Industries / Industrial equipment and services / Industrial machinery / Bearings
Lubrication and seals
Lubrication and seals is the work of choosing and putting in the grease or oil that keeps a thin film between a bearing's rolling parts, and the seals and shields that keep that lubricant in and dirt out.
7
Milestones achieved, from 1927 to 2014.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1927
Sealed at the factory
New Departure of Bristol, Connecticut, then part of General Motors, says in its company history that it introduced the first integrally sealed single-row ball bearings in 1927, for cars' rear wheels and propeller shafts, along with the first integral fan and water pump bearing. Seals built into the bearing keep its lubricant in and dirt out. The Smithsonian's National Museum of American History holds a sealed New Departure ball thrust bearing from about 1950, collected by the Anti-Friction Bearing Manufacturers Association for a public relations exhibit.
Credit
New Departure, a General Motors division; the "first" is the company's own claim
1942
Lithium thickens grease
Grease is oil held in a thickener, usually a soap. Clarence E. Earle of Washington, D.C., applied in April 1940 for a patent on greases thickened with a lithium soap, granted on 3 March 1942 as US 2,274,673. He wrote that his greases resisted water, which greases made with other alkali metal soaps did not, and withstood temperatures "greatly in excess" of those greases.
Credit
Clarence E. Earle
1959
Computing the oil film
Where a ball or roller presses on its ring, the steel flattens slightly and the oil becomes far more viscous under the pressure; this regime is called elastohydrodynamic lubrication (EHL). A. M. Ertel (1939) and A. N. Grubin (1949), in the Soviet Union, explained the film in such a contact by combining the two effects, assuming a film of even thickness. D. Dowson and G. R. Higginson published the first numerical solution for a long roller in 1959 and found a second pressure peak near the outlet of the contact, close to where the film is thinnest. A journal of the Institution of Mechanical Engineers published a review of their paper in 2008, and Semantic Scholar counts more than 500 citations of it.
Credit
D. Dowson and G. R. Higginson; earlier work by A. M. Ertel and A. N. Grubin
1963
Seeing the oil film
R. Gohar and A. Cameron reported in Nature in November 1963 a way to measure the oil film under elastohydrodynamic lubrication by optical interferometry: a steel ball runs against a glass disc, and the colors of the light reflected through the film show its thickness. Later work by Gohar and colleagues credits such studies with confirming the numerical predictions, including a horseshoe-shaped narrowing of the film at the outlet of a ball's contact. The method became a standard laboratory instrument: PCS Instruments sells a fully automated bench-top machine that measures the film between a 19.05 mm steel ball and a rotating glass disc down to 1 nm, by the company's figures.
Credit
R. Gohar and A. Cameron; PCS Instruments
1989
A standard test for grease
DIN, the German standards body, first issued DIN 51821-2 in March 1989, a method that runs grease in test bearings on the FAG FE9 apparatus to find the grease's life "under conditions similar to those encountered in practice"; the current edition dates from July 2016. In 1992 E. Kleinlein described FE9 tests at 100 to 200 °C in which grease life rose more steeply as the temperature came down, and found starved lubrication, too little lubricant reaching the contacts, the main cause of failure at moderate temperatures, especially where parts slide, such as between the cage and the rolling elements. Grease makers now print an FE9 result on their data sheets: Schaeffler's sheet for its Arcanol LOAD150 grease reports "F50 ≥ 100 h".
LessonAt moderate temperatures, a greased bearing failed mainly because too little lubricant reached the places where parts slide.
Credit
FAG; DIN; E. Kleinlein
2013
Kepler's wheels give out
Kepler, NASA's planet-hunting space telescope, pointed itself with reaction wheels, flywheels spun on ball bearings. One wheel failed in July 2012 after showing erratic friction, and another developed friction problems in early 2013; NASA rested it so that its lubricant might spread back, which did not help, and stopped trying in August 2013, according to the trade magazine Lubes'n'Greases. Christopher DellaCorte, NASA Glenn's senior technologist for tribology, listed Hubble, the control moment gyroscopes of the International Space Station and Kepler as recent high-value wheel failures in which bearings were implicated. He named the bind: wheel bearings run for years under high preload with minimal lubricant, to keep power loss low, in a vacuum that evaporates most oils and greases.
LessonA bearing that can never be relubricated lasts only as long as the lubricant put in when it was built.
Credit
NASA's Kepler mission; Christopher DellaCorte, NASA Glenn Research Center
2014
Grease claims put to test
A grease's data sheet states the highest temperature it can work at, but suppliers base that figure on different tests, or on rules of thumb. Chuck Coe of ExxonMobil Lubricants led tests of nine commercial lithium-complex and polyurea greases, all with public data sheets and identified only by number, in two static heat tests and two bearing rig tests: FAG's FE9 and a high-temperature wheel bearing test. Their FE9 lives at 140 °C ranged from 60 to 276 hours. Converted to temperatures, the riskiest data-sheet claim sat 91 °C above what its FE9 result supported, and the two bearing tests ranked the greases differently: one near the best by FE9 was near the worst in the wheel bearing test. The trade magazine Lubes'n'Greases reported the program in May 2014.
LessonA grease's stated top temperature means little unless the test behind it is named, and one bearing test does not predict another.
Credit
Chuck Coe and colleagues, ExxonMobil
Watch for
Hub seals with less drag
NSK says a new hub unit bearing, the bearing in a car's wheel hub, cuts friction by 40% compared with its conventional products while still keeping out muddy water. It uses a seal with one contact lip in place of three, a curved lip tip and a U-shaped slinger, the turning ring the lip runs on, shaped to drain water; a seal grease with a thinner base oil, plus a polymer to keep its oil film thick enough; and a seal surface smoothed by laser. NSK puts the gain at about 0.5% in fuel efficiency, or about 4 km more range per charge for an electric car, both by its own internal calculations.
Achieved whena carmaker or an outside test lab publishes measured friction torque and muddy-water test results for the new bearing beside the previous design.
Claimed by
NSK, March 2026
Open
01
Grease life, side by side
A public table of grease life results from bearing tests, starting with FE9 tests to DIN 51821-2, giving each grease's thickener, base oil, test conditions and result, and a model fitted to it that predicts how grease life changes with temperature, tested on greases it has not seen.
The briefCredit · unclaimed02
Measure the oil film cheaply
An open-hardware version of the ball-on-glass-disc instrument Gohar and Cameron introduced in 1963: a 3D-printed frame, a steel ball loaded against a glass disc turned by a small motor, a microscope camera, and software that turns the interference colors into film thickness, with full drawings and a parts list.
The briefCredit · unclaimed
Open 01 · the brief
Grease life, side by side
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of grease life results from bearing tests, starting with FE9 tests to DIN 51821-2, giving each grease's thickener, base oil, test conditions and result, and a model fitted to it that predicts how grease life changes with temperature, tested on greases it has not seen.
Why it is still open
Makers print one line per grease, as a lower bound and under differing test codes: Schaeffler's data sheets give "F50 ≥ 100 h" for Arcanol LOAD150 under a test code ending in 140 and "F50 ≥ 200 h" for Arcanol LOAD400 under one ending in 130. Kleinlein's 1992 FE9 study found that life rose steeply as temperature fell, so one figure at one temperature says little about life at a user's temperature. P. Lugt and Frank Berens proposed a "grease life factor" concept for ball bearings in Tribology International in 2022. The largest public comparison found, ExxonMobil's test of nine greases, identifies them only by number. Searches of the web, GitHub and Hugging Face in November 2026 found no open table of grease life results.
santent.inignera.ltosti.govapi.crossref.orglube-media.ambrit.com
Built with what exists
Grease data sheets and published papers; WebPlotDigitizer for results given only as charts; an AI document reader to pull the test lines out of hundreds of data sheets, every value checked by hand (the Note "Seven more flagged" covers an AI reviewer of long documents); Python with pandas and SciPy.
First step
Copy the FE9 lines and test codes from the data sheets of three grease and bearing makers into one spreadsheet.
Done looks like
FE9 or comparable results for at least 150 greases from at least five makers, each linked to its source, and a temperature model that predicts the measured life of greases left out of the fitting within a stated error, with every error published.
Open 02 · the brief
Measure the oil film cheaply
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open-hardware version of the ball-on-glass-disc instrument Gohar and Cameron introduced in 1963: a 3D-printed frame, a steel ball loaded against a glass disc turned by a small motor, a microscope camera, and software that turns the interference colors into film thickness, with full drawings and a parts list.
Why it is still open
The commercial instrument from PCS Instruments is a fully automated bench-top machine sold on quotation. Open-hardware tribometers exist, such as a pin-on-disc machine from the University of Bayreuth (2024) that can be 3D-printed whole and whose friction results were reproducible to 0.4% and 1.1% on two test systems, but they measure friction and wear, not the oil film. Searches of GitHub and the web in November 2026 found no open design for an optical film-thickness rig.
Built with what exists
A glass disc and a steel ball from a bearing; a stepper motor and an Arduino-class controller; a USB microscope camera and a white light; Python with OpenCV for the color analysis; the Bayreuth design as a starting frame.
First step
Press a steel ball against a glass flat under a USB microscope with white light, and check whether interference fringes show at the contact.
Done looks like
Central film thickness of one reference oil at five speeds and two loads, each within 15% of the same oil measured on a commercial instrument at a university lab, with drawings, code and parts list published.