The short answer: tightening torque is an indirect measure of bolt preload, and friction decides most of it. Titanium's high sliding friction and tendency to gall make torque-only installation less predictable than it is on steel, so a torque value copied from a steel chart can leave a titanium joint under-clamped, seized, or both. The fix is procedural: define the lubricant or surface treatment on the order, develop the torque value from tests on the actual fastener pair, and verify critical joints with a method that measures preload more directly than torque.
Choosing the material first? See Choosing Grade 2 or Grade 5 Titanium Fasteners for strength and corrosion selection. This article assumes the grade is settled and deals with what happens on the wrench.
Torque measures friction as much as preload
The familiar relationship T = K·F·d connects applied torque to preload through a nut factor K that bundles thread friction, bearing friction and geometry. Most of the applied torque is consumed overcoming friction at the threads and under the turning face; only a small remainder stretches the bolt. K is therefore not a material constant. NASA's Fastener Design Manual records that the coefficient of friction in a threaded assembly "can vary from 0.04 to 1.10, depending on the materials and the lubricants," notes that 0.2 is the commonly assumed value of K, and suggests 0.15 as a more realistic typical value for steel on steel — none of which transfers to titanium.[1]
NASA-STD-5020B shows what a conservative programme does when no test data exist: under torque control, and in the absence of applicable data showing otherwise, the preload uncertainty factor is set at 25 percent for fasteners lubricated at assembly and 35 percent for non-lubricated, as-received fasteners.[2] Read that again as a buyer: a torque-only installation of a titanium fastener in an undefined surface condition should be treated as accurate to roughly a third — before galling is even considered.
Galling is the failure mode a torque wrench cannot see
Titanium combines a high coefficient of sliding friction, low thermal conductivity and high surface reactivity. Under pressure, mating titanium surfaces transfer material and cold-weld. NASA's fastening study on titanium states it directly: "Titanium surfaces should never be used in sliding applications unless a lubricant is present" — and a tightening thread is a sliding application.[3] TIMET's design handbook lists the tendency toward galling among the properties a titanium design must accommodate, not an occasional defect.[4]
The field sequence is consistent: friction climbs as the thread flanks begin to weld, the wrench reaches its target torque early, and the joint ends with far less preload than the reading implies. On disassembly, the damaged threads can lock solid. This is a surface-system problem, not a workmanship excuse — the same installer, tool and torque value produce different results on different finishes and lubricants.
A titanium nut on a titanium bolt is the hardest case
Galling between titanium fasteners and titanium nuts is the documented reason steel and aluminium nuts became the general practice for titanium fasteners.[3] The pairing decision deserves the same care as the grade decision:
- Dissimilar nut material (for example a corrosion-resistant steel or A286 nut) breaks the titanium-on-titanium adhesive pair. Check strength matching, temperature limits and galvanic compatibility with the structure.
- Silver-plated nuts are a long-standing aerospace anti-galling practice under AMS 2410.[5] The same NASA study records why the practice became restricted with titanium: silver accelerates the salt corrosion of titanium. Dry, controlled service is one thing; marine hardware is another.[3]
- Filmed or lubricated titanium nuts keep a matched-material joint possible when dissimilar nuts are unacceptable, at the cost of a controlled coating process.
The lubricants that are actually specified
"Use anti-seize" is not a specification. These are the treatments procurement language can name, each with a standard behind it:
| Treatment | Governing specification | Where it fits | Watch out for |
|---|---|---|---|
| Molybdenum-disulfide anti-seize paste | MIL-PRF-83483[6] | Threaded fasteners and fittings below 427 °C (800 °F) | Field-applied; film thickness varies between installers, so the nut factor varies with it |
| Heat-cured MoS₂ solid film | SAE AS5272 / MIL-PRF-46010[7] | Factory-applied, repeatable friction on production parts | A supplier process capability, not an assembly-line step; NASA notes sulfide lubricants are unsuitable for vacuum or liquid-oxygen service[3] |
| Cetyl alcohol | SAE AS87132[8] | The standard aerospace lubricant for fastener installation | Essentially a single-installation film; define the reuse policy around it |
| Silver plating on the nut | AMS 2410[5] | High-temperature aerospace joints with dissimilar nuts | The titanium–silver salt-corrosion restriction above |
Whichever system is chosen, it defines K. A torque value developed for a filmed nut is wrong for a dry one, and a value measured on new threads does not transfer to threads on their fifth reuse.
How much preload scatter should you expect?
The Fastener Design Manual, citing Industrial Fasteners Institute data, puts typical preload accuracy by installation method as follows:[1]
| Installation method | Typical preload accuracy |
|---|---|
| Operator feel (uncontrolled) | ±35% |
| Torque wrench | ±25% |
| Turn of the nut | ±15% |
| Fastener elongation measurement | ±3 to 5% |
Two conclusions follow. Torque control is a mid-precision method even when it is done well, and when preload genuinely matters — fatigue-loaded joints, gasketed closures, clamping of composites — the procedure should verify preload by elongation or angle rather than by torque alone.
Developing a defensible installation procedure
- Fix the surface system first. Finish, lubricant, nut material and plating belong in the specification, not on the bench.
- Measure the nut factor on the actual fastener pair in the actual condition — new, and after the intended number of reuses. Frameworks such as VDI 2230-1 structure the calculation well, but the standard's own scope is steel property classes 8.8 to 12.9 and comparable stainless grades; titanium coefficients must be measured and then inserted into the framework.[9]
- Control the installation: clean, undamaged threads; low and steady tightening speed; no impact tools; stop if torque climbs abnormally — a partially seized thread does not recover under force.
- Define reuse. Adhesive transfer starts before it is visible, so "looks fine" is not an acceptance criterion for a critical joint.
What to put on the RFQ
- Fastener standard and material grade — for inch-series nonferrous bolts and cap screws, ASTM F468 where its scope applies[10]
- Surface treatment or lubricant, with its specification: AS5272 film, AS87132 cetyl alcohol, MIL-PRF-83483 paste, or AMS 2410 plating on the nut
- Nut material and finish, stated explicitly
- Who develops the installation torque, and on what test basis
- Tightening method, speed limits and tool-calibration requirement
- Reuse policy and the inspection criterion for threads
- Environment: vacuum or oxygen service, salt exposure and temperature — MoS₂ and silver each carry restrictions
Our titanium fastener range ships in plain natural finish as standard; filmed, lubricated or plated supply to a named specification is available on order. Send the drawing and service conditions and we will confirm the surface system before quotation.
Still selecting a grade? Start with Titanium Fasteners: Grades, Benefits and Applications, then go deeper on the two common grades with Choosing Grade 2 or Grade 5 Titanium Fasteners.
References
- NASA RP-1228 — R. T. Barrett, Fastener Design Manual (1990).
- NASA-STD-5020B — Requirements for Threaded Fastening Systems in Spaceflight Hardware (2021).
- NASA TM X-53442 — Mechanical Fastening of Titanium and Its Alloys (Battelle Memorial Institute, 1966).
- TIMET — Titanium Design and Fabrication Handbook.
- SAE AMS 2410 — Silver Plating, Nickel Strike, High Bake.
- MIL-PRF-83483 — Thread Compound, Antiseize, Molybdenum Disulfide-Petrolatum (search the spec number in the DoD ASSIST Quick Search portal).
- SAE AS5272 — Lubricant, Solid Film, Heat Cured, Corrosion Inhibiting (supersedes MIL-PRF-46010).
- SAE AS87132 — Lubricant, Cetyl Alcohol, 1-Hexadecanol, Application to Fasteners.
- VDI 2230 Blatt 1:2015-11 — Systematic calculation of highly stressed bolted joints, Part 1.
- ASTM International — ASTM F468-23, Nonferrous Bolts, Hex Cap Screws, Socket Head Cap Screws, and Studs for General Use.
From the Catalog
Products covered in this guide
View products
Grade 5 Ti-6Al-4V Titanium Hex Bolts UNC UNF ASME B18.2.1 Plain Natural Finish High Strength Lightweight Fasteners Aerospace
View Product
Hex Socket Head Cap Screws UNC UNF Grade 1 Grade 2 Grade 3 Grade 4 CP Titanium ASME B18.3 Plain Natural Finish Fully Threaded
View Product
Titanium Hex Bolts Grade 1 Grade 2 Grade 3 Grade 4 CP Titanium UNC UNF Thread ASME B18.2.1 Plain Finish Lightweight Fasteners
View Product