Bolt Torque Calculator

Tightening torque for metric bolts to ISO 898-1 property classes, calculated from proof load and the nut factor for your coating and lubrication. Made by a bolt manufacturer, so it also tells you where the number stops being reliable.

Figures are for a bolt tightened in tension against a plain washer, threads and bearing face in the stated condition. Nut and bearing surfaces are assumed clean and undamaged.

How tightening torque is calculated

A bolt does its work by being stretched. The clamping force it holds is called preload, and torque is only the means of producing it. The short-form relationship used throughout industry is:

T = K × F × d

• T — tightening torque, N·m
• K — the nut factor, an empirical number covering thread friction, bearing-face friction and thread geometry
• F — target preload, N
• d — nominal bolt diameter, m

Preload itself is set as a percentage of the bolt’s proof load, which comes from the tensile stress area and the proof stress of the property class in ISO 898-1. Most general engineering practice targets 65–75%.

Worked example: M20, property class 8.8, plain finish

• Tensile stress area A_s = 245 mm²
• Proof stress for class 8.8 above M16 = 600 N/mm²
• Proof load = 245 × 600 = 147,000 N = 147 kN
• Target preload at 70% = 103 kN
• T = 0.20 × 103,000 N × 0.020 m = 412 N·m (about 42 kgf·m)

Class 8.8 uses a proof stress of 580 N/mm² up to and including M16, and 600 N/mm² above M16. The calculator applies the correct value automatically.

Nut factor K by finish and lubrication

K is not a property of the bolt. It is a measured average for a particular combination of coating, lubricant and surface condition, and it is the single largest source of error in torque control. Published ranges in common use:

The galvanised rows are worth reading twice. The same bolt at the same torque, hot-dip galvanised and dry versus galvanised with a waxed nut, can differ in achieved preload by close to a factor of two. This is why HSFG assemblies are supplied with lubricated nuts and why the bolt, nut and washer are treated as a matched set rather than three separate items.

Bolt torque chart: M6 to M64

Tightening torque in N·m, calculated at K = 0.20 and 70% of proof load, coarse pitch threads to ISO 898-1. Use it as a starting point and adjust with the calculator above for your actual finish.

To convert: 1 N·m = 0.102 kgf·m = 0.738 lbf·ft.

Why torque is only an indirect measure of preload

Of the torque applied to a fastener, roughly half is consumed by friction under the nut bearing face and roughly forty per cent by friction in the threads. Only about ten per cent goes into actually stretching the bolt. You are controlling friction and hoping preload follows.

Torque control alone typically achieves preload within about ±25–30% of target. That scatter is not a defect in the calculation. It is what happens when the measured quantity is separated from the wanted quantity by two friction interfaces. The things that move it:

Thread manufacturing method

Rolled threads are formed by displacing material rather than cutting it. The flanks come out burnished and work-hardened, with a consistent surface finish, so thread friction is repeatable from piece to piece. Cut threads leave a torn surface with a rougher and more variable finish, and the friction goes with it. Two bolts to the same drawing and the same property class can behave differently on the wrench purely because of how the thread was produced. At S R Forgings threads are rolled on hydraulic thread rolling machines up to ø100 mm for this reason, and because a rolled thread is stronger in fatigue.

Everything else that shifts the number

• Coating type and thickness, and whether the nut carries a lubricant
• Condition of the nut face, washer and joint surface — burrs, rust, paint overspray, weld spatter, site dirt
• Reuse of a bolt or nut; galling on the second tightening, common with stainless and with damaged galvanised threads
• Impact wrench versus a calibrated torque wrench, and the speed of tightening
• Grip length, number of plies, and embedment relaxation in the first hours after tightening
• Thread damage in transit, storage or handling on site
• Temperature, and whether the assembly is being tightened wet

None of this makes torque useless. It makes it a control method with a known tolerance, which is fine for most general engineering joints and not fine for a joint whose design depends on a specific clamping force.

Structural bolting: what the codes actually require

For preloaded structural connections, the codes do not ask you to rely on a torque figure. IS 4000, EN 1090-2 and ASTM F3125 all specify methods that either measure or bypass the friction problem:

• Turn-of-nut — bring the joint to snug tight, mark the nut, then apply a specified further rotation. Because bolt extension in that rotation is geometric, friction largely drops out.
• Calibrated wrench — the wrench setting is established each shift on a tension calibrator using bolts from the same lot, in the same condition, rather than taken from a table.
• Direct tension indicators — a washer with formed protrusions that flatten to a measurable gap at the specified tension. This measures bolt tension itself instead of inferring it.
• Tension control bolts — a splined end that shears off at a preset torque.

We manufacture DTI washers to EN 14399-9, which is the standard called up by RDSO for Indian Railways work, along with HSFG bolt assemblies and high tensile nuts. If your specification names a preload rather than a torque, a DTI is usually the simplest way to demonstrate you achieved it.

Common questions

What torque should I use for an M20 8.8 bolt?

About 412 N·m, or 42 kgf·m, for a plain finish bolt tightened to 70% of proof load with a nut factor of 0.20. If the same bolt is hot-dip galvanised and dry, the torque needed for the same preload rises to roughly 620 N·m; if it is galvanised with a waxed nut it falls back to about 370 N·m. The finish matters more than most people expect.

Why does the same bolt need a different torque after galvanising?

Hot-dip galvanising leaves a rough zinc-iron alloy surface on the threads with high and inconsistent friction. More of the applied torque is lost overcoming that friction, so more torque is required to reach the same clamping force, and the spread between one bolt and the next widens. This is why galvanised HSFG assemblies are supplied with lubricated nuts.

Should the torque be applied to the nut or the bolt head?

Turn the nut wherever possible. The published nut factors assume the nut is the rotated element against a washer. Turning the head instead changes the bearing geometry and the friction radius, and generally needs a higher torque for the same preload.

Can I reuse a bolt and apply the same torque?

Not reliably. Thread surfaces change on first tightening, coatings and lubricants are damaged, and galling may have started. Preloaded structural bolts are generally not reused at all. For general engineering, treat a reused fastener as having unknown friction.

Does the calculator work for fine pitch threads or UNC and UNF?

Select “Other size / fine pitch” and enter the nominal diameter and the tensile stress area for your thread. The calculation is the same; only the stress area changes.

Related: Bolt Weight Calculator · HSFG Bolt Standards Guide · Bolt Material Grades