The Integration of Mechanical Energy Absorbers into Rollover Protective Structures to Improve the Safety of Agricultural Tractors in the Event of Rollover

Abstract

The combination of safety belts and rollover protective structures (ROPSs) is key in improving the safety of agricultural tractors in the event of rollover. However, we also have the opportunity to enhance the security provided by each ROPS; one such example is the combination of this safety device with adequate mechanical energy absorbers (MEAs). Inexpensive disc-shaped MEAs can be included in the anchoring points of a ROPS onto the chassis of a tractor. Three configurations of ROPS combined with MEAs were tested during the application of loads that simulated the effects of side rollover in the vehicle. The tested configurations included a blank MEA as a reference case alongside a single MEA and a stack assembly containing both elements. The results of the tests show that both the deformation of the ROPS itself and the strain energy are larger in the case of blank MEAs; thus, there is also a risk that the clearance zone will be infringed upon and that the protective structure will collapse. We can conclude that the implementation of an appropriate MEA in ROPS reduces the deformation of the ROPS itself and its strain energy in cases of vehicle rollover; hence, the safety provided by such protection systems may be improved at a low cost.

1. Introduction

The agricultural sector suffers one of the highest rates of fatal accidents among all the productive sectors [1,2,3,4], and this rate is decreasing at the slowest pace of all the sectors, though it increases on occasion [3,5]. Tractor overturns are the cause of the majority of deaths and severe accidents [1,2,5]. Most fatal accidents occur when old tractors operate without ROPSs [1,2,3] or with ROPSs that have not been configured or maintained properly [4,6,7]. Accidents also occur when ROPSs are used without having passed official tests; such ROPSs account for 5.5% of those installed in agricultural tractors in Spain [3]. Several measures have been proposed and considered for the improvement of safety measures taken by tractor users, including the promotion of programs that retrofit old tractors with ROPSs [4]; training farmers to follow safety recommendations involving safety belts and a ROPS with an appropriate configuration [8,9,10]; devoting public funds to hel** farmers retrofit their tractors with a ROPS [2,4]; and develo** regulations to make safety belts and ROPSs compulsory in tractors [1].

Additional developments have focused on reducing the probability of tractors overturning through analyzing [11,12] and improving the stability of tractors [13], ensuring drivers are conscious of the risk of overturn [14,15], and hel** drivers to avoid risky situations [12]. For example, active front wheel steering control can enhance the lateral stability of a tractor and restore the tractor in the event of a potential rollover, thereby preventing accidents [13]. The influence of ROPSs on tractor stability has also been studied, leading to the conclusion that the stability of cabs is worse than that of two-post ROPSs for narrow-track tractors [16].

In the particular case of narrow-track tractors, the vehicles most suited to agricultural operations in vineyards and orchards in which a ROPS could collide with surrounding vegetation, foldable ROPSs (FROPSs) are commonly installed. Some studies have proved that rollover accidents in this context have happened with the FROPS in the folded position, leading to tragic consequences. In order to overcome the difficulties of unfolding and blocking FROPS in the upright position, some lift-assisting mechanisms have been proposed. Among them, different devices have been used within the lift-assisting mechanism, such as electric actuators [6], gas springs [7], and torsional springs [8]. Additionally, deployable ROPSs have been developed to assist drivers in unfolding the FROPS manually or automatically when the stability of their tractor is compromised [17,18].

In order to reduce the cost and time required to design ROPSs and have them pass official tests, a significant amount of research has been devoted to applying computer-based techniques to such tasks. Finite element analysis has proven to be a successful method of designing ROPSs using virtual tests, the results of which may be close to those obtained in real tests [19,20,21]. Bullet physics engines have also been successfully applied in simulating the behavior of agricultural tractors with a rollover protective structure (ROPS) during rollover events [22].

Studies on rollover accidents involving agricultural tractors show that there are a non-negligible number of fatal accidents during which the ROPS is installed in the tractor and configured in an upright position. However, many reports do not detail the use of safety belts in fatal rollover accidents involving tractors equipped with a ROPS; that said, they will usually state if the ROPS has successfully passed official tests, if the ROPS has collapsed, or if the clearance zone has been compromised. Jarén et al. [3] reported that in Spain, during the period from 2010 to 2019, 91% of 595 deaths caused by tractor rollover either did not involve a ROPS or the ROPS was folded down; this means that 53 deaths could be attributed to a ROPS in the upright position. Facchinetti et al. [4] found that in Italy, in the period from 2008 to 2019, the type of ROPS fitted in the tractor was known in only 30.7% of cases of fatal overturning, which accounted for 434 cases. Among them, 143 of such events involved tractors with a ROPS installed in the upright position: 63 tractors with a cab, 22 with a two-post front-mounted folding roll-bar, and 15 with two- or four-post rear-mounted roll-bars. Arana et al. [23] reported one fatal overturning accident in which the tractor was equipped with a ROPS. In the United States, in 2015, 27 percent of accidents involving tractors already equipped with a ROPS caused injury to an operator [22]. One fatal rollover in which the ROPS collapsed was reported by Rondelli et al. [9], alongside nine cases in which the driver hit the deformed ROPS.

Improving the design and enhancing the safety of ROPSs, as well as implementing stricter tests, is therefore a just and important research objective. Additionally, any reduction in the cost of ROPSs could contribute to an increase in tractors retrofitted with ROPSs and incentivize farmers to install ROPSs in old tractors [2].

ROPSs should be stiff enough to protect the clearance zone and survival volume of the driver, and should be able to absorb enough strain energy to avoid collapse and additional overturn; such requirements may be seen as conflicting objectives [24]. Thus, looking at deformed post-rollover structures, the improvement of ROPSs may be achieved through the fulfilment of the following objectives:

• Incrementing the distance to the clearance zone;
• Augmenting the strain energy;
• Reducing the stress;
• Diminishing the cost, for example, by reducing the mass or reducing the cost of design and validation using computer-based methods, such as FEM.

In this context, some recent trends in the improvement of ROPSs are described hereafter.

Stereovision may help tractor drivers detect obstacles in orchards and vineyards, meaning they can drive safely with unfolded ROPSs; stereovision may also help to stipulate an appropriate shape for the ROPSs made for these environments [25]. In order to overcome the limitations of steel-based ROPSs (in relation to corrosion resistance and bending strength), other materials, such as reinforced polymer composites, have been studied and proposed for building ROPSs [26]. Bhargava et al. [27] studied the best conditions for the bending of profile tubes to minimize wall thinning in ROPSs and, therefore, maintain high force requirements for the collapse of ROPSs. A combination of FEM analysis and a method for the optimization of ROPSs may produce structures with lower mass (and cost). Such a combination may also increase the distance between the deformed ROPS and the clearance zone for similarvalues of absorbed strain energy [24]. FEM analysis has also been combined with an optimization process to define the thickness of the tube of a ROPS for a heavy mining dump truck; a double-nested genetic algorithm search was used to minimize the deformation and weight of the structure [28].

Adding components to the ROPS that can increase its strain energy without compromising its safety zone or inducing a collapse in the protective structure is a promising objective of the current research. An increase in the strain energy of ROPSs can be attained by designing appropriate cab support bolts [29]. To this end, disc-shaped devices have interesting properties, such as a small size, reduced cost, and the capacity to absorb energy under the application of axial loads [30,31,32,33]. Shinde et al. [34] produced a compilation of disc-shaped devices that they called flexural cartridges, analyzing their geometrical characteristics and main applications.

With the purpose of their integration into the anchors of a ROPS in an agricultural tractor, disc-shaped mechanical energy absorbers (MEAs) have been studied and characterized by tensile tests and finite element analysis [35,36]. Their integration in a tractor cabin has also been analyzed by FEM, which involves increasing the potential and kinematic energy of the tractor that a ROPS can absorb as strain energy in case of overturn [37]. This promising approach is investigated in more depth in the present study.

The present study represents the first time that the behavior of an MEA in combination with a ROPS has been analyzed through real tests. In this context, the main contributions of the research presented in this manuscript are as follows:

• The study of elastic and plastic deformation through the tensile testing of an MEA specifically designed to be integrated with a ROPS.
• An analysis of the deformation of a combined MEA and ROPS in order to improve the protection of agricultural tractor drivers in cases of vehicle overturn, achieved by means of a standard static test of ROPSs according to the OECD codes 6 and 7.
• The design of anchoring points for a combined MEA and ROPS in order to isolate the effect of the axial deformation of the MEA in static tests.
• A comparison of the behavior of a combined MEA and ROPS with the deformation of a single conventional ROPS in the same static test.
• An analysis of the effect of two MEAs combined in the same anchoring point of the ROPS, so that we might profit from manufacturing the same MEA in two different configurations of the combined ROPS and MEA.
• A comparison of the behavior of a single MEA combined with two MEAs and the ROPS alone, in order to assess improvements made to the safety of agricultural tractor drivers, which are achieved through the integration of an MEA into the protection system.
• The conclusion that the two tested combinations of ROPS and MEA outperform a ROPS alone in terms of the safety provided to agricultural tractor drivers (as assessed via two criteria: the infringement of the clearance zone and the collapse of the ROPS).

2. Materials and Methods

The main objective of this research was to assess the influence of the combination of a ROPS and MEA on the safety of agricultural tractor drivers in the event of vehicle overturn. The MEA was designed to be installed in the anchoring points of the ROPS onto the chassis of the tractor. The installed MEA was expected to alleviate the effects of tractor rollover on the ROPS, thus improving the capacity of this safety device to protect the driver.

A specifically designed MEA was combined with the ROPS. The performance and behavior of a ROPS alone and two different configurations of combined ROPS and MEA were investigated by means of standard tests [38,39]. Consequently, the deformation and strain energy of the ROPS and MEA were analyzed alongside their ability to protect the clearance zone (in which the tractor driver should remain) in case of overturn.

2.1. Model of the ROPS

The ROPS considered in this research is a U-shaped S275JR steel pipe arc with 1.5 mm wall thickness and a diameter of 37 mm. This arc is similar to the two-post ROPS installed in narrow-width tractors. For the purpose of simplifying the tests, the protective structure tested in this research is smaller than the ones installed in most agricultural tractors, but the qualitative conclusions are general enough to be applied to this same structure, scaled to different sizes. Figure 1 represents the safety zone protecting the driver of a tractor, considering the ROPS.

2.2. Model of the MEA

The MEA that has been chosen to develop the present investigation was manufactured by laser cutting from an S275JR steel plate of 2 mm thickness. This MEA, shown in Figure 2, presents a diameter of 100 mm and is composed of four elements. The two outer elements contain 8 arms, while the two inner stages have 4 arms in a shape cut into its flat face. The aforementioned geometry is succinctly described in reference MEA100W20A8B8C4D4, according to the methodology proposed by Latorre-Biel et al. [36].

The choice of this particular MEA was made at a previous iterative stage, in which finite element simulations of a range of MEAs were carried out, and followed by real tests using a universal testing machine in order to identify the most promising designs. This process has been explained in detail in previous works [35,36,37]. The aim of this design stage was to develop an MEA that is able to absorb a significant amount of energy before breaking, while still presenting a small size. The optimization of the characteristics of an MEA, according to the requirements of a given application, is a future objective of this research group.

The process of manufacturing an MEA, based on laser cutting, melts a volume of steel around the cutting trajectory. As a consequence, the material making up the MEA can reach locally high temperatures, which could produce stress in the material and structural changes in the steel. With the purpose of reducing the influence of the manufacturing process on the behavior of the MEA, an annealing process (normalization) following ISO (2018) [40] was performed. During this procedure, the MEA was heated at a low rate until reaching 825°C, and it was then cooled by forced convection with air. The normalization process was developed in the furnace HOBERSAL MOD. HD-150.

The resulting MEA was tested using a universal testing machine that applied a progressive tensile load in the axial direction. Figure 3 shows the obtained curve of displacement vs. applied load. This information was utilized to calculate the strain energy of the MEA in the tests of different combinations of ROPS and this MEA model.

2.3. Tested Configurations of the Combination of ROPS and MEA

The aforementioned ROPS and MEA were combined in three different configurations, to assess how the introduction of MEA to the anchoring point of the ROPS onto the chassis of the tractor influences the safety of drivers in cases of vehicle overturn.

In the first test, the ROPS integrated one unit of MEA100W20A0B0D0C0, which is a steel disc with a thickness of 10 mm and a diameter of 100 mm without any shape cut inside it. This component is depicted in Figure 4a. It can be considered a rigid or blank MEA whose deformation we expect to be negligible, as was confirmed by testing. This first configuration was developed as a reference case of MEAs that are not deformed; it was used to compare the behavior of the other two configurations tested in this research.

Two units of MEA100W20A8B8C4D4 were used in a second test. Both MEAs were installed in tandem, being separated by an inner spacer and an outer spacer (both 2 mm thick steel rings), to prevent an unwanted interaction between the MEAs that could affect their respective performances. The final configuration, called a stack assembly [34], is depicted in Figure 4b. The purpose of this second case was to investigate the behavior of a combination of two MEAs with identical characteristics, as opposed to the behavior of a single device, in order to profit from the design and manufacture of the same energy absorber.

A single unit of MEA100W20A8B8C4D4, shown in Figure 4c, was combined with the ROPS in the third test.

2.4. Tests to Analyze the Behavior of the Combination of ROPS and MEA

In this research, with the purpose of assessing the safety provided by two different combinations of MEA and ROPS in case of vehicle overturn, static tests were developed. The applied tests followed the guidelines provided by the OECD codes 6 and 7 [38,39], which are based on the early and influential works of Chisholm [41].

In contrast with the static tests, dynamic tests were developed some decades ago to assess the behavior of ROPS in agricultural tractors. In dynamic tests, a load proportional to the reference mass of the tested tractor was applied in a process that was considered unsafe and difficult to implement. The practice of these dynamic tests has almost stopped, as the danger inherent to them increased with the mass of commercial tractors over the years. For this reason, in many countries, such as the USA and Spain, static tests that follow international standards have replaced dynamic ones in official tractor test laboratories, based on the fact that they are considerably safer and have proven effective in saving lives.

In the present research, a single type of protection arc was combined with MEA in three different configurations, with the purpose of analyzing the absorption of strain energy and the protection of the clearance zone (in which the tractor driver should remain) in cases of overturn.

To test these combinations, specific tools were built and installed in a hydraulic press. This press operates in accordance with the requirements applied to the standard tests of ROPS, as detailed by codes 6 and 7 of the OECD [38,39]; it does so through the specific design of the elements anchoring the ROPS to the test bench, and the appropriate selection of the point at which the force is applied.

The ROPS were anchored to the testing bench at both ends, as shown in Figure 5. In the case of rollover, only the anchoring point on the side to which the force was applied suffered a tensile load, while at the opposite anchoring point, compression force was applied. Subsequently, the only MEA that deformed was the one under the tensile load. In a real application, the design of the anchoring system that contains the MEA would prevent deformations under compression forces in the axial direction [37]. Given all these considerations (and even though in a real application, both anchoring points would incorporate an MEA), a simplified arrangement was designed for this test, where only the anchoring point of the ROPS on the side to which the external force was applied would contain an MEA, as can be seen in Figure 5.

The development of the tests of the arcs was based on a description of the load from the side test in codes 6 and 7 of the OECD standards [38,39]. These codes prescribe that the load application point shall be placed at the point on the roll-over protective structure that is likely to hit the ground first during a sideways overturning accident. Figure 5 depicts the force application point considered in the tests carried out in this research. The application of the side load produced a displacement of 1 mm/s, which is lower than the value of 5 mm/s defined in the aforementioned standards; hence, the tests performed can be considered static tests.

The metrics measured in the tests were applied force and displacement at regular time intervals. With this information, it was possible to calculate the strain energy and display the applied load vs. displacement and the strain energy vs. displacement curves. A key objective of the tests was to compare the behavior of each combination of ROPS and MEA in terms of displacement and strain energy.

The tests were performed to determine the deformation of an arc caused by the application of a side load. Both anchoring points of the arc with the test bench could rotate freely, and one of them was combined with an MEA. Taking this approach, it was possible to combine two different effects produced by the load applied to the arc, as depicted in Figure 6:

• A rotation induced by the deformation of the MEA (depicted by a red arrow in Figure 6);
• A deformation of the arc produced by the application of the external load (represented by two green arrows in Figure 6).

2.5. Mechanical Test Bench

The mechanical test bench upon which the safety arcs combined with MEA were tested was composed of a hydraulic piston (which was able to apply an external load to the tested ROPS) and a resistant structure (which was able to withstand the application of the side force with negligible deformation). The hydraulic press is depicted in Figure 7. It was manufactured by TRECALSA in Parma, Italy, and it is a SICMI model PSS 40, reference 7279, with 40 tf (metric tons of force), which corresponds to 3.92 × 10⁵ N. The press contains a double-acting hydraulic cylinder with a stroke length of 500 mm and a maximal pressure of 35 MPa.

The hydraulic cylinder was used to apply the external load to the tested arc following the procedure described in code 6 and code 7 of the OECD standards. The adaptation of the press to perform the tests required the design and construction of specific tools, as shown in Figure 7, which are composed of the following elements:

(a) A frame to anchor the ROPS and MEA to the test bench:

This is a steel frame composed of a hollow structural section (HSS) made of structural steel of 100 mm × 50 mm and 4 mm thickness, which presents high resistance. The deformations suffered by this frame and by the press structure during the development of the tests were found to be negligible when compared to the deformations of the tested safety arcs.

(b) A hydraulic cylinder to apply the external side load to the tested arc:

The hydraulic piston of the press was used to apply an external load to the tested ROPS. The force was applied very slowly; hence, the deformation of the safety arc under the influence of the mentioned load was around 1 mm/s. The specifications of the hydraulic cylinder state a maximal force of 3.92 × 10⁵ N, which is much higher than the load required by the developed tests.

(c) Instrumentation to measure the applied load and the displacement of its application point in the ROPS:

The applied force was measured by a strain gauge manufactured by the company UTILCELL, Mod. 350 Son:1541508(17)I, which has a precision of 0.9807 N (0.1 kgf or kilopond).

The displacement of the tested arcs was measured by a laser distance meter commercialized by LEICA, model DISTO d3A, and presents 1 mm precision.

(d) Anchoring devices:

The fixed ends within the anchors of the ROPS onto the chassis of the vehicle were designed specifically for these tests. Their design allowed the researchers to study the behavior of the protection arc, combined with the corresponding MEA and the ROPS anchors, avoiding the influence on the axial deformation of the MEA. The real shear stress that could appear in the anchoring points of a ROPS during static tests or vehicle overturn could not influence the behavior of the MEA, because the anchoring device (in which the MEA would be installed) would absorb the shear stress [37].

One of the anchoring elements exhibited free rotation, while the other allowed for rotation and the axial displacement produced by the deformation of the MEA. The same anchoring devices were considered in the reference test with a blank MEA, as well as in tests with coupled MEAs and a single MEA, respectively.

The anchoring mechanism, shown in Figure 8, allows for the free rotation produced by the deformation of the MEA placed in the other anchoring device, and the rotation produced by the deformation of the safety arc itself. By avoiding, in this way, the fixed end of the arc, it is possible to independently analyze the strain energy absorbed by the arc and the MEA.

The second anchoring device allows for the free rotation of the arc to avoid the impact of a fixed end, and it also facilitates the displacement produced by the deformation of the MEA included in this anchoring mechanism, as shown in Figure 9.

Furthermore, the mechanical device in which the MEA was installed featured a limit on displacement in order to prevent the excessive deformation of the MEA, which could produce fracturing. This displacement limiter is shown in Figure 10.

2.6. Application of the External Load

The mechanism designed for the application of force follows the prescriptions of code 6 and code 7 of the OECD standards. In order to apply such force, a steel cable was implemented to pull the arc in a direction parallel to the line containing both anchoring points of the arc with the test bench. Figure 11a represents the aforementioned mechanism.

2.7. Measurement of the Displacement and the Applied Load

The strain gauge provides a continuous measure of the applied force. Additionally, the laser distance meter measures the distance from the fixed point to the plumb, as it is designed to show its lower face equidistant from the load application point. Figure 11b shows part of the instrumentation implemented in the test bench to carry out the aforementioned measurements, as well as the plumb at which the point’s displacement was measured. The resulting data were exported to a computer file and analyzed in a subsequent stage of the research process.

The developed tests concern three different combinations of ROPS and MEA. Hereafter, they are called tests 1 to 3.

2.8. Test 1

A ROPS combined with a blank or rigid MEA, which experiences negligible deformation, was tested.

Figure 12a shows the deformation of the ROPS combined with a blank MEA, and a simplified and approximated representation of this deformation as an orange polygon.

As it can be seen in Figure 12b, the blank MEA did not experience any deformation; hence, in test 1, there was no rotation of the whole arc around the other anchoring point as a consequence of MEA deformation.

We stored a set of more than 300 measured values of displacement and applied force that resulted from test 1. The measured applied force and the subsequent calculated strain energy values were lower than those reached in the usual tests for most commercial tractors with ROPS. These values are a consequence of the use of a ROPS of reduced dimensions. However, the developed devices and procedures, as well as the test results, led to qualitative conclusions that validate the methodology and the concepts involved; hence, they allow for the further development of this methodology in the future for its subsequent application to commercial models of heavier tractors.

2.9. Test 2

In this test, we assessed a ROPS combined with two units of MEA100W20A8B8C4D4 installed in a stack assembly.

Figure 13 shows the deformation of the ROPS combined with an arrangement of two identical MEAs, alongside a simplified and approximated representation of this deformation as a green polygon. After this test, the two MEAs and the ROPS were deformed, as seen in Figure 13. As a result, both the ROPS and the set of two MEAs absorbed strain energy. As can be seen in Figure 13b, the stack assembly of both MEAs did not exhibit the maximal deformation allowed by the displacement limiter of the anchoring device.

The strain gauge and the laser distance meter were used to carry out measurements, and these data were registered; we thus obtained a curve representing the applied load vs. the displacement of the load application force.

2.10. Test 3

A ROPS combined with a single MEA100W20A8B8C4D4 was tested.

Figure 14a shows the deformation of the ROPS combined with an MEA, and a simplified and approximated representation of this deformation as a white polygon. In Figure 14b, the final axial deformation of the MEA can be seen in greater detail. The MEA in test 3 experienced the maximal deformation allowed by the displacement limiter installed in the anchoring device, thereby preventing the breaking of the MEA.

The data obtained from the measurements in test 3 were registered, and these values were then transformed into an applied load vs. deformation curve.

3. Results

During the tests carried out to investigate the behavior of three different configurations of ROPS and MEA, the applied load and the displacement of the force application point were determined.

In test 1, the blank MEA did not show any influence on the deformation of the ROPS, and its strain energy was negligible in proportion to its negligible deformation. In the other two tests, axial deformation of the MEA took place. Furthermore, the axial deformation of the single MEA integrated in test 3 was larger than that of the stack assembly of the two MEAs analyzed in test 2.

3.1. Applied Load vs. Deformation Curves

The three tests were carried out until maximal displacement (300 mm) of the force application point was reached. Using the data derived from each of the tests, an applied force vs. displacement curve was obtained. A comparison of the curves corresponding to the three tests is shown in Figure 15.

From Figure 15, the following conclusions can be drawn:

(a) The data obtained from test 1 show an applied load vs. displacement curve similar to those from tests of standard two-post ROPS;

(b) The arrangement of a ROPS and a single MEA (as in test 3) presents larger values of displacement for a given applied force.

(c) Test 2, involving a combination of two MEAs, produced intermediate values of displacement.

We can, therefore, conclude that the compliance of the combination of ROPS and MEA is higher for a single MEA and lower in the case of a blank MEA. The stack assembly of two MEAs shows intermediate compliance.

3.2. Shape of the Deformed Combination of ROPS and MEA

The deformed shapes of the combined ROPS and MEA produced by the tests present us with important information about the behavior of safety devices and the protection they provide to tractor drivers. These deformed arcs (after the development of the different tests in which the same displacement of 300 mm was reached) are considered in this section in our assessment of the safety devices’ behavior. Furthermore, with the purpose of clarifying the behavior of the arcs and facilitating the measurement of geometric parameters, simplified representations of the deformed and rotated arcs are depicted in Figure 16. This simplification is achieved by drawing straight lines for each side and the top and bottom of the deformed ROPS, following the axial direction of the different sections. In all three cases, the resulting shapes show different approximations of the clearance area.

As can be seen in Figure 16, in the configurations that include a more rigid MEA (i.e., tests 1 and 2), the deformed ROPSs are closer to the clearance area than the combination of a single MEA and an arc (i.e., test 3), thus increasing the risk of the tractor driver’s protection zone being infringed upon.

3.3. Strain Energy vs. Displacement Curves

The area under the applied force vs. displacement curve, as shown in Figure 15, is equal to the strain energy of the combined ROPS and MEA in the three investigated configurations. From Figure 15, it can be concluded that the strain energy is different for each configuration:

(a) For a given displacement value, the strain energy is larger for the combination of a blank MEA and safety arc (test 1).

(b) For a given applied load, the strain energy is larger for the combination of a single MEA and arc (test 3).

(c) The combination of two MEAs and an arc (test 2) presents intermediate values of strain energy that fall between the values of the other two configurations.

A curve representing the strain energy of the tests vs. the displacement of the force application point is depicted in Figure 17.

In order to compare the effectiveness of each configuration as it pertains to the strain energy of the combination of ROPS and MEA, the integrity of the ROPS, and the protection provided in case of rollover, the common value of strain energy to be reached in all tests was fixed to 500 J. It has to be taken into account that OECD codes 6 and 7 [38,39] prescribe a certain energy to be reached in the static tests that assess the behavior of ROPSs.

From the strain energy vs. displacement curve shown in Figure 17, one can observe the displacement corresponding to this reference value of 500 J strain energy in each test. These values of displacement can be found in

Table 1. Displacement and strain energy of the ROPS and MEA after tests 1 to 3, as assessed until a total strain energy of 500 J was reached by the combined ROPS and MEA.

	Total Energy (J)	Arc Displacement (mm)	MEA Axial Displacement (mm)	MEA Strain Energy (J)	ROPS Strain Energy (J)	ROPS Strain Energy (%)
Test 1	500	227	0	0	500	100
Test 2	500	244	42 *	60.1	439.9	88
Test 3	500	283	64	60.9	439.1	87.8

* Displacement of each one of the two MEAs included in the stack assembly.

. Table 1 also displays the final axial displacement of the MEA, which was integrated into the anchoring point of the ROPS with the test bench. From this information, the strain energy of the MEA could be calculated; therefore, the strain energy of the ROPS alone could be obtained for each one of the three configurations. Table 1 contains all these values.

3.4. Geometry of the Arcs for a Common Value of Strain Energy

In the next step of the analysis, the geometries of the different configurations of ROPS and MEA were studied, considering the reference value of strain energy, which was 500 J.

Figure 18 represents a comparison between the shape of the deformed combination of ROPS and MEA, given a strain energy of 500 J, at the end of tests 1 to 3.

The geometric result of the deformation experienced by the ROPS is presented by the angle between the side and top bar of each one of the polygons portraying the shape of the ROPS. The different measured angles for each one of the tests (and a reference value of 500 J strain energy) can be found in Figure 19.

4. Discussion

An MEA is a device conceived to be inserted into the anchoring points of a ROPS, with the purpose of alleviating the effects of tractor overturn by absorbing the kinematic and potential energy of the tractor in the form of strain energy.

We performed several tests in which three different MEAs were installed in one of the anchoring points of an arc-shaped ROPS; these MEAs were smaller than those implemented in most commercial tractors. Our qualitative conclusions validate the original conceptual methodology, being applicable to commercial tractors for which the involved forces and strain energy may be larger.

Two key factors determine the protection that a ROPS provides to agricultural tractor drivers in cases of rollover: the preservation of the clearance zone through avoiding its infringement and the integrity of the ROPS that prevents its collapse.

4.1. Preservation of the Clearance Zone

In tests 2 and 3, where flexible MEAs were implemented, rotation of the ROPS around one of the anchors of the ROPS took place, as can be seen in Figure 13, Figure 14 and Figure 16. The aforementioned rotation occurred around the anchor opposite the ROPS (where the force was applied during the tests). This rotation did not occur in test 1, in which a rigid disc or blank MEA was installed at the anchor of the ROPS, as shown in Figure 12 and Figure 16. This rotation may lead to an improvement in the safety provided by the ROPS to agricultural tractor drivers, as is justified in the next paragraph.

Figure 16 shows that the final geometry of the deformed arc is closer to the safety zone in the case of ROPS alone than the combined ROPS and MEA. Consequently, the introduction of the MEA in the ROPS means there is less risk that the deformed protection structure will infringe upon the clearance area (when compared to the risk of using only a single ROPS). Preserving the integrity of the safety area is one of the key factors in the safety of agricultural tractor drivers in the event of vehicle overturn. Therefore, as can be seen in Figure 16, the combination of the MEA and ROPS may improve the safety of drivers.

4.2. Prevention of ROPS Collapse

Figure 18 shows that the final position of the deformed arc, and therefore its distance to the clearance zone, were similar in the three configurations tested. However, the energy absorbed by the ROPS was smaller in the configurations of tests 2 and 3 (into which flexible MEAs were integrated). This claim is justified by the following observations:

• The final geometries of the three cases, as shown in Figure 18, were reached in the instant that the strain energy of the combined ROPS and MEA reached the reference value of 500 J. In the case of the blank MEA, in test 1, the ROPS itself absorbed all of this energy. Nevertheless, in the other cases, tests 2 and 3, in which a single MEA or a couple of MEAs were deformed, the MEA absorbed a proportion of this energy; hence, we know that the strain energy of the ROPS was less than 500 J. Table 1 shows that the strain energy of the MEAs in tests 2 and 3 is around 12% of the total strain energy of 500 J, meaning the strain energy of the ROPS was around 88%, representing a large reduction in the energy absorbed by the ROPS alone in test 1.
• Figure 19 shows the angles defined by the side bar containing the force application point and the top of the ROPS, when the strain energy reached 500 J in the three configurations. As can be seen in Figure 19, the ROPS alone exhibited a larger angle, meaning that the joint between the side and the top of the ROPS with a blank MEA showed larger deformation. Moreover, configurations with a single MEA or a couple of MEAs, as obtained via tests 2 and 3, present smaller values. This joint is a critical point because the concentration of stress at this point favors the breakage of the structure.

From previous findings, it can be stated that the ROPS alone absorbs more strain energy than the combination of a ROPS and MEA, which has several implications:

• Larger values of strain energy imply that the ROPS alone (test 1) reached larger stress values than the combined ROPS and MEA (tests 2 and 3).
• Since the ROPS alone reached larger stress values, it came closer to collapse than the combination of ROPS and MEA.
• The amount of strain energy in the combined MEA and ROPS before its collapse was larger than that in the ROPS alone; hence, the former will be better equipped to protect heavier tractors than the latter, in accordance with OECD codes 6 and 7 [38,39].
• A configuration with a ROPS and a MEA can offer protection against rollover events, during which a tractor has greater kinetic and potential energy than the configuration with the ROPS alone; this is because the former can absorb more strain energy than the latter.
• When combined with an appropriate MEA, it is possible to reduce the thickness of the wall of a ROPS and, therefore, use a cheaper ROPS with a similar level of protection, as opposed to using only a more expensive ROPS with thicker walls.

The reduced risk of ROPS collapse is a second key factor in the improvement of tractor drivers’ safety in the event of vehicle overturn. This reduction can be achieved through the integration of an MEA in the anchors of a ROPS.

4.3. Comparison of MEA Arrangements

Using the present research, it is possible to compare the performance of the two configurations of MEA integrated with ROPS. The performance of both the single MEA and the stack assembly of two MEAs was very similar. Nevertheless, slight differences appeared in the magnitude of their deformation as well as in the strain energy. The combination of a single MEA and ROPS came closer to the clearance zone when the total strain energy was fixed to a common value, as can be seen in Figure 18. However, if the common factor is the maximal displacement of the force application point in static tests, Figure 16 shows that the arrangement of two MEAs was the configuration that came closest to the safe zone.

Regarding absorbed strain energy, Table 1 shows that the ROPS itself absorbed a slightly smaller amount of energy in the arrangement with a single MEA than with a stack assembly of two MEAs, at 88.7% of the total energy vs. 88%.

As shown in Figure 19, the angle between the side and top bars of the deformed ROPS was slightly more favorable and smaller when utilizing the combination of a couple of MEAs with a ROPS than when utilizing a single MEA: 104.5° vs. 105.3°.

This research represents a step forward in our mission to select the optimal MEA that will engender the most improvements in the safety of any given ROPS and tractor. The combination of MEA arrangements and ROPSs reached a technological maturity level of 5, according to the TRL (Technology Readiness Levels) scale of the U.S. Government [42]. A possible next step could be to study the behavior of different configurations of MEA and ROPS as well as materials, following the work of several researchers using similar devices for other applications [43,44].

5. Conclusions

5.1. Experimental Evidence of the Safety Provided by a Combined ROPS and MEA

We compared the behaviors of two configurations that combine a single MEA and a stack assembly with two MEAs with the behavior of a ROPS alone (as a reference case following OECD codes 6 and 7).

Two key factors that influence the protection that a ROPS provides to the driver of an agricultural tractor were investigated via the following two metrics: the preservation of the clearance zone and the prevention of ROPS collapse.

The three sets of results from the tests of these metrics provide experimental evidence that the combination of ROPS and MEA improves the safety of agricultural tractor drivers in the event of vehicle overturn when compared with a conventional ROPS alone.

(a) A ROPS alone is more likely to infringe upon the clearance zone in cases of vehicle overturn because its final geometry grew closer to the safety zone than any of the tested configurations with an MEA. Hence, integrating an MEA in ROPSs may reduce the risk of the clearance zone being infringed upon in the event of an overturn.

(b) The ROPS combined with an MEA absorbed 12% less strain energy than the ROPS alone, which implies less deformation of the ROPS and, therefore, less stress. Higher values of stress in the ROPS alone will increase the probability of collapse.

(c) The locations of U-shaped ROPSs in which the geometry changes abruptly produce concentrations of stress; hence, they are more likely to initially fail in cases of tractor rollover. The deformed ROPS presented an angle between the side and top bars that was larger in the case of ROPS alone; this joint suffered greater deformation (and, subsequently, higher stress), leading to a higher risk of the ROPS alone collapsing.

The tests did not show significant improvements in the performance of the ROPS as a result of integrating a single MEA or an arrangement of two MEAs in a ROPS. The parameters and criteria considered to accurately reflect the safety provided by tractor drivers were similar.

5.2. Expected Impact of This Research

Even though the number of fatal accidents involving agricultural tractors has been largely reduced in recent decades due to the use of safety measures such as safety belts and ROPSs, such fatal accidents, many caused by tractor overturns, continue to occur. The development of safety measures (for example, improving ROPS through their combination with MEA) will allow for further reductions in fatal injuries sustained in tractor accidents.

Additionally, implementing inexpensive MEAs at the anchoring points of ROPSs to the chassis of agricultural tractors could have the following consequences:

(a) The ROPS itself could be manufactured using fewer materials and, therefore, at a lower cost.

(b) The installation of an MEA in combination with a ROPS could allow the complete protective structure to pass official tests.

(1)

A ROPS without an MEA may still fail such tests.

(2)

These complete protective structures may also pass exigent tests that demand a higher mass than the reference mass, thereby providing users of these protective devices with greater safety. These more challenging tests could be designed by specific manufacturers or be a component of future regulations.

We can conclude from the present study that the improvement of ROPSs using MEAs in agricultural tractors is a very promising area of investigation that may enhance the safety provided by safety structures and allow for agriculture tractors to be protected by less costly ROPS. The manufacture of disc-shaped MEAs, such as those described in this research, is inexpensive, as well as the adaptation of the anchoring devices of ROPS to include one or several MEAs. It is common that present commercial ROPSs are suspended on elastomer silent blocks, and their substitution by MEAs would not require significant or costly modifications to the anchoring points.

5.3. Future Research Lines

Future research studies could undertake methods of optimizing the design of MEAs; such studies should consider the materials, manufacturing processes, geometry, and particular applications indicated by specific ROPS and tractor models. Moreover, FEM analyses and real tests of commercial ROPS for heavy tractors could be carried out in order to obtain more experimental data and grow both our knowledge base and the range of applications of this technology.