A threadlocker’s data sheet number and its performance on your actual assembly are two different things, and the only way to close that gap is to test the joint the way it will really be used — not trust the label.
Why a Data Sheet Number Isn’t Enough
Most threadlocker comparisons stop at reading a strength rating off a data sheet, but strength and vibration resistance are measured differently and don’t move together in a predictable way. A compound can post an impressive static breakloose torque and still lose a meaningful share of its prevailing torque after a few hundred cycles of transverse vibration — the exact loading pattern that loosens fasteners in the field. Running your own comparison, even a small one, is the only way to know how a specific compound performs on your specific fastener, torque spec, and duty cycle.
Setting Up a Fair Test
A useful comparison holds every variable constant except the compound itself: identical fastener material and size, identical torque specification applied with a calibrated torque wrench, identical surface preparation, and identical cure time before testing begins. Testing with mismatched variables — one sample torqued slightly differently, one given extra cure time — produces a result that looks like a compound difference but is actually a setup difference, and that mistake is common enough to be worth double-checking before drawing any conclusion from the data.
The Transverse Vibration Test
The standard method for measuring fastener self-loosening under vibration mounts the test joint so that vibration is applied transverse to the fastener axis — perpendicular to the bolt, not along it — because this loading direction is what actually drives self-loosening in real equipment, unlike axial vibration which rarely loosens a properly torqued fastener at all. The fixture cycles the joint through a set number of transverse displacement cycles, typically in the thousands, while continuously logging clamp load or prevailing torque. A joint that holds its clamp load through the full cycle count passes; one that shows progressive loss, even if it doesn’t fully back off, is telling you something the static strength number never could.
Reading Prevailing Torque Before and After
Prevailing torque — the torque required to continue turning a fastener that’s already engaged — is measured before the vibration test to establish a baseline and again afterward to quantify loss. A compound that retains 80–90% of its original prevailing torque after the full vibration cycle is performing well; one that drops to 30–40% has effectively failed even if the fastener technically never came loose during the test window, since that same degradation would likely continue with more cycles or with the added stress of thermal cycling in real service.
Sample Size and Statistical Confidence
A single test joint tells you almost nothing reliable — normal manufacturing variation in fastener finish, hole tolerance, and surface prep can produce a wide spread of results even with one compound tested twice. Running a minimum of five to ten samples per compound and per condition, then comparing the distribution rather than a single data point, is what separates a genuine comparison from an anecdote. Email Us if you want help designing a sample size and test matrix appropriate to your specific fastener and duty cycle.
Building in the Real Operating Environment
A vibration test run at room temperature tells you less than one that also incorporates the thermal range the joint will actually see in service, since thermal cycling changes both the polymer’s mechanical properties and the differential expansion at the fastener interface. Where possible, running the vibration protocol inside an environmental chamber that cycles through the equipment’s real temperature range gives a result that’s far more predictive of field performance than a single ambient-temperature pass.
Turning Test Data Into a Specification Decision
Once test data exists for two or three candidate compounds under identical conditions, the decision stops being about which data sheet reads more impressively and becomes about which compound actually held clamp load on your fastener, at your torque spec, under your vibration profile. That data is also worth keeping on file — the same protocol becomes the reference point the next time equipment duty cycles change or a new supplier is evaluated, rather than starting the comparison from scratch.
Incure’s applications engineers run this same kind of transverse-vibration comparison when helping manufacturers benchmark a fastening spec, and the same rigor — testing the actual failure mechanism rather than trusting a headline number — applies just as much to bonded joints; see how CTE mismatch drives adhesive bond failure for the equivalent failure mechanism in adhesive assemblies, and which bonding approach delivers higher joint strength for cases where a structural adhesive bond might outperform a threaded fastener entirely.
Testing before specifying costs a modest amount of setup time; discovering a mismatch after a product ships costs considerably more. Contact Our Team to talk through a vibration-testing protocol appropriate to your equipment and duty cycle.
Visit www.incurelab.com for more information.