UK Bolt Torque Chart -- Tightening Torques for Class 8.8 and 10.9 Bolts per BS EN 1090-2

Establishing the correct tightening torque for preloaded structural bolts is fundamental to achieving the design preload force Fp,Cd. The torque-tension relationship is not a material property but a tribological one -- it depends on the friction coefficient between the nut and bolt threads, the friction coefficient under the nut face, and the geometric parameters of the thread. BS EN 1090-2 Method T requires the torque coefficient k to be determined through pre-installation verification testing (PIVT) on the specific batch of bolt assemblies to be used. This reference presents the standard UK torque values, the underlying torque-tension theory, and worked specifications for UK structural steel connections.

Torque-Tension Theory

The relationship between applied torque and induced preload derives from the mechanics of the inclined plane (the thread helix) and friction at two interfaces: the thread contact surface between the nut and bolt, and the bearing surface under the nut face. The classical Short Formula is:

T = k x Fp,Cd x d

Where:

• T = applied tightening torque (N.mm or N.m)
• k = torque coefficient (dimensionless, empirically determined)
• Fp,Cd = design preload force (N or kN)
• d = nominal bolt diameter (mm)

The torque coefficient k aggregates the effects of thread geometry and friction into a single empirically determined value. Typical k values for structural bolting range from 0.14 to 0.24 depending on the lubrication condition, surface finish, and whether washers are used.

For a metric ISO thread, the theoretical k can be decomposed as:

k = (d2/d) x (P/(2 x pi x d2) + mu_t / cos(beta)) + 0.625 x mu_n

Where:

• d2 = pitch diameter of the thread
• P = thread pitch
• mu_t = thread friction coefficient
• mu_n = nut face friction coefficient
• beta = thread flank angle (30 degrees for ISO metric threads)

In practice, k is determined experimentally because mu_t and mu_n are assembly-specific and cannot be predicted from material properties alone.

Target Torque Values for UK Structural Bolts

The following torque values assume k = 0.20, which is a reasonable mid-range assumption for new, clean, lightly oiled structural bolts. These values must be verified by PIVT for the specific bolt assembly before use on site.

Class 8.8 Bolts (fub = 800 MPa), k = 0.20

Class 10.9 Bolts (fub = 1000 MPa), k = 0.20

These torque values are indicative. The actual target torque must be determined through PIVT per BS EN 1090-2 Clause 6.3.2, using a representative sample of the actual bolt assemblies, tested in a Skidmore-Wilhelm or equivalent tension calibrator. The PIVT typically tests 10-15 assemblies and establishes the mean k and its standard deviation. The target torque is then set at the value corresponding to the mean k.

Effect of k Variability

The torque coefficient k is sensitive to assembly conditions. A bolt with k = 0.14 (well-lubricated) will develop 1.4 times the preload of a bolt with k = 0.20 at the same torque. Conversely, a bolt with k = 0.24 (dry or slightly corroded) will develop only 0.83 times the design preload. This variability is why BS EN 1090-2 mandates PIVT and why torque control (Method T) requires the tested assemblies to be from the same production batch, with the same lubrication condition, as those used on site.

The consequences of k variability flow in both directions. Under-torqueing (k high) results in preload below the design value, reducing slip resistance and potentially compromising the connection. Over-torqueing (k low) can cause bolt fracture during installation, particularly in Class 10.9 bolts where the preload is already close to the yield point.

Pre-Installation Verification Testing (PIVT) per BS EN 1090-2

BS EN 1090-2 Clause 6.3.2 requires PIVT for all preloaded bolt assemblies tightened by Method T. The test procedure is:

1. Select a representative sample of 10-15 bolt-nut-washer assemblies from the delivery batch.
2. Assemble each bolt in a Skidmore-Wilhelm tension calibrator (direct tension measurement) or equivalent load cell.
3. Tighten to the target torque and record the induced preload.
4. Calculate the mean k and standard deviation.
5. Verify that the mean preload >= 1.0 x Fp,Cd and that no individual bolt develops less than 0.9 x Fp,Cd.
6. If the verification fails, adjust the torque and re-test, or reject the batch.

PIVT must be repeated if the bolt batch changes, the lubrication condition changes, or more than 1000 bolts have been tightened since the last test.

Inspection and Verification on Site

BS EN 1090-2 requires that the tightening process be verified on site through:

• Daily torque wrench calibration: The calibrated wrench is checked at the start and end of each shift against a torque analyser. The wrench must indicate within plus/minus 5% of the analyser reading.

• Bolt assembly sampling: For every 200 bolts tightened, a 10% sample is checked with a check torque wrench at the inspection torque (typically 1.1 x target torque). If the nut rotates, the bolt has lost preload and must be re-tightened.

• Visual inspection: All bolts must be checked for the presence of washers (both head and nut sides for oversized/slotted holes), correct projection of the bolt thread through the nut (minimum one full thread, maximum 10 mm), and absence of gaps between connected plies.

Worked Example -- Tightening Specification for a Portal Frame

A UK portal frame building specifies M20 Class 8.8 preloaded bolts (Category E) for the eaves and apex moment connections. There are 120 M20 bolts in the frame. The designer must prepare the tightening specification.

Step 1 -- Determine design preload: Fp,Cd = 0.7 x 800 x 245 / 1.10 = 124.7 kN

Step 2 -- PIVT: A sample of 12 M20 bolt assemblies from the delivered batch is tested in the Skidmore-Wilhelm calibrator. The mean k is 0.185 with a standard deviation of 0.015. All 12 assemblies achieve a preload >= 124.7 kN at the target torque.

Target torque: T = k_mean x Fp,Cd x d = 0.185 x 124.7 x 20 = 461 N.m

The target torque is rounded to 460 N.m for practicality.

Step 3 -- Specification on drawings: "All M20 bolts at eaves and apex connections to be preloaded Class 8.8 assemblies to BS EN 14399-3, tightened by torque control Method T to BS EN 1090-2. Target torque: 460 N.m. Tightening sequence: from the most rigid part of the connection outward, in two stages -- first to 300 N.m (snug-tight), then to 460 N.m. PIVT required before site bolting commences."

Step 4 -- Site verification: The calibrated torque wrench is checked daily. A 10% sample (12 bolts) is checked at 1.1 x 460 = 506 N.m -- no nut rotation observed, confirming that preload >= Fp,Cd is maintained.

Design Resources

• UK Bolt Capacity Tables -- Fv,Rd and Ft,Rd for Classes 8.8 and 10.9
• UK Bolt Pretension Guide -- Preloaded bolt design Categories D/E
• UK Bolt Grades Guide -- Grade selection and material properties
• UK Connection Design Guide -- EN 1993-1-8 bolted and welded joints
• All UK Steel Design References -- complete library

Frequently Asked Questions

What torque should be applied to an M20 Class 8.8 bolt in UK steelwork?

For a standard assumption of k = 0.20, the tightening torque for M20 Class 8.8 is T = 0.20 x 124.7 x 20 = 499 N.m. This is an indicative value -- the actual target torque must be determined through pre-installation verification testing (PIVT) on the specific bolt batch per BS EN 1090-2. The torque should never be applied without PIVT verification because the actual k value for the specific assembly may differ by plus/minus 30% from the assumed 0.20, resulting in either insufficient preload or over-tightening to fracture.

Why does the torque coefficient k vary between bolt assemblies?

The torque coefficient k depends on the friction at the thread interface and under the nut face. Lubrication condition is the dominant variable: well-lubricated threads can give k as low as 0.10-0.12, while dry or slightly corroded threads can give k as high as 0.30. Even within a single production batch, the standard deviation of k is typically 0.01-0.02. Electro-galvanised, hot-dip galvanised, and sherardized coatings all produce different k values for the same nominal bolt. This is why PIVT is mandatory -- the actual k must be measured, not assumed.

Is torque control suitable for all preloaded bolt applications in UK practice?

Torque control (Method T) is suitable for the majority of UK building connections. However, for large-diameter bolts (M30 and above), bolt lengths exceeding 500 mm, or connections where the consequences of under-preload are severe (e.g., bridge fatigue-critical connections), the Combined Method (CM) or Direct Tension Indicator (DTI) method is recommended because torque variability increases with bolt diameter and length. For wind turbine tower flange connections (typically M36-M64), hydraulic tensioning or DTI washers are the standard methods, as torque wrench calibration at the required torque levels (3000-8000 N.m) becomes impractical on site.

Does the UK NA specify tightening torque values for bolts?

The UK National Annex to BS EN 1993-1-8 does not specify torque values. It references BS EN 1090-2 for the tightening method and requires that the tightening procedure be documented in the execution specification. The torque values must be determined by PIVT for each batch of bolt assemblies. The UK NA emphasises that the torque values in general reference tables (based on k = 0.20) are for preliminary design only and must not be used as the basis for site tightening without PIVT confirmation.

Educational reference only. All design values are per BS EN 1993-1-8:2005 + UK National Annex and BS EN 1090-2:2018. Verify all torque values through pre-installation verification testing on the specific bolt assembly batch. Designs must be independently verified by a Chartered Structural Engineer registered with the Institution of Structural Engineers (IStructE) or the Institution of Civil Engineers (ICE). Results are PRELIMINARY -- NOT FOR CONSTRUCTION without independent professional verification.