Industries / Industrial equipment and services / Industrial machinery / Bearings
Counterfeits
A counterfeit bearing is one sold under a brand that did not make it, usually a cheap unbranded bearing marked and boxed to look like a premium maker's product.
6
Milestones achieved, from 2010 to 2025.
1
To watch for: announced, not yet confirmed outside.
3
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2010
Makers campaign together
The World Bearing Association (WBA), a non-profit group founded in 2006 by the American, European and Japanese bearing makers' associations (ABMA, FEBMA and JBIA), launched its "Stop Fake Bearings" campaign in late 2010, with a website, emails and social media planned for three years. Its anti-counterfeiting committee also works with police and customs to prosecute offenders. NSK, announcing its part in December 2010, said that spotting a counterfeit usually takes a trained expert.
Credit
the World Bearing Association; WBA president James W. Griffith
2012
A fake that passed inspection
In July 2012 the Canada Border Services Agency stopped a pallet of bearings from China at Toronto's Pearson airport, marked as Timken. The boxes lacked Timken's hologram and had the wrong barcode, but the bearings matched Timken's specifications for size, surface hardness and weight. Scoring marks on a raceway sent them to Timken's metallurgical laboratory in Canton, Ohio, where cutting them open showed they were through-hardened, hard all the way through, and not case-hardened, hard only in a surface layer as Timken makes them, and made of the wrong materials. Timken said it could not trace the maker.
LessonChecks made from the outside passed the fake; only cutting a bearing open showed the difference.
Credit
Canada Border Services Agency; Timken (Evan Boere, Daniel Szoch), as reported by MRO Magazine in April 2013
2017
Eight years to crush fakes
In 2009 a raid on the Piraeus warehouse of an unauthorized dealer in Greece seized 17,000 counterfeit bearings marked as SKF, weighing 15 tonnes and with a market value of over €1 million. Once Greek courts had ruled them counterfeit, they were crushed in a scrap press at a metal recycler in early 2017. SKF reported one customer who spent €100,000 on large bearings from an unauthorized supplier that failed after a few months, against an expected life of several years.
LessonA seizure is the start of a case, not the end; the bearings were held for years while courts decided.
Credit
SKF Hellas and SKF Group Brand Protection; the Greek courts
2020
A phone checks the box
NSK began printing a symbol on the 2D barcode of each packing label of its Japanese-made bearings in 2018, and in August 2020 extended its free NSK Verify app from machine-tool bearings to its general and very large industrial bearings. The app checks the scanned code against NSK's database and shows how many times that product's identification number has been looked up, so a code copied onto many boxes stands out. It checks the label, not the bearing, and older stock without the symbol cannot be checked.
Credit
NSK
2023
One app, seven makers
The WBA Bearing Authenticator 2.0, built by the German firm oneIDentity+ for the WBA and released in February 2023, checks the code on the packaging of bearings from JTEKT (Koyo), NACHI, NTN, NSK, Schaeffler, SKF and Timken. The WBA's campaign site says it gives one of three answers: the code is in the maker's database; the code is known but has been scanned many times; or the code is unknown and the bearing is a suspected fake. In November 2026 the US App Store listing showed nine ratings averaging 2.2 out of 5, and no accuracy figures were found.
Credit
World Bearing Association; oneIDentity+ GmbH
2025
Marking machines seized
Raids in Delhi and Bengaluru, India, led by police on complaints from National Engineering Industries (NBC Bearings), seized fake NBC bearings worth over INR 2 crore (20 million rupees), with printing dies, empty boxes and a laser-marking machine "used to falsify product codes", the company reported in December 2025. A month earlier Schaeffler India's brand protection team and local police raided five sites in Delhi and seized thousands of fake FAG wheel and ball bearings with fake packaging.
LessonCounterfeiters now mark bearings and print codes with the same kind of machines the makers use, so a mark or a label alone proves little.
Credit
Delhi Police (DIU North District) and Bengaluru enforcement teams; NBC Bearings; Schaeffler India
Watch for
Security labels on auto parts
JTEKT, maker of Koyo bearings, says it is extending the KURZ SCRIBOS anti-counterfeiting technology it has used on industrial bearings for a decade to its wider automotive aftermarket catalog, through a TRUSTSEAL ID system that lets buyers and mechanics check a part digitally when it arrives. JTEKT's own app reads a QR code on a hologram label.
Achieved whena customs agency, distributor or testing laboratory outside JTEKT and SCRIBOS reports how many fakes the labels caught, or that copied labels were rejected.
Claimed by
JTEKT, September 2026
Open
01
Fakes run to failure
A bench endurance test that runs confirmed counterfeit and genuine bearings of one common size, for example the 6205 deep-groove ball bearing, side by side under the same load and speed until a fatigue spall shows in the vibration signal or a fixed time runs out, then cuts a sample of each open to compare hardness depth and steel cleanliness.
The briefCredit · unclaimed02
A tap test for fakes
A way to tell a counterfeit bearing ring from a genuine one without cutting it, by tapping it and recording how it rings. A ring's natural frequencies and how fast its ringing dies away depend on its shape, steel and heat treatment, so a through-hardened fake should sound different from a case-hardened original.
The briefCredit · unclaimed03
What the code apps catch
A public tally, made with a maker's brand protection team, of how many seized counterfeit packages the WBA Bearing Authenticator and the maker's own app flag as fake, how many they pass, and how many cannot be read, including boxes carrying codes copied from genuine stock.
The briefCredit · unclaimed
Open 01 · the brief
Fakes run to failure
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A bench endurance test that runs confirmed counterfeit and genuine bearings of one common size, for example the 6205 deep-groove ball bearing, side by side under the same load and speed until a fatigue spall shows in the vibration signal or a fixed time runs out, then cuts a sample of each open to compare hardness depth and steel cleanliness.
Why it is still open
Makers' claims rest on single cases and on general points about steel. NSK's 2020 white paper says steel quality can change bearing life "up to 20 times" but gives no test of counterfeits; SKF reports one customer's bearings failing in months; Timken's 2012 analysis found the wrong hardening but did not run the bearings. Searches of Crossref, Europe PMC, GitHub, Hugging Face and the web in November 2026 found no published endurance test of counterfeit against genuine bearings.
Built with what exists
Confirmed counterfeits from a maker's brand protection team or from seized stock awaiting destruction, and genuine bearings bought from authorized distributors; a guarded bench rig with an electric motor, a shaft on two support bearings and a lever or spring loading the test bearing; low-cost MEMS accelerometers and thermocouples logged by a Raspberry Pi; Python with SciPy for vibration analysis and the open-source reliability package for Python for Weibull fits; a commercial metallurgy lab for sectioning and hardness profiles.
First step
Ask the brand protection teams of two WBA members whether they will release 20 seized bearings of one size for testing, and price a two-station rig.
Done looks like
At least 20 counterfeit and 20 genuine bearings run under the same conditions, with every bearing's hours to failure or to cut-off published, Weibull plots for both groups, and hardness-depth profiles for at least five of each.
Open 02 · the brief
A tap test for fakes
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A way to tell a counterfeit bearing ring from a genuine one without cutting it, by tapping it and recording how it rings. A ring's natural frequencies and how fast its ringing dies away depend on its shape, steel and heat treatment, so a through-hardened fake should sound different from a case-hardened original.
Why it is still open
Resonance testing is already sold to bearing factories: The Modal Shop says its resonant acoustic system checks inner and outer races before assembly and lists "through hardened vs case hardened" among the faults it catches, and Foerster sells eddy-current testing that flags hardness deviations in bearing balls. These are production-line tools set up for one maker's own parts. The Timken case of 2012 was settled only by cutting the bearings open, and no published test of resonance or eddy-current screening on counterfeit bearings was found in November 2026.
Built with what exists
A small impact hammer or a steel ball dropped from a fixed height; a measurement microphone or a piezo contact sensor and a USB audio interface; a soft foam support; Python with SciPy to find resonance peaks and damping; scikit-learn for a simple classifier; sectioning of a sample by a metallurgy lab to confirm which rings are case-hardened.
First step
Record the ringing of ten genuine 6205 outer rings of one brand and ten of an unbranded make, and see whether their main resonance frequencies separate at all.
Done looks like
Rings from at least 40 genuine and 40 confirmed counterfeit bearings of one size, tested blind by someone who does not know which is which, with accuracy, false alarms and missed fakes published with their confidence intervals, and the raw recordings released.
Open 03 · the brief
What the code apps catch
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public tally, made with a maker's brand protection team, of how many seized counterfeit packages the WBA Bearing Authenticator and the maker's own app flag as fake, how many they pass, and how many cannot be read, including boxes carrying codes copied from genuine stock.
Why it is still open
The apps check the code printed on a label against the maker's database. The NSK and WBA apps warn when one code has been scanned many times, to expose copies, and the December 2025 Indian raids found a laser-marking machine used to falsify product codes. None of the apps publishes a detection rate, and the WBA's own site advises buyers not to try to identify fakes on their own.
Built with what exists
Seized packages held by brand protection teams awaiting destruction; two or three phones on iOS and Android; the free WBA, NSK Verify and Schaeffler OriginCheck apps; a spreadsheet recording each scan's result, time and place.
First step
Write a one-page test plan and send it to the World Bearing Association through its Stop Fake Bearings site, asking which member will supply seized stock.
Done looks like
At least 300 seized packages from at least two brands and two countries scanned with every relevant app, with the share flagged, passed and unreadable published for each app, and every package photographed.