Preload: The Only Pivot of Joint Reliability

A bolted joint does not rely on how hard the bolt itself is, but on how tightly it presses two parts together. This clamping force is the preload (also called clamping force). The joint's resistance to slip, separation, fatigue, and loosening all rest on the preload.

A common saying in the industry is: "Eighty percent of bolt failures are caused by incorrect preload." This is no exaggeration. Insufficient preload lets the joint loosen rapidly under vibration; excessive preload can stretch the bolt to yield during tightening, or even fracture it on the spot. Therefore, determining preload is not "the tighter the better," but a quantitative calculation process with clear upper and lower limits.

Step 1: Determine the Target Preload

The target preload is determined by two bounds: the upper limit allowed by the material and the lower limit required by function. The target value is taken at a conservative position between them.

Upper Limit: Determined by Yield Strength

During tightening, the bolt is simultaneously under tensile stress and torsional shear stress. To control the combined stress, engineering practice limits the assembly preload as a fraction of the bolt material's yield strength. The common simplified estimate is:

Different tightening strategies correspond to clearly different utilization factors ν:

As you can see, the more precise the tightening strategy, the higher the fraction of preload that can be used safely, and the smaller the scatter. This is the fundamental reason high-safety-class joints prefer the torque-angle method or even yield-point control.

Lower Limit: Determined by Three Functional Conditions

1. Anti-slip condition: the friction between the mating surfaces must exceed the transverse working load; typically the friction reserve factor should be no less than 1.2～1.5.
2. Anti-separation condition: the portion remaining after preload is reduced by the load share must still be greater than zero and retain margin, to prevent the mating surfaces from opening.
3. Anti-fatigue condition: the bolt stress amplitude caused by alternating load must not exceed the safe range of the material's fatigue limit.

Step 2: Back-Calculate Tightening Torque from Preload

Preload cannot be measured directly; on the shop floor it is achieved indirectly by controlling the tightening torque. The relationship between the two is given by the classic torque-preload formula:

The torque coefficient K compactly reflects the friction state; it consists of two parts: thread pair friction and bearing-surface friction. It can be approximated as:

Typical values from engineering practice:

Step 3: Verify the Actual Margin After Assembly

After tightening is completed, preload drops for two reasons: embedding loss and relaxation loss.

Embedding Loss

The microscopic asperities on the thread flanks and bearing surfaces are flattened under pressure, reducing the bolt's actual elongation, and the preload drops accordingly. The embedding amount depends on surface roughness, hardness, and the number of mating surfaces:

Relaxation Loss

At high temperature and under long-term load, both the bolt and the clamped parts creep, and the preload decays slowly. For applications where the service temperature exceeds the material's creep threshold, the residual preload must be calculated against the service life.

In engineering practice, the residual preload ratio is commonly used to assess this: after deducting embedding and relaxation losses, the initial preload should still meet the minimum anti-slip and anti-separation requirements. As a rule of thumb, the short-term residual ratio after assembly should be no less than 80%, and after long-term service no less than 60%.

Step 4: Write Preload into the Process Document

The final step of the calculation is to produce an executable shop-floor document. It should include the following elements:

• Target torque and tolerance: for example "M10×1.25, property class 10.9, with lubricating coating, torque 48 N·m ± 8%".
• Tightening strategy description: torque control / torque-angle / yield-point control, and whether staged tightening is used.
• Friction state definition: whether oil is applied, coating type, and whether reuse is permitted.
• Tightening sequence: for multi-bolt joints, provide a diagonal-cross staged sequence.
• Re-inspection requirements: sampling ratio and re-inspection method (e.g. torque-method re-inspection introduces error; marking method or ultrasonic method is preferable).

Conclusion: Preload Can Be Calculated Precisely

The quantitative calculation of preload is not mysterious; it is a clear chain:

1. Determine the upper limit from the material yield strength and the lower limit from anti-slip/anti-separation/anti-fatigue;
2. Select the utilization factor between the upper and lower limits according to the tightening strategy, to obtain the target preload;
3. Back-calculate the tightening torque from the torque coefficient;
4. Deduct embedding and relaxation losses, and verify whether the residual preload still meets the functional requirements;
5. Write the target torque, friction state, and tightening sequence into the process document.

When these five steps are done properly, joint reliability moves from "by experience" to "designable, verifiable, and traceable."