Fasteners & Bolted Joints — Engineering Reference

See also (Tier 3 family index): Fasteners Taxonomy

1. At a glance

Bolted joints are the dominant demountable engineering connection — the only structural joint that can be disassembled for service, inspection, and re-use without destroying the parent material. They are present in essentially every machine built since the 19th century: bridges, aircraft, engines, pressure vessels, machine tools, robots, electronics enclosures, satellites. Welding bonds members permanently; adhesives bond them lighter; rivets bond them at one assembly cost. Only bolts give you all four of removability, controllable preload, predictable fatigue life, and field-installability with hand tools.

There are three things to get right in any bolted joint, and they have to be done in order:

1. Select the right strength class. Pick from a recognised system — ISO 898-1 (4.6 / 5.6 / 8.8 / 10.9 / 12.9), SAE J429 (Grade 2 / 5 / 8), or ASTM F3125 (A325 / F1852 / A490 / F2280) — and match the nut and washer class so the assembly fails by bolt-tension yielding, not thread-stripping, not nut splitting, not washer crush.
2. Apply the right preload. Target ~75 % of bolt proof load; verify by torque, turn-of-nut, tension indicator, or ultrasonic. Friction owns the torque-to-tension conversion; ±25 % preload scatter from torque alone is normal.
3. Design the joint geometry so the preload stays put. Adequate joint stiffness, controlled clamp length, prevented self-loosening (Junker-validated locking), and seating-loss budget so that after settle-in, gasket creep, and thermal cycling, the remaining preload still exceeds the maximum external load times the joint-stiffness ratio.

Failure modes split into two camps that an analyst must look at separately:

• Bolt failure — yielding, fatigue at the first engaged thread or under the head, fracture, stripping, galling, hydrogen embrittlement, stress-corrosion cracking.
• Joint failure — slip in bearing-type shear joints, separation in tension joints, fretting at the faying surface, embedment of asperities, gasket creep, loss of preload from vibration or thermal mismatch.

Where it sits in the design stack: bolted joints are the engineering interface between statics (load path, reactions), mechanics of materials (combined tension / shear / bending), fatigue analysis (cyclic stress amplitude), and standards-driven design (AISC 360-22 Chapter J for structural steel, VDI 2230 for high-performance machine joints, Eurocode 3 EN 1993-1-8 for European structural design).

---

2. First principles

2.1 The threaded fastener as a wedge

A screw thread is a helical inclined plane wrapped around a cylinder. Tightening converts torque applied at the head into axial tension in the shank by the mechanical advantage of that wedge. The helix angle (lead angle) λ for an ISO metric coarse M10×1.5 is

λ = arctan( P / (π · d_p) ) = arctan( 1.5 / (π · 9.026) ) ≈ 3.03°

with P = pitch (mm) and d_p = pitch diameter. For comparison the friction angle of plain-steel-on-steel is arctan(0.15) ≈ 8.5° — that is, the friction angle exceeds the lead angle for every standard coarse thread, which is precisely why threaded fasteners do not back out under axial load alone. They are self-locking in the static sense. Vibration loosening (Junker) is a different mechanism (transverse shear at the contact) and not addressable by helix geometry.

2.2 Why preload exists

Tightening a bolt against an unloaded joint puts the bolt into tension F_i and the clamped members into equal-and-opposite compression F_i. Both spring elements are now pre-stressed and connected in parallel from the standpoint of any subsequent external load.

When an external tensile load F_ext is applied to the joint, it is shared between bolt and clamped members in proportion to their stiffnesses:

ΔF_bolt   = C       · F_ext         (additional tension picked up by the bolt)
ΔF_member = (1 − C) · F_ext         (clamp force relieved in the members)
F_bolt    = F_i + C · F_ext         (total bolt tension)

with C = K_b / (K_b + K_m) the joint stiffness ratio — bolt stiffness K_b over the sum of bolt and member stiffnesses K_m.

For a typical steel-on-steel joint with steel cap-screw clamping plates, C ≈ 0.1 – 0.4. Members are usually much stiffer than the bolt (K_m ≈ 3–10 × K_b), so most of an external load relieves clamp force in the members and only a small fraction reaches the bolt. This is the central insight of bolted-joint design.

2.3 Why preload extends fatigue life

If the bolt were not preloaded, every cycle of F_ext = 0 → F_max would apply the full range to the bolt. With high preload, the bolt sees almost-constant tension F_i and only a small additional swing ΔF_bolt = C · F_ext. Since bolt fatigue scales roughly as the stress amplitude, suppressing the swing by a factor (1/C) ≈ 3–10 multiplies fatigue life by orders of magnitude.

The penalty: the mean stress in the bolt rises to F_i. The bolt now lives at high mean stress with low amplitude — the most favourable corner of a Haigh/Goodman diagram for endurance-limited steels.

2.4 Torque → tension: where the energy goes

Tightening converts wrench torque T into bolt tension F_i through three friction paths and one useful path:

Path	Fraction of T (typical)
Friction under the bolt head / nut face	~50 %
Friction in the threads	~40 %
Useful axial work stretching the bolt	~10 %

Only 10 % of the torque you apply actually goes into stretching the fastener. The other 90 % is dissipated in friction at two surfaces — each of which has a coefficient of friction that varies from ~0.08 (waxed) to ~0.20 (rusty dry steel). That is the origin of the empirical nut factor K in the short-form torque equation:

T = K · F_i · d_nom

with d_nom = nominal bolt diameter. Typical K values: 0.20 plain steel dry, 0.18 plated/oily, 0.15–0.16 lubricated, 0.10 waxed or PTFE-coated. A ±25 % uncertainty in K is normal, which is why torque-only tightening produces ±25 % scatter in preload. This is the #1 source of preload variation in industry.

2.5 Hierarchy of clamp-force control accuracy

Method	Preload accuracy (typical 1σ)	Cost / complexity
Calibrated torque wrench (click)	±25 %	low
Torque + angle (yield-control)	±5–10 %	medium
Turn-of-nut (per AISC)	±10–15 %	low, no calibration
Hydraulic tensioner	±2–5 %	high; only big bolts
Ultrasonic bolt extensometer	±1–3 %	high; instrumented
Strain-gauged smart bolt	±1–2 %	very high

---

3. Practical math / design equations

3.1 Bolt tensile stress area

The effective cross-section of a threaded fastener under tension is not the nominal diameter — it is the tensile stress area A_t that averages the minor and pitch diameters (ISO 898-1, ASME B1.1):

A_t = (π / 4) · [ (d_2 + d_3) / 2 ]²

with d_2 = pitch diameter and d_3 = minor diameter (root). Standard tabulated values — memorise the common ones; reach for the table for the rest.

Thread	Pitch P (mm)	A_t (mm²)
M3	0.5	5.03
M4	0.7	8.78
M5	0.8	14.2
M6	1.0	20.1
M8	1.25	36.6
M10	1.5	58.0
M12	1.75	84.3
M14	2.0	115
M16	2.0	157
M20	2.5	245
M24	3.0	353
M30	3.5	561

UN (Unified Inch) equivalents:

Thread	A_t (in²)	A_t (mm²)
1/4-20 UNC	0.0318	20.5
5/16-18	0.0524	33.8
3/8-16	0.0775	50.0
7/16-14	0.1063	68.6
1/2-13	0.1419	91.5
5/8-11	0.226	146
3/4-10	0.334	215
7/8-9	0.462	298
1-8	0.606	391

3.2 Proof load, yield, ultimate

For ISO 898-1, the class designation is (σ_u / 100) . (σ_y / σ_u × 10). So 8.8 means σ_u = 800 MPa, σ_y = 0.8 × 800 = 640 MPa.

Class	σ_p (MPa)	σ_y (MPa)	σ_u (MPa)	Note
4.6	225	240	400	low-carbon, mild
5.6	280	300	500	low-carbon
8.8	580	640	800	medium-C, Q&T — most common high-strength
10.9	830	940	1040	alloy steel Q&T
12.9	970	1080	1220	alloy steel Q&T, high HRC

Proof stress σ_p is the stress the bolt must reach without measurable permanent set. Practically: σ_p ≈ 0.90 · σ_y for classes 8.8 / 10.9 / 12.9 and 0.85 · σ_y for 4.6 / 5.6.

SAE J429 equivalents:

Grade	σ_p (ksi / MPa)	σ_y (ksi / MPa)	σ_u (ksi / MPa)	Markings
Grade 2	55 / 380	57 / 393	74 / 510	(none)
Grade 5	85 / 585	92 / 634	120 / 827	3 radial lines
Grade 8	120 / 827	130 / 896	150 / 1034	6 radial lines

ASTM F3125 (consolidated, 2015) replaces A325/A490 with marking-compatible Grades A325, F1852, A490, F2280:

Grade	σ_p (ksi / MPa)	σ_u (ksi / MPa)	Note
A325 / F1852	85 / 585	120 / 827	F1852 = tension-control “TC bolt” twist-off
A490 / F2280	120 / 827	150 / 1034	F2280 = TC-bolt variant of A490

Cross-system mapping (approximate, not interchangeable for code use):

ISO	SAE	ASTM
4.6 / 4.8	Grade 2	—
8.8	Grade 5	A325 / F1852
10.9	Grade 8	A490 / F2280
12.9	—	— (no SAE/ASTM equivalent; aerospace uses 220 ksi NAS)

3.3 Joint stiffness ratio C

The exact computation is laid out in VDI 2230; an engineering approximation that gets within ~10 % for steel-on-steel cap-screw joints:

K_b = (E_b · A_b) / L_b                 (bolt stiffness, axial spring)
K_m ≈ (E_m · A_m) / L_m                 (member stiffness, frustum approximation)

C = K_b / (K_b + K_m)

with A_m the effective clamp-cone area (Rotscher / VDI frustum-cone model, half-angle 25–30°). Typical results: C ≈ 0.2 for a steel cap-screw clamping steel; C ≈ 0.3–0.4 with a gasket; C → 0.5 with a soft gasket; C ≈ 0.1 for a long thin bolt clamping a thick stack.

Joint separation (clamp force goes to zero) occurs when the external load exhausts the preload reserve in the members:

F_ext,sep = F_i / (1 − C)

Design rule of thumb: keep F_ext,max ≤ 0.5 × F_ext,sep, i.e. preserve ≥ 50 % of preload as residual clamp at peak external load.

3.4 Bolt fatigue

Rolled threads (the standard manufacturing process for grades ≥ 8.8) develop a compressive residual stress at the root and a fine-grain microstructure that elevates endurance dramatically over cut threads. Even so, the thread root is a sharp stress concentrator (K_f ≈ 3–5 for rolled, 4–6 for cut) and the endurance limit referenced to A_t is much lower than the parent steel:

Thread type	σ_e at thread root (MPa)
Rolled, then heat-treated	~60
Rolled after heat treat (better)	~85
Cut threads	~45

These are σ_a (alternating component) values — Shigley §8.11 / Bickford Ch. 13. Design rule: keep ΔF_bolt / A_t < 60 MPa for rolled-then-HT class 8.8/10.9 to achieve infinite life. Since ΔF_bolt = C · F_ext, that ties back to the joint-stiffness ratio.

3.5 Bolt shear and bearing

For a bolt acting in shear (e.g. bearing-type structural connection):

τ_allow = 0.60 · σ_u           (bolt body in shear, Shigley/AISC)
F_v     = τ_allow · A_s        (shear strength of one shear plane)

A_s = body area πd²/4 if the shear plane is in the unthreaded shank, or 0.875·A_t if the shear plane is in the threads (“threads in shear plane” — N suffix in AISC).

Bearing stress at the hole:

σ_bearing = F / (d · t)

with t = plate thickness. AISC J3.10 limits bearing to 2.4·d·t·F_u (with tear-out check 1.2·L_c·t·F_u).

Slip-critical joint (AISC 360-22 §J3.8):

R_n = μ · D_u · h_f · T_b · N_s

— μ slip coefficient (0.30 Class A faying surface, 0.50 Class B blasted), D_u = 1.13 mean-to-spec preload factor, h_f = filler factor, T_b = minimum pretension (Table J3.1), N_s = number of slip planes.

3.6 Torque target

Tightening torque for a target preload F_i = 0.75 · F_p = 0.75 · σ_p · A_t:

T = K · F_i · d_nom

3.7 Worked example 1 — M12 class 10.9 socket-head cap screw clamping 25 mm Al plate to steel

Given: M12 × 1.75, class 10.9 (σ_p = 830 MPa), A_t = 84.3 mm². Plain steel bolt threads against zinc-plated nut, light oil — K ≈ 0.18. Aluminum plate t = 25 mm, steel base t = 25 mm, total grip ℓ_g ≈ 50 mm.

Target preload (75 % of proof):

F_p = σ_p · A_t = 830 MPa × 84.3 mm² = 69,970 N ≈ 70.0 kN
F_i = 0.75 · F_p = 0.75 × 70.0 = 52.5 kN

Required torque:

T = K · F_i · d_nom
  = 0.18 × 52,500 N × 0.012 m
  = 113.4 N·m  (~84 lb·ft)

Manufacturer table for M12-10.9 lubricated typically lists 95–110 N·m — matches.

Joint stiffness check (quick frustum estimate):

K_b  = E_steel · A_t / ℓ_g  = 200 GPa × 84.3 mm² / 50 mm = 337 kN/mm
K_m  ≈ 1500 kN/mm  (mixed Al + steel frustum, VDI 2230 §5.1)

C    = 337 / (337 + 1500) = 0.184

Separation threshold:

F_ext,sep = F_i / (1 − C) = 52.5 / 0.816 = 64.3 kN

Design max external load ≤ 0.5 × 64.3 ≈ 32 kN for adequate residual clamp.

Fatigue check (assume external cyclic load 0 → 20 kN):

ΔF_bolt   = C · F_ext = 0.184 × 20 kN = 3.68 kN
σ_a = ΔF_bolt / (2 · A_t) = 3680 / (2 × 84.3) = 21.8 MPa

Below 60 MPa rolled-then-HT endurance — infinite life. Pass.

3.8 Worked example 2 — 8 × M8 class 8.8 in a flange under 50 kN tensile

Given: Flanged connection, 8 bolts on a circle, total external axial load F_ext_tot = 50 kN. Class 8.8: σ_p = 580 MPa, A_t = 36.6 mm². Joint stiffness ratio estimated C = 0.25.

Per-bolt external load:

F_ext = 50 / 8 = 6.25 kN per bolt

Proof load and preload target:

F_p = 580 × 36.6 = 21.2 kN
F_i = 0.75 · F_p = 15.9 kN per bolt

Bolt force at peak external:

F_bolt = F_i + C · F_ext = 15.9 + 0.25 × 6.25 = 15.9 + 1.56 = 17.5 kN

Factor of safety vs proof:

FoS = F_p / F_bolt = 21.2 / 17.5 = 1.21

1.21 is below the stated target of 1.5 — design action: either move to class 10.9 (F_p = 30.4 kN, FoS = 1.74 ✓) or use 8 × M10 8.8 (F_p = 33.6 kN, FoS = 1.92 ✓). Either choice works; M10-8.8 preferred if envelope allows because lower-class fastener is more forgiving on assembly variance and cheaper.

Verify joint does not separate:

F_ext,sep = F_i / (1 − C) = 15.9 / 0.75 = 21.2 kN  >  6.25 kN  ✓

3.9 Worked example 3 — ASTM A325-N, 3 bolts in single shear, 100 kip

Given: 7/8” diameter ASTM A325-N bolts (Threads iN shear plane), single shear, 3 bolts. F = 100 kip = 445 kN total. σ_u = 120 ksi = 827 MPa.

Per-bolt shear:

V = 100 / 3 = 33.3 kip = 148 kN per bolt

Nominal shear strength (AISC J3.6 Table, A325-N): F_nv = 54 ksi = 372 MPa on full body area A_b = π(0.875)²/4 = 0.601 in² = 388 mm² (AISC’s tabulated F_nv embeds the 0.75 threads-in-shear factor on A_b).

R_n = F_nv · A_b = 54 ksi × 0.601 in² = 32.5 kip per bolt

Design shear strength (LRFD, φ = 0.75):

φR_n = 0.75 × 32.5 = 24.4 kip per bolt
3 · φR_n = 73.1 kip  <  100 kip applied  ✗

Connection is under-strength at A325-N × 3. Options: upgrade to A490-N (F_nv = 68 ksi → φR_n = 30.7 kip, 3 bolts = 92 kip, still short), use 4 bolts (97.4 kip ≥ 100 kip ✓), or move to A325-X (threads eXcluded, F_nv = 68 ksi → 30.7 kip, 3 bolts = 92 kip — still 8 % short).

Cleanest fix: 4 × A325-N or 3 × A490-X (F_nv = 84 ksi → 37.9 kip, 3 bolts = 114 kip ✓). The 3 × A490-X version saves a hole but increases bolt cost and demands tighter installation control (turn-of-nut or TC bolts).

---

4. Reference data

4.1 Recommended tightening torques (75 % proof, K assumed)

Metric coarse, target T (N·m) for 0.75 · F_p:

Size	4.6, K=0.20	8.8, K=0.18	10.9, K=0.18	12.9, K=0.18
M4	0.9	2.5	3.6	4.2
M5	1.9	5.0	7.3	8.5
M6	3.3	8.7	12.7	14.7
M8	8.0	21	31	36
M10	16	42	61	71
M12	28	73	107	124
M14	44	117	171	200
M16	68	180	263	308
M20	133	350	514	600
M24	230	605	890	1040

Values are rounded; always check the latest VDI 2230 table or the bolt manufacturer’s lubricated-K chart. Add ±25 % uncertainty for torque-only methods.

4.2 Common metric coarse threads (ISO 261 / ISO 262)

M3 × 0.5 / M4 × 0.7 / M5 × 0.8 / M6 × 1.0 / M8 × 1.25 / M10 × 1.5 / M12 × 1.75 / M14 × 2.0 / M16 × 2.0 / M20 × 2.5 / M24 × 3.0 / M30 × 3.5 / M36 × 4.0.

Fine pitches exist (M10 × 1.25, M12 × 1.5) — use where you need slightly higher tensile stress area, finer adjustment, or thin-wall thread engagement.

4.3 Common Unified threads (ASME B1.1)

4-40 UNC / 6-32 / 8-32 / 10-24 / 10-32 (fine) / 1/4-20 / 5/16-18 / 3/8-16 / 7/16-14 / 1/2-13 / 5/8-11 / 3/4-10 / 7/8-9 / 1-8

4.4 Bolt head and drive types

Head style	Standard	Drive	Use case
Hex cap screw	ISO 4014 (partial thread), ISO 4017 (full thread), ASME B18.2.1	external hex	universal; visible-head structural and general assembly
Socket-head cap screw (SHCS)	ISO 4762, ASME B18.3	internal hex	machinery; counterbored installation; high-torque tightening
Button-head cap screw	ISO 7380	internal hex	low-profile, cosmetic; lower strength than SHCS
Flat / countersunk (CSK)	ISO 10642, ISO 14581	internal hex / Torx	flush installation, low profile
Flange-head bolt	ISO 4162 / DIN 6921	external hex	integrated washer; preferred for production assembly (auto)
Shoulder bolt (stripper bolt)	ISO 7379	internal hex	precision shaft with bearing-fit shoulder; mechanism pivot
Set screw (grub)	ISO 4026/27/28/29	internal hex / slotted	retain pulley/collar on shaft; cup / cone / dog point
Carriage bolt	ASME B18.5	square-shoulder, no head drive	wood-to-metal; square shoulder prevents rotation
Lag screw / coach bolt	ASME B18.2.1	external hex	wood; coarse fast-pitch self-tapping into timber
Threaded rod / stud	ISO 898-1 (class)	none	studded flanges, anchor bolts

4.5 Washer types

Washer	Standard	Function
Plain flat	ISO 7089 / DIN 125	distribute clamp force; protect surface
Heavy plain	ISO 7093 / DIN 9021	larger OD for soft materials
Thick / hardened	ISO 7090 / DIN 7349, ASTM F436	structural; resist crushing under high-class bolt
Split lock (“spring lock”)	DIN 127, ASME B18.21.1	vibration loosening: ineffective per Junker test — avoid for new design
Internal/external tooth	DIN 6797, 6798	electrical bonding; mild anti-rotation
Belleville (conical disc)	DIN 6796	spring stiffness; maintains clamp through gasket creep / thermal swing
Nord-Lock wedge-locking	technical bulletin (no ISO yet)	passes Junker DIN 65151; preferred for vibrating joints
Schnorr safety washer	DIN 6908	conical with serrations; anti-rotation by bite

4.6 Material / coating designations

• Plain carbon steel + plating: ZP (zinc plate / clear chromate), HDG (hot-dip galvanised, +1 thread oversize), GeoMet / Dacromet (zinc-aluminum flake, ~120 h salt-spray rating per standard), DeltaProtekt KL100.
• Stainless steel (ISO 3506-1): A2-70 (austenitic 304-equivalent, 700 MPa σ_u), A4-70 (316-equivalent), A4-80 (316 cold-worked, 800 MPa σ_u). Magnetic permeability ≈ 1.01 (cold work slightly magnetises austenitic).
• Aerospace / high-strength: A286 (precipitation-hardened austenitic, NAS standards), Inconel 718 (NAS1351 etc.), MP35N, H-11 / 4340 / 8740 hex-head (NAS6603 series), titanium 6Al-4V (NAS6303). Cost 10–100× of commercial.
• Property class for nuts (ISO 898-2): 6 / 8 / 10 / 12 — must match or exceed the bolt class. A class-12.9 bolt with a class-8 nut will strip the nut before reaching proof.

---

5c. Variants & topologies

• Machine screws / cap screws. General-purpose threaded fastener for assembly into a threaded hole or through a clearance hole into a nut. The bulk of mechanical-design fastener selection.
• Self-tapping screws (DIN 7970, ISO 7049 cross-rec): cut/form their own threads in sheet metal or plastic. Type B (sharp point) for sheet metal; trilobular (Taptite, Plastite) for plastic without splitting.
• Thread-forming for plastic (PT, PlastiteR48, EJOT Delta-PT): trilobular cross-section, asymmetric thread angle (30° leading / 60° trailing) that displaces rather than cuts the plastic — preserves strength.
• Studs / threaded rod (ASTM A193 B7, B8, B8M for high-temp pressure-vessel use). Engine head bolts, flanges, anchor rods, pressure-vessel cover bolts.
• Anchor bolts (ASTM F1554 Grade 36/55/105): cast-in-place into concrete; ductile-grade for seismic.
• Shoulder bolts: precision-ground unthreaded shoulder between head and thread for use as a low-friction pivot in mechanisms.
• Set screws: cone, cup, flat, dog, oval point. Hardness HRC 45–53. Use on a flat, dimple, or D-shaft — never against round shaft without dimple, or it walks.
• Aerospace structural fasteners:
  • Hi-Lok / Hi-Lite (Hi-Shear / Lisi): pin + collar; collar shears off at preset torque; ~10–50× the cost of a commercial bolt.
  • Lockbolts (Huck, Avdel): pin with grooves, swaged collar; permanent.
  • Blind rivets (Cherry / Avdel): one-sided installation; structural variants (Cherry MAX, Composi-Lok).
• Threaded inserts (for soft parent material — Al, plastic, wood):
  • Helicoil (wire-formed insert) — original 1937 patent.
  • Time-Sert (solid one-piece thin-wall) — better fatigue behaviour for repair.
  • Keenserts / Keysert (key-locking solid insert) — high-vibration aerospace.
  • PEM nuts / PEM studs — press-fit clinching into sheet metal.
  • Heat-set / ultrasonic inserts (for thermoplastic) — installed by melting brass into the host.
• Tension-control (TC) bolts (ASTM F1852 / F2280): splined-end variant of A325/A490; calibrated wrench shears off the spline when correct preload is reached. Eliminates torque-method uncertainty for steel structural work.

---

6c. Selection criteria

Decisions, in order of impact on the design:

1. Joint duty. Structural slip-critical → ASTM F3125 (A325/A490) with controlled faying surface. Structural bearing → same bolt, simpler installation. Machine joint → ISO class 8.8 or 10.9 SHCS. Mechanism pivot → shoulder bolt. Aerospace primary → Hi-Lok / NAS / MS. Cosmetic → button head / oval head.
2. Loading mode. Pure tension → preloaded class 10.9/12.9; tension + fatigue → preloaded class 8.8/10.9 with high C-control; shear (static) → bearing-type, A325-N or X; shear (cyclic with possible reversal) → slip-critical A325-SC / A490-SC.
3. Material compatibility. Don’t put a class-12.9 steel bolt into pure-aluminum threads — strip imminent. Don’t put a 4.6 bolt in a class-10 nut — bolt yields before clamping. Match nut and washer class to bolt class.
4. Galvanic compatibility. Stainless bolt + aluminum plate + marine atmosphere = aluminum corrodes at the hole. Mitigate: isolating washer (nylon, fibre), coating (passivation, zinc-rich primer at interface), or aluminum-compatible bolt (passivated A2 stainless is the standard aerospace choice — accept the slow galvanic loss). Magnesium + steel is the worst common pairing.
5. CTE match. Steel α ≈ 12 × 10⁻⁶ /K, aluminum 23 × 10⁻⁶, titanium 8.8, magnesium 26. A steel bolt clamping a thick aluminum stack from cold to hot: aluminum tries to expand 2× more than the bolt, increasing clamp force; cold soak does the reverse and loses clamp. Engine-block design budgets explicitly for this.
6. Environment. Salt spray → 316 / A4 stainless or hot-dip galvanised. Marine submersion → Monel K500 or duplex. Chemical (chloride / sour gas) → Inconel 625 / 718. High temperature (>300 °C) → A193 B7 / B16 / B8. Cryogenic → austenitic stainless (no DBTT).
7. Access / tooling. SHCS needs hex-key clearance equal to the key length + handle; flat-head allows flush. Structural A325 hex needs a 1-3/8” impact gun with 8” clearance. Designers under-account for this — wrench-access drawings exist for a reason.
8. Reusability. Torque-to-yield (TTY) bolts are single-use. Engine head bolts, conrod bolts, many CV joint bolts: tightened past proof into plastic — guarantees uniform preload across the production fleet at the expense of one-shot life. Reuse → broken bolt, almost certainly inside the engine. Always replace per OEM service manual.
9. Cost / availability. M3–M16 class 8.8 zinc-plated is commodity (¢ each from McMaster, Bossard, Fastenal). M20+ in 10.9 or stainless 316 ramps quickly; aerospace NAS / MS bolts are USD 5–50 each. Lead-time on non-stock sizes can dominate a project schedule.

---

7c. Datasheet / head-marking decoding

You can identify a bolt class in the field by the markings stamped on the head:

Marking	System	Class / grade
8.8, 10.9, 12.9 (numeric)	ISO 898-1	as labelled
Plain head, no marks	SAE J429	Grade 2 (low-carbon)
3 radial slashes	SAE J429	Grade 5
6 radial slashes	SAE J429	Grade 8
A325 + maker symbol	ASTM A325 / F3125	structural, σ_u = 120 ksi
A490 + maker symbol	ASTM A490 / F3125	structural, σ_u = 150 ksi
BC, BD, B7, B8, B8M	ASTM A193	high-temp service
A2-70, A4-80	ISO 3506-1	stainless
NAS1351, MS21250	aerospace military spec	high-strength alloy

Nut markings follow ISO 898-2 (4, 5, 6, 8, 10, 12 — must ≥ bolt class) or ASTM A563 (Grade A, C, DH).

Coating callouts on drawings or packaging:

• ZP / ZN — zinc plating, ~5–13 μm; ~96 h salt-spray to red rust.
• CR3 / TR — trivalent chromate (RoHS-compliant, replaced hex-chromate CR6).
• HDG — hot-dip galvanised, ~45–85 μm, +1 thread oversize required.
• GeoMet 321 / Dacromet 320 — zinc-aluminum flake, 480–1000 h salt-spray, low hydrogen embrittlement risk (no acid pickling).
• PA / passivated — stainless steel chemical passivation per ASTM A967.

---

8c. Drive / interface — tightening tools

Bolt installation is half the design problem. Tooling determines preload accuracy, throughput, and cost.

• Manual torque wrench, click type. ±4–10 % accuracy when calibrated, used correctly (smooth pull, hand on the marked grip, single click). Sensitive to operator technique. Click is acknowledgement of applied torque, not of resulting preload.
• Manual torque wrench, beam type. ±2–4 % accuracy; no setting to dial in incorrectly. Slower; visual reading.
• Pneumatic / electric torque-controlled drivers. ±5–10 %. High throughput (production assembly). Often combined with angle monitoring (yield-control). Atlas Copco, Bosch, Cleco, Ingersoll-Rand.
• DC-electric “smart” tools (Atlas Copco Tensor, Cleco Tappex): torque + angle + curve-fit; can detect cross-thread / no-thread / soft joint and flag rejects. ±2–5 % preload in production.
• Hydraulic torque wrench (Hytorc, Enerpac, Plarad). Reaction-arm tool driven by ~70–700 bar hydraulic. ±3–10 %. Field tool for flange bolts up to M100.
• Hydraulic tensioner / bolt stretcher (Hydratight, Enerpac, Boltight). Stretches the bolt axially via a hydraulic ram pulling on a pull-rod screwed onto an exposed stud thread; nut is then run down to seat against the now-tensioned bolt. ±2–5 % preload, friction-independent. Standard on large flanges (M30+ / 1¼”+), wind-turbine tower bolts, pressure-vessel covers.
• Ultrasonic bolt extensometer (Norbar USM, BoltCheck). Measures actual elongation by time-of-flight of an ultrasonic pulse through the bolt length. ±1–3 %; friction-independent. Requires acoustic-grade flat ends and a baseline-length measurement. Used in critical aerospace, nuclear, race-engine builds.
• Turn-of-nut method (AISC 360-22 §J3.7, RCSC Specification). After “snug-tight” (full effort by ironworker with a spud wrench), rotate the nut an additional 1/3 turn (bolts ≤ 4d long) up to 1 full turn (bolts > 8d long). Repeatability ±10–15 % on preload, better than calibrated-wrench torque method in field conditions. Standard practice for steel construction.
• Tension-control (TC) bolts (F1852 / F2280). Splined end is engaged by a coaxial inner socket that resists while the outer socket spins the nut. When the preload reaches the calibrated breakoff (torque × angle), the spline shears off at a machined groove. Visual confirmation that preload was achieved. Cannot be re-tightened or re-checked once spline gone.

---

9c. Real parts & sourcing

Distributors / catalogue houses:

• McMaster-Carr (US) — the dominant US industrial fastener catalogue. Same-day shipping; mid-range price; near-comprehensive coverage M2 – M30, 00-90 – 2-1/2” UN, every common class/coating/head/length.
• Bossard (CH/global) — engineering-grade supplier with kitting / VMI services; strong in EU automotive supply.
• Fastenal (US) — branch-network industrial distributor; lower price than McMaster, smaller catalogue depth.
• Würth (DE/global) — EU industrial-supply leader; strong in trade / installer market.
• Hilti (LI/global) — construction anchors specialty (concrete, post-installed); proprietary HUS-3 / KH-EZ / HSC series.
• Bondhus, Wera, Wiha — premium hex / Torx key and torque-tool brands.
• Nord-Lock Group — wedge-locking washers and Superbolt (multi-jackbolt tensioner).

Part number decoding (ISO):

ISO 4762 M6 × 30 — 10.9 ZP means:

• ISO 4762 → socket-head cap screw standard
• M6 → metric coarse, 6 mm nominal × 1.0 mm pitch
• × 30 → 30 mm long (under-head length for SHCS)
• — 10.9 → ISO 898-1 strength class
• ZP → zinc-plated

Common stocked range:

• M2 – M2.5: electronics / instrument scale; usually only in 4.8 or A2 stainless.
• M3 – M16: machine-design common range; class 8.8 and 10.9 ZP plus A2 / A4 stainless are commodity.
• M16 – M30: heavy machinery, structural; lead time on non-standard lengths.
• M30 – M100+: bespoke / pressure-vessel / wind-turbine; weeks of lead time, hydraulic-tensioner installation.

Aerospace fasteners (NAS / MS / AN designations — Hi-Lok HL18-/HL19-, Hi-Lite, lockbolts, A286 nuts, titanium NAS6303) are stocked by Wesco, KLX, B/E Aerospace, Aviall, and command 10× – 100× the price of equivalent commercial bolts because of lot traceability, source approval (NADCAP), and special-process inspection (X-ray, magnetic-particle, dye-pen on every lot).

---

10c. Failure modes & derating

10.1 Bolt failure modes

1. Tensile yielding under monotonic overload. Caused by preload exceeding proof or addition of external load past separation. Mitigation: 75 %-of-proof preload target, joint-stiffness margin to keep F_bolt below F_p at peak external load.
2. Fatigue failure at the first engaged thread root. The dominant failure mode for cyclic-load joints. The thread root is K_f ≈ 3–5; combined with low base-metal endurance limit at the root (~60 MPa for rolled-then-HT), it sets the design constraint. Mitigation: high preload (suppresses ΔF_bolt via C), rolled threads after heat-treat, generous fillet under the head (no sharp transition).
3. Fatigue under the bolt head. Sharper than thread-root for cut threads, similar for rolled. Often the second-most-common fatigue site. Mitigation: under-head radius per ISO 4014 / 4017; hardened washer to prevent corner-bite into the clamped surface.
4. Thread stripping. Either internal (nut / parent) or external (bolt) thread fails in shear. Length of engagement L_e ≥ 1·d for steel-on-steel is rule of thumb; L_e ≥ 2·d for steel-into-aluminum; L_e ≥ 2.5·d for steel-into-magnesium / plastic — or use a threaded insert. Stripping is a benign failure if the joint is single-bolt (you’ll notice during install); catastrophic if it happens at one bolt in a multi-bolt redundant joint where the others now overload.
5. Galling / cold-welding. Stainless A2/A4 nut on stainless A2/A4 bolt is the classic — the protective oxide film breaks during sliding under load and the bare austenitic surfaces cold-weld. The joint then seizes mid-tightening and the bolt twists off. Mitigation: dissimilar grades (A4 nut on A2 bolt is better than same-grade), nickel-based anti-seize (Permatex 80078, Loctite N-1000) for stainless, copper anti-seize (Permatex 31163) for carbon steel.
6. Hydrogen embrittlement. Class 12.9 / A490 / ≥39 HRC bolts can absorb atomic hydrogen during electroplating (acid pickling, alkaline cleaning, electrodeposition step) and develop delayed brittle fracture days-to-weeks after assembly — characteristic “weekend failure” of overnight-installed fasteners. Mitigation: avoid acid-electroplate finish on high-strength bolts (use mechanical zinc, GeoMet, or oil-blackening); bake at 200 °C × 4 h post-plate per ASTM F1940; or restrict high-strength bolts to chromate-only or as-machined finishes.
7. Stress corrosion cracking (SCC). Class 12.9 / A490 / 7xxx aluminum / high-strength stainless in chloride or sulphide environments. Time-to-failure: days to years. Mitigation: lower hardness (limit to HRC 39 for A490; use A325 instead where chloride exposure is likely); ASTM F2329 / F2833 zinc coatings.
8. Fracture from over-torque + side load. A bolt loaded to ~80 % proof in tension and then bent by misaligned mating surfaces sees combined stress > σ_u. Common in field-assembled flanges where bolt holes don’t quite line up and operator reefs on it. Mitigation: drift pins, slightly oversized holes, spherical washers (Belleville-on-cone) for known angular tolerance.

10.2 Joint failure modes

1. Slip in bearing-type shear joint. Joint moves until bolt body contacts hole wall, then takes load in shear. Acceptable in static-load bearing joints (AISC bearing-type); unacceptable in fatigue-loaded or cyclically-reversing joints (use slip-critical, AISC §J3.8). Slip + reversal → cyclic bolt-body bending → fatigue failure.
2. Preload loss from embedment / settle-in. Asperities on the faying surface flatten under clamp force in the first hours of service. Typical loss: 5–15 % of preload for machined surfaces; up to 30 % for as-rolled, painted, or gasketed joints. Mitigation: re-torque after 24 h to 72 h; budget for the loss in initial preload selection (target 80 % proof if you can’t re-torque).
3. Gasket creep / paint compression. Soft gaskets (rubber, PTFE, compressed-fibre) creep under sustained clamp — preload bleeds off over days-to-months. Mitigation: live-loading with Belleville stacks (provides ~1–2 mm of spring travel to absorb creep), spiral-wound metal-graphite gaskets (less creep), re-torque schedule.
4. Vibration self-loosening (Junker). Transverse vibration of the joint creates a brief slip at the head/nut interface that lets the helix unscrew under preload. The classic Junker test (DIN 65151 / DIN 25201) sinusoidally shears the joint and counts cycles to detectable loosening. Defeats split-lock washers, defeats simple toothed washers. Effective mitigation: Nord-Lock wedge-locking washers (cams under the head bite into the bolt, the wedge angle > thread helix angle so the bolt cannot rotate without first lifting against preload — and it can’t), prevailing-torque nylock or all-metal lock nuts (deformed thread + nylon insert), chemical locking (Loctite 243 medium-strength, 263 high-strength), lock wire / safety wire for aerospace, double-nut (jam-nut classic method, low-cost), or achieve and maintain a high enough preload that the joint never slips in the first place — the design ideal.
5. Fretting at the faying surface. Micro-slip at high contact pressure abrades the faying surface, generates oxide-rich black debris, accelerates fatigue of the parent (not the bolt). Mitigation: higher clamp force to suppress slip; harder faying-surface treatment (carburise, nitride); dry-film lubricants (MoS₂).
6. Galvanic corrosion of dissimilar metals. Bolt and clamped member separated by aqueous electrolyte → anode (more active in galvanic series) dissolves. Marine aluminum hulls with stainless bolts: Al dissolves around the bolt. Mitigation: isolating washer / sleeve (nylon, fibre, fluoropolymer); sacrificial anode; or accept slow loss and budget. Magnesium with anything: always isolate.
7. Thermal-cycling preload drift. CTE mismatch between bolt and stack. Steel bolt clamping aluminum stack: hot → preload rises (Al expands faster); cold → preload falls (Al shrinks faster), can go to zero. Engine cylinder head case: M11 TTY studs torqued to ~95 % proof at room temp; at 90 °C operating, stack expansion adds ~5 kN; at -40 °C cold start, preload drops ~10 kN. Design margin must cover both extremes.

10.3 Engineering judgement on derating

• For safety-critical fatigue joints (turbine flanges, lifting equipment, pressure vessels), always tighten by tensioner or angle-control, never torque-only. Torque-only’s ±25 % preload scatter eats the entire fatigue margin.
• For structural steel (AISC) bolts in bearing-type connections: bolt threads in the shear plane reduce nominal shear strength by ~25 % vs threads excluded — design accordingly.
• For high-temperature service > 300 °C: bolt stress relaxation (creep at constant strain) drops preload over time. Use A193 B7 / B16 for hot service and design for the relaxed preload after the design life.
• For single-bolt clamping joints, the joint cannot redistribute load — apply higher reliability margin (FoS ≥ 2 on proof, vs ≥ 1.5 for multi-bolt redundant joints).
• For wood, plastic, soft aluminum parent: design length-of-engagement for the female-thread shear strength, not the bolt. The bolt is usually overkill; the host is the limit.

---

11. Cross-references

• statics-fundamentals — reactions at joints; transfer of load through a connection
• mechanics-of-materials — combined-loading analysis (tension + shear + bending) of bolt shanks
• beam-theory — flange / plate bending under bolt clamp load; gasket compression as a beam-on-elastic-foundation problem
• materials-steel — chemistry / heat treatment underlying class 8.8 (medium-C Q&T), 10.9 / 12.9 (alloy steel Q&T)
• materials-aluminum — galvanic corrosion of Al with stainless / steel bolts; CTE mismatch in mixed-metal clamped stacks
• structural-analysis — joint connections in determinate / indeterminate frames
• steel-design — AISC 360-22 Chapter J connection design (slip-critical, bearing, block shear)
• vibration-dynamics — Junker test, transverse-vibration loosening mechanism
• joining-welding — alternative joining method comparison (weld vs bolt vs adhesive vs rivet)
• fatigue-analysis — endurance-limit referencing, mean-stress effect on Goodman / Haigh diagrams
• manipulator-design — robot joint and link fasteners; preload management for cyclic-loaded arm joints
• industrial-automation — ISO 898 / 261 / 262 / 965 / 3506 standards family

---

12. Citations

1. Shigley, J. E.; Budynas, R. G.; Nisbett, J. K. “Shigley’s Mechanical Engineering Design,” 11th ed., McGraw-Hill, 2020 — Chapter 8 “Screws, Fasteners, and the Design of Nonpermanent Joints.” Canonical machine-design reference; the joint-stiffness frustum-cone model and the C-based external-load equations follow Shigley §8.5–8.8.
2. Bickford, J. H. “An Introduction to the Design and Behavior of Bolted Joints,” 4th ed., CRC Press, 2008. The specialist text: ~1000 pages exclusively on bolted joints, including stress relaxation, ultrasonic measurement, hydrogen embrittlement, gasketed-joint creep.
3. Norton, R. L. “Machine Design: An Integrated Approach,” 6th ed., Pearson, 2019 — Chapter 14 “Fasteners and Connections.”
4. VDI 2230 Part 1: 2015 “Systematic calculation of highly stressed bolted joints — Joints with one cylindrical bolt.” The European gold standard for analytical bolt-design. Defines the frustum-cone (Rotscher) approach, embedment / settle-in coefficients, fatigue calculation, and the R0-R13 calculation procedure used across German automotive and machinery design.
5. VDI 2230 Part 2: 2014 “Systematic calculation of highly stressed bolted joints — Multi-bolt joints.” Extension to load distribution in flanges / bolt circles.
6. AISC 360-22 “Specification for Structural Steel Buildings,” American Institute of Steel Construction, 2022 — Chapter J “Design of Connections.” LRFD and ASD shear / tension / slip-critical bolt strength.
7. Research Council on Structural Connections (RCSC). “Specification for Structural Joints Using High-Strength Bolts,” 2020. Detailed turn-of-nut procedures, inspection of installed pretension, faying-surface classification (Class A / B / C).
8. Eurocode 3: EN 1993-1-8:2022 “Design of joints.” European equivalent of AISC J; partial-factor format γ_M2 for bolt resistance.
9. ISO 898-1:2013 “Mechanical properties of fasteners made of carbon steel and alloy steel — Part 1: Bolts, screws and studs with specified property classes — Coarse thread and fine pitch thread.” Defines the X.Y class system (4.6 / 8.8 / 10.9 / 12.9) used worldwide.
10. ISO 898-2:2022 “Nuts with specified property classes — Coarse thread and fine pitch thread.”
11. ISO 3506-1:2020 “Mechanical properties of corrosion-resistant stainless steel fasteners — Part 1: Bolts, screws and studs with specified grades and property classes.” Defines A1 / A2 / A4 / C1 austenitic and ferritic stainless grades.
12. SAE J429: 2014 “Mechanical and Material Requirements for Externally Threaded Fasteners.” US/Imperial bolt grades (Grade 2, 5, 8 etc).
13. ASTM F3125 / F3125M-22 “Standard Specification for High Strength Structural Bolts and Assemblies, Steel and Alloy Steel, Heat Treated, Inch and Metric Dimensions.” Consolidated standard replacing A325, A325M, A490, A490M, F1852, F2280.
14. ASTM F1554-22 “Standard Specification for Anchor Bolts, Steel, 36, 55, and 105-ksi Yield Strength.” Cast-in-place foundation anchors.
15. ASTM A193 / A193M-22 “Alloy-Steel and Stainless Steel Bolting for High Temperature or High Pressure Service.” Grades B7, B8, B8M etc.
16. ASME B18.2.1-2022 “Square, Hex, Heavy Hex, and Askew Head Bolts and Hex, Heavy Hex, Hex Flange, Lobed Head, and Lag Screws (Inch Series).”
17. ASME B18.2.2-2022 “Nuts for General Applications: Machine Screw Nuts, Hex, Square, Hex Flange, and Coupling Nuts (Inch Series).”
18. ASME B1.1-2019 “Unified Inch Screw Threads (UN and UNR Thread Form).”
19. ISO 261:1998 “ISO general purpose metric screw threads — General plan.”
20. ISO 262:1998 “ISO general-purpose metric screw threads — Selected sizes for screws, bolts and nuts.”
21. ISO 965-1:2013 “ISO general purpose metric screw threads — Tolerances — Part 1: Principles and basic data.”
22. Junker, G. H. “New Criteria for Self-Loosening of Fasteners under Vibration.” SAE Technical Paper 690055, 1969 — original transverse-vibration loosening study. DIN 65151 / DIN 25201 are the modern standardised test procedures.
23. McMaster-Carr Catalog (current edition, https://www.mcmaster.com) — the standard real-parts sourcing reference for US machine design.
24. Bossard Assembly Technology Manual, Bossard AG — frustum-cone calculations, K-factor tables, embedment factors for common surface combinations.