Use of FEA to Improve the Design of Suspension Springs for Reciprocating Compressor
4 likes2,024 views
The document presents a study on using Finite Element Analysis (FEA) to optimize the design of suspension springs for reciprocating compressors, focusing on methodologies such as static structural analysis and modal analysis to ensure that rigid body frequencies are distinct from operational frequencies. Experimental tests were conducted to measure spring rates and validate numerical models, resulting in findings that suggest beam188 and sol187 elements provide better accuracy with lower computational demands. The conclusion emphasizes the utility of this methodology for robust spring design and dynamic response evaluation.
![0
100
200
300
400
500
600
time [s] vs. element
cpu time [s]
Results
•Focus on the last 3 modes
error % (rigid body frequencies)
experimental
spring joint
beam 188
hexa (sol186)
tetra (sol187)
hexa (sol185)
tetra (sol185)
mode1
*
0.8%
0.2%
1.9%
1.9%
3.7%
5.7%
mode2
*
2.8%
4.8%
3.1%
3.1%
1.5%
10.0%
mode3
*
4.8%
4.9%
7.0%
6.9%
8.6%
2.1%
mode4
*
2.3%
0.9%
1.5%
1.5%
2.7%
4.3%
mode5
*
5.7%
0.7%
3.3%
3.2%
4.6%
4.0%
mode6
*
4.6%
0.5%
2.2%
2.1%
4.0%
5.2%
DOF
*
3.77E+03
1.59E+04
1.04E+06
8.84E+05
2.88E+05
5.01E+05
cpu time [s]
*
14.00
5.00
531.00
282.00
112.00
88.00](https://image.slidesharecdn.com/fea-improves-design-of-reciprocating-compressor-140922113634-phpapp01/85/Use-of-FEA-to-Improve-the-Design-of-Suspension-Springs-for-Reciprocating-Compressor-14-320.jpg)
Use of FEA to Improve the Design of Suspension Springs for Reciprocating Compressor
- 1. 1 © 2013 ANSYS, Inc. September 22, 2014 ANSYS Confidential Use of FEA to Improve the Design of Suspension Springs for Reciprocating Compressor 2014 ANSYS Convergence Conference Mattias da Silva Castro Marcos Giovani Dropa de Bortoli Daniel Henri Bedatty Hofmann Whirlpool S.A. - Compressors Unit - Embraco
- 2. Use of FEA to improve the design of suspension springs for reciprocating compressor Mattias da Silva Castro Marcos Giovani Dropa de Bortoli Daniel Henri Bedatty Hofmann Whirlpool S.A. - Compressors Unit - Embraco
- 3. PRESENTATION TOPICS •Company Overview; •Problem Description; •Methodology; •Goals; •Conclusion and next steps.
- 7. •Rigid body frequencies of the kit rules compressor dynamics. •Design goal is to have rigid body frequencies far from compressor operating rotation Problem Description Suspension springs Customer view Inside compressor Lugs
- 8. Methodology •Experimental tests: Impactation to identify rigid body frequencies; spring rate measurement. •Static Structural analysis – FEA –To validate spring rate measured with CAD spring simulated, and as input for simulation with spring joints. –Spring coils self-contact is important in structural analysis •Beam188 coils self-contact is created using APDL script –Output: Axial and transversal spring rate. •Modal analysis – FEA –All parts rigid, except springs and discharge tube. –Coils self-contact disregarded –Parts connected with joints (fixed body-body). •Tools: ANSYS Workbench and Mechanical APDL 14.5
- 9. Goals •Validate numerical model with experimental data. •Evaluate different elements: spring joints, beam, solid. •Define the most efficient element type for this kind of analysis: good precision & less computationally expensive. •With this model well defined, it is possible to assure that rigid body frequencies of compressor are far from its operating rotation.
- 10. Results – Spring rate •Axial and Transversal Spring rate was measured to compare with FEA data •Also, spring rate is input for spring joints •Number of active coils was modified to achieve experimental spring rate
- 11. Results - Experimental •Rigid body frequencies experiment •Impact compressor crankcase with an impact hammer •Acceleration response (triaxial accelerometer) Position 2 Position 4 Position 1 Position 3
- 12. Results – FEA – Modal analysis Spring joints: 1 joint for each direction (x,y,z), total 3x4=12 joints to define axial and transversal stiffness. Ky Kx Kz No contact needed
- 13. Results – FEA – Modal analysis Solid •Spring rate defined by geometry/material properties •Parametrized CAD •Contacts needed •Obs: all cases consider beam188 element on the discharge tube Beam 188
- 14. 0 100 200 300 400 500 600 time [s] vs. element cpu time [s] Results •Focus on the last 3 modes error % (rigid body frequencies) experimental spring joint beam 188 hexa (sol186) tetra (sol187) hexa (sol185) tetra (sol185) mode1 * 0.8% 0.2% 1.9% 1.9% 3.7% 5.7% mode2 * 2.8% 4.8% 3.1% 3.1% 1.5% 10.0% mode3 * 4.8% 4.9% 7.0% 6.9% 8.6% 2.1% mode4 * 2.3% 0.9% 1.5% 1.5% 2.7% 4.3% mode5 * 5.7% 0.7% 3.3% 3.2% 4.6% 4.0% mode6 * 4.6% 0.5% 2.2% 2.1% 4.0% 5.2% DOF * 3.77E+03 1.59E+04 1.04E+06 8.84E+05 2.88E+05 5.01E+05 cpu time [s] * 14.00 5.00 531.00 282.00 112.00 88.00
- 15. Conclusion •Beam188 and sol187 elements showed lower errors compared with experimental data. •Spring joints have the disadvantage of requiring axial and transversal stiffness as input, so experiments or simulation must be done to acquire stiffness. •For a robust design, this methodology is useful to evaluate dimensional variations of springs and verify dynamic response of compressor. In this case, the use of beam elements showed to be very interesting due to lower error and low required computational time to solve simulations, mainly on transient structural analysis.
- 16. Thank you Mattias da Silva Castro mattias.s.castro@embraco.com