Industrial tightening processes play a central role in ensuring the reliability, repairability, and resource efficiency of products that rely on threaded fastener joints. The primary functional quantity in such joints is the clamp force, which governs joint reliability but is normally not measured directly during industrial tightening. Instead, tightening systems are predominantly controlled using torque and angle, so that the achieved clamp force must be inferred indirectly. This relation is strongly influenced by friction in the threads and under-head contact, as well as by structural compliance, embedment, and other internal process conditions. Friction varies between specimens and across operating conditions, which limits the robustness of conventional tightening strategies and poses a central challenge for clamp-force-oriented control. These challenges create a need for physically grounded, dynamic simulation models that explicitly account for the coupled behavior of the tightening system and remain suitable for control-oriented analysis.

Dynamic, control-oriented simulation models are developed and assessed for threaded fastener tightening. The tightening system is represented as a coupled rotational and translational dynamic system, in which thread kinematics, frictional losses, joint compliance, phase-dependent joint behavior, and embedment interact during the tightening process. Particular attention is given to the transition from rundown to seating and elastic clamping, and to transient operating conditions such as acceleration, deceleration, dwell phases, break-away, and motion near zero relative velocity. A structured system-level model is first established to capture the dominant torque–angle–clamp-force behavior of the tightening process, including embedment as an explicit internal state. This framework is then used to compare representative friction-model formulations under tightening-relevant conditions, with emphasis on transient dynamic response, numerical robustness, computational effort, and the practical distinguishability of model responses.

The results show that the dynamic model reproduces the principal qualitative features of realistic tightening processes, including the build-up of torque and clamp force across process phases and the influence of phase transitions and friction on the overall response. The friction-model comparison demonstrates that the influence of model formulation is strongly regime-dependent: differences between formulations are small during continuous relative motion, but become pronounced near zero relative velocity, during break-away, and in stick–slip-prone conditions. Several friction-model formulations can reproduce experimentally observed tightening behavior with comparable accuracy when appropriately parameterized. This indicates that the practical predictive limits of dynamic tightening simulations are constrained not only by friction-model structure, but also by parameter uncertainty, calibration ambiguity, process variability, and limited measurement-based distinguishability. The models are therefore most effective as structured simulation environments for analysis, comparison, and method development rather than as exact specimen-specific predictive twins. Overall, the results provide a foundation for future work on clamp-force estimation, observer design, and model-based tightening control.

The basic mechanics of a bolted joint are well-known and have been studied for a long time. The dominating principle is to represent the parts in a joint as a series connection of linear compression and tension springs. However, traditional models often neglect the tightening dynamics and their interrelation with, for instance the friction or embedment. To study these phenomena further and determine their impact on the tightening process and dynamics, and for developing new tightening control strategies, it is necessary to model a threaded fastener and implement it in a suitable simulation environment.Existing models and experimental data have been studied to find equations that fit the observed behavior. Novel models were combined with standard Modelica components to form a threaded fastener model. The simulation results were compared with tightening data from experiments. This work proposes new models for the first three tightening phases, embedment, and threaded fastener friction. These models are implemented in the modeling language Modelica. The results show that it is possible to resemble a typical threaded fastener tightening with power tools. The friction and tightening phases show the expected behavior, while the embedment model needs further experimental verification. During modeling, the model is susceptible to the chosen parameters. Parameters for the joint stiffness, obtained via the VDI guidelines, needed to be reduced by 30% to resemble the joint in a dynamic simulation.

The assembly of threaded fasteners may seem straightforward. However, there are many factors to consider to achieve quality tightened joints, including the joint material, threaded fastener, and coatings. Additionally, there are many assembly tool types and torque application strategies to choose from. This investigation studies the tightening speed dynamics when using torque as a control method. The clamp force obtained in the joint changes when tightening at high speed or when the speed varies greatly during tightening. This type of tightening is called highly dynamic. Highly dynamic torque control tightening strategies are studied, such as impact, pulse, and inertia-controlled methods, and compared with the continuous drive strategy, which is a standard dynamic torque tightening method. The clamp force and its scatter caused by the torque accuracy in the assembly tool type are investigated for the abovementioned torque application strategies. The study also focuses on the different results obtained from the International Organization for Standardization’s (ISO) 16047:2005 (Fasteners-Torque/clamp force testing) standard compared to a production-like setup.