Functional failure analysis of a Pandrol clip.
The traditional rigid fastenings like the nut and bolt, etc. were used by the railways all over the world. But this could not safeguard the track parameters and dampen the vibrations. The primary purpose of a fastening is to connect or fix the rail with the sleeper. The fixing of the rail to the sleeper may be done directly or indirectly with the help of the fastenings, but in the process the fastening is subjected to severe vertical, lateral and longitudinal forces. The forces which are predominantly dynamic increase rapidly with the increasing loads and speeds. Due to shocks and vibrations caused by moving loads, the rigid fastenings get loose, an interplay between the components of the track develops, track parameters get affected and rapid deterioration of the track starts. There is a need, as such, of a fastening which can safeguard the track parameters and dampen the vibrations. An elastic fastener like the PANDROL clip is possibly an answer for the problem.
This work describes the stress analysis in Pandrol clip due to moving loads and comparison with the functional failure. To evaluate the stresses induced a standard MK-III elastic rail clip was used and the 3D model was developed in PRO-E software as per the drawings. Since the load due to moving train wheels is a periodic load with varying frequency and amplitude ANSYS package is used to study the vibrational characteristics and Harmonics response analysis of the Pandrol clip with a defined boundary conditions there by stresses induced are studied.
Pandrol Clips & Its Functioning
Pandrol clips of British origin - also called elastic clips- are extensively used in the Indian Railways. These are "fit-and-forget" type of fastenings. Once fixed in position, the pandrol clip is expected to maintain its desired toe-load without any subsequent attention. It is applied parallel to the rail and is driven and removed with an ordinary hammer.
When driven, one leg of the clip is housed into a groove, and the clip deflects from its original shape to exert a heavy toe-load on the rail. The friction grip of the clip in the housing is two to three times that of clip on the rail , so that rail creep forces are unable to dislodge the clip. The creep is resisted in both directions, an essential requirement of long welded rail fastenings. The pandrol clip PR 40, standardized in the Indian Railways, IS manufactured from 20.6 mm dia silicomanganese steel rods heat treated to proper specifications. It weighs about 1 kg. The static toe-load of 710 kg. per clip gives a total rail to sleeper load of 2840 kg. Assuming 0.5 as the coefficient of rail to pad friction, this provides a rail to sleeper resistance of about 1420 kg, which is well above the average sleeper to ballast resistance of about 1000 kg per sleeper in the direction of traffic. The chances of relative rail to sleeper movement are therefore less.
The up-to-date experience with standard pandrol clips of PR 401 series has not been very satisfactory. Over the years, they tend to loose their toe-load, thus allowing the rails to creep. The life of the rubber pads and liners in the assembly has also been poor. By following a different space curve, new series of improved elastic rail clips have been designed, using about the same length and diameter of the steel rod, and thus within the same weight of steel. ERC mark III and ERC mark IV versions of the clip have higher toe-loads of 800-1000 kg, and 1100-1200 kg. respectively providing a creep resistance of about 1000 kg per rail seat. Round toe has been modified to flat toe to distribute the point load on a wider area. This reduces the indentation on the 21 liner and enhances its life. It also helps in maintaining the toe-load of the clip.
* To develop a finite element model of a Pandrol clip.
* To perform model and harmonic analysis of the Pandrol clip.
* To validate the results (computed values).
* To study the concurrence of the results with the functional failure and suggest the suitable modifications based on the analysis in order to avert failure.
Based on the drawing that is provided for the MK-33 Pandrol clip, a 3-D model has been developed in Pro-E and is imported to ANSYS by creating an intermediate file in Para-solid form. Solid45 element is used for finite element modeling and both the analysis have been done on the component. 3-D model that has been developed in Pro-E and meshed solid model with solid 45 elements is shown below.
The model analysis is to determine the vibration characteristics (natural frequencies and mode shapes) of a structure or a machine component while it is being designed. It also can be a starting point for another, more detailed, dynamic analysis, such as a transient dynamic analysis, a harmonic response analysis, or a spectrum analysis. The important feature of modal analysis is model cyclic symmetry, which allows reviewing the model shapes of the cyclically symmetric structure by modeling just a sector of it. The model analysis in the ANSYS family of products is a linear analysis. Any nonlinearity, such a plasticity and contact (gap) elements are ignored. Some assumptions are made in the present analysis.
* Load acting on the clip is a point load.
* Straight rail track is considered.
* Distance between two axles of the same wagon and adjacent wagons are kept the same.
* The toe load applied on the rail by the clip is kept constant at 1000kgs.
* In view of the seating of the clip (on the rail and on the sleeper) on the either side of the housing, the friction grip in the housing it is treated as fixed.
With the assumed boundary conditions the Modal analysis has been carried out on the pandrol clip and natural frequencies are found to be 17.176Hz, 20.98 Hz, 33.436 Hz, 56.432 Hz, 60.253 Hz, 118.357 Hz, 139.961 Hz, 243.605 Hz, 282.65 Hz and 371.799Hz. The assumed boundary conditions and the mode shapes at every natural frequency are given below.
The technique of Harmonic response analysis is used to calculate the structure's response at several frequencies and obtain a graph of some response quantity (usually displacements) versus frequency. "Peak" responses are then identified on the graph and stresses reviewed at those peak frequencies. From this we can verify the designs which successfully over come resonance, fatigue and other harmful effects of forced vibrations.
When the train is moving on the rails, the loads are applied at various frequencies. Thus it is necessary to determine the response of the pandrol clip at those frequencies. Harmonic analysis under the toe load of (1000kg) and the frequency range of 0-500Hz is performed with 25 sub-steps and displacement of 0.05mm.
From the Amplitude Vs frequency graph it is found that the maximum amplitude of 1.125mm is obtained at frequency of 140 Hz. So the response of the clip at 140 Hz has been observed.
Amplitude Vs frequency
The maximum stress obtained is 14457N/sq mm and the location is found to be at portion of the Pandrol clip just outside of the housing.
Von-mises Stresses at 140 Hz
Further the maximum deformation at 140 Hz of the clip is observed at two different areas as shown in the figure and they do not have any correlation with the location where maximum stress is observed.
The portion of the clip in the housing appears to have hinge condition. But in reality due the contour of the clip and points of the support that the clip has, there exist a momentum in the vertical plan containing the axis of the housing. This is because, the line of action load on either side of the axis of the housing is not normal to the axis of the housing. This momentum and the grip friction in the housing make the portion of the clip in the housing as a fixed element. This will result in induction of max stress at portion of the clip just coming out of the housing. The conditions are shown in the figure.
The maximum stress that has been induced though is less than the permissible limit; it has been observed that these Pandrol clips are mostly facing brittle failure at the same spot. The possible reason could be that because of the swinging action of the clip about the fixed portion in the housing, the area which has maximum stress will be subjected to twisting action and ultimately result in fatigue loading. This leads to strain hardening and end up in brittle failure. The following picture clearly shows that the functional failure of the clip is located at the same area where max stress is observed during the analysis.
For the two different analysis that have been performed on the Pandrol clip viz. Modal analysis and Harmonic analysis, observations were studied and analyzed. From the modal analysis the value of the first natural frequency is determined to be 17.676Hz and the corresponding mode shapes have been presented. It is observed that the maximum displacement of 2.137mm was obtained at seventh frequency of 139.961Hz. The Opertating frequency of the clip was observed to be 4Hz for all speed of the passenger and cargo trains and it is well below the first natural frequency. This confirms that the clip is being operated in safe mode.
Harmonic analysis has been performed in-between the frequency range of 0-500Hz and the maximum stress obtained is 14457N/mm2. This is very much below the allowable stress limits (1) for the clip material. The functional failure of the clip which was observed to be brittle is in agreement with the expected failure due to strain hardening at the same point at which the maximum stress is induced. The concurrence of functional failure with the results of the analysis reveals that the boundary conditions are assumed are close to the actual conditions and hence this will facilitate further extension of the work with regard to the design modifications leading to prevention of failure and to have more functional life for Pandrol Clip.
 M.M. Agarwal,"Indian Railway Track" Prabha & co publications.
 Yann Bezin, Simon Lwniski, Julion stow, Stephen Blain,"development of a method to predict stresses in rails using ADAMS/Rail & ABAQUSW".
 "FINITE ELEMENT ANALYSIS OF 3 WHEELER VEHICLES", by Sravan Kumar.G and K. Sivaprasad.
 "STRESS ANALYSIS OF PRE-STRESSED RAILWAY CONCRETE SLEEPER", By A.Sireesha, D.Amar, E.V. Ashok kumar, B .Madhuri and G.Srinivas.
(1) C.L.V.R.S.V. Prasad and (2) S. Srikiran
Mechanical Engg. Department, GMR Institute of Technology, Rajam, India
(1) Professor, E-mail: Prasad.email@example.com
(2) Associate Professor, E-mail: firstname.lastname@example.org
Tabulated results of modal analysis: S.No Mode No Freq. (Hzs) Def.(mm) 1 1 17.676 1.604 2 2 20.98 1.85 3 3 33.436 2.158 4 4 56.432 1.644 5 5 60.253 2.265 6 6 118.367 1.585 7 7 139.961 2.837 8 8 243.605 1.624 9 9 282.65 1.735 10 10 371.799 1.966
|Printer friendly Cite/link Email Feedback|
|Author:||Prasad, C.L.V.R.S.V.; Srikiran, S.|
|Publication:||International Journal of Applied Engineering Research|
|Date:||Nov 1, 2009|
|Previous Article:||Power quality improvement by suppressing harmonics using power active filters.|
|Next Article:||Analytical analysis of fluid flow and heat transfer through microchannels.|