Fretting Fatigue Study of Dovetail Joints with Retaining Pin

 

ABSTRACT:

The turbine blade-disc joints in a jet engine often fail prematurely against predicted life due to fretting. Fretting and fatigue related failures are around 70-80% of total mechanical failure. This paper presents the simulation of the effect of coefficient of friction, rotational speed on fretting fatigue of dovetail joint attached with retaining pin. Contact pressure and sliding distance influences the fretting phenomenon. Higher the values of contact pressure and sliding distance higher will be the fretting phenomenon. The dovetail interface was modeled using Unigraphics-8 software and nonlinear static structural analysis was performed in ANSYS WORKBENCH 15.0. The analysis was performed at rotational speeds ranging from 2000 rpm to 10,000 rpm and coefficient of friction was varied from 0.1 to 0.3.  From the simulation it was observed that, the peak contact pressure reduced by 30% and 10% for straight and skew dovetail joint respectively by increasing COF from 0.1 to 0.3. The sliding distance reduced by 26% and 39% for straight and skew dovetail joint respectively by increasing COF from 0.1 to 0.3. From the simulation the fretting wear index for the given condition was 1.25 for straight dovetail joint and 6.009 for the skew dovetail joint. Higher fretting wear index indicate higher fretting damage in the joint.

 

 

INTRODUCTION:

Fretting phenomenon is a mechanism which appears in contacts, where relative oscillations of small amplitude occurs at the contacting surface or at the edge of the contacting surfaces which results in the forms of wear, surface damage and lead to life reduction of dovetail region in the blade/disc joint in a gas turbine engine. The design or Geometry tends to play a role in dissipating high stress areas in fan blade body since it is costly to produce a single blade with blade joint¹ hence efforts are made to reduce the cost and increase the fatigue life of the blade joints. Fretting fatigue is one of the principal problems affecting the life of turbine blades, gears, splines and other bearing surfaces subjected to vibration loads. However, the fretting fatigue damages should be minimized or completely eliminated by the appropriate fretting fatigue design. In this report, different dovetail joints were studied, designed, analyzed and compared with each other to find the most feasible and reliable dovetail joint among them.

 

The coefficient of friction influences the contact pressure and the sliding distance for both straight and skew dovetail joint and also the peak contact pressure for the skew dovetail joint was found to be twice that of straight dovetail joint². The root cause for the failure of a compressor blade with retaining pin as a locking mechanism had proven that the stress concentration at the retaining pin under fretting condition leads to crack initiation at the contact region³. By focusing numerical analysis method, it was concluded that it was more probable to encounter multiple cracks at the notch area of blade disk interface in an aero engine⁴. Due to improper riveting of the connecting pins, the failure resulted in the overloading of bending stress as per the results given by metallography, microscopy and Fractographic examinations⁵. The study revealed that the maximum stress occurred at the bottom of the tooth and magnitude of the induced stress depends on the factors like number of teeth, contact angle and flank length of a turbine disc⁶. Experimental and numerical simulation for retaining pin revealed that cracks were produced at an angle to the contact surface of test specimen⁷. Various methods like Fractography, electron microscope etc. were employed to study the surface damages of the blade of a jet engine that caused fracture⁸. It was predicted that the wear volume for the gross sliding state was less for variable co efficient of friction when compared with constant co efficient of friction⁹. Shot peening proved to improve the fatigue life of the Ti-6Al-4V specimen¹⁰. Analysis showed that crack propagation life and contact stresses consumes a majority of the total life and is insensitive to a large range of intial crack sizes¹¹.A specimen when subjected to high cycle Axial fatigue test, fretting cracks were observed over varying contact conditions¹².

 

As per the literature review there is fewer studies based on retaining pin with blade disc joint. So the objective of this work was to simulate the fatigue analysis and investigate the fretting fatigue parameters for the blade attachment with retaining pin through variation of rotational speed and friction co-efficient. The methodology involved modeling, analysis of straight, skew dovetail joints and simulation results were compared with cited research for validation. This gave confidence to perform analysis and simulation to blade disk with retaining joint. The nonlinear static structural analysis was performed using appropriate boundary conditions and responses such as Von-Mises stress, peak contact pressure, sliding distance were measured. The responses were measured for Von-Mises stress at different rotational speed (2000 to 16000 rpm) at 0.1 and 0.3 coefficient of friction¹³. The fatigue life for the straight dovetail joint, skew dovetail joint and the blade attachment with retaining pin was evaluated for rotational speed and fretting wear index is calculated for are three joints and compared.

 

Modeling of dovetail joints:

In this work, three types of blade disc joints such as the straight dovetail joint, skew dovetail joint and blade disc with retaining pin are modelled using the Unigraphics-8 based on the dimensions taken from the literature¹. The straight dovetail joint was modelled using dimensions¹ as shown in the Figure 1. For ease of analysis a sector of the disc was considered for modeling and analysis. The skew dovetail joint having almost the same dimensions as straight dovetail joint with addition of 20⁰ skew angle was modeled as shown in Figure 2.

 

                                                      

                     Figure 1: Straight dovetail joint                                                             Figure 2: Skew dovetail joint    

 

                 

The blade disc with retaining pin has three components blade, disc and a retaining pin². Using the same straight blade joint dimensions, the lower part of the blade is redesigned by adding retaining pin shown in Figure 3.

 

                                  

Figure 3: Retaining Pin dovetail joint

 

FRETTING STUDY:

A non-linear static analysis was done to the dovetail blade disc interface. The blade disc was fastened in place by means of dovetail slot which then used to create a part file called Para solid. With the use of para solid model, the straight dovetail joint, skew dovetail joint and blade disc with retaining pin models were individually imported to ANSYS environment without any loss. The program analyses the data and the element type SOLID 186 was selected since it has a higher order element consisting of 20 nodes and exhibit quadratic displacement behavior. The meshing of those models were done on ANSYS Workbench where they are meshed with hex dominant mesh because of its quadratic displacement behavior. The models globally meshed with the element size of 2 mm and the refinement of the meshing is provided at the critical regions as shown in the Figure 4, Figure 5 and Figure 6 in order to obtain more accurate results. The straight dovetail shown in the Figure 4 has 20 mm thickness and the generated mesh has total node of 4,76,073 and number of element is 11,620. The skewed dovetail has a skew angle of 20° and the mesh generated has 86,478 number of nodes and 19,596 number of elements. The mesh for skew dovetail is shown in the Figure 5. The Figure 4 shows the meshing for straight dovetail joint while Figure 5 shows meshing of skew dovetail joint.

 

 

 

Figure 4: Meshing for straight dovetail              Figure 5: Meshing for skew dovetail       Figure 6: Blade attachment with retaining pin

 

The mesh generated for The blade with retaining pin had 70,162 nodes and 17,272 number of elements as shown in Figure 5. The analysis is performed to simulate dovetail in an aero engine compressor. But in actual analysis it is not possible to consider all real life conditions as it will complicate the study, hence some parameters can be neglected. The assumptions made in the loading conditions are: bending load, aerodynamic loads, other factors like residual stresses are neglected for ease of analysis. The centrifugal load on blade root is simulated by applying load on blade top face and the effect of vibrations is also neglected.

 

As the contact occurs between disc and dovetail at the flank region, the faces in the blade is treated as TARGET and the faces of the dovetail slot are considered as CONTACT. Thus, the element can be easily adapted to model the complex shape of the target surface. The dovetail disc does not show any deformation, whereas the blade undergoes deformation upon loading which exhibits the rigid to flexible type of contact behavior. The contact pair generated for straight and skew dovetail joints in the ANSYS Workbench is shown in the Figure 7. Similarly, contacts are generated for blade with Retaining Pin joint as shown in Figure 9.

 

                   

Figure 7: Frictional contact at the flank region of dovetail joints   Figure 8: Boundary conditions for straight and skew dovetail joints

 

 

 

The boundary conditions considered for straight and skew dovetail joints shown in Figure 8 were as follows, since it is difficult to analyze whole model, as part of symmetry only one sector out of 24 blades was considered. The displacements on faces A and B were zero, on face C existed frictionless support, rotational velocity applied at D was 1050 rad/s and the force applied on face E was 2000N.  

 

Figure 9: Boundary condition for blade with Retaining Pin

 

Similarly, contacts were generated for blade with Retaining Pin joint. The displacements on faces A and B were zero, the rotational velocity applied was 1674 rad/s and the loading applied to it was static in nature as shown in Figure 9.

 

FATIGUE LIFE ANALYSIS:

In order to proceed with the retaining pin simulation, in this research the confirmation experiment has been conducted in the lines of Anandavel¹, which is shown in Table 1. As the error is less than 10% for both straight and skew dovetail joints, it was clear that the results of simulation done was close to the cited results which gave confidence to proceed towards retaining pin simulation.

 

Table 1: Comparison of Cited and simulated results

Parameter	Type of joint	Cited results	Simulated results	Error %
Peak von-Misses stress	Straight Dovetail Joint	232 MPa1	226.76 MPa	2.25 %
Peak contact Pressure		211 MPa1	192.93  MPa	8.56 %
Peak von-Misses stress	Skew Dovetail Joint	425 MPa1	408.16 MPa	3.96 %
Peak contact Pressure		313 MPa1	301.08 MPa	3.80 %

 

 

During simulation various parameters such as Von-Mises stress, principal stress, co-efficient of friction, contact pressure and sliding distance was found out for both the dovetail joints. The analysis is then done to each components of the model of retaining pin on the same software for many parameters such as Von Mises stress, total deformation, equivalent stress, maximum and minimum principal stresses. The values found are then used to find the fretting wear indexes for straight dovetail, skew dovetail and blade disc with retaining pin joints whose values are then compared with one another which is depicted in the next subtopic. 

 

4. RESULTS AND DISCUSSION:

The Figure 10 shows the von Mises stress distribution for retaining pin, where it experiences less stress (176.46 MPa) when compared with the blade lug (434.71 MPa from Figure 14). This can be attributed that different materials were used for the retaining pin. The Figure 11 shows the maximum principal stress distribution of the pin.

                                                      

Figure 10: Von-Mises stress distribution for the retaining pin              Figure 11: Maximum principal stress for the retaining pin                                                                                                                                  

                                                                     

 Figure 12: Minimum principal stress for the retaining pin                    Figure 13: Deformation of the retaining pin 1   

 

 

The Figure 12 shows the minimum principal stress for the retaining pin while Figure 13 depicts the total deformation of retaining pin. The central region experiences the maximum deformation of 0.11867 mm as shown in the Figure 13. The principal stress was minimum at the center with respect to the sides, as the maximum deformation occurred at the center. From the Figure 14 it is clear that the maximum von-Mises stress (434.71 MPa) occurs at the area of contact between retaining pin and the disc. The stress distribution for the blade root is shown in the Figure 15.

 

                                     

Figure 14: von-Mises tress distribution for disc                                           Figure 15: von-Mises stress distribution for blade   

 

 

The Table 2 shows the von Mises stress values for different components of blade attachment with a retaining pin.

 

Table 2: von Mises stress for blade disc attachment with a retaining pin

 

 

The disc and blade are made out of same material that is Titanium alloy Ti-6Al-4V and the retaining pin is made out of nickel chromium based alloy called Inconel. Inconel has ultimate tensile strength of 1190 MPa whereas Titanium alloy Ti-6Al-4V has ultimate tensile strength of 875 MPa¹¹. The Table 3 gives the fatigue life for different component in case of blade disc attachment with a retaining pin.

 

Table 3: Fatigue life for components in retaining pin blade disc attachment

Component	Disc	Retaining pin	Blade
Fatigue life  (number of cycles)	2090	40582	3395

 

 

The fatigue life is shown in the Figure 16 for the pin and the fatigue life for the disc in Figure 17. In both the figures the red colored region indicates the Minimum Fatigue Life which is critical because above this number the part might sustain micro cracks. The crack originated will propagate and results in the fatigue failure of the component.

 

                                            

Figure 16: Fatigue life for the retaining pin                                                   Figure 17: Fatigue life for disc

 

 

The blade attachment with retaining pin configuration was simulated at different rotational speeds. The blade and the disc material was considered to be Titanium alloy while pin material was considered to be Inconel 718. In Figure 18, von-Mises stress for the retaining pin attachment are plotted on the graph at different rotational speeds. The blade and the disc endures almost same von-Mises stress level (434.71 MPa), whereas the retaining pin made out of Inconel experienced less von-Mises stress (176.46 MPa).

 

Figure 18: Von-Mises stress plot for blade disc with retaining pin

 

The Table 4 describes the fatigue life for the various component of the straight dovetail joint, skew dovetail joint and blade attachment by retaining pin at coefficient of friction 0.1 and 0.3 and rotational speed of 16000 rpm.

 

Table 4: Fatigue life for straight and skew dovetail joint and retaining pin at coefficient of friction 0.1 and 0.3 and rotational speed of 16000 rpm.

 

From Table 4 it is clear that, for maximum rotational velocity of 16000 rpm the skew dovetail joint has less fatigue life compared to that of straight dovetail joint and blade attachment with pin. The straight dovetail joint had fatigue life of 982 cycles and 1276 cycles at COF 0.1 and COF 0.3 respectively. The change in the coefficient of friction from 0.1 to 0.3 increased the fatigue life by 129.33 % for the straight dovetail joint. The skew dovetail joint had fatigue life of 222 cycles and 514 cycles at COF 0.1 and COF 0.3 respectively increasing the fatigue life by 181.13 %. The retaining pin has fatigue life of 40852 cycles and 114990 cycles at COF 0.1 and COF 0.3 respectively increasing the fatigue life by 283.35%.

 

The Table 5 quantifies the fretting wear in straight dovetail, skew dovetail joint and blade attachment with retaining pin.

 

Table 5: Fretting wear index for the straight, skew and retaining pin joint

 

From the Archard’s equation¹⁴, the fretting wear index is calculated for the straight dovetail joint which was resulted as 6.3726, for the skew dovetail it was 43.7114 while for blade attachment with retaining pin it was 2.6705. Higher the index number, higher will be the fretting wear. Thus the blade attachment with retaining pin experiences least fretting wear among the blade disc joints.

 

5. CONCLUSIONS:

The maximum von-Misses stress (434.71 Mpa) in the retaining pin blade assembly was induced at the disc which was less than the ultimate yield stress of 875 MPa for Ti-6Al-4V, and for retaining pin it was 176.46 MPa which was less than the ultimate yield stress 1190 Mpa for Inconel 718.

The retaining pin in the blade attachment with retaining pin assembly has a fatigue life of 40582 cycles, the disc has 2090 fatigue cycle and the blade tang has fatigue life of 3395 cycles at COF 0.1.

The fatigue life increased on an average of 272% with change in coefficient of friction from 0.1 to 0.3 for a blade attachment with retaining pin.

The fretting wear index of retaining pin is less when compared to straight and skew dovetail joints.

 

 

6. REFERENCES:

1.             Leye M. Amoo, “On the design and structural analysis of jet engine fan blade structures”, Progress in Aerospace sciences vol. 60, 2013, pp. 1-11

2.             K. Anandavel, Raghu V. Prakash, “Effect of three-dimensional loading on macroscopic fretting aspects of an aero-engine blade–disc dovetail interface”, Tribology International, vol. 44, 2011, pp. 1544–1555,

3.              Bok-Won Lee, Jungjun Suh, Hongchul Lee, Tae-gu Kim, “Investigations on fretting fatigue in aircraft engine compressor blade”, Engineering Failure Analysis, vol. 18, 2011, pp. 1900–1908,

4.             M.M.I. Hammouda, R.A. Pasha, A.S. Fayed, “Modelling of cracking sites/development in axial dovetail joints of aero-engine compressor discs”, International Journal of Fatigue, vol. 29, 2007, pp. 30-48

5.             A.K. Ray, G. Das, V.R. Ranganath, B. Goswami, R.N. Ghosh, “Failure of connecting pins of a compressor disc in an aero engine”, Engineering Failure Analysis, vo.11, 2004, pp. 613-617

6.             S.A. Meguid, P.S. Kanth, A. Czekanski, “Finite element analysis of fir-tree region in turbine discs”, Finite Elements in Analysis and Design, vol. 35, 2000, pp. 305-317

7.             Zeeshan Anjum, Faisal Qayyum, Shahab Khushnood, Sagheer Ahmed, Masood Shah, “Prediction of non-propagating fretting fatigue cracks in Ti6Al4V sheet tested under pin-in-dovetail configuration: Experimentation and numerical simulation”, Materials and Design, vol.87, 2015, pp.750 –758

8.             Myounggu Park, Young-Ha Hwang, Yun-Seung Choi, Tae-Gu Kim, “Analysis of a J69-T 25 engine turbine blade fracture”, Engineering Failure Analysis, vol.9, 2002, pp.593–601,

9.             Tongyan Yuea, Magd Abdel Wahabb, “Finite element analysis of fretting wear under variable coefficient of friction and different contact regimes”, Tribology International, vol. 107, 2017, pp. 274–282,

10.          Qi Yang, Wenlong Zhou, Pengtao Gai, Xinhua Zhang, Xuesong Fu, Guoqing Chen, Zhiqiang Li, “Investigation on the fretting fatigue behaviors of Ti-6Al-4V dovetail joint specimens treated with shot-peening”, Wear, vol.372-373, 2017, pp.372-373,

11.          P. J. Golden, J. R. Calcaterra, “A fracture mechanics life prediction methodology applied to dovetail fretting”, Tribology International vol. 39, 2006, pp. 1172-1180

12.          W.A. Glaeser, Bernard H. Lawless, “Behavior of alloy Ti–6Al–4V under prefretting and subsequent fatigue conditions”, Wear, vol.250, 2001, pp.621–630

13.          Appaji Gowda B.M, Dr. H R Yeshovanth, C.Siddaraju, “Investigation and efficient modeling of an Dovetail Attachment in Aero-Engine”, Procedia Material Science, vol. 5, AMME 2014, pp. 1873-1879

14.          Toshio Hattori, Vu Trung Kien, Minoru Yamashita, “Fretting fatigue life estimations based on fretting mechanisms”, Tribology International, vol. 44, 2011, pp. 1389-1393

 

 

 

Received on 23.05.2018            Accepted on 18.06.2018            ©A&V Publications all right reserved Research J. Engineering and Tech. 2018;9(3): 233-240. DOI: 10.5958/2321-581X.2018.00032.6