055, (Shimizu (10))
[16]
Table 7 shows the P-S-N regressions corresponding to
the lognormal and the three-parameter Weibull distribution
function for both SUJ2 and S55Cx materials. From this, it is
clear that the stronger material has a larger A value and also
larger E

value than the weaker material. The Weibull slope

is independent of the stress-life exponent, the strength of the
material, and a constant value m = 3/2 for alternating the tor-
sion fatigue life test.
5uHHARY
Probabilistic stress-life (P-S-N) alternating torsion life test
studies were conducted with the bearing steel JIS SUJ2/AISI
52100 having a hardness range of 58-62 HRC to determine
the existence of a fatigue limit for this through-hardened
steel. One hundred and fty alternating torsion tests were
performed at six torsion stress amplitudes of 0.5, 0.63, 0.76,
0.80, 0.95, and 1.0 GPa. The number of specimens that were
run at each stress amplitude varied from 19 to 33. The fol-
lowing results were obtained:
1. There is no observable fatigue limit for JIS SUJ2/AISI
52100 bearing steel heat-treated to a hardness of 58-62
HRC. On log-log plotted data, life is linear and is inverse-
ly proportional to shearing stress to a 10.34 exponent.
2. The failure mode for the experiment is due to the surface
conjugate tensile and compressive alternating stresses in
accordance with the alternating shearing stresses. The
ridge marks oriented from the surface are observed, but a
sheye and/or a nonmetallic inclusion is not observed at
the top origin point.
3 Corresponding to the life distributions of six stress ampli-
tudes, the three-parameter Weibull life distribution func-
tion introducing a minimum life as the third parameter
with the Weibull slope m = 3/2 = constant is found to have
the best conformity. This is followed by the lognormal
distribution and the two-parameter Weibull distribution
function in the decreasing order of conformity. The shear
stress-life exponent (A = 10.34) for JIS SUJ2/AISI 52100
bearing steel was independent of the Weibull slope.
ALkNwL6HN15
The authors thank E. V. Zaretsky of the NASA Glenn Re-
search Center; Dr. C. S. Sharma of Shinrai Preci-Tech Systems
in Mumbai, India; and Dr. K. Hiraoka of Sanyo Special Steel
Co., Ltd. This work is supported by the Academic Frontier
Project for Private Universities: matching fund subsidy from
MEXT (Ministry of Education, Culture, Sports, Science and
Technology) 2007-2009. The authors would like to extend
their heartfelt gratitude to all concerned.
RFRNL5
1. Palmgren, A. (1924), Die Lebensdauer von Kugellagern [The Service
Life of Ball Bearings], Zeitschrift des Vereines Deutscher Ingenieure, 68,
14, pp 339-341 (NASA TT-F-13460, 1971).
2. Palmgren, A. (1937), Om kullagrens barformaga och livslangd [On
the Carrying Capacity and Life of Ball Bearings], Teknisk Tidskrift, Mek
1936, h.2, (in Swedish) and The Ball Bearing Journal, 3, pp 34-44.
3. Lundberg, G. and Palmgren, A. (1947), Dynamic Capacity of Rolling
Bearings, IVA Handlingar, 196, 13, pp 16-19.
4. Ioannides, E. and Harris, T.A. (1985), New Fatigue Life Model for
Rolling Bearings, J. Tribol. Trans. ASME, 107, 3, pp 367-378.
5. ISO, Amendment 2. (2000), Rolling BearingsDynamic Load Rat-
ings and Rating Life, Life Modication Factor, ISO 281.
6. ISO (2002), Rolling BearingsDynamic Load Ratings and Rating
Life, Amendment 2, ISO-281/2.
7. Barnsby, R. (2003), Life Ratings for Modern Rolling BearingsA Design
Guide for the Application of International Standard ISO 281/2. ASME In-
ternational and Co-published with STLE, New York,
8. ISO 281(2007), Rolling BearingsDynamic Load Ratings and Rating
Life, 2nd Ed. 2007-02-15, Ref. No. ISO 281(E).
9. Ioannides, E., Bergling, G. and Gabelli, A. (1999), An Analytical For-
mulation for the Life of Rolling Bearings, Acta Polytechnica Scandi-
navica, Mechanical Engineering Series, 137, pp 9-12, 21-24, Finland,
1999.
10. Shimizu, S. (2005), Fatigue Limit Concept and Life Prediction Model
for Rolling Contact Machine Elements, STLE, Tribology & Lubrication
Technology, 5, pp 32-41.
11. Shimizu, S. (2005), P-S-N/P-F-L Curve Approach Using 3-Parameter
Weibull Distribution Function for Life and Fatigue Analysis of Struc-
tural and Rolling Contact Component, STLE, Tribology Trans., 48, 4,
pp 576-582.
12. Zaretsky E.V., Poplawski, J.V. and Root, L.E. (2007), Reexamination
of Ball-Race Conformity Effects on Ball Bearing Life, STLE, Tribology
Trans., 50, 3, pp 336-349.
13. Tosha, K., Ueda, D., Shimoda, H. and Shimizu, S. (2007), A Study on
PS-N Curve for Rotating Bending Fatigue Test for Bearing Steel, STLE,
Tribology Trans., 51,2, pp 166-172.
14. Sakai, M. (2007), Review and Prospects for Current Studies on Very
High Cycle Fatigue of Metallic Materials for Machine Structural Use,
5 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Proceedings of VHCF-4, TMS (The Minerals, Metals & Materials Soci-
ety), pp 3-12.
15. Shimizu S. (2000), P-S-N Curves Model for Rolling Contact Ma-
chine Elements, JAST, Proceedings of the International Tribology
Conference,Nagasaki, 3, pp 1767-1772.
16. Wohler A. (1870), ber die Festigkeits-Versuche mit Eisen und
Stahl, Zeitschrift fr Bauwesen, 20, pp 74-106.
17. Abernethy B. (1966), The New Weibull Handbook. (1966) 2nd ed. Gulf
Publishing Co., Houston, TX.
18. Weibull, W. (1939), A Statistical Theory of the Strength of Materials,
IVA Handlinger, 151, pp 2, 28, 95.
19. Lundberg, G. and Palmgren, A. (1952), Dynamic Capacity of Roller
Bearings, IVA Handlingar, 210, pp 6.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 5 3
APPNIL5
Figure A1 | Lognormal plot where ordinate is
graduated as a probability of failures corre-
sponding to the percentage point K
n
for the
standardized normal distribution N(0, 1).
Figure A2 | Data plot N where Weibull slope or tangent of line is m; probability of failures
n = 63%, reliability R = 0.37 at which = N
63
. (a) Three-parameter Weibull plots N and
N-. (b) Two-parameter Weibull approximation N
2
, usually m
2
> m.
LognormuI Puper und
Lineur Regreaaion
weibuII Puper und Lineur Regreaaion
2139 High Tech Road State College 16803 814-353-800 800-676-623 Fax 814-353-8007
[email protected] www.cannoninstrument.com
Its here true ASTM D445
automation at a price guaranteed
to make your lab manager smile.

10 or 100-fold tube ranges

KinVis up to 5000 cSt
Selectable temperature 20
to 100C
Smart table technology save
space, save time, save money!
Small and Affordable
Low Solvent/Sample Volume
Faster Through-Put
miniAV-X
NEW! miniQV-X
miniAV
NEW! miniAV-LT
Ideal Testing
for Jet Fuel
at -20C
1P 51RI5
LhHIk I5 Nw INNvAk
Lbemidex, a leading business-to-business search engine for
lubricant product developers worldwide, has changed its
name to Innovudex.
Innovadex speeds innovation by connecting product de-
velopers worldwide with material suppliers during the earli-
est phases of the development process. Half of Innovadexs
customers are outside of the U.S., and the search engine is
available in seven languages. In the last 24 months, Innova-
dex usage has more than doubled.
Using Innovadex, lubricant product developers can search
for information based on a products type, end-use applica-
tion and performance criteria, as well as Eco-Certication
criteria that locate materials that have achieved certications
such as CleanGredients and Ecocert.
Innovadexs proprietary Cybrary houses the worlds larg-
est repository of product technical literature serving the food
and beverage, adhesives and sealants, graphic arts and inks,
paint and coatings, household, institutional and industrial
cleaners, lubricant and metalworking uids, plastics and
elastomers and personal care and cosmetics industries.
To access the information, product development profes-
sionals must register as Cybrary members. Membership is
free to all qualied industry professionals.
Both STLE and ILMA are proud partners of Innovadex.
For more information, log on to www.innovudex.com.
uAkR LhHILAL ALuIR5 .A. 51uAR1 u5IN55
uuker LbemicuI Lorp. has announced the company has ac-
quired .A. 5IuurI'a U.S. aluminum hot rolling oil business
from hougbIon InIernuIionuI, which includes the rights to
sell the product portfolio internationally.
Net sales for the D.A. Stuart business was $7 million in
2009.
This business is a strategic t for Quaker because it is
complementary to our existing business, is aligned with our
core competencies of lubrication and surface treatment and
will facilitate growth in our primary metals business, both
within and outside of North America, said Michael F. Barry,
Quakers chairman, CEO and president. The aluminum hot
rolling market has strong long-term growth prospects, espe-
cially in emerging economies such as Russia, South America,
Eastern Europe, the Middle East and Asia. Quaker currently
has a presence in those countries and is well positioned for
growth.

1RIA LuRILA1IN LHPNN15, LLL
5L 1 PALHR hLLAN
PuImer hoIIund, Inc. has announced the company has ac-
quired 1riud LubricuIion LomponenIa, LLL, a distributor of
lubricant raw materials.
Founded in May 2005, Triad has historically supplied lu-
bricant additives and basestocks to territories in the south-
ern United States. All current Triad sales and administrative
management personnel will continue to work in similar po-
sitions within Palmer Holland. In addition, RuIpb eurd,
former president and founder of Triad, was named vice pres-
ident-lubricants for Palmer Holland.
hNR5 6 AwAR5
RHIR NAH A5H FLLw
STLE-member r. AIi rdemir, Ar-
gonne Distinguished Fellow of the
U.S. Department of Energys Argonne
National Laboratory in Argonne,
Ill., has been named a fellow of the
American Society of Mechanical En-
gineersthe highest elected grade of
membership in ASME.
Erdemirs research interests in-
clude surface engineering and tribol-
ogy. He is responsible for pioneering
the development of new materials
and coatings, providing ultra-low friction and wear coef-
cients.
He has authored more than 230 publications in archi-
val journals, scientic books and handbooks, and currently
holds nine U.S. patents. Erdemir has received the R&D 100
Award four times and the Discovery Magazine Award.
Ali currently serves on STLEs board of directors and has
been actively involved in numerous STLE and ASME com-
mittees.
LNLPhILLIP5 LuRILAN15
RLIv5 NA1INAL RL6NI1IN
LonocoPbiIIipa LubricunIa has announced that a Kendall
Motor Oil digital commercial created to promote the com-
panys premium passenger car motor oil featuring Liquid Ti-
tanium

was recognized by the American Advertising Fed-

eration with the 2010 Silver ADDY Award.
The video, titled Metal on Metal was acknowledged
by the ADDY committee for Sound Design at a recent na-
tional awards reception in Orlando. The ADDY Awards is a
national competition, which recognizes and rewards creative
excellence in the art of advertising. Over 50,000 entries are
submitted each year for the advertising industrys largest and
most representative competition, which must win at the lo-
cal and regional levels to be eligible for this national recogni-
tion.
We are honored to receive an award for the production
quality and creativeness of this commercial, said Marshall
Cohen, director, brand management at ConocoPhillips Lu-
bricants. This highlights, in an entertaining manner, the

Dr. Ali Erdemir

5 4 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
friction-reducing performance of our new Kendall products
with Liquid Titanium, and the video has helped to generate
increased interest and demand for our offerings.
PRH1IN5 6 1RAN5I1IN5
FL1LhR }IN5 5N1INL AHRILA, INL.
5enIineI Lunudu of Montreal, Que-
bec has appointed STLE-member
Luciun FIeIcber, LL5, to managing
director of Sentinel America, Inc.,
according to Iqbal Ahmed, executive
vice president of Sentinel Canada.
Fletcher has 28 years of experi-
ence in sales, marketing and techni-
cal management with specialty syn-
thetic lubricants for aerospace and
industrial applications. His respon-
sibilities includes sales and market-
ing, technical liaison and product
development for North America.
INu51RY LALNAR
Au6u51
- 2nd uk-Lbinu 5ummer 5cbooI on
1riboIogy und 5urfuce ngineering,
Aug. 23-26, University of Southamp-
ton, Southampton, UK. For more in-
formation, visit www.aoIon.uc.uk/
ncuIa/venIa/upcoming/2nd-uk-
Lbinu-5ummer-5cbooI-2010.bImI.
- ReIiubIe PIunI 2010 Lonference 6
xbibiIion, Aug. 31-Sept. 2, Nashville
Convention Center, Nashville, Tenn.
For more information, visit bIIp://
conference.reIiubIepIunI.com/.
5P1HR
- A5H 2010 InIernuI LombuaIion
ngine iviaion FuII 1ecbnicuI Lon-
ference, Sept. 12-15, Hilton Palacio
Del Rio, San Antonio, Texas. For
more information, visit www.uame.
org.
- Lumbridge 1riboIogy Lourae {Fric-
Iion, weur und LubricuIion], Sept.
15-17, Mller Centre, Cambridge, UK.
For more information, visit www.ifm.
eng.cum.uc.uk.
- AIuminum AaaociuIion 2010 AnnuuI HeeIing, Sept. 20-
22, Hyatt Regency OHare, Chicago, Ill. For more infor-
mation, visit www.uIuminum.org.
- uropeun uae iIa 6 LubricunIa 2010, Sept. 22-23,
Pestana Chelsea Bridge Hotel, London, UK. For more in-
formation, visit bIIp://www.uciua.neI/wiki.uapx/Lon-
ferencea/upcoming?view=overview6id=162.
- AHA 2010 ducuIion Progrum: "aaenIiuI LoncepIa
of euring 1ecbnoIogy," Sept. 29-Oct. 1, Chicago, Ill.
For more information, visit www.umericunbeuringa.
org.
Lucian Fletcher
Want to be recognized in TLT?
TLT is interested in hearing from our readers. Let us
know whats happening in your company. If you have
news about a new employee or if someone in your com-
pany has been recognized with an award or any other
interesting items, let us know. Please send us your news
releases and photos for publication in Newsmakers to
TLT Magazine, Attn: Karl Phipps, 840 Busse Highway,
Park Ridge, IL 60068, kpbippaQaIIe.org.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 5 5
Herguth Laboratories, Inc., a leader in oil analysis,
has over 25 years of client dedication and service with
firsts to prove it.
First independant oil analysis lab to:
Achieve 10CFR50 Appendix B Certified
Achieve Radiological Certification
Publish Limits on Reports
Introduce Micro Grease Analysis
Analyze Refrigerants
Own an SEM for Failure Analysis
Develop the One Spot Oil Test
CALL US FIRST
For all of your oil analysis and testing solutions!
800-OIL-LABS (645-5227)
or visit our website www.herguth.com
Herguth Laboratories, Inc.

SOUNDING BOARD
Frofess|oae|
NaticnaI Lubricatin Crease
!nstitute and !ndependent Lubri-
cant Manufacturers Asscciaticn.
Belong to network and learn about
activities in industry.
Scciety cf Maintenance and
ReIiabiIity PrcfessicnaIs, !nter-
naticnaI CcunciI fcr Machinery
Lubricaticn and Asscciaticn cf
Maintenance PrcfessicnaIs.
American PetrcIeum !nstitute to
offer expertise.
!nstitute cf CcunciI fcr Machin-
ery Lubricaticn. Mainly for techni-
cal certication in machinery lubrica-
tion and oil analysis because in my
language and my country (Spanish
and Colombia), I could not access
other types of certication in specic
skills of lubrication.
American Scciety fcr Materi-
aIs, MateriaIs Research Scciety,
American Vacuum Scciety.
American ChemicaI Scciety. I
am a chemist, and this is the leading
organization for chemists.
xc|ad|a S1L, Wh|ch profess|oae| ead persoae|
orea|zet|oas do yoa be|oa to ead Why?
1 23%
2 30%
3 26%
4 12%
5 4%
More than 5 5%
Based on responses from 138 TLT readers.
!ncIudin STLE, hcw many
prcfessicnaI/technicaI
asscciaticns cr sccieties
are ycu a member cf?
5 6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
5TLE members are |oiners, according to
the !3S TLT readers who responded to this
month's survey. 77 oI them belong to
more than one proIessional or technical or
ganization, and 2! oI them belong to Iour
or more such groups. Among the list oI or
ganizations readers belong to are Iamiliar
names such as ILMA, NLCI, A5ME and 5AE.
But also included on our readers' lists were
regionalbased groups, including the Cen
tral Kentucky Main Council, the Dallas Paint
& Coatings Association, the Polish Tribology
5ociety and even the Forum oI Lithuanian
5cientists Abroad. Cn the personal side,
many 5TLE members belong to organiza
tions dedicated to charitable work such as
the Ped Cross and Habitat Ior Humanity.
Cthers belong to hobbybased groups, in
cluding motorcycling, stamp collecting and,
yes, home brewing.
Scciety cf Autcmctive Enineers.
I am active in the technical commit-
tee activities for grease, ATF and gear
oils. Asscciaticn cf !rcn and SteeI.
On the lubes committee. American
Scciety cf Testin MateriaIs. Help
to set the testing standards. Ameri-
can Cear Manufacturers Assccia-
ticn. On the lubes committee.
American ChemicaI Scciety,
which offers resources for industry
news and analytical chemistry and
helps me as a lab manager. American
Scciety fcr 0uaIity, which offers
resources for designing and imple-
menting quality systems. American
!nstitute cf ChemicaI Enineers,
which helps me with Chem-E basics
and relates to my plant.
I am a member of the CentraI
Kentucky Main CcunciI. It is a
local council primarily made up of
trucking companies, truck dealer-
ships and other related service type
companies serving the trucking
industry. We meet monthly and bring
in guest speakers to discuss upcoming
industry/maintenance practices. Good
for me to keep up on whats new and
what problems the industry is facing
at the local level.
Scciety cf PIastic Enineers.
American Scciety fcr Testin and
MateriaIs, which provides technical
and industry information.
Scciety cf Manufacturin Eni-
neers.
Fcrum cf Lithuanian Scientists
Abrcad. Being Lithuanian, I get to
meet fellow Lithuanian scientists who
work in various technical elds. Since
tribology is a very multidisciplinary
eld, we often nd a common topic
and discuss it in great detail. This ad-
ditional knowledge sometimes is very
helpful.
NaticnaI Lubricatin Crease
!nstitute. I belong to this organiza-
tion to further immerse myself in the
lubricant industry and to provide the
best possible opportunities to learn
and keep myself in the forefront of
lubricating technologies.
Precisicn MetaIfcrmin Assccia-
ticn.
American Scciety cf Testin and
MateriaIs. We use many of their pro-
cedures in our lubricant test lab.
DaIIas Paint & Ccatins Assc-
ciaticn, NaticnaI Asscciaticn cf
ChemicaI Distributcrs.
American Vacuum Scciety and
American ChemicaI Scciety.
These along with STLE benet me
because I dont just identify with one
discipline. I am a chemist and surface
scientist as well as a tribologist.
!nternaticnaI !sraeI TribcIcy
Scciety.
Asscciaticn fcr !rcn and SteeI
TechncIcy, because I used to work
in steel and still deal with the indus-
try in my current role.
American ChemicaI Scciety
because I get free coffee mugs and
subscriptions to some journals which
otherwise cost inordinate sums per
article.
PcIish TribcIcy Scciety.
!'m cnIy a member cf STLE. As
part of my OMA II certication, Ive
presented at, participated in and at-
tend Vibration Institute and SMRP
organization meetings and seminars.
American ChemicaI Scciety and
American !nstitute cf ChemicaI
Enineers. My degree is in chemistry
and I have graduate hours in chemi-
cal engineering. I am retired, so my
professional associations enable me to
keep current in the eld.
American Scciety cf Heatin Re-
frieraticn and Air Ccnditicnin
Enineers.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 5 7
I pay 35%
My employer pays 65%
Based on responses from 138 TLT readers.
Think cf the prcfessicnaI/
technicaI scciety that ycu
ccnsider tc be ycur primary
membership cranizaticn~dc
ycu cr ycur empIcyer pay the
membership fee?
American Marketin Asscciaticn.
Excellent monthly speakers from both
local and large international companies.
ASTM. I like to work on the com-
mittees reviewing the test methods.
SAE. I like to keep abreast of engine
technology requirements.
Scuth African !nstitute cf Tri-
bcIcy. I can share my knowledge,
contribute to develop the knowledge
and gain knowledge. Actually I try to
actively participate in as many tribo-
logical issues as possible.
RcyaI Scciety cf Chemistry.
Sinapcre NaticnaI !nstitute cf
Chemistry. They organize talks/
lectures/courses related to chemistry
issues which may be related to work
and interest.
TribcIcy Scciety cf !ndia. They
are a group of professionals working
toward solving industry problems in
India.
American and Eyptian ChemicaI
Sccieties. My life experience was
in lubricating oils, and this helps me
keep in touch with what is running
worldwide.
American Scciety cf MechanicaI
Enineers, because it includes me-
chanical designs and material proper-
ties. SAE, because it has industry-
specic activities.
0ntaric Asscciaticn cf Certi-
hed Enineerin Technicians and
TechncIcists. Certihed !ndustri-
aI MiIIwriht. One or the other are
requirements or improve the chances
for certain jobs in the maintenance
eld.
Nierian Scciety cf Enineers and
CcunciI cf ReuIaticn cf Eni-
neerin in Nieria. Benets include
professional membership, industry
networking and exchange of technical
ideas.
Wire Asscciaticn !nternaticnaI
& Asscciaticn cf !rcn and SteeI
TechncIcy. Main reason for both
is they are more relevant to specic
markets in which I am involved.

Rctary CIub. I get personal satisfac-

tion in doing charity work that is
appreciated by others.
ExperimentaI Aircraft Assccia-
ticn. I receive information related to
personal aircraft including custom-
built vehicles, antiques, classics,
warbirds, ultralights, etc.
Bcy Sccuts cf America. The ben-
et I receive from this organization is
to witness the growth of young men,
both physically and mentally, through
opportunities of outdoors activi-
ties along with community and civic
service.

Licns !nternaticnaI, which is devot-
ed to community service.
Vintae ChevrcIet CIub cf
America. Great magazine and an
opportunity to visit with other old-car
enthusiasts.
United Synacue cf Ccnserva-
tive Judaism. Cultivation of Jewish
living among conservative Jews.
The NaticnaI PhiIateIic (stamp
ccIIectin) CIub in CcIcmbia. This
gives me the opportunity to meet
people with my same hobby and to
share experiences with them.

A IccaI antique scciety that deals
with the preservation and restoration
of antique farm machinery, tractors,
engines, etc. It is a great way to meet
some of the older people in the area,
learn from them and keep my interest
in not losing this part of our history.

LccaI ccuntry cIub, a great tool for
networking locally.
5 8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
STLE is the most important
professional organization
I belong to 57%
STLE is the second most
important professional
organization I belong to 32%
STLE is the third or
fourth most important
professional organization
I belong to 11%
Based on responses from 138 TLT readers.
Where dces STLE rank
amcn aII the prcfessicnaI
cranizaticns ycu are a
member cf?
Editors Note: Sounding Board is based on an
e-mail survey of 7,200 TLT readers. Views
expressed are those of the respondents and do not
reect the opinions of the Society of Tribologists
and Lubrication Engineers. STLE does not vouch
for the technical accuracy of opinions expressed in
Sounding Board, nor does inclusion of a comment
represent an endorsement of the technology by
STLE.
! beIcn tc L.A. Fitness, a gym be-
cause I want to stay t and sexy and
get rid of that smart, nerdy guys cant
be sexy stereotype. Im like uber
smart and even uber-er sexy.
CentraI Bucks Bike CIub, which
has organized outings.
! beIcn tc a ccmmunity crches-
tra. The main benet that I derive
from this organization is the cama-
raderie that I share with adults who
truly love to perform great music.

Red Crcss. Were ready when the
time comes and help support hurri-
cane relief efforts while contributing
to the community

ABC Mentcrin !nitiative, which
provides one-on-one mentoring with
high school students in all aspects of
their life choices (school, hobbies,
interests). I benet from seeing situa-
tions from a wider perspective when
with my mentee.
Habitat fcr Humanity Fccd Bank
cf CentraI !cwa. Lets me assist oth-
ers less fortunate than myself.
American MctcrcycIist Assccia-
ticn. Protects motorcyclists rights on
the road.
Bearded CcIIie CIub cf America.
Social interaction with other owners
of the breed.
NaticnaI RiBe Asscciaticn,
which supports my second amend-
ment rights. Heritae Fcundaticn,
which supports constitutional rights.
Pheasants Fcrever, which supports
hunting rights.
LccaI baIIrccm dance rcup
exercise and entertainment.
Churchso I can get a little help
from above!

Nuts 'N' BcIts Fcundaticn. This is
a non-prot organization designed to
provide volunteer services to schools
and communities, thereby enhanc-
ing the lives of children and senior
citizens.
!nternaticnaI Scciety fcr Krishna
Ccnscicusness. I am getting puri-
ed of all my bad tendencies and
habits from the meditation program
they conduct.

Michian VcIkswaen Enthusi-
asts. Gives me the opportunity to
teach my hobby (building race and
show cars) to others and also to trade
parts around.
A vcIuntary rcup cf my schccI
buddies. Through this group we
provide support to underprivileged
children to enable them to have ac-
cess to education. We also provide
support to retired school teachers. We
do not expect to get any benet from
this endeavor; we just try to give back
a little to the society for all that we
have received.
American Hcmebrewers Assccia-
ticn. Networking with other home
brewers, sharing techniques and
recipes.
! am invcIved in a hcme-based
BibIe feIIcwship afhIiated with
The Way !nternaticnaI. It teaches
me how to read, understand and
apply the Bible in my life and also
have a base of strength for facing lifes
situations.
NaticnaI McdeI RaiIrcad Assccia-
ticn. I play with trains.
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 5 9
If you are looking to make your metalworking
fluids greener with vegetable oil-based formulations,
Huntsmans new SURFONIC

MW-100 additive can help.

SURFONIC

MW-100 is a single component emulsifier

designed to enhance the metalworking properties of
vegetable oil-based lubricants. MW-100 is inherently low
foaming, compatible with phosphate esters, and can be
used to formulate very stable macro or micro emulsions. In
addition, MW-100 has a low pour point for easy handling
and features low toxicity.
For more information, contact a Huntsman representative and
learn how SURFONIC

MW-100 can turn heavy metal green.

Heavy Metal
Goes Green.
SURFONIC is a registered trademark of Huntsman Corporation or an affiliate thereof in one or more, but not all, countries.
www.huntsman.com/metalworking
North America +281.719.6000
Europe +44.1946.694108
Asia Pacific +65.6390.0371
P1R151 LLA 5 Au1HA1IL
LLvLAN PN LuP FLA5h PIN1 151R
AH1k PeIroIub Lo., introduces the PeIroIeaI LLA 5 LIeveIund pen Lup
FIuab PoinI und Fire PoinI AnuIyzer that automatically tests both liquid and high-
ly viscous liquid and solid samples, according to the ASTM D92 and ISO 2592 standards.
The benchtop instrument comes completely ready for testing with automatic test rou-
tines, user-denable programs and self-testing/
servicing routines, as well as a rapid search pro-
gram when fast results are required. The versatile
analyzer features the latest electronics, includ-
ing a 5.7-inch color graphics display and Pmove
jog wheel control. It offers a ash point scanning
function and comes equipped with a barometric
pressure sensor for corrected results, according
to ASTM. It has selectable output, either C or F,
and a measuring temperature testing range from
ambient to 400 C (752 F) without requiring any ad-
ditional equipment.
AH1k PeIroIub Lo.
Broken Arrow, Okla.
(918) 459-7170
www.peIroIub.com
IAPhRA6H vALv5 FR
RAL1Iv FLuI5
1riduk LLL introduces the com-
panys new 1riduk HodeI 475
5eriea iupbrugm vuIvea that
are designed to be used with low-
to-medium viscosity uids. The
internal design of these dispensing
valves prevent uid contact with
the actuating components, making
them ideal for use with reactive
uids such as cyanoacrylates and
solvent-based adhesives and coat-
ings. Minimal valve maintenance is
required, which reduces costs asso-
ciated with spare parts and produc-
tion downtime. Tridaks diaphragm
valves offer stroke adjustment to
precisely dial-in the desired shot
volume for consistent and repeat-
able performance. The Model 475 is
available in three congurations,
which allows the valve to be com-
patible with any number of materi-
als. All three congurations of the
Model 475 are actuated using the
Tridak Model 345 valve controller.
1riduk LLL
Torrington, Conn.
(860) 626-6700
www.Iriduk.com
Nw HL 5Y51H 5I6N FR LNL PRA1IN5
uNI51 Inc. introduces the latest product in MQL (Minimum Quantity Lubrication) technol-
ogy, uni-Hux RevoIuIion, which is the rst MQL system designed to integrate seam-
lessly with CNC machines and gives users the ability to customize the lubricant output
volume for each tool in a machine directly from the CNC program. The Uni-Max Revolution
also includes an optional pulsed air blow-off nozzle for removing chips in difcult opera-
tions like pocket milling. This new system was developed specically for OEM installations
on CNC mills and turning centers, but can also easily be retrotted on existing machines
with available M-code outputs. MQL technology is a state-of-the-art lubrication technique
that carefully and accurately applies small amounts of
a high quality lubricant to the interface where the cut-
ting tool meets the material being worked. This process
reduces friction, removes heat and keeps the part cool.
MQL has been adopted by several manufacturers because
it saves from 50% to 90% on uid-related costs, keeps
them manufacturing plant cleaner and greatly
reduces uid-related environmental concerns by
eliminating much of the mess, waste and cleaning
associated with traditional lubrication methods.
uNI51 Inc.
Grand Rapids, Mich.
(800) 253-5462
www.uniaI.com
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 6 1
Send us your new product news with color photos to:
TLT Magazine, Attn: Karl Phipps, 840 Busse Highway,
Park Ridge, IL 60068, kpbippaQaIIe.org
I5LP LA1ALY51 PRFu5IN6 51A6 INLuA1R
NAL5 AvANL LN6-1RH LIv LLL 51uI5
veeco InaIrumenIa, a leading provider of scanning probe
microscopes to the nanoscience community, introduces the io-
5cope LuIuIyaI Perfuaing 5Iuge IncubuIor, which is de-
signed to facilitate cell biology studies using simultaneous AFM
and optical methods. This accessory enables researchers to main-
tain ideal cell culture conditions for long-duration live cell stud-
ies with Veecos Catalyst AFM, signicantly extending cell viability
over hours or even days. Featuring a broad range of incubation and
perfusion options, including continuous or intermittent liquid per-
fusion, heating and maintenance of carbon dioxide buffered media,
the Perfusing Stage Incubator accessory optimizes the Catalysts
ideal integration between atomic force microscopy and optical
microscopy techniques. Researchers are able to observe dynamic,
long-duration events without compromising the full capabilities
of either technology or the samples. The accessory also supports
Veecos exclusive Microscope Image Registration and Overlay fea-
ture, which automatically imports and rescales light microscope
images to then be used to direct the location of AFM imaging and
force measurements. In addition to the Perfusing Stage Incubator,
the BioScope Catalyst AFM incorporates many features addressing
the unique requirements of life science samples, including heat-
ing capabilities and exible mounting options. The system utilizes
a top-down laser path that permits uncompromised use of phase
contrast, differential interference contrast and brighteld optical
microscopy, while offering unmatched optical and physical access
from the top of the
AFM head. Compat-
ibility with the top
inverted optical mi-
croscope brands, as
well as with their
standard condens-
ers and a full tur-
ret of objectives,
ensures nonlimited
optical capabilities.
veeco
InaIrumenIa
Plainview, N.Y.
(516) 677-0200
www.veeco.com
Nw FuL 1hANL AI1Iv hLP5
FI6h1 PIPLIN 51R55 LRR5IN LRALkIN6
HidLonIinenIuI LbemicuI Lo. Inc., introduces a new fuel
ethanol additive designed to combat stress corrosion cracking.
According to MCC, this new chemistry potentially paves the way
for ethanol to be shipped throughout North America using the
nearly 100,000-mile pipeline network that delivers the nations
petroleum products (gasoline, diesel fuel, home heating oil, etc.)
reliably, safely, efciently and economically. Despite governmen-
tal regulations requiring increased use of ethanol in gasoline,
most ethanol is transported by truck and rail. As a result, the cost
savings of pipeline transportation is critically important. SCC is
the slow growth of cracks along a pipeline, which are caused by
mechanical stress and exposure to a corrosive material. MCCs new
product family consists of lm formers combined with other func-
tional additives. These products are
clear, nonhazardous, mobile liquids,
which mean they can provide both
protection from corrosion and do
not harm the fuel product.
HidLonIinenIuI
LbemicuI Lo., Inc.
Olathe, Kan.
(913) 390-5556
www.mccbemicuI.com
Formulate Green
With Fruits & Beans
Palm Based Oleochemicals & Derivatives:
Fatty Acids Glycerine
Fatty Alcohols Fatty Acid Esters
Fractionated Methyl Esters
(C610, C8, C10, C12, C1214, C14, C16, C1618, C18)
Castor Oil & Derivatives:
12 Hydroxy Stearic Acid CO Fatty Acid
Hydrogenated Castor Oil Ricinoleic Acid
Methyl 12 Hydroxy Stearate Sebacic Acid
Visit our New Website: www.acme-hardesty.com
Call us for a free Oils & Fats Composition Chart
Acme-Hardesty Co.
450 Sentry Parkway East
Blue Bell, PA 19422
866.226.3834
6 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Increasing uptime through
lubricant condition monitoring

Your partner in online and ofine monitoring
of critical fluids, lubricants and machinery

+1 (678) 905-5630 x801
www.kittiwake-americas.com
[email protected]
On-site
Online
Real Time
Remote
lh /|c-Pcc|f|c Bce C|| Ccnference
lcrch 7-8, 2011
17lh /nnuc| lue| & Luce /|c Ccnference
lcrch -11, 2011
F+I Week |n $|ngopore
Creating More with Less underlines the issue of
energy efficiency/savings on the agenda, as we all face
the challenge of resource limitation and serious concerns
about the environment. Hence, there is a need to explore
all possible avenues to reach the goals/aspirations/
targets of energy savings in both the transportation
and industrial sectors.
Technology & innovation are the keys to success,
as we move forward in the transition from the traditional
mobile sources of energy to new and renewable forms of
energy. Partnership and collaboration can help to deliver
more efficient solutions. The SAE F&L Councils Steering
Committee for Asia will meet during F+L Week to try to
establish a more structured form of collaboration among
all key stakeholders in Asia-Pacific.
For more deIo||s, v|s|I our webs|Ie
oI www.fue|sond|ubes.com.
lf ycu cre |nlerelec |n mck|ng
cn crc| preenlcl|cn, ycu mcy
prcpce c 00-wcrc cclrccl
cy enc|ng cn e-mc|| lc ec|lcr
fue|cnc|uce.ccm nc |cler lhcn
/ugul 31, 2010. Ycur cclrccl w|||
ce rev|ewec cy cur ccnference
ccv|cry cccrc. lf ycur pcper |
cmcng lhce e|eclec fcr crc|
preenlcl|cn, ycu w||| rece|ve cn
e-mc|| ncl|f|ccl|cn nc |cler lhcn
Seplemcer 30, 2010.
How Io presenI o poper
"Iechno|ogy & InnovoI|on.
CreoI|ng More w|Ih Iess"
IHM.
Co|| for
Popers
TLT A0VERTISERS IN0EX |uuu' ?|`| - \0|. ||. N0. |
Spread the wcrd with TLT!
Put the strong marketing power of TLTs print and digital editions behind your sales message.
Your ad will reach more than 13,000 leading lubricant industry professionals, including
manufacturers, additive suppliers and end-users. For more information on our competitive ad
rates, contact: Tracy Nicholas VanEe, National Sales Manager, (630) 922-3459, [email protected].
Company Page Contact Phone E-mail Web
Acme-Hardesty Co. 62 Bryan Huston (215) 591-3610 [email protected] www.acme-hardesty.com
Afton Chemical Co. IFC Lauren Ereio (804) 788-6081 [email protected] www.aftonchemical.com
Analysts Inc. 7 Cary Forgeron (310) 210-5000 [email protected] www.analystsinc.com
Anton Paar USA 71 Christen Hangey (804) 550-1051 [email protected] www.anton-paar.com
Cannon Instrument Co. 53 Patricia Argiro
(814) 353-8000
ext. 267
[email protected] www.cannoninstrument.com
Compass Instruments 69 Jim Hepp (630) 556-3669 [email protected] www.falex.com
Condition Monitoring
International
63 Jack Poley (305) 669-5181 [email protected] www.conditionmonitoringintl.com
CRC Press - Taylor and Francis
Group
IBC
Maikala del
Castillo
(561) 998-2562 [email protected] www.crcpress.com
ExxonMobil Chemical Co. 9 Seth Barron (281) 368-5926 [email protected] www.exxonmobilchemical.com
Fuels & Lubes Asia 64
Hannah Eliza
Alvaro
632-772-4731 [email protected] www.fuelsandlubes.com
Herguth Laboratories 55 Linda Perry (707) 554-4611 [email protected] www.herguth.com
Huntsman 60 James Machac (281) 719-4743 [email protected] www.huntsman.com
International Joint Tribology
Conference
49 Merle Hedland (630) 323-7212 [email protected] www.stle.org
LANXESS Corp. 41 Cindy Williams (412) 809-3558 [email protected] www.lanxess.com
The Lubrizol Corp. 3 James Baldwin (440) 347-1291 [email protected] www.lubrizol.com
Monson Co. 33 Doug Hiple (609) 773-0031 [email protected] www.monsonco.com
Mnzing OBC Dana Sultan (973) 233-6791 [email protected] www.munzing.com
NCeed Enterprises Inc. 24 Noreen Cherry (352) 750-5670 [email protected] www.nceed.com
Polaris Laboratories LLC 25 Jacque Powers
(317) 808-3750
ext. 229
[email protected] www.polarislabs.com
RheinChemie 19 Julie Bradler (440) 285-3547 [email protected] www.rheincheime.com
Sea-Land Chemical Co. 18 Joseph Clayton (440) 871-7887 [email protected] www.sealandchem.com
Soltex Inc. 40 Steve Case (281) 587-0900 [email protected] www.soltexinc.com
STLE 2011 Annual Meeting 5 Judy Enblom (847) 825-5536 [email protected] www.stle.org
O 8 : 2 07
a t I f ur
g e t eb te
o ow t e mo ey
W en c e ce nd
po t s c sh
P e i en s R po t
n p i e o S LE
c r o a e m mbe s
Rate ef chane
alarms
A b t er m t ic er x en i l be i e
ISO 281 2007
Part II of our
g eat debate
Follow the money
When science and
poli ics clash
Presidents Report
In praise of STLE
corpo a e members
Rate-ef-chane
alarms
A better metric fer extendin lube ife.
Ch ck out ur d gi al LT sp nso ed by Po ar s L bor to i s t www t e rg
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 6 5
k5
HLIN6 AN ANALY1ILAL H1h5 IN 1RIL6Y
Authors: Ilya I. Kudish, Michael J. Covitch
Publisher: CRC Press
Improving our understanding of fric-
tion, lubrication and fatigue, Mod-
eling and Analytical Methods in Tri-
bology presents a fresh approach to
tribology that links advances in ap-
plied mathematics with fundamen-
tal problems in tribology related to
contact elasticity, fracture mechanics
and uid lm lubrication. The au-
thors incorporate the classical tenets
of tribology while providing new
mathematical solutions that address various shortcomings
in existing theories. From contact interactions to contact fa-
tigue life, this book connects traditionally separate areas of
tribology research to create a coherent modeling methodol-
ogy that encompasses asymptotic and numerical techniques.
The authors often demonstrate the efcacy of the models by
comparing predictions to experimental data. In most cases,
they derive equations from rst principles. They also rigor-
ously prove problem formulations and derive certain solu-
tion properties. Solutions to problems are presented using
simple analytical formulas, graphs and tables. In addition,
the end-of-chapter exercises highlight points important for
comprehending the material and mastering the appropriate
skills. Unlocking the secrets that govern the physics of lubri-
cated and dry contacts, this book helps tribologists on their
quest to reduce friction, minimize wear and extend the oper-
ating life of mechanical equipment. It provides a real-world
industrial perspective so that readers can attain a practical
understanding of the material. Available at www.crcpreaa.
com. List Price: $119.95 (USD)
HALhINRY vIRA1IN AN R1RYNAHIL5
Authors: John Vance, Fouad Zeidan, Brian Murphy
Publisher: Wiley
Whether its a compressor on an off-
shore platform, a turbocharger in
a truck or automobile, or a turbine
in a jet airplane, rotating machinery
is the driving force behind almost
anything that produces or uses en-
ergy. Counted on daily to perform
any number of vital societal tasks,
turbomachinery uses high rotational
speeds to produce amazing amounts
of power efciently. The key to in-
creasing its longevity, efciency and
reliability lies in the examination of rotor vibration and
bearing dynamicsa eld called rotordynamics. A valuable
textbook for beginners, as well as a handy reference for ex-
perts, Machinery Vibration and Rotordynamics provides rich
technical detail and real-world examples geared toward the
study of machine vibration. A logical progression of informa-
tion covers essential fundamentals, in-depth case studies and
the latest analytical tools used for predicting and preventing
damage in rotating machinery. This book combines rotor-
dynamics with the applications of machinery vibration in a
single volume, includes case studies of vibration problems
in several different types of machines, as well as computer
simulation models used in industry and contains fundamen-
tal physical phenomena, mathematical and computational
aspects, practical hardware considerations, troubleshooting
and instrumentation and measurement techniques. Available
at www.wiIey.com. List Price: $175 (USD)
HuL1IHIA
5H RLA55 5Ik1h I1IN F HANuFAL1uRIN6
1k1k LLA55IL FuNAHN1AL5 F 1L 5I6N
1be 5ocieIy of HunufucIuring ngineera has released the
sixth edition of its textbook classic Fundamentals of Tool De-
sign, which for more than 40 years manufacturing practitio-
ners, designers, educators and students have used to gain
an in-depth understanding of all the factors that impact tool
success.
The table of contents for the fully illustrated sixth edition
includes: General Tool Design; Materials Used for Tooling;
Cutting Tool Design; Workholding Concepts; Jig Design;
Fixture Design; Power Presses; Die Design and Operation;
Inspection and Gage Design; Tool Design for Joining Pro-
cesses; Modular and Automated Tool Handling; Computer
Applications in Tool Design and Geometric Dimensioning
and Tolerancing.
Following are updates that readers will nd in the sixth
edition:
Rcviscdgcomciric dimcnsioning andiolcrancing (GD&T)
information ASME Y14.5-2009.
An insirucior's guidc wiil answcrs io ilc rcvicw qucs-
tions in the book and information on how to incorporate
the DVDs into classroom or online instruction.
Lxpandcd lroacling cxamplcs and conccpis.
Rcviscd iool maicrial, mcclanical dclniiions and rclcr-
ence tables.
In-dcpil rcvicw ol lusling applicaiions.
Ncwappcndiccs wiil sourccs lor iool dcsign componcnis,
related Web sites and handy GD&T reference charts.
Morc rcvicw qucsiions ai ilc cnd ol cacl clapicr.
In addition, the book and nine DVDs are available as a
package or individually to serve both educational and profes-
sional needs. For more information, log on to www.ame.org.
RESOURCES
6 6 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
5eeIburumun, 5. und kubrumun, A.,
A Windage Power Loss Model for Spur
Gear Pairs, pp. 473-484.
zckogIu, h., uIik, A., Imrek, h.
und kurukua, H.5., Examination of
Pitting and Wear in Borided, Carbur-
ized and Borocarburized AISI 8620
Gears, pp. 485-490.
Ruy, 5. und Lbowdbury, 5.k.R., An
Analysis of Surface Temperature Rise at
the Contact between Sliding Bodies with
Small-Scale Surface Roughness, pp.
491-501.
Lbung, .6., wung, k.F., Yung, F..
und wung, }.Z., Lubricating Properties
of Tripod Sliding Universal Joints
Considering Thermal Effect, pp.
502-510.
Leveaque, 6.A. und Arukere, N.k.,
Critical Flaw Size in Silicon Nitride
Ball Bearings, pp. 511-519.
iawua, 5. und 5uIupuIby, A., A
Study on Tribological Behavior of
Alumina-Filled Glass-Epoxy Compos-
ites Using Taguchi Experimental
Design, pp. 520-532.
5ubu, 5. PR., 5uIupuIby, A., Hiabru,
., PuInuik, A. und 5reekumur, k.P.,
Tribo-Performance Analysis of Fly
Ash-Aluminum Coatings Using Experi-
mental Design and AMN, pp. 533-542.
}uckaon, R.L., An Analytical Solution
to an Archard-Type Fractal Rough
Surface Contact Model, pp. 543-553.
Li, 5. und kubrumun, A., Prediction of
Spur Gear Mechanical Power Losses
Using a Transient Elastohydrodynamic
Lubrication Model, pp. 554-563.
wung, }., Lben, h., hun, Z. und Lben,
., Investigation of the Effect of
Milli-Scale Dimples on Planar Contact
Lubrication, pp. 564-572.
ireauw, 6. und unIcbev, 6..,
Elastohydrodynamic (EHD) Traction
Properties of Seed Oils
1
, pp. 573-583.
1bomua, ., hudeId, H. und
AuaIen, 5., Wear and Friction Model-
ing on Lifeboat Launch Systems, pp.
584-599.
weaIerberg, L.6., LundaIr6m, 1.5.,
h6gIund, . und LugI, P.H., Investiga-
tion of Grease Flow in a Rectangular
Channel Including Wall Slip Effects
Using Microparticle Image Velocimetry,
pp. 600-609.
6reco, A., HurIini, A., Liu, Y., Lin, L.
und wung, .}., Rolling Contact
Fatigue Performance of Vibro-Mechani-
cal Textured Surfaces, pp. 610-620.
Leveaque, 6. und Arukere, N.k.,
Empirical Stress Intensity Factors for
Surface Cracks under Rolling Contact
Fatigue, pp. 621-629.
jiIe, }.., 1eixeiru, }.A. und Lur-
mody, L., Mechanical Seal Failure
Analysis, pp. 630-635.
A bimcnthIy Icck at tribcIcy's Ieadin-ede jcurnaI
Review
Tribeley Transactiens
' ' ' . ' | | . 0 | u | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u | u u u ' ? | ` | - 6 7
nIine ucceaa Io 1riboIogy 1runaucIiona uvuiIubIe
free Io 51L membera!
STLE has reached an agreement with journal publisher partner Taylor &
Francis to provide full access for all STLE members to Tribology Transac-
tions. This includes the current volume plus a full searchable archive of 52
years worth of tribology research. Downloads of the full papers are free and
available 24/7.
To access the journal, log on to the Members Only section of the
STLE Web site (www.aIIe.org) and enter your ve-digit membership ID
and password, then click the menu button on the left column labeled
Tribology Transactions Journal.
Th rs
are featured in the latest issue
of STLE Tribology Transactions
(Vol. 53, No.4/July-August 2010):
n the June TLT we discussed learning to love the sound of no.
Having someone tell you no, faster and more often, is critical
to a salespersons suc-
cess. So how can you get
comfortable with hear-
ing no?
First and most im-
portant, you must mas-
ter the proper mindset.
Why? Developing the
right mindset helps you
handle the inevitable
rejection involved in
selling for a living. For
example, have you ever
wondered how great
salespeople can get re-
jected all day long and
get up the next day to
do it all over again? Its
because they know this
little secret: No doesnt
really mean noit just
means not now.
If youre talking to
a prospective customer
(and if hes not a pros-
pect, why are you talking to him in the rst place?), the fact that
he tells you no doesnt suddenly turn him into a non-prospect, does
it? Of course not. It just means you werent able to either uncover
his triggers and present the right solution or connect with him on
some level. Or maybe he wasnt interested in buying right now.
And thats OK.
It happens to the best of us and doesnt change the fact that
you still have a valid prospect. If you played your cards right and
made a favorable impression, youll probably get another shot at
some point in the future. In the meantime, move on. This isnt the
end of the world. Put this prospect back in the queue and check in
with him periodically to see if circumstances have changed. When
the time is right, make another run at it.
While youre at it, how about offering a sincere thank you to
the prospect that had the guts to tell you no instead of dragging
you through that hell known as maybe? Remember, maybe is the
natural enemy of the
salespersona word to
be avoided at all cost!
Salespeople typi-
cally develop an aver-
sion to hearing no. They
sometimes do stupid
things to avoid hear-
ing it such as not mak-
ing calls, asking for the
business or playing the
maybe game with pros-
pects who wouldnt buy
from them. Sound fa-
miliar? If so, its time
to get comfortable with
hearing no again.
Dont get me wrong.
Im not encouraging you
to return to the old days
of sticking your foot in
the door and refusing to
take no for an answer.
Let me assure you that
nothing could be further
from my mind. On the contrary, Im encouraging you to take no for
an answer, as often as possible, so that you can move on more
quickly to prospects who want to engage with you.
So what should you do when a customer says no? Shake his or
her hand and say, Thanks very much! and get back in the saddle
as quickly as possible. Otherwise, this may not be the right profes-
sion for you!
SELLING POINTS
Jerry L. Kennedy
'ho' ree||y meeas 'aot aoW'
Jerry Kennedy, CLS, is owner of Inside Out
Business Solutions, a sales and customer service
training organization. You can reach him at
jkennedyQinaide-ouI-aoIuIiona.com.
Developing the right mindset can help you handle the
inevitable rejection involved in selling for a living.
I
Accepting no for an answer allows you to move on more
quickly to prospects who want to engage with you.
6 8 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
physical properties
and performance
characteristics of
fuels, lubricants
and materials
(630) 556-4835 | fax: (630) 556-3679 | www.compass-instruments.com
are the proud distributors of: We are the proud distributors of:
Condition monitoring is like a responsible pet owner who makes sure
his dog doesnt leave unwelcome presents on the shop oor.
SHOP FLOOR
Peter A. Oglevie
aforc|a the |eesh |eW
y local police department
is enforcing the leash law
in an effort to force people to
clean up their doggy dodo. Not a
bad idea. In fact, knowing how to
use the leash of condition moni-
toring can prevent the equivalent
of shop oor dodo.
I remember two maintenance
personnel who knew how to use
that leash well. Brad, the mainte-
nance supervisor, instructed Ron,
the technician, to monitor the
condition of the plants lubri-
cants.
I worked with both men and
learned a lot from them. My part
in the operation was to monitor
machine oils and specify lubri-
cants for the entire operation.
Their job was to take the informa-
tion I supplied and direct their
maintenance program for maxi-
mum results.
The problem came when two
things happened. First, we ran
into an economic downturn (much
like todays). Second, the plant
manager retired.
The new plant manager soon
put a choke collar on the maintenance budget. Unfortunately, Brad
had to lay off Ron and endure cuts in the lubricants purchasing
budget. It was purely a cost-savings measure directed from man-
agement.
Brad got discouraged and eventually left the company. I lost
the business to a lower-priced oil company that didnt care if con-
dition reports were followed.
Later management replaced Brad with a man who either did not
believe in condition monitoring or did not want to buck the system.
He ignored oil condition reports leading to, you guessed it, failing
pumps, sticking valves and oil on the oor.
The maintenance departments budget initially went down but,
of course, started to rise. The
condition of the shop deteriorat-
ed further.
Two years later, parts rejec-
tion was up 200%-300%, delivery
times were missed, and the shop
oor was often unsafe for walk-
ing. Upper management nally
stepped in and removed the plant
manager.
Brad was called back to the
plant. He called me and we re-
started the condition monitoring
program. A year later, the plant
was back in shape. We had to
make up for three years of ne-
glect, but the costs of rejected
parts were cut to our original lev-
els, deliveries were back on time
and the shop oor was oil free.
Meanwhile, Ron accepted a
position as maintenance director
in another plant and implemented
a condition monitoring program. I
got the business.
Lessons? Sometimes youll
have a hard time convincing se-
nior managers that a proper con-
dition monitoring program saves
bundles of money and a ton of
headaches. Urge them to view the maintenance department as a
prot center. Without a good program, scrap rates, machine down-
time and delayed deliveries will take huge bites out of their prof-
its.
When it comes to condition monitoring, know how to use the
leash.
M
Pete Oglevie is president of International
Production Technologies in Port Washington, Wis.
You can reach him at [email protected].
Urge management to view condition monitoring as a
leash that prevents expensive equipment failures.
7 0 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Tribology and Rheology on the Physica M
Universal Testing Platform in the Tempera
Range of -40 C to 200 C with Peltier C
Tribology applications such as:
Stribeck curve
Static friction
Ball bearing testing
Rheology applications such as:
Yield stress
Viscoelastic properties
Viscosity of oils and greases
High pressure up to 400 bar
Anton Paar

USA
800-722-7556
[email protected]
www.anton-paar.com
::: Intelligence in Rheometry
Rheometry
Focusing on Solutions.
here are many medical devices such as
stents, contact lenses and catheters
where the sliding of cells against man-made
materials is important. Additionally, there is
also growing evidence that a number of dis-
ease states may be reected in changes in
the measured mechanical properties of cells,
including frictional response. This opens the
possibility that such measurements could be
used diagnostically. Moreover, there are nu-
merous reports that cells react biochemically
to mechanical inuencesa fact that could
be of potential relevance in contact lens
use.
In a previous column,
1
we reported on
work by professor Greg Sawyers group at
the University of Florida where tribometer
measurements were used to determine the friction coefcients of
various materials sliding on aortic endothelial cells.
2

In a recent study at the same institution, professor Scott Per-
rys group has measured and analyzed, in collaboration with Saw-
yer, friction on single corneal epithelial cells in culture medium by
means of an atomic force microscope (AFM), incorporating a spher-
ical, colloidal silica tip.
3
The paper is signicant for a number of
reasons. By using a local-probe method such as AFM, they are able
to distinguish the intrinsic behavior of the outer surface of the cell
from the collective behavior of a cell layer. This provides mechani-
cal data that could be directly related to the environment of the
individual cell in terms of physiology, disease or drug stimuli.
From a mechanical-engineering standpoint, the paper raises an
interesting issue concerning tribological measurements of soft
systems such as cells. By employing an elastic foundation (bed of
springs) model rather than the conventional Hertz-Sneddon ap-
proach, the authors were able to extract the load-dependent shear
stress at the cell-probe interface.
This interface includes the membrane components, the trans-
membrane proteins and the proteins adsorbed from the cell-cul-
ture medium onto the probe tip. The shear stress is the conse-
quence of the friction and the stimulus to which the cell can react
biochemically. The ability to measure this for a living cell poten-
tially opens up a new area of tribological-biochemical investiga-
tions.
The Hertz-Sneddon model makes numerous assumptions such
as a at, non-frictional contact, innitesimal
deformation, innite sample thickness and
isotropic properties and is, therefore, less
suitable than a model that takes a more real-
istic approach to cell-material interactions.
A demonstration that the elastic-founda-
tion approach seems to work for the cells
investigated by the Florida groups is that
the AFM indentation behavior of the cells can
be very well tted by the model. It is inter-
esting to note that an elastic-foundation
model predicts a signicantly higher (by a
factor of ve) elastic modulus than that pre-
dicted by the Hertz-Sneddon model.
Tribological investigations are sensitive
to many mechanical properties and chemical
effects. As such they provide a valuable win-
dow onto the way in which biological systems react to the stresses
of the outside world. The challenges, as these authors have shown,
lie both in the measurements and in the data analysis.
FR FuR1hR RAIN6:
1. Tysoe, W.T. and Spencer, N.D. (2007), Fenders and Stents, TLT,
63(8), p. 80.
2. Dunn, A.C., Zaveri, T.D., Keselowsky, B.G. and Sawyer, W.G.
(2007), Macroscropic Friction Coefcient Measurements on Living
Endothelial Cells, Tribology Letters, 27(2), pp. 233-238.
3. Straehla, J.P., Limpoco, F.T., Dolgova, N.V., Keselovsky, B.G., Saw-
yer, W.G. and Perry, S.S. (2010), Nanomechanical Probes of Single
Corneal Epithelial Cells: Shear Stress and Elastic Modulus, Tribol-
ogy Letters, 38(2), pp. 107-113.
CUTTING EDGE
Drs. Wilfred T. Tysoe & Nicholas D. Spencer
Eddy Tysoe (left) is a distinguished
professor of physical chemistry at
the University of Wisconsin-
Milwaukee. Nic Spencer is
professor of surface science and
technology at the ETH Zurich,
Switzerland. Both serve as
editors-in-chief of STLE-afliated
Tribology Letters journal.
You can reach them at
[email protected] and
[email protected].
T
Friction measurements on single cells may provide
insights into cell-membrane behavior.
S||d|a oa ce||s
7 2 - | u u u ' ? | ` | | | 0 | 0 u t | u | | | | 0 N | | | N 0 | 0 u ' ' ' . ' | | . 0 | u
Jhere are aumerous reports that
ce||s react b|ochem|ca||y to
mechaa|ca| |a0ueaces~a fact that
cou|d be of poteat|a| re|evaace |a
coatact |eas use.
STLE MEMBERS!
*Log in to STLE.org to receive a
special member savings of 20%
Shows how submicron powder
lubricants share similar characteristics
with conventional fluid films
Explains how, instead of being
harmful, debris produced by wear
can act as a lubricant
Discusses the new technology of
powder lubrication and its depend-
ence on hydrodynamic lubrication
ffers experimental data on the
rheology of triboparticulate films,
as well as design guidelines for
power lubrication
Catalog no. DK2449
May 2010, 464 pp.
ISBN: 978-0-8247-5832-5
$139.95 / 89.00
$111.96 / 71.20
(After member savings)
Not an STLE member?
Go to CRCPress.com and use promo
code 866FE at checkout to save 15%
2010 ACS National Exposition
August 22-26, 2010
Boston Convention Center, MA
M
ark Your
C
alen
d
ar!
EXPLORE THE
PHENOMENON
OF CONTINUUM
SPECTRUM
as Applied to New Powder
Lubrication Technology
Search for:
Best defoamer for metalworking coolants, hydraulic fuids and non-aqueous lubricants
Search Results (1) match:
Mnzing. The Perfect Solution. The Only Solution.
When looking to perfect a product formulation, chemists need to rely on a company that understands their needs
and has the expertise to deliver results. With over 175 years of industry experience and the willingness to develop
the precise defoaming additive to solve any problem, Mnzing is committed to delivering optimum performance
and expert technical support
The Industry Standard in Defoamers
To try our FOAM BAN defoamers and take advantage
of our unlimited technical service,
call
+1 973-279-1306
www.munzing.com [email protected]
FOAM BAN

HP 710
FOAM BAN

HP 720
FOAM BAN

HP 730
FOAM BAN

HP 732
FOAM BAN

HV-810G
FOAM BAN

HV-820G
FOAM BAN

130B
FOAM BAN

149
FOAM BAN

154
FOAM BAN

159
FOAM BAN

3633E