Bolt Preload and Clamp Force: Fastener Engineering Fundamentals

Bolted joints fail more often than they should — usually not because the bolt broke, but because the engineer didn’t understand how preload works. If you’ve ever seen a bolt “tighten to spec” and then loosen under vibration, this article explains why.

What Preload Actually Is

When you torque a bolt, you’re stretching it elastically. That stretch creates tension in the bolt — this is preload, also called clamp force. The nut and clamped parts are compressed in equal and opposite reaction.

Preload is what keeps the joint together. The compressive force between the clamped parts prevents them from separating under external load and creates friction that resists lateral sliding.

A bolt that isn’t preloaded is just a pin with a nut on it — it doesn’t perform like a fastener.

The Torque-Preload Equation

The relationship between applied torque and resulting preload is:

T = K · F_i · d

Where:

T = applied torque (lb·in or N·m)
K = torque coefficient (also called nut factor), dimensionless
F_i = preload / clamp force (lbs or N)
d = nominal bolt diameter (in or m)

Solving for preload:

F_i = T / (K · d)

The K Factor Problem

The torque coefficient K accounts for friction — under the bolt head, in the threads, and between the nut and mating surface. Its value depends on:

Thread condition (clean, oiled, dry, coated)
Surface finish of the bearing faces
Bolt plating or coating
Whether a washer is used

Typical K values:

Condition	K
Dry, clean steel	0.20
Light machine oil	0.15
Waxed threads	0.10
Galvanized	0.25
Cadmium plated	0.12

This 2.5× range in K means a torque spec applied to “dry” bolts delivers 2.5× more preload than the same torque on waxed bolts. That’s why torque specs must always specify lubrication condition — and why published specs that don’t are incomplete.

In practice, K varies ±25–30% even under controlled conditions. Plan for this scatter.

Stress Area: Why You Don’t Use Nominal Area

Bolt stress is calculated on the tensile stress area, not the nominal circular area. The stress area accounts for the reduced cross-section at the thread root:

A_t = (π/4) · [(d_p + d_r) / 2]²

Where d_p = pitch diameter and d_r = root diameter. For standard UNC/UNF threads, ASME B1.1 tabulates these values directly.

Example: A 1/2-13 UNC bolt has a nominal area of 0.196 in² but a stress area of only 0.142 in² — 28% less. Using nominal area would overestimate strength by 38%.

PartCalc automatically uses the ASME B1.1 tabulated stress area when you enter a fastener URL — not a geometric approximation.

Proof Load vs. Yield Strength

Standard practice targets bolt preload at 70–90% of proof load. Proof load is defined as the maximum load a bolt can sustain without permanent set — it’s roughly 85–95% of yield strength for most grades.

Common Grade Proof Loads:

Grade	Proof Strength	Yield Strength	UTS
SAE Grade 5 (1/4”–1”)	85 ksi	92 ksi	120 ksi
SAE Grade 8	120 ksi	130 ksi	150 ksi
ISO 8.8	600 MPa	660 MPa	800 MPa
ISO 10.9	830 MPa	940 MPa	1040 MPa

Target preload = 0.75 × Proof Strength × Stress Area is a common default for non-critical joints. Safety-critical joints often target 0.90 × proof load and use torque-angle or direct-tension indicators for verification.

The Joint Diagram: What Keeps the Bolt from Loosening

A bolted joint under external tensile load F_ext behaves differently depending on how it’s analyzed:

Below joint separation: The bolt sees only a fraction of the external load because the clamped parts are still in compression. The preload is what makes this work — it acts like a spring that absorbs most of the external load.
At joint separation: The clamp force reaches zero. Beyond this point, the bolt carries 100% of additional load.
Above separation: The bolt now sees the full external load — plus impact, vibration, and whatever else the assembly experiences. Fatigue life drops dramatically.

The takeaway: preload is not wasted. A highly preloaded bolt in a properly designed joint sees very little variation in bolt load under external loading — and lasts much longer in fatigue.

Vibration Loosening

Vibration-induced loosening (self-loosening) is a separate failure mode from preload relaxation. It occurs when lateral slip between clamped surfaces causes the nut to rotate backward. The fix is:

Sufficient preload to prevent slip in the first place (most common solution)
Thread locking compound (Loctite) — chemical prevailing torque
Lock washers — limited effectiveness, not recommended for critical joints
Prevailing torque nuts (nylon insert, all-metal) — good for vibration environments
Wire locking — aerospace standard for critical applications

Loctite is not a substitute for proper preload — it’s a supplement for joints where the correct preload simply can’t prevent slip.

Putting It Together

For any bolted joint, the workflow is:

Know your bolt grade — get the proof strength and UTS
Find the stress area — from ASME B1.1 (or PartCalc)
Set target preload — 75% of proof load × stress area is a reasonable default
Calculate required torque — T = K × F_i × d, using the actual K for your lubrication condition
Check joint separation — make sure preload > external load with adequate margin
Choose locking method — based on vibration environment and consequence of failure

Paste any McMaster fastener URL into PartCalc to get the stress area, proof load, and a full preload calculation for your specific diameter and grade.