Skip to content
STEELTOETOOLS

Bolt Torque Calculator — Torque, Preload and the K Factor

Torque is only a way of reaching a clamp load, and the relation between the two is T = K x D x F: the nut factor, the nominal bolt diameter and the preload you are aiming for. This calculator converts either way. It asks you for K and for the target preload because both belong to your joint specification — there is no torque table and no assumed nut factor on this page.

in
The nominal size of the fastener — 1/2 in, M12. Not the thread pitch diameter and not the hole size.
Dimensionless. It comes from your joint specification, your fastener supplier's data for that exact finish and lubricant, or a test on your own assembly. This page deliberately offers no value — a nut factor guessed from the internet is the usual reason a torque number is wrong.
lb
The clamp load the joint is designed to have. It comes from the engineer, the assembly drawing or the equipment manual — never from a general table.

Results are estimates for planning purposes and do not replace a competent person's evaluation or an engineer's design. Follow the OSHA standards that apply to your work and your employer's written program.

Why this page gives you no torque value and no K

A deliberate choice, not an omission

Most torque calculators open with a nut factor already filled in and a table of torques by grade underneath. Both are dangerous defaults. The tables are copyrighted, and worse, every one of them is valid only for the friction conditions it was computed with: change the plating, the lubricant, the washer or whether the nut or the head is turned, and the same torque produces a different clamp load in the same bolt.

So this page asks you for K and for the target preload, and it will sit there showing nothing until you supply them. They belong to your joint specification, your fastener supplier's data for that exact finish and lubricant, or a test on your own assembly in a tension calibrator. A number that looks authoritative and does not match your joint is worse than no number at all.

What is actually inside K

The nut factor is not a property of the bolt. It is a single number standing in for everything that resists turning, and NASA's Fastener Design Manual opens its torque section by calling the determination of proper torque the biggest problem in fastener installation. The variables it lists first are:

  • Friction between the mating threads, and separately between the turning head or nut and its bearing surface — where most of the applied torque is actually consumed.
  • Coatings and lubricants, which move both of those. The same manual notes the coefficient of friction can run anywhere from 0.04 to 1.10 depending on the materials and lubricants in contact — a range of more than twenty-five to one.
  • What fraction of the fastener's strength the design wants as preload, and how to hit that value reliably.
  • The relative stiffness of the bolt and the joint, and whether the running torque of a locking feature has to be added on top.

FHWA's structural fastener guidance shows how concrete that is in practice: from experience, wax-based lubricants give lower torque values than oils and greases for the same result. Which is why a torque figure is only valid together with the lubrication condition it was derived for.

How close any of this gets you to the real clamp load

The Industrial Fasteners Institute's comparison of preload measuring methods, reproduced in NASA RP-1228, is the most useful thing on this page. It puts numbers on what everyone suspects: torque is a cheap and inaccurate way to reach a tension.

Accuracy and relative cost of preload measuring methods — Industrial Fasteners Institute, as published in NASA RP-1228 Table VII
MethodAccuracy on preloadRelative costWhat it is
Feel — the operator's judgment±35 %1A wrench and experience. Still how a great many fasteners are actually tightened.
Torque wrench±25 %1.5What most people mean by "torqued to spec". A quarter either way on the clamp load.
Turn of the nut±15 %3Rotation measured from snug tight, which steps around the friction problem instead of trying to model it.
Load-indicating washers±10 %7The washer deforms at a known tension and is checked with a feeler gauge.
Fastener elongation±3 to 5 %15Measuring how much the bolt actually stretched — the tension itself, not a proxy for it.
Strain gages±1 %20Instrumented fasteners. Reserved for the joints that justify the cost.

Read the first two rows together: a calibrated torque wrench buys you ten points of accuracy over a good mechanic's feel, and still leaves a quarter of the clamp load in doubt. That gap is the whole reason structural and pressure work uses other methods.

The other ways a joint gets tightened

FHWA's bolting procedures name four installation methods besides plain torque, each with its own calibration and verification testing:

  • Turn-of-nut. Snug the joint, mark it, then turn a specified fraction of a rotation. It measures rotation rather than torque, so most of the friction uncertainty drops out.
  • Calibrated wrench. A torque method, but with the wrench calibrated daily against actual fastener assemblies from the lot being installed rather than against a table.
  • Alternate design (twist-off) fasteners. A splined end shears off at a tension built into the fastener.
  • Direct tension indicators. A washer with formed protrusions that flatten to a measurable gap at the required tension.

On a structural connection the specification says which method applies, and the verification testing that goes with it is part of the method — not an optional extra.

Before you pull the wrench

  • More torque is not a safer joint. Past yield the fastener stretches permanently, then relaxes and loses the clamp load you were trying to create.
  • Specify the lubricant with the torque. Dry, oiled and waxed are three different jobs at the same wrench setting.
  • Watch what the torque is being reacted against — a gasket that keeps creeping, a soft flange, threads in aluminium, a nut bottoming out on the thread run-out.
  • Reused fasteners are a specification question, not a judgment call. Some are single-use by rule; all of them have different surfaces the second time around.
  • Tighten in the pattern and the passes the manual gives, especially on a flange. The last bolt changes the load in the first one.

How it's calculated

Torque from preload
T = K × D × F
With D in inches and F in pounds the result is pound-inches; divide by 12 for pound-feet. In metric, N·m = K × D (mm) × F (kN), because the millimeter and the kilonewton cancel to the newton-meter exactly.
Preload from torque
F = T / (K × D)
The same relation read backwards. It tells you what clamp load a torque value is supposed to produce — supposed to, because everything uncertain in the joint sits inside K.
What K contains
K = f(thread friction, bearing-face friction, thread geometry)
It is not a property of the bolt. NASA RP-1228 lists the friction between mating threads, the friction between the bolt head or nut and its bearing surface, and the effect of coatings and lubricants as the first three of eight variables that make fastener torque the hardest part of installation — and notes that the coefficient of friction alone can vary from 0.04 to 1.10 depending on materials and lubricants.

Frequently asked questions

T = K × D × F — the nut factor, the nominal bolt diameter and the preload you want. With D in inches and F in pounds you get pound-inches, so divide by 12 for pound-feet. In metric it is even simpler: newton-meters = K × diameter in millimeters × preload in kilonewtons.

The one that belongs to your joint, and this page will not offer you another. K is not a property of the bolt: it wraps up the friction in the threads, the friction under the turning head or nut, the plating, the lubricant and how many times that fastener has been run down before. NASA RP-1228 notes that the friction coefficient alone ranges from 0.04 to 1.10 depending on materials and lubricants. Take K from the assembly specification, from your fastener supplier's data for that exact finish and lubricant, or from a test on your own joint in a tension calibrator.

Two reasons, and both matter. The published tables are copyrighted, and more importantly a torque table is only valid for the friction conditions it was calculated with — change the plating, the lubricant or the washer and the same torque produces a different clamp load. A number that looks official and does not match your joint is worse than no number at all.

As a way of hitting a clamp load, not very. The table published by the Industrial Fasteners Institute and reproduced in NASA RP-1228 puts a torque wrench at about ±25 % on preload, against ±35 % for a mechanic's feel, ±15 % for turn-of-nut, ±10 % for load-indicating washers, ±3 to 5 % for measuring the bolt's elongation and ±1 % for strain gauges. The cost climbs the same way. That spread is why critical joints are not controlled by torque alone.

Substantially, and it is one of the most common reasons a joint ends up under- or over-tightened. FHWA notes from experience that wax-based lubricants give lower torque values than oils and greases for the same result. A torque figure is only valid for the lubrication condition it was derived with, so specify the lubricant along with the torque, and re-lubricate as the specification says rather than as the weather suggests.

FHWA's structural bolting procedures name turn-of-nut, calibrated wrench, alternate design (twist-off) fasteners and direct tension indicators, each with its own calibration and verification testing. Turn-of-nut works past the friction problem by measuring rotation after snug tight rather than torque; direct tension indicators and instrumented bolts measure the tension itself. On a structural connection, the specification says which method applies — it is not the installer's choice.

Not without checking the specification. Running a fastener down changes its surfaces and often its lubricant, so the same torque produces a different preload the second time, and a fastener tightened into the plastic range has stretched permanently. Some structural and high-performance fasteners are single-use by rule. When the specification says replace, replace.

No. Over-torquing yields the fastener, which then relaxes and loses the clamp load you were trying to create, and it can crush gaskets, distort flanges and strip threads in softer material. The target is the preload the designer chose. Torque is only the crude instrument used to reach it.

Sources & references

Content checked against these sources — last reviewed September 1, 2026.

By — Editorial team of SteelToeTools.com (published by LSEA SAS) Updated v1