The “Abnormal bolt pretension is detected during the modifying pretension cycle” warning stops your simulation run and tells you nothing useful. The docs say check material and geometry. Forums say add contact sets. None of that is wrong, but none of it explains the actual failure mechanism.

This post covers how bolt connectors work internally, why the two-step pretension analysis exists, what “abnormal” actually means in this context, and the specific setup errors that trigger it.

What Bolt Connectors Actually Are

SolidWorks Simulation does not model the bolt as a solid body with threads. Instead, it creates a simplified assembly of finite element primitives:

  • Beam element: Represents the shank. Tension-resistant only — the axial rotational degree of freedom is released, meaning the connector provides no bending torque resistance. This is intentional and appropriate for most bolted joint loading.
  • Rigid bar elements: Connect the beam endpoints to the faces at the head and nut locations. The number of rigid bars equals the number of nodes on those faces — which means mesh density directly affects the connector’s force distribution.

For loose-fit bolts (the default): rigid bars connect only to the top and bottom bearing faces.

For tight-fit bolts: additional rigid bars connect to the cylindrical surface of the bolt hole, constraining the shank concentrically. This improves accuracy under shear loading but over-constrains the middle component in sandwich (multi-layer) joints, creating stress concentrations on the hole edges that should not be used directly for strength calculations.

The bolt connector formulation is one-directional for compression: it correctly models bolt loosening under tensile loads, but when external loads cause the bolt to go into compression, the force distribution can be inaccurate. For any scenario involving significant compressive loading on the bolt axis — cyclic loads, seismic, impact — model the actual bolt geometry with no-penetration contact instead of using a connector.

The Two-Step Pretension Analysis

When you define a bolt connector with a preload (axial force or torque × friction factor), the solver runs two sequential analyses automatically:

Step 1 — Transfer (Pretension Only): All external loads are suppressed. The solver applies only the user-specified preload to each bolt connector and runs a static analysis to find the equilibrium tensile forces in each connector. This step establishes how much the bolts actually stretch under preload before any external load is applied.

Step 2 — External Loads: The solver computes the difference between the target preload and the tensile force from Step 1 (the “shortfall”). It uses that shortfall as an internally adjusted preload input and runs the full analysis with all external loads (forces, pressures, gravity, thermal) plus the adjusted preload.

This two-step approach ensures that the pretension is self-consistent with the geometry and material stiffness before the external loads are added. Without it, applying the preload and external loads simultaneously produces inaccurate force distribution — especially when flange stiffness is significantly different from what a simple spring model would predict.

One practical consequence: any loads that existed before the bolts were tightened (gravity, spring preloads from adjacent components) should be present in your study when Step 1 runs. They’re included in the pretension equilibrium calculation.

What “Abnormal Pretension” Actually Means

The error fires when the internally adjusted preload values computed after Step 1 are anomalously high relative to what you specified — typically 50 to 200% higher, though the threshold isn’t published. It indicates the solver had to apply an unrealistic correction to achieve equilibrium.

The root causes, in rough order of frequency:

Geometry mismatch: The hole diameter is significantly larger than the bolt connector’s specified shank diameter. The solver creates rigid bar elements based on the face area; a large mismatch creates excessive compliance in the rigid bar assembly, requiring a larger internal correction to maintain tension.

Unintended gaps: Parts that appear touching in the assembly are actually separated by a small gap (sub-millimeter tolerances from assembly constraints). The solver needs large deformations to establish contact before pretension can develop, which reads as abnormal.

Soft material assignments: An incorrect or missing material assignment — especially if a part defaults to a very low modulus — reduces local stiffness to the point where the pretension correction overshoots. Check the material assignment on every component, not just the structural parts.

Sandwich bolt without “Bolt Series” defined: When a bolt passes through three or more components, the automatic Toolbox fastener conversion often misses the middle components. The solver then sees an unconstrained floating component in the load path, which creates force instability. See the Toolbox section below.

Contact definition gaps: If the only contact defined is on the primary load-bearing faces (top and bottom flanges) and you have vertical faces at the bolt location that are also in contact, those unmodeled contact conditions create stiffness discontinuities. The solver tries to compensate through the bolt adjustment.

Mixed solid/shell meshes at the joint: Stiffness discontinuities at the interface between solid and shell elements adjacent to the connector cause the same problem.

Diagnosing the Failure

Before changing anything, extract the information you have:

  1. Run List Pin/Bolt Force after the failed study (if it ran far enough to produce connector forces). The reported axial tension relative to your specified preload shows the magnitude of the correction the solver applied. If the reported tension is 3× your preload, you have a compliance problem in the connector geometry.

  2. Check the mesh around bolt locations. Right-click Mesh > Properties > Aspect Ratio to visualize element distortion. Elements with aspect ratios above 50 at bolt faces should be remeshed.

  3. Enable Contact Visualization before running and confirm that all contact pairs show green (touching) status. Any yellow or red interfaces indicate gaps.

  4. Compare the bolt shank diameter in the connector definition against the hole diameter in the part. If the hole is large enough for a wrench (not just a bolt), the geometry mismatch is causing the error.

Toolbox Fasteners in Sub-Assemblies

Right-click the Connectors folder and select “Convert Toolbox fasteners to bolt connectors.” This works reliably for single-shear joints (bolt through two parts). For three-or-more-component joints (the sandwich case), the conversion fails to populate the Bolt Series parameter for the middle layers.

The result: the solver treats the middle component as unconstrained. In extreme cases this produces displacement values in the billions (the canonical “172 trillion mm” result from forum threads) because the unconstrained component finds equilibrium at an arbitrary location.

Fix: edit each converted multi-component bolt connector individually.

  1. Open the connector definition
  2. Expand Advanced Options
  3. Check “Bolt Series”
  4. Select the cylindrical face of the hole in each middle component (one selection per intermediate layer)
  5. Apply

The Bolt Series parameter tells the solver to constrain those cylindrical faces concentrically with the beam element. Without it, the rigid bar elements from the beam only connect to the outer bearing faces, and the middle components are effectively floating.

For symmetrical models where you’re using half-bolt connectors: the force reported in List Pin/Bolt Force is for half the bolt. Multiply by 2 before comparing against bolt strength specifications. A uniform radial displacement pattern around the bolt flange in the results indicates correct half-bolt setup; a non-radial pattern means the symmetry plane selection is wrong.

Contact Set Strategy

The standard bolt connector creates an implicit bonded contact on the bearing faces (bolt head and nut). For most tension-dominated loading, this is sufficient.

Where it fails:

Compressive loads on the joint plane: If the joint experiences loads that try to separate or slide the flanges, bonded contact at the bearing faces overstates the clamping force contribution. Add a no-penetration contact between the two main flanges so the solver can capture realistic gap opening behavior.

Vertical faces at bolt locations: If parts have features that create vertical contact surfaces near the bolt (bosses, shoulders, ribs), define contact there as well. An undefined contact interface creates a free surface that the solver sees as a stress-free boundary, producing inaccurate load paths near the connector.

For no-penetration contact setup:

  • Source face (Set 1): finer mesh, ideally the smaller face
  • Target face (Set 2): coarser mesh, should be larger and flatter than source
  • Faces must be touching (zero gap) before meshing — confirm with Contact Visualization
  • Avoid grouping 90-degree face pairs in a single contact set; define each pair separately

Changing from local no-penetration contacts to global bonded contact (Connections > Component Contact > Bonded) is a useful debugging step when a study with many contact definitions fails to converge. If the study runs clean with global bonded, the failure was in a local contact definition. Work backwards from there.

Pre-Study Checklist

Run through these before committing to a full mesh and solve:

Material:

  • Every part has a material assigned (check the Part tree, not just the feature tree)
  • Bolt material is set in the connector definition (default is Alloy Steel; verify for non-standard fasteners)
  • No soft materials from imported geometry with default material applied

Geometry:

  • Bolt shank diameter in connector definition matches actual bolt diameter (not hole diameter)
  • All faces that should be in contact show as touching in the assembly
  • For sandwich bolts, middle component Bolt Series is configured
  • Parts that are suppressed or in envelope mode are excluded from the study

Mesh:

  • At least 8 nodes per edge on bolt-contact faces (this determines rigid bar count and force distribution accuracy)
  • Aspect ratios below 50 in the connector region
  • No mixed solid/shell element interfaces at bolt locations
  • Run a draft mesh first, check quality, then refine locally at connectors before full solve

Connections:

  • All bolt connectors are listed in the Connectors folder
  • Contact Visualization shows green (touching) on all defined contact pairs
  • Symmetry boundary conditions are consistent with load direction (axisymmetric loads require axisymmetric fixtures)
  • Fixtures don’t over-constrain the model (rigid body modes check before adding connectors)

Bolt Check Plot: Define the Bolt Check parameters before running — yield strength, shear strength, thread friction factor — so the factor-of-safety plot is available in the results without a re-run.

ICWBoltConnector API

For automated study setup or batch analysis, ICWBoltConnector (in SolidWorks.Interop.cosworks) provides programmatic access. The interface exposes preload values, geometry references, and material assignments. Creation requires persistent reference IDs for face selections.

The two-step analysis runs automatically when bolt connectors with preload are present — there is no separate API call to trigger it. The internal preload adjustment calculation is handled by the solver and is not exposed for direct query.

For parametric studies (sweeping preload values, varying bolt sizes), the standard approach is to modify the connector definition via API, re-run the study, and collect List Pin/Bolt Force results programmatically. This is the same pattern used for automated SolidWorks workflows where study setup follows a template and results are logged rather than reviewed interactively.

When Not to Use Bolt Connectors

The connector is the right tool for:

  • Static studies with tension-dominated bolt loading
  • Studies where modeling the actual bolt geometry would add excessive mesh complexity
  • Parametric preload studies where the bolt geometry is standard

Use actual modeled bolt geometry instead when:

  • External loads cause bolt compression (connectors provide inaccurate force distribution here)
  • Cyclic or fatigue loading where joint separation matters
  • Friction grip joints where pretension magnitude is critical and must be verified against a physical test
  • The bolt itself is the failure mode of interest (bearing area, thread shear)

For the modeled-bolt approach, use no-penetration contact under the head and nut, and either bond or no-penetrate the thread engagement depending on whether the study needs to capture bolt pull-out behavior.

The simulation pretension workflow connects naturally to how parts are prepared upstream. Clean geometry, correct material assignments, and explicit contact definitions matter as much in Simulation as they do in part preparation for DXF export — stale or incorrect properties propagate through every downstream step.