Engineering guide

Torque & preload guide

At M1 the difference between a joint that holds and one that strips is often 0.04 N·m. This guide covers the calculation, the friction data behind it, and how to verify what you actually applied.

T = K × F × dK factor tableM1 – M6Strip torque data

The calculation

Two equations, one assumption

Torque is converted into clamp load through friction. That is the whole model — and the friction coefficient is where most of the uncertainty lives.

Preload

F = σ × A_s

σ is the target stress (typically 0.6–0.75 × proof strength), A_s is the thread stress area.

Torque

T = K × F × d

K is the nut factor (friction), F the preload in newtons, d the nominal diameter in metres.

Worked example — M2.0, class 8.8, zinc-nickel with lubricant:
A_s = 2.07 mm², proof strength ≈ 640 MPa, target 70% → F = 0.7 × 640 × 2.07 ≈ 930 N.
With K = 0.15, T = 0.15 × 930 × 0.002 = 0.28 N·m. With K = 0.22 (dry zinc), the same torque produces only 630 N — a 32% loss of clamp load from friction alone.

Rule of thumb

If you cannot measure K, assume ±25% scatter

An unmeasured joint does not have a preload — it has a preload range. For non-critical joints that is acceptable. For a sealed, thermal or safety-relevant joint, measure K with a torque-tension test on the actual parts and finish.

Our laboratory runs torque-tension to ISO 16047.

Ask for a joint test with your parts.

Request a joint test

Friction

Nut factor by joint condition

K is not a material constant. It depends on the finish, the lubricant and the surface roughness of both threads. The scatter column matters as much as the mean.

Joint condition	K factor	Clamp load scatter	When to use
As-received, plain steel	0.20	±25%	General estimation only
Zinc plated, dry	0.22	±25%	Low-criticality joints
Zinc-nickel with lubricant	0.14 – 0.16	±12%	Controlled preload, automotive
Black oxide, lightly oiled	0.18 – 0.21	±18%	Internal cosmetic joints
PVD DLC	0.10 – 0.13	±10%	Low friction, high-cycle
Stainless, passivated, dry	0.24 – 0.30	±30%	Galling-prone; use anti-seize
Molybdenum disulphide paste	0.12 – 0.15	±12%	High preload, controlled
Wax / dry-film lubricant	0.13 – 0.17	±15%	Clean assembly, no liquid

Scatter values are indicative for production parts. A controlled finish with a specified lubricant can hold ±12%; an unspecified dry joint can exceed ±30%.

Reference data

Stress area and recommended torque

Calculated at 70% of proof strength with K = 0.20 for 8.8 and 12.9, and K = 0.24 for A2-70 stainless. Use these as a starting point, then verify on the actual joint.

Size	Stress area As (mm²)	8.8 preload (N)	8.8 torque (N·m)	12.9 torque (N·m)	A2-70 torque (N·m)
M1.0	0.46	210	0.04	0.06	0.03
M1.2	0.65	290	0.07	0.10	0.05
M1.4	0.88	395	0.11	0.16	0.07
M1.6	1.27	570	0.18	0.27	0.12
M2.0	2.07	930	0.37	0.55	0.24
M2.5	3.39	1,520	0.76	1.14	0.49
M3.0	5.03	2,250	1.35	2.03	0.88
M4.0	8.78	3,930	3.15	4.72	2.04
M5.0	14.2	6,360	6.36	9.54	4.13
M6.0	20.1	9,000	10.8	16.2	7.02

Values are for unlubricated, as-received parts. A lubricated or coated joint with lower K will need proportionally lower torque for the same preload — do not apply these figures to a DLC-coated screw.

Upper bound

Strip torque in the mating material

The joint fails at whichever is weaker: the screw or the female thread. In aluminium and brass the female thread usually governs, so the torque target must sit below it.

Size	Brass insert (N·m)	Aluminium 2 mm wall (N·m)	Steel tapped (N·m)	Plastic boss (N·m)
M1.0	0.18	0.11	0.25	0.09
M1.2	0.26	0.16	0.36	0.13
M1.4	0.36	0.22	0.50	0.18
M1.6	0.50	0.31	0.70	0.26
M2.0	0.90	0.55	1.25	0.45
M2.5	1.60	0.95	2.20	0.78
M3.0	2.60	1.55	3.60	1.25
M4.0	5.50	3.30	7.60	2.70

Strip torque values are typical for the stated material and engagement length. They vary with engagement, wall thickness and alloy condition. Measure in the actual mating part before setting a torque specification.

Setting a window

From calculation to a line specification

Define the functional clamp load

Sealing, thermal contact or vibration resistance each imply a minimum preload. Start there, not from a torque table.

Measure K on the actual parts

Ten assemblies, torque-tension test, record the preload scatter. This gives the K range, not a single number.

Measure strip torque in the mating part

Ten drives into the real material. Take the lower bound and apply a safety factor of at least 1.4.

Set the target inside both bounds

The driver's upper tolerance must stay below 70% of strip torque; the lower tolerance must still meet the functional clamp load.

Verify on the line, at production speed

A hand driver is not a production tool. Check torque with the actual driver, at cycle rate, on the first three parts of every lot.

Pan head micro screws ready for torque testing

Micro screws · torque-tension test samples

Questions

Torque FAQ

Common questions from design and process engineers.

Which formula should I use for preload?

For a quick estimate, F = 0.7 × proof strength × stress area, then T = K × F × d. For a critical joint, replace the 0.7 factor with the utilisation your design requires (typically 0.6–0.75 of yield) and use a measured K from a torque-tension test on the actual parts.

Why is my calculated torque different from the supplier’s recommendation?

Almost always because of the friction coefficient. A recommendation based on K = 0.20 and a joint with DLC coating (K = 0.11) will over-tighten by nearly a factor of two for the same torque. Ask what K the recommendation assumes, or measure it.

Is a higher property class always better?

No. A 12.9 screw in an aluminium thread will strip the aluminium before the screw reaches its preload. The joint strength is set by the weaker member. Higher class only helps if the mating thread and bearing surface can carry the load.

How do I verify torque on the line?

Use a current-controlled electric driver rather than a mechanical clutch, verify the tool with a torque transducer at the start of each shift, and check the first three parts of every lot. For critical joints, add a torque audit with a break-loose tester on sampled assemblies.

What causes a screw to break under the head?

Usually a combination of too-high preload and a sharp under-head radius. At micro sizes the fillet is a large fraction of the shank, so a sharp corner concentrates stress. A radiused transition and a lower torque target solve it more reliably than a stronger material.

Ask for a torque-tension test

Send a sample of the screw and the mating part. We will measure K, strip torque and the preload range, and recommend a line specification.

Request a Quote Talk to Engineering