Bolt Torque Calculation Guide | Industrial Engineering Studio

Why Torque Matters

Proper bolt tightening creates sufficient preload (clamping force) to hold joints together under operating loads. Too little torque — joint leaks or separates. Too much torque — bolt yields, strips threads, or fractures.

The relationship between torque and preload depends on thread friction, under-head friction, thread pitch, and bolt diameter.

Torque-Tension Relationship

The standard short-form equation:

T = K × D × F

Where:

T = torque (N·m)
K = nut factor (dimensionless, accounts for friction)
D = nominal bolt diameter (m or mm — use consistent units)
F = preload/bolt tension (N or kN)

The Nut Factor K

The nut factor K is the key variable. It lumps together thread friction, collar friction, and the pitch torque component. Typical values:

K = 0.20 for lubricated or coated bolts
K = 0.30 for dry steel-on-steel (most common reference value)
K = 0.15-0.18 for PTFE-coated or moly-lubed
K = 0.35-0.45 for hot-dip galvanized (high friction) Always use K consistent with actual lubrication conditions.

Detailed Torque Formula (Long Form)

For more accuracy, break torque into three components:

T = T_pitch + T_thread + T_collar

T = F × [P/(2π) + (µ_t × r_t)/cos(α/2) + µ_c × r_c]

Where:

P = thread pitch (mm/rev)
µ_t = thread coefficient of friction (0.10-0.20 for lubricated; 0.15-0.25 dry)
r_t = effective thread radius ≈ (D − P/2)/2
α/2 = half thread angle (30° for standard 60° threads)
µ_c = collar/nut face friction coefficient
r_c = mean collar radius ≈ 1.25 × D/2 for standard hex nuts

For 60° threads (UNC/UNF/ISO metric): cos(30°) = 0.866

∑

Bolt Torque Calculator

Open bolt-torque-calculator →

Bolt Grades and Strength

Metric Grade	Property Class	Proof Strength (MPa)	Ultimate (MPa)	SAE/Imperial
Low carbon	4.6	225	400	Grade 2
Medium carbon	5.8	420	520	Grade 2 (higher)
High strength	8.8	600	800	Grade 5
Alloy steel	10.9	900	1000	Grade 8
High alloy	12.9	1080	1200	Grade 8 (higher)

Preload Target

Typical preload is 75% of bolt proof load for permanent connections:

F_preload = 0.75 × S_proof × A_tensile

Tensile stress area for metric bolts (ISO 898-1):

Diameter	Pitch	Stress Area (mm²)
M8	1.25	39.7
M10	1.5	64.1
M12	1.75	92.1
M16	2.0	167
M20	2.5	258
M24	3.0	369
M30	3.5	580
M36	4.0	840

Standard Torque Values (Metric, K=0.20 Lubricated)

Bolt	8.8 (N·m)	10.9 (N·m)	12.9 (N·m)
M8	25	37	44
M10	49	73	88
M12	85	127	152
M16	210	315	378
M20	425	630	755
M24	725	1100	1320
M30	1450	2100	2520
M36	2530	3780	4540

Lubrication Drastically Changes Torque

The same bolt torqued to 300 N·m with K=0.20 (lubricated) produces ~50% MORE preload than with K=0.30 (dry). If you use lubricated torque values on dry bolts, you'll UNDER-tighten. If you use dry torque on lubricated bolts, you'll OVER-tighten and may break bolts. ALWAYS specify and control lubrication condition.

Worked Example — M20 Grade 8.8

M20 bolt, stress area A_s = 258 mm²
8.8 grade proof strength = 600 MPa
Target F = 0.75 × 600 × 258 = 116,100 N = 116 kN
Lubricated (K = 0.20): T = 0.20 × 0.020 × 116,000 = 464 N·m
Dry (K = 0.30): T = 0.30 × 0.020 × 116,000 = 696 N·m

Same bolt, same preload, different torque values depending on K.

Tightening Methods

Method	Accuracy	Cost	Applications
Torque wrench	±25-35%	Low	General purpose
Calibrated torque wrench	±15-25%	Medium	Most industrial
Turn-of-nut (angle)	±10-15%	Low	Structural bolts
Hydraulic tensioner	±5-10%	High	Critical flanges, large bolts
Bolt elongation (ultrasonic)	±1-5%	Very high	Turbines, pressure vessels
Direct tension indicator (DTI)	±10%	Medium	Structural steel

Why Such Poor Accuracy?

Simple torque control has ±25-35% scatter because friction accounts for 80-90% of the torque applied. Only 10-20% actually stretches the bolt. Small changes in friction (rust, lubrication, surface finish) cause large preload variations. Critical joints use hydraulic tensioning or ultrasonic measurement.

Tightening Sequence (Flanges)

For flange joints, tighten bolts in a crisscross (star) pattern in at least three passes:

First pass: 30% of target torque
Second pass: 60% of target torque
Third pass: 100% of target torque
Optional final pass: clockwise around flange at 100%

This ensures even gasket compression and prevents flange bowing or leaks. Follow ASME PCC-1 guidelines.

Bolt Torque for Pressure Vessels and Flanges

For ASME B16.5 flange joints per ASME PCC-1:

Calculate required gasket seating stress (from gasket manufacturer)
Determine total bolt load to seat gasket and resist internal pressure
Select torque that achieves target bolt stress
Apply in crisscross pattern per PCC-1

For spiral-wound gaskets: target gasket stress ≈ 50-70 MPa (7-10 ksi), which sets the bolt load.

Bolt Relaxation and Re-Torquing

Embedding: Bolts and gaskets relax in the first 24 hours; re-torque hot bolts after heat cycle
Gasket creep: Spiral wound and compressed fiber gaskets relax; some specs require re-torquing
Thermal effects: Different thermal expansion between bolt and flange changes preload

Hot Bolting

For high-temperature flanges (steam, process), it is common practice to: torque at ambient, heat to operating temperature, shut down and re-torque after cooling to address relaxation. Some specs require this re-torque cycle.

Common Mistakes

Using same torque value regardless of lubrication
Not calibrating torque wrenches (should be calibrated annually)
Tightening in a circle instead of crisscross pattern
Using impact wrenches for final torque (no control, causes overload)
Ignoring bolt material differences (Grade 5 vs Grade 8 torque values differ)
Overlooking thread condition (dirty/rusty threads = high friction = low preload)
Mixing bolt grades in same joint (uneven load sharing)

Summary

Use T = K × D × F for bolt torque, with K = 0.20 lubricated or K = 0.30 dry. Target preload at 75% of proof load. Standard metric torque tables assume K = 0.20 — adjust for actual lubrication. Friction dominates (80-90% of torque), so control surface condition and lubrication. Always tighten in crisscross pattern per ASME PCC-1 for flanges. For critical joints (pressure vessels, large flanges), use hydraulic tensioners or ultrasonic measurement instead of torque-only control.

Related Guides & Tools