Metric Bolt Torque & Preload Calculator
Tightening torque for M3–M30 in classes 8.8, 10.9 and 12.9 — computed from the real chain (stress area → proof load → preload → torque) with every assumption on show and adjustable, because the K friction factor moves the answer more than anything else. Reverse mode: a torque you have been given → the preload it produces and how close to proof that is.
The single biggest uncertainty — see the note below.
75% is the common conservative default; many published charts use 85%.
Where a torque figure actually comes from
Every bolt torque chart is three steps stacked together. First the tensile stress area of the thread, As = (π/4) × (d − 0.9382·P)² — for M8×1.25 that is 36.6 mm². Then the proof load: stress area times the proof stress of the property class (580 MPa for 8.8 up to M16, 600 above; 830 for 10.9; 970 for 12.9) — 21.2 kN for that M8. Then a chosen preload, typically 75–85% of proof, turned into torque by T = K × F × d. Same three lines, every chart, every supplier.
The number doing the heavy lifting is K. It bundles all the friction — under the head and in the threads — and typical values run from 0.10 well-lubricated to 0.30 for rough dry surfaces. Since torque is proportional to K, the same 25 Nm on an M8 can produce anywhere from about 10 kN to 31 kN of clamp depending on lubrication — a factor of three. That is why two published charts rarely agree, why “dry” and “oiled” columns differ so much, and why critical joints are specified by preload (torque-and-angle, stretch measurement) rather than torque alone.
Worth knowing when checking someone else’s chart: at K = 0.20 and 85% of proof, this calculator lands within a digit of the widely circulated tables (M8 8.8 ≈ 29 Nm, M10 ≈ 57, M12 ≈ 100, M16 ≈ 248, M20 ≈ 500). Fastenal’s published methodology uses 75% and K 0.20 dry / 0.15 lubricated, which is the conservative default here. Neither is “right” — they are different assumptions, and the point of showing the chain is that you can match whichever your drawing or customer specifies.
Two cautions from the machining side. Tapped holes in soft materials (aluminium, brass) usually fail by thread stripping long before the bolt reaches proof — engagement length, not bolt strength, becomes the limit. And a 12.9 bolt in a joint designed around 8.8 torque figures is not an upgrade; it is an overloaded thread in the parent material. If a joint matters, tell us what it clamps — thread engagement is a design conversation, not a torque number.
Bolt torque — FAQ
What is the tightening torque for an M8, M10 or M12 bolt?
Class 8.8 dry (K = 0.20) at 85% of proof load: M8 about 29 Nm, M10 about 57 Nm, M12 about 100 Nm. At the more conservative 75% preload: roughly 25, 50 and 88 Nm. Lubrication reduces the required torque substantially — the calculator shows the chain so you can match your assumption.
How do you calculate bolt torque?
T = K x F x d: friction factor times preload times nominal diameter. Preload comes from the stress area As = (pi/4)(d - 0.9382P)^2 times the proof stress of the class (580/600 MPa for 8.8, 830 for 10.9, 970 for 12.9) times a utilisation of typically 75-85%.
What K factor should I use?
About 0.20 for dry steel, 0.15 lightly oiled, 0.10-0.13 for well-lubricated or plated-and-waxed fasteners. K is the dominant uncertainty in the whole calculation — for critical joints it should come from the fastener supplier or testing, not a generic chart.
What preload should a bolt have?
Common practice is 75% of proof load for reused or conservative joints and up to about 85-90% for permanent high-performance joints. Above 90% you are relying on very good control of friction and technique — torque alone does not deliver that accuracy.
Why do published torque charts disagree?
Different K factors and different preload percentages. A chart built on K 0.15 and 85% proof gives very different numbers from one at K 0.2 and 75% — both are internally consistent. Always check the assumptions printed at the bottom of the chart.
Does a stronger bolt class mean I should torque it higher?
Only if the joint — especially a tapped hole in softer parent material — can take it. In aluminium the thread usually strips before a 12.9 bolt yields, so engagement length and parent material set the real limit, not the bolt.
Send Your Drawing for a Same-Day Quote
Talk directly to the engineers who will machine your parts — no account managers, no trading-company markups. Complimentary DFM review with every enquiry.