Bolt preload is the tension created in a bolt when you tighten it, and it equals the clamping force that squeezes the joined parts together. When you turn a nut or bolt head, you’re stretching the bolt like a spring. That stretch pulls the joint members together, and the resulting compressive force is what actually holds the joint tight. Without adequate preload, bolted joints are vulnerable to loosening, fatigue failure, and eventually coming apart entirely.
How a Bolt Works Like a Spring
A tightened bolt is an elastic system. When you apply torque, the bolt shank stretches slightly while the clamped parts compress slightly. Both the bolt and the joint materials behave like springs, and the preload is the force at which those two “springs” reach equilibrium. This relationship follows Hooke’s Law: the preload equals the stiffness of the compressed zone multiplied by its displacement. As long as the bolt stays within its elastic range, it will keep pulling the joint together with a predictable force.
This spring-like behavior is what makes bolted joints work. The bolt wants to return to its original length, so it constantly pulls the joint members toward each other. If an external load tries to pull the joint apart, the bolt stretches a little more while the clamped parts decompress a little. The joint doesn’t separate until the external load exceeds the preload, which is why getting the right preload matters so much.
Why Preload Matters
Preload does three critical things for a bolted joint: it prevents the parts from separating under load, it resists vibration-induced loosening, and it dramatically improves fatigue life. A bolt with proper preload sees only a small fraction of any external cyclic load because most of that load is absorbed by the decompression of the clamped parts. Higher preload means smaller stress fluctuations in the bolt, which translates directly to longer fatigue life.
Without sufficient preload, the consequences compound. The joint faces can slip, causing fretting wear at the contact surfaces. Research on bolted joints shows that fretting wear at the threads can cause self-loosening even without nut rotation, and the damage worsens with each loading cycle, gradually reducing clamping force until the joint fails. In vibrating environments, this creates a dangerous feedback loop: less preload leads to more movement, which leads to more wear, which leads to even less preload.
How Much Preload to Target
The general engineering target is to preload a bolt to somewhere between 50% and 90% of its yield strength, depending on the application and how accurately you can control the tightening process. API specifications for high-pressure flange bolts, for example, call for preload between 67% and 73% of yield strength, with a maximum in-service load (preload plus working loads) of 83% of yield. Surface wellhead bolts use a more conservative 50% of yield for preload.
These percentages balance two competing goals. You want preload high enough to keep the joint secure under all expected loads, but low enough that the bolt stays safely in its elastic range and doesn’t yield (permanently deform). The right target depends on how precisely you can hit it, which brings us to tightening methods.
Methods for Achieving Preload
There are two fundamental approaches: rotating the fastener or stretching the bolt directly without rotation.
Torque tightening is by far the most common. You apply a measured torque to the nut or bolt head, and friction converts some of that rotational energy into axial tension. The problem is that friction is unpredictable. Roughly 85% to 90% of the applied torque goes to overcoming friction in the threads and under the bolt head, leaving only a small fraction to actually stretch the bolt. Because of this, torque tightening limits you to about 70% of the bolt’s yield strength. The remaining 30% is consumed by torsional stress from the twisting action.
Hydraulic tensioning bypasses friction by pulling the bolt directly with a hydraulic tool, then spinning the nut down to hold that stretch. This allows you to reach 90% of yield strength during the stretching phase, but when the hydraulic pressure is released, the bolt springs back and you lose roughly one-third of the applied load. The net result is about 66% of yield for short bolts, improving to around 80% for longer tie rods where the proportional spring-back is smaller.
Mechanical multi-jackbolt tensioners use small jackbolts arranged around the main fastener to apply force directly without rotating the primary bolt. This avoids both the friction losses of torque tightening and the spring-back losses of hydraulic tensioning.
The Torque-Preload Relationship
The standard equation connecting torque to preload is T = K × F × D, where T is the applied torque, F is the resulting preload force, D is the bolt’s nominal diameter, and K is the “nut factor” that captures all the friction variables. For a 1/2-inch Grade 5 bolt with a typical K factor of 0.2, applying 75 foot-pounds of torque produces about 9,000 pounds of preload.
The nut factor K normally ranges from 0.15 to 0.28, but it can swing from 0.1 to 0.5 under unusual conditions like damaged threads or unknown lubricants. That range is enormous. A K factor of 0.15 versus 0.25 means the same torque produces nearly 70% more preload in the low-friction case. For a 7/16-inch Grade 5 bolt, the recommended torque ranges from 37 foot-pounds at K = 0.15 to 62 foot-pounds at K = 0.25, all targeting the same preload.
Variables that affect K include bolt material and size, plating type, surface finish, corrosion, the presence and type of lubricant, and even how many times the bolt has been previously tightened.
How Lubrication Changes Everything
Lubrication has an outsized effect on the torque-preload relationship, and getting it wrong in either direction is dangerous. NASA testing found that a carefully lubricated bolt achieved preload tensions nearly 70% greater than what the standard formula predicted. At the same service torque, the lubricated bolt generated 1.67 times the expected tension. Even a degreased, dry bolt produced about 20% more preload than the dry-assembly prediction.
This is why consistent lubrication matters. If a torque specification assumes dry threads and you apply lubricant without adjusting the torque, you can easily overshoot the target preload and yield the bolt. Conversely, if the spec assumes lubricated threads and you install dry, you’ll fall well short of the intended preload. The K factor for lubricated fasteners is roughly 0.1, compared to 0.2 for dry fasteners, meaning lubricated bolts need about half the torque to reach the same preload.
Preventing Vibration Loosening
In environments with vibration or cyclic loading, maintaining preload over time is just as important as achieving it during installation. Several anti-loosening strategies exist, each working differently. Wedge washers use a clever negative feedback mechanism: if the bolt starts to loosen, the upper and lower washer halves misalign, which actually stretches the bolt and increases preload. Wedge-locking nuts push bolt tips against internal ramps, creating linear contact between threads that resists rotation. Eccentric double nuts generate strong frictional forces between their convex and concave faces after tightening.
Testing shows that bolted joints with these anti-loosening devices maintained nearly constant preload after 30 vibration cycles, while regular double nuts showed a small preload drop after the first cycle due to stress redistribution. For double nut configurations, the upper nut should carry the larger share of preload. The lower nut gets tightened with a small torque, and the upper nut gets tightened with a large torque, because larger preload in the upper nut correlates with better anti-loosening performance and improved fatigue resistance.
Preload Loss Over Time
Even in a properly assembled joint, preload doesn’t stay perfectly constant. Embedding relaxation occurs in the first hours or days after assembly as microscopic surface roughness on the bolt head, threads, and joint faces flattens under load. This slight settling reduces the bolt’s stretch and drops the preload. Gaskets and soft materials between the joint faces make this worse because they compress and creep over time.
Thermal cycling can also change preload if the bolt and joint materials have different rates of thermal expansion. A steel bolt clamping an aluminum housing will see preload increase as temperature rises (because the aluminum expands more, stretching the bolt further) and decrease as temperature drops. In critical applications, these effects are calculated during the design phase and the initial preload is set high enough that the joint remains secure at the worst-case condition.