Shape memory alloys (SMA) are part of the smart materials group. The major quality characteristics of SMAs are the shape memory effect (SME) and superelasticity (SE). Due to these two main effects, the material is able to mechanically deform and maintain the deformed shape until a thermal load is applied. These capabilities make them particularly suitable as actuators.
In the frame of preliminary design analysis, the present study investigates the behavior of a bistable actuator based on shape memory alloys that are characterized by two springs operating in opposition of the phase. At the reference temperature, the SMA springs are subjected to a compressive load to initiate spring transformation in the martensitic phase. However, after removing the mechanical load, the material can only recover a small portion of the residual deformation (3%), which is significantly less than the applied mechanical load. Full recovery is achieved only when thermal loading is applied to trigger the transition from the martensitic to the austenitic phase.
Bistable actuators are characterized by a key feature that addresses a major challenge in SMA actuators. The ability of these actuators to switch between configurations and maintain stability without the requirement of keeping the SMA elements consistently above the activation threshold temperature has been utilized in a skid plate system. The designed actuator can switch from one configuration to another by heating one or the other spring and maintain the balanced position by the mechanical locking system based on a bias spring.
4.1. Aerodynamic and Operative Requirements
The selected case study involves a segment-B vehicle equipped with a front air dam. This component, typically in the form of a vertical plate on the front bumper, helps to separate the airflow at the rear of the car, significantly reducing the vehicle’s drag coefficient. Specifically, it decreases the overpressure area on the underbody mechanics. However, using such a component has some drawbacks, particularly at low speeds. To be effective, the air dam must be positioned very close to the ground, which increases the risk of it hitting the ground and breaking on steep or rugged roads. Therefore, it is advantageous to propose a movable geometry that replaces this component, providing aerodynamic performance comparable to the front air dam at high speeds, while eliminating the risks associated with low speeds by changing its configuration.
Therefore, CFD simulations have been carried out to define the geometries of the above-mentioned configurations. The numerical method consists of steady RANS, uses k-ε for turbulence modeling, and considers an inlet velocity of 140 kph to simulate the behavior of the vehicle at high speeds.
The optimal shape has been achieved by eliminating the front air dam (
Figure 4a) and assuming that a part of the skid plate is mobile (
Figure 4b). This mobile part in the non-actuated configuration reproduces the aerodynamic performances of the front air dam.
The skid plate has been rotated around the hypothesized hinge point, obtaining the two configurations shown in
Figure 5. The red line represents the section of the movable part, while the blue one represents the fixed component.
The three configurations were then analyzed to verify the aerodynamic performances. The aerodynamic results are reported in
Table 1.
Two considerations can be made from the obtained results:
To briefly summarize, at low speeds, the aerodynamic behavior of the activated solution does not differ from the basic model, but it has the advantage of removing the fixed component to avoid frequent failures, given its proximity to the ground. At low speeds, on the other hand, the Cx coefficient is higher; however, due to the decreased speed, the aerodynamic force acting on the skid plate is significantly reduced, thus making the substitution of a fixed skid plate with a mobile one a beneficial option.
4.2. Skid Plate Geometrical, Material, and Boundary Conditions Description
A hinge system is used to connect the actuator to the skid plate and convert the finite linear displacement of the actuator into the rotation of the plate. The assembled system is shown in
Figure 7.
This system has two stable configurations (actuated and deactivated), and no external energy is required to keep the device in one of the other stable positions.
The actuator is required to guarantee a cover rotation of 12°, i.e., an upward displacement of 42.9 mm.
As a preliminary analysis, it has been verified in the CAD software that the 5 mm linear displacement of the actuator is enough for the rotation of the skid plate.
In
Figure 8, the kinematic motion is shown in a section view to better appreciate the details of the mechanism. To rotate the gear wheel attached to the cover, the actuator’s central body moves 5 mm toward the left (green arrow).
The actuator’s linear motion results in a 12° rotation (red arrow) around the hinge (orange in the figure), thereby enabling the cover to be elevated by approximately 42 mm in relation to the vertical axis (
Figure 9).
The individual components of the SMA-actuated skid plate are described hereafter in detail.
4.2.1. Skid Plate
The system consists of gear wheels, cranks, and hinges that allow the bottom plate to be rotated (
Figure 10).
The linear movement induced by the actuator rotates the gear wheel mechanism. The latter moves the crank that moves the plate and rotates it around the center of rotation of the brackets (
Figure 11).
In
Figure 12, the main dimensions of the actuation system are shown.
4.2.2. Bistable SMA Actuator
Bistable SMA actuators are much more efficient in comparison to other actuators, as can be seen in
Table 2.
In terms of the forces exerted, SMA actuators need low current values. Current values required to activate the springs range from 10 ampere to 30 ampere and influence the response time of the actuator because activation times range from 40 s to 3 s. Having a bistable actuator means that the SMA springs do not deteriorate. In fact, once actuation has taken place, the current is removed, and the actuator remains stationary in the second equilibrium position.
The bistable actuator, previously introduced, consists of a central body that provides a guide for the springs and establishes the maximum explicable stroke, the two SMA springs that operate in phase opposition, and two lateral springs that represent the locking system. The actuator assembly including the outer case is shown in
Figure 13.
All actuator geometric properties are shown in
Figure 14. The central body of the actuator has two antibuckling guides that ensure proper compression and expansion of the springs along a given axis. The total stroke of the actuator is 5 mm, and the two equilibrium positions can be seen in
Figure 14a. The two SMA springs have the same dimensions and are shown in
Figure 14b. The locking system consists of two hemispheres, which are shown in
Figure 14c.
The locking system consists of two metal hemispheres connected to two traditional metal springs (
Figure 14c). These springs have a stiffness of 14 N/mm, allowing the actuator to lock and release.
Figure 15 shows a front view of the operation of the locking system to highlight the movement of the hemispheres in the x-y plane. Activation of the SMA springs requires the central body to move along the y-axis. When the SMA springs are activated, the translation and special shape of the central body cause the hemispheres to be pushed outward against the action of the metal spring until they reach their new position. The metal springs then push and lock the hemispheres into the second horse saddle-shaped area until the other SMA spring is activated.
The operation of the entire actuator is based on the transformation of the SMA springs from the austenite to the martensite phase and vice versa.
The thermomechanical properties of the adopted materials, NiTiNOL (SMA springs), steel (central body, hemispheres, and hinges), and ABS (skid plate) are, respectively, shown in
Table 3,
Table 4 and
Table 5.
An image of the FEM model and the applied boundary conditions are described in detail in
Figure 16. The lower cover of the underbody shield is attached to the car’s frame and is allowed to rotate on a 12° axis. By heating one of the two SMA springs (the one on the right), the actuator moves toward the left and rotates the connection with the cover through the gear wheel (red arrow). The connecting rod is only constrained to rotate around the two axes indicated in the figure.
Moreover, an applied displacement preload of 55 mm and a constant temperature of 100 °C has been applied to the SMA springs to allow the complete transition to the austenitic phase.
4.3. Results and Discussion
The objective of this feasibility study was to conduct a preliminary exploration of actuator designs and performance metrics.
In this section, the numerical results of the work conducted are shown. The simulations of the SMA-actuated skid plate thermomechanical behavior can be subdivided into three stages. The first step concerns the force computation required to move the mechanism.
A preliminary analysis has been performed on the mechanism without the use of the SMA springs and, therefore, without UMAT. The analysis has been conducted by imposing a displacement of 5mm, which is the value of the actuator stroke. In this way, the force required to move the skid plate mechanism has been computed.
As can be seen in
Figure 17, the force required for the motion has several peaks, with the maximum value being 35 N. These peaks result from the gearwheel rotation mechanism, which can be seen in the detail on the right.
This analysis has been performed on the model without the introduction of the SMA spring in order to evaluate the load transmitted by the cranks. In
Figure 18, the contour plot of stress distribution on the deformed shapes of the skid plate is shown. The maximum stress value is on the actuator axis at the lateral spring transition.
The second stage has been focused on the evaluation of the maximum force developed by the SMA springs at the martensite–austenite phase transition.
To ensure the required force, a customized sizing of the shape memory alloy spring’s geometry was conducted. Dynamic effects were left out of the simulation since they are assumed to have negligible influence on the global behavior of the actuator.
The analysis has been performed on the SMA spring fully constrained at an edge. A first mechanical load has been applied to the SMA spring in terms of applied displacement (55 mm) in order to bring the spring to the stress value to allow the transformation in the martensite phase. Then, a constant temperature of 373 K has been applied to the SMA springs in order to allow the transition to the austenite phase.
Figure 19 clearly shows the starting temperature (298.15 K) and the heating temperature (373.15 K) to recover part of the deformation. Moreover, the maximum SMA spring force is 740 N, which is higher than the locking spring-induced force. In addition,
Figure 19 illustrates a single shape recovery cycle since the task, at this stage of the design and analysis process, has been to ensure the successful implementation of the bistable actuator using shape memory alloys.
In
Figure 20, the evolution of the volume fraction of martensite in the SMA spring (variable SDV7) is shown for four different timesteps of the analysis (for a spring that is unloaded and cold, for a spring hat is half preloaded and cold, for a spring that is fully preloaded and cold, for a spring that is unloaded, and for a hot spring). These results are a further confirmation of the accurate prediction provided by the UMAT SMA routine.
Figure 20 shows that the transformation to austenite is not fully completed, which enables an actuator to release from its first stable position with less energy input. Moreover, it highlights that the force responses of the SMA springs can be significantly improved by optimizing their geometry.
An experimental study has been performed on the SMA springs to assess their actuation force. Specifically, the designed spring has been subjected to a test campaign to determine the active point, which represents the optimal point of actuation. The actuation system allows the net force to be evaluated.
The fixture designed for testing is depicted in the
Figure 21. The implementation requirements must ensure that during the heating phase, the SMA spring performs a
.
Tests were carried out using a universal testing machine equipped with a 5 kN load cell and an 800 mm stroke at a crossbeam speed of 0.2 mm/s. Three different types of K thermocouples are positioned on the top, middle, and bottom parts of the SMA spring to control the temperature.
Figure 22 displays the experimental set of springs utilized for the active point evaluation. The design of the fixture permits the accommodation of two bias springs and one SMA spring, with the SMA spring positioned between the two bias springs to minimize the moments resulting from its actuation. The bias springs and the SMA spring positioned inside the fixture are represented by yellow arrows, while the three K-type thermocouples are identified by a blue arrow. The compressions of the bias springs and the SMA springs are indicated in red.
To evaluate the performance of the SMA actuation device as a function of increasing temperature, various combinations of the SMA and bias springs, as well as active points, are examined. The SMA spring temperature has been recorded by thermocouples. In addition to these, a fourth external thermocouple, placed approximately 1 cm from the spring, has been inserted to measure the ambient temperature during the test. This measurement makes it possible to quantify the shape memory effect in the SMA. The force values measured and shown in the table represent the force produced by the SMA spring net of the two bias springs.
The evaluation of the SMA force involved examining three different configurations. The SMA spring has been subjected to varying displacements of 55 mm, 59 mm, and 64 mm, while a compression of 45 mm was computed for the bias spring. The tests are performed for four load and unload cycles.
Table 6 presents the findings of the experiment, including the maximum values observed by both the thermocouples and the load cell. After testing all the configurations, it has been determined that they all meet the required force specification. However, the first configuration stands out as the most favorable due to its small footprint, making it the most space-efficient option.
In
Figure 23, the load and temperature histories of the best configuration is shown.
In the third stage, a comprehensive verification of the SMA-actuated skid plate has been conducted. A full simulation with all the components and the complete set of boundary conditions and loading conditions has been performed in seven steps. First, a preload in the SMA spring has been applied up to the full transition to the martensite phase. In these conditions, the skid plate system is kept blocked by the locking mechanisms. Then, the two side springs arranged radial to the actuator axis by 180° have been precompressed by a displacement of 0.529 mm. Once the displacement of the lateral springs has been applied, a step has been created in which the release of the lateral springs takes place.
The fourth step concerns the heating phase. A heating temperature of 373 K has been considered. A lower temperature could not be considered because it would allow the springs to be activated even when not necessary by the temperature reached in the car in operative scenarios. To evaluate the entire mechanism, the functioning of both springs has been validated. Hence, the heated spring is cooled down, followed by a stabilization step of the entire mechanism. Finally, the second spring is heated.
In
Figure 24, the main four configurations of the SMA-actuated skid plate resulting from the seven steps are introduced. It can be observed that the actuator is able to guarantee the necessary displacement of the skid plate during the heating phase of the first SMA spring. Additionally, it is capable of reverting to its initial configuration after the thermal load has been applied to the second spring.
In
Figure 25, the main four configurations of the SMA-actuated skid plate resulting from the seven steps are introduced. The maximum value is computed on the central body of the actuator throughout the entire FEM analysis.
Based on the results of the preliminary analyses, it is evident that the designed SMA bistable actuator can rotate the skid plate system 12° around the fixed hinge, resulting in a 42 mm displacement at the end in the upward and downward vertical directions. According to experimental validation of the SMA springs and actuator dimensioning [
54,
55], the main properties of the SMA springs for the implementation of the complete numerical model have been derived. In conclusion, the force exerted by the SMA springs is 21% greater than the force required by the system. This is an indication of how the actuator can still be optimized by working on the volumetric fraction of martensite, which is indicative of material transformation. However, at the same time, it shows how it can be used for various engineering applications and investigations.