Home Background.htm Design.htm Analysis.htm

Text Box: An ANSYS model with fixed outer boundary conditions simulated a cracked choke ring stuck within the water brake cylinder.  As discussed, the ANSYS model was utilized in testing the various wedge-shaped cut patterns.  Various simplifications were included in this model.  Although the wedges remain attached, they were modeled as completely removed.  This variation is acceptable because the cuts themselves provide for a space greater than the overall deformation of the choke ring.  As an additional simplification, the variable inner diameter choke ring was modeled as cylinder.  However, this cylinder uses the greatest thickness of the choke ring as the constant cylinder thickness and is thus a very conservative estimate.  Concerning material properties, the model assumed that the choke ring is made of Aluminum Bronze (B-148-97, C95500, heat-treated).  Accordingly, the modeled assumed an elastic modulus of 16,700 ksi and a Poisson’s Ration of 0.32 as listed in Appendix M.  The choke ring may also be made of a Copper Based Alloy, having a smaller modulus of elasticity.

 

Text Box:  When inputting the models in ANSYS it was assumed that 1000 ft*lbf of torque could be created using the existing torque wrench as well as a three foot long cheater bar. The model was constrained by placing pin boundary conditions near the circumferential cut of each wedge.  The model has been scaled for illustrative purposes, allowing the reader to view the uniform inward deformation of the choke ring. Various cut patterns were tested in order to optimize the design.  The design was optimized with the intention of providing substantial deflection with the least amount of cutting possible.  Both symmetric and slanted wedges were tested in this optimization effort.  The symmetric wedges are composed of two 10-degree cuts.  The slanted wedges are composed of one 10-degree cut and one vertical cut.  The wedges are composed of two 10 degree cuts.  The severity of these angles was chosen due to practical considerations discovered during testing.  Table 1 shows the results of these tests, wherein maximum deflection denotes the maximum inward deformation at any wedge.  Initial two dimensional tests revealed that similar deflections occurred for both symmetric and slanted wedges.  However, the symmetric wedges caused a uniform inward deflection whereas the slanted wedges caused an uneven deformation.  Therefore, symmetric wedges were chosen for the final design.  Initial testing also revealed that the deformation was substantially enhanced by increasing the cut pattern from two to four wedges. 

Text Box: Based on these results, a symmetric four wedge pattern was selected as the optimized design.  In order to further validate this selection, a three-dimensional model of the choke ring was tested.  This model included similar simplifications to those discussed previously.  The model is shown below in Figure 9.   The model verified previous results, showing a uniform deformation of 3.31E-4 inches.  Although these deformations appear relatively low, true deformations will likely be larger due to the conservative nature of the analysis.  Additionally, it is assumed that the expansion/ binding of a cracked choke ring is of a similar small scale.

Text Box: Dynamic Analysis Dynamic analysis was performed to determine the Metal Removal Rate of the material as well as the forces required to deflect the cantilever arm resulting from the eight, ten degree cuts.  This analysis determined that it would take roughly 30 minutes per cut, or four hours for cutting time.  This neglects the time required for chip removal and lubrication.  The force required to yield the cantilevered wedge is approximately 760 lbf.  However, as previously discussed in Section 4.1,  yielding of the cantilevered wedges will not be required given the expected degree of overall deformation.