Computational Study of Fatigue Fracture in Rivet Housing of an Aeronautical Aluminum Alloy 7075-T6

This article describes computerized traction testing through the ANSYS software in aircraft materials, aluminum 7075-T6, which leads to variations of the strength limits with the stress concentration factor in the rivet cavity. However, was performed with a change in the radius of transition from the head to the body of the rivet in order to alleviate the stress concentrations within the rivet bore and improve the strength of the part by 5.25% with no relevant change in part displacement. The fatigue strength limit was higher in the samples without of the burr operation, observing that the geometric trunk of the section generated a decrease of 18.27% in the maximum plate stresses, with better uniformity in the tension distribution of the rivets, with a displacement of 3.125% and a minimum equivalent stress of 2.39% and a maximum of 24%, justifying the cause of the fracture in the riveting region.


INTRODUCTION
Technology in this industry is growing every year and one of the main focuses of research on aircraft is in relation to its weight, since fuel consumption is directly linked to the weight of the airplane. Therefore, a reduction of this aspect would entail a more competitive price of tickets since fuel expenses would be optimized. The importance of reducing the weight of aircraft is noticeable, so they need to be the lightest and most durable, built of materials with high mechanical strength, so the use of aluminum in aeronautics has become essential because of its advantage in comparison to other materials when it comes to the relation weight as a function of the resistance [1]. These materials have as a positive point the resistance to corrosion and electrical and thermal conductivity, have good machinability and be a fragile material, yet all these characteristics do not deprive it of suffering fractures, particularly fatigue fractures when the efforts that the aircraft are submitted are so oscillating, whether due to loading, pressurizing or even the external environment that is exposed. In the study of Aluminum 7075 -T6 you can see a great range of benefits, compared to other materials, it admits thermal treatments, is of excellent toughness, weldable and recyclable in the general industrial environment. Used in the manufacture of aircraft structure and other requirements of high strength, strong corrosion resistance of the high stress structure, such as aircraft, under the wing panel, stringer frame, etc.. After solution treatment, good plasticity, heat treatment strengthening effect is particularly good, in 150ºC high strength and low temperature strength particularly good, poor welding performance, stress corrosion cracking, double aging can improve the resistance to SCC. As mentioned above, its choice for the aeronautical industry is based mainly on its weight versus resistance ratio, where in Table 1 we can see some physical and mechanical properties of this material. The plate used in the study is riveted with five rivets in a linear form with a center-to-center distance of 28.6 mm each, the rivets are of HI-LITE (HL) type made with a titanium alloy with 6% aluminum and 4% vanadium, produced by HI-SHEAR Company a subsidiary of LISI S/A Aeronautical, automotive and medical supplement development company, its activity is focused on fasteners of all kinds. produced by HI-SHEAR Company a subsidiary of LISI S/A Aeronautical, automotive and medical supplement development company, its activity is focused on fasteners of all kinds. The HI-LOK rivets, abbreviated to HL, are not ordinary rivets and yes created by the HI-SHEAR Company and have according to the company the combination of the benefits of a rivet with those of a threaded screw. The HL (HI LOK) has quick and easy installation mainly compared to a rivet or bolt, where its fastening necklace, as shown in Figure 1, has an excess of material that serves as a torque regulator so that the thread does not crush, and the HL can be fixed in the correct mode.

Fig.1: Rivet torqued control and pressing systems [2]
The HL compared to the other rivet models, and has a smaller working area section and therefore a shorter transition area enters the rod and the load bearing segment, with optimized transition from working area to the screw and considerably increasing the fatigue resistance. HL type rivets account for 85% of aircraft structural joints according to LISI S/A Company, combining high strength materials and innovative design. The research and development of technologies in this area are of extreme importance seeing that the current processes are becoming obsolete, and details become very important in the structural and behavioral analysis of the structures, which led this work to accurately analyze fatigue fracture behavior of the 7075-T6 aluminum riveted with and without the scarification and burr process.
The installation of the fastening rivets involves milling and cold expansion of the main hole to the limit of fatigue crack growth is due to the compressive zone transmitted to the cold expansion orifice [3] II.

MATERIALS AND METHODS
The test specimen (CP) used in the experimental tests is in sheet form with a useful area of 4278.24 mm² and with 5 centered and aligned rivets whose CP dimensions are shown in Figure 2 and the rivet housing in Figure 3 . On the right in Figure 4 we have the measurements on the plate currently used and to its left the measures to be proposed for better distribution of stresses in the plates. The HST11 6 rivet is from the Hi-Lite family as shown previously, it is composed of 6% aluminum and 4% vanadium and 90% titanium, this material has the following mechanical properties located in Table 2 [4].

Computational analysis of the plate
After previous behavior of the plates by laboratory traction test, a computational test was performed for a better understanding of the isolated components, where the specific plate 1 (in faces 1 and 2) was studied because of the presence of rupture characteristics and concentration of loads with higher incidence, certainly due to its geometry presenting area difference between its upper and lower areas, since the geometry of the housing bore section of the rivet showed change, according to The software computational test was done by applying a fixed geometry to one of the faces of the plate and applying a load of 1x 106 N on the opposite face of the fixed geometry applied as shown in Figure 6. [5] The computational analyzes were performed on the upper plate of the assembly referred to herein as plate 1 and the lower plate as plate 2 to facilitate discussion of the results obtained.  The study was based on a mesh of the solid mesh type, according to data and Figure 7, with mesh generator based on curvature, 4 Jacobian points, maximum and minimum size of elements = 0 mm, total nodes 16424, total elements 8462, maximum proportion 6,1643,% of elements with proportion <3 = 98 and completion time of the mesh = 00:00:02 [6]. After the measurements of the test body and HL were taken, they were drawn in the SOLIDWORKS where the first computational tests of tensile strength were made to generate real data of the transition part of the rivet head and the body, which was the region of greatest fragility [7]. Figure 9, the maximum and minimum stress points of the VON MISES are determined by the software and shown on the plate for better understanding, however in Figure 10are shows the maximum and minimum points of the resulting displacement -URES.

Fig.9: Points of concentration of maximum and minimunstresses of VON MISES inthe computational study
In turn in the Figure 11 we have the points of maximum stress concentration also showing 6082.198 MPa, the yield strength that showed 505 MPa. In this image, it is noticed that the tension concentration was inside the hole where the fifth rivet is housed, in Figure 8 with the hole highlighted, we can see where the tension was concentrated [8]. The point highlighted this figure is a "living corner" produced by the shape of the rivet, it is in the transition from head to body, where there is a radius of 0.75mm, but it is not able to relieve concentration at this point When we analyze the ray, we can see that it only exerts a smooth transition between the head of the rivet and the body and does not help in a better distribution of force and less concentration, so acting on this ray can represent much more expressive results. Thus, by analyzing different rays the following result is shown in Figure 9.
As the test demonstrates, with the change of radius to the value of 0.39, there was an increase of 319.503 MPa in the tensile strength somewhat around 5.25% increase, something much larger than the 1.76% of the transition radius. [9,10] Complementary to the computational tests, fatigue tests were carried out on the test bodies at the InstituteTechnological Aeronautics(ITA-BRAZIL) at the request of EMBRAER, the company interested in the study, which allowed us to verify that the fatigue strength limit of the material was maximum of 16.6 Ksi at 2x10 6 cycles with burr operation and maximum of 19.1 Ksi at 2x10 6 cycles without burr operation. For comparative purposes, the maximum tensions and fatigue strength of the respective specimens [11] were highlighted in Table 3 and Table 4. On the other hand, in Figure 12, the equivalent ESTRN equivalent maximum and minimum points are shown on the plate for a better understanding.Equivalentstrain (ESTRN) isdefinedon the X, Y, Z axesas:  The test specimens of the fatigue test behaved as expected and predicted in the computational traction tests where the rupture occurred at that point by the charge concentration due to the geometry of the section of the hole of the upper plate as shown in Figure 13, thus there being thus a "tearing" of the plate.
Results of these tests extend information previously reported fromtests on unnotched specimens and tests on specimens more severelynotchedand afford data on the variation of fatigue-strength reduction with notch severity. Under rotating bending fatigue tests thehigh stress zones are located at the fracture surface perimeter and decrease to the fracture surface center: the size ofgrooves decreases from fracture surface perimeter to the center [12]. Furthermore, for this type of specimen thecenter of fracture surface is characterized by fast crack growth zone where the plasticdeformation is small or zero. This rupture occurred at that point by the charge concentration due to the geometry of the section of the bore of the upper plate as shown in Figure 14, thus there being a "tearing" of the plate as shown in Figure13 . The fatigue cracking growth is associated with the growth of cracks in the riveted fracture mechanics. In turn, small cracks and grain size limitations in the microplastic regions, cause deceleration gradients and consequent slower cracking evolution [13,14]. The stress intensity is used to predict the stress state near the hole and crack tipcaused by the applied loading; as the crack tip radius is effectively zero, the stresses at the crack tip would be infinite [15][16] After changing the geometry of the housing section of the rivet there was a reduction of approximately 18.27% in the maximum stresses in the plate plus a better uniformity in the stress distribution in the housings as shown in Figure 15 and 16, whereas the displacement only had an increase of 3.125% and the minimum equivalent strain of 2.39% and the maximum of 24%.The stress distribute drafts of the riveted plates obtained so that it is convenient to check the bending strength of the root of the hole and the fatigue contact strength of the union faces [16]. However, the fatigue strength limit was higher in the test specimens without burr operation than the operation, noting that the geometric change of the section generated a decrease of 18.27% in the maximum stresses in the sheet, in addition to a better uniformity in the distribution of the tension in reaming of the rivets, with displacement increasing by 3.125% and the minimum equivalent strain of 2.39% and the maximum of 24%, often explaining the cause of the fracture.

ACKNOWLEDGMENT
This work was made possible through an interaction with the ITA -Aeronautical Technological Institute/EMBRAER that helped us in the interactions of aircraft component fixation projects and provided us with data for observations and analyzes. We should also thank the UTFPR -Federal Technological University of Paraná that subsidized us for the works.