Ultimate axial load capacity of v-band clamp joints.

Matthias A. Müller, Simon BarransUniversity 2009 · DOI · OpenAlex

Abstract

V-section band clamps are critical fastening components in high-pressure systems across aerospace and automotive industries. Predicting their Ultimate Axial Load Capacity (UALC) is essential for safety and performance, yet it traditionally requires expensive physical testing. This study utilizes finite element analysis (FEA) to model the failure mechanics of V-band joints. The results demonstrate that failure modes transition from plastic bending in small-diameter clamps to elastic radial deformation in larger diameters. The developed numerical model provides a reliable predictive tool, validated by experimental data, offering a cost-effective alternative to empirical testing in joint design.

Introduction

V-section band clamps, often referred to as Marmon clamps, are ubiquitous in high-pressure applications, ranging from turbocharger housings to satellite launch shrouds. Despite their importance, engineers have historically struggled to accurately predict the Ultimate Axial Load Capacity (UALC)—the point at which the clamp fails and the joint separates. Currently, determining this limit necessitates costly physical "pull-apart" testing. The interaction between the V-shaped band and the circular flanges involves complex contact mechanics, high localized pressures, and a combination of elastic and plastic deformation. This research aims to bridge this gap by establishing a reliable numerical model to predict failure across various clamp sizes and geometries.

Methods

The study employed a 2D axisymmetric finite element analysis (FEA) using ABAQUS software. The material, AISI 304 stainless steel, was modeled as an elastic-plastic material with linear hardening. To simulate real-world assembly, a "thermal contraction" technique was used to induce an initial tightening load of 6 kN by cooling the band onto rigid flanges. Failure was then simulated by moving one flange axially until the resistance force peaked, representing the UALC. The model was validated through physical experiments on 114 mm diameter clamps using an Instron tensile test machine.

Key Results

The FE model revealed a non-linear relationship between clamp diameter and load capacity. Interestingly, for small diameters (up to 250 mm), the load capacity increases with diameter before beginning a steady decline.

Ultimate Axial Load Capacity vs. Flange Diameter

Experimental validation for a 114 mm clamp showed an average UALC of 75.5 kN, while the numerical model predicted approximately 68 kN. This conservative ~10% difference is attributed to the model's assumption of frictionless contact, whereas real-world friction provides additional load resistance.

Discussion: Shifting Failure Modes

A primary finding of this research is the transition in failure mechanisms based on clamp size. Small clamps (<250 mm) typically fail via plastic deformation, where the V-section bends significantly until the flange slips out. Conversely, larger clamps fail through elastic deformation; the increased material in the circumference allows the band to resist bending but causes it to deform radially (stretching outward) until failure occurs, often without reaching the material's yield strength.

Comparison of failure modes for small (55mm) and large (1500mm) diameter clamps

The model also identified high plastic strain concentrations on the inside of the flat section of the band. This was validated by examining a failed turbocharger clamp, which exhibited a crack in the exact location predicted by the FEA, suggesting the model's utility in predicting fatigue-related failures.

Conclusion

This research successfully developed a finite element model capable of predicting the ultimate axial load capacity of V-band clamp joints. By identifying the shift from plastic to elastic failure modes as diameter increases, the study provides a new predictive framework for engineers. Future work will focus on incorporating friction coefficients and investigating the impact of dynamic loading and thermal cycling on joint integrity.

Summary updated

Abstract

We developed a cellular Bioluminescent Resonance Energy Transfer (BRET) assay based on the interaction of TrkB fused to Renilla luciferase with the intracellular adaptor protein Shc fused to Enhanced Yellow Fluorescent Protein (EYFP). The TrkB agonist Brain Derived Neurotrophic Factor (BDNF) induced a maximum BRET signal as of 10 min with an EC(50) value of 1.4 nM, similar to the other endogenous agonists NT-3 and NT-4/5, 1.5 nM and 0.34 nM, respectively. Interestingly, measure of the BRET signal with increasing expression of Shc-EYFP, in the presence or absence of BDNF, suggested a conformational change of preformed TrkB/Shc complexes rather than Shc recruitment. Furthermore, the Y516F TrkB mutant deficient to bind Shc as well as the kinase-dead K572R TrkB mutant was unable to respond to BDNF and exhibited a lower basal BRET signal than that of the wild-type TrkB receptor, again suggesting a preformed complex with constitutive activity. The double YY706/707FF TrkB mutant in the kinase activation loop also showed reduced basal activity but surprisingly kept its capacity to enhance BDNF-induced interaction with Shc, though with less efficacy. The Trk selective kinase inhibitors K252a and BMS-9 blocked BDNF-induced BRET signal with similar potency (100-150 nM), the preferential c-Met inhibitor PF-2341006 being one order of magnitude less potent. Remarkably, in the absence of BDNF, K252a and BMS-9 also reduced basal activity to the level of the Y516F TrkB mutant, suggesting that these compounds were able to reduce the TrkB constitutive activity. BRET responses of mutants and to kinase inhibitors thus reveal a complex level of interaction between TrkB and Shc and suggest that this BRET assay could be of great utility to test blockers of TrkB signalling in a physiologically relevant context.
BibTeX
@inproceedings{muller2009ultimate,
  author = {Matthias A. Müller and Simon Barrans},
  title = {{Ultimate axial load capacity of v-band clamp joints.}},
  booktitle = {University},
  year = {2009},
  doi = {10.1016/j.cellsig.2009.09.027},
}

Loading related content...

Figure