1、 WORLDWIDE ENGINEERING STANDARDS Test Procedure GMW16970 Guidelines for Conducting Axial Load-Controlled Fatigue Testing of Neat, Filled, and Short Fiber Reinforced Thermoplastic Polymeric Materials Copyright 2013 General Motors Company All Rights Reserved July 2013 Page 1 of 13 1 Scope Note: Nothin
2、g in this standard supercedes applicable laws and regulations. Note: In the event of conflict between the English and domestic language, the English language shall take precedence. 1.1 Purpose. To provide a common, concise, accurate method of obtaining fatigue information for polymeric materials. 1.
3、2 Foreword. Most components and structures are subjected to cyclic loading; therefore, fatigue is a major consideration in their design and failure analysis. Several ASTM and ISO standards exist for characterizing fatigue behavior and determining fatigue properties of metallic materials. One such st
4、andard is ASTM E466 on conducting force controlled constant amplitude axial fatigue tests for metallic materials which has existed for 40 years. Polymeric materials are increasingly used in many structural applications where they are often subjected to cyclic loadings. Therefore, fatigue behavior ch
5、aracterization and properties of polymeric materials are also of prime consideration in many of their applications. However, fatigue test standards for polymeric materials have not been available from major standard writing organizations such as ASTM and ISO until recently. In 2012, a test standard
6、was issued by ASTM as ASTM D7791, Standard Test Method for Uniaxial Fatigue Properties of Plastics. This brief test standard lacks many details such as specimen geometry and cycling frequency selection. It is also limited to loading conditions for which stresses do not exceed the proportional limit
7、of the material. These guidelines are intended to provide detailed axial load-controlled fatigue testing of polymeric material. This addresses testing equipment, specimen geometry, testing procedures, data acquisition and analysis, and material property determinations. Such guidelines can be used in
8、 support of activities such as research and development, process and quality control, and product performance and failure analysis. 1.3 Applicability. This Fatigue Testing Procedure is applicable for all polymeric materials. 2 References Note: Only the latest approved standards are applicable unless
9、 otherwise specified. 2.1 External Standards/Specifications. ASTM D7791 ASTM E466 ASTM E739 ASTM E1012 2.2 GM Standards/Specifications. GMW16652 2.3 Additional References. O. Krause, 2002, “Frequency Effects on Lifetime,” Optimat Blades Project, DLR, doc. OB_TX_N003 rev. 1, Wieringerwerf, The Nether
10、lands. R. I. Stephens, A. Fatemi, R. R. Stephens, and H. O. Fuchs, 2000, Metal Fatigue in Engineering, 2nd Ed., John Wiley and Sons, New York, NY. 3 Resources 3.1 Facilities. Copyright General Motors Company Provided by IHS under license with General Motors CompanyNot for ResaleNo reproduction or ne
11、tworking permitted without license from IHS-,-,-GM WORLDWIDE ENGINEERING STANDARDS GMW16970 Copyright 2013 General Motors Company All Rights Reserved July 2013 Page 2 of 13 3.1.1 Calibration. The test facilities and equipment shall be in good working order and shall have a valid calibration label. 3
12、.1.2 Alternatives. Alternative test facilities and equipment may also be used. However, all measuring variables as specified in this standard shall be determined correctly with respect to their physical definition. 3.2 Equipment. 3.2.1 Testing Equipment and Verification. A variety of testing machine
13、s can be used to perform the tests. These include a variety of mechanical machines (such as power screw machines), hydraulic or electrohydraulic machines, and electromechanical or magnetically driven machines. Regardless of the type of machine used, the form and magnitude of the applied load signal
14、should be closely controlled and monitored for the duration of the test. It is recommended that the load be maintained within 2% of the intended value for the duration of the test. An effect to be most avoided is misalignment of the specimen and the load train components (i.e., load cell, grips, loa
15、ding actuator, etc.) in the testing machine. Misalignment can significantly affect test results and cause large data scatter. It results from tilt and/or eccentricity of the test specimen in the loading machine from a variety of sources. These sources include lack of repeatability of the grips durin
16、g gripping/un-gripping actions, misalignment of the load train, insufficient lateral stiffness of the load frame, and lateral movement of the loading actuator in the case of electrohydraulic machines during the cyclic loading process. To achieve satisfactory alignment, it is recommended to use a sti
17、ff load frame (both in axial and lateral directions), a hydrostatic actuator to minimize lateral actuator movement, as short as possible of a load train length, and repeatable grips (typically hydraulic or pneumatic), in conjunction with an alignment fixture. An alignment fixture enables adjusting t
18、ilt in two perpendicular orientations, as well as adjusting eccentricity in x-y plane perpendicular to the loading axis. An anti-rotation device may also be necessary to prevent twisting of the loading actuator, and in turn the test specimen, during axial cyclic loading. If measurement or monitoring
19、 of axial deformation during fatigue test is desirable (for example, to evaluate cyclic softening or cyclic creep or ratcheting during tensile mean stress tests), a Linear Variable Displacement Transducer (LVDT) or an extensometer (non-contact video or laser, or contact mechanical) can be used. An e
20、nvironmental chamber with an electronic heating element and a coolant system (such as liquid nitrogen) with flow circulation is typically used for testing at cold or at elevated temperatures. To monitor any temperature rise during fatigue tests due to self-heating at the applied load and testing fre
21、quency, a thermal imaging camera or a thermocouple can be used. 3.3 Test Vehicle/Test Piece. Not applicable. 3.4 Test Time. Reference material frequency dependence. 3.5 Test Required Information. See explanation via this document. 3.6 Personnel/Skills. Engineers and technicians should be familiar wi
22、th testing equipment and behavior of polymers. 4 Procedure 4.1 Preparation. 4.1.1 Specimen Design, Fabrication and Storage. Test specimens can be machined from molded plaques (sheets or plates). The GM material engineer will provide guidance for injection molded specimens. Machining specimens from p
23、laques are recommended as the preferred method and should be followed. Note that the thickness of the plaques shall be similar to the thickness of the actual components. See Appendix A. A CNC milling machine with an appropriate cutting tool can be used for machining the samples. Prior to specimen ma
24、chining, the plaques should be stored under a controlled environment (i.e., temperature, humidity, ultraviolet light, etc.) It is important to identify and record the orientation of the test specimen to be machined with respect to the plaque mold flow direction (for example, longitudinal versus tran
25、sverse), as well as with respect to the plaque location (for example, edge versus middle). This is particularly important for fiber reinforced polymers, since fiber length and/or orientation distribution may not be uniform in different regions. For this reason, it is recommended not to use the begin
26、ning and the end of a molded plaque for specimen manufacturing (see GMW16652). An example of specimen identification method is shown in Figure 1. Copyright General Motors Company Provided by IHS under license with General Motors CompanyNot for ResaleNo reproduction or networking permitted without li
27、cense from IHS-,-,-GM WORLDWIDE ENGINEERING STANDARDS GMW16970 Copyright 2013 General Motors Company All Rights Reserved July 2013 Page 3 of 13 Specimens should be machined to a suitable geometry, such as that shown in Figure 2 (a). The geometry shown in this figure has been designed as an optimum g
28、eometry considering several important factors. These considerations include: a. A large enough gage section volume sufficient to represent realistic fiber distribution in short glass fiber filled polymers. b. A sufficiently long gage section to represent uniform uniaxial stress condition, but short
29、enough to avoid or minimize buckling during compression loading. c. A sufficiently wide gage section to accommodate stress concentrations, if testing to evaluate notch effects is desired. Typical notches include a central circular hole and a U or a V notch at the edge(s) of the gage section. d. A ge
30、nerous transition radius for the gage to the grip sections to minimize stress concentration. The geometry shown in Figure 2 (a) has a stress concentration factor of 1.06. Dimensional tolerances of about 0.025 mm are recommended. Of particular importance is symmetry of the machined specimens with res
31、pect to the specimen centerline. It is also important to avoid scratches during the machining process. After specimens are machined, any excess material and burrs should be removed, while maintaining the integrity of the specimen finish. Appendix A provides recommended guidelines on the specimen mac
32、hining process. Specimens should be tested dry as molded (DAM) unless otherwise indicated. Specimens should be stored in a controlled environment subsequent to machining. If moisture has an effect on test results, such as for glass-filled polymers, dry conditioning should be obtained prior to testin
33、g. This is accomplished by heating the specimens at a specific temperature and for a specified duration as per material supplier guidelines. (Any questions, contact GM Materials Engineer.) Dried specimens should be stored in a desiccator with desiccant packs to avoid moisture exposure and absorption
34、 prior to testing. Specimens should also be stored away from ultraviolet (UV) exposure. Visual inspection with unaided eyes or with low magnification should be used to note any obvious abnormalities such as fillet undercutting, cracks, or machining marks. Specimen dimensions including thickness and
35、width should be measure using method(s) or instruments in a manner not to introduce any visible marks or scratches. 4.2 Conditions. 4.2.1 Environmental Conditions. 4.2.2 Test Conditions. Deviations from the requirements of this standard shall have been agreed upon. Such requirements shall be specifi
36、ed on component drawings, test certificates, reports, etc. Note: GM Materials Engineering shall be consulted on any proposed deviation to this standard. 4.3 Instructions. 4.3.1 Specimen thickness and width should be measured to within 0.025 mm or smaller at several locations along the gage section.
37、The minimum measurement should be used for the stress calculations. The specimen should be positioned in the grips precisely to have good alignment consistently from specimen to specimen. Grip pressure producing the normal force on the specimen grip ends should be adequate to avoid slipping during c
38、yclic loading, while not too high to cause visible deformation of the specimen grip ends. Tabs between the specimen surface and the grip faces may be necessary to reduce non-uniform grip force distribution and avoid grip end failures. 4.3.2 Alignment should be verified prior to starting the testing
39、program. A common method for verifying alignment is using a strain gaged round or rectangular cross section bar with two or three series of four strain gages positioned across from each other (left-right and front-back), and each series positioned at a constant axial position. ASTM E1012 provides gu
40、idelines for verification of test frame and specimen alignment under tensile and compressive axial loads. It is recommended that alignment be checked at loads similar in nature (tension and/or compression, static and cyclic) and in magnitude to those applied during the testing program. Bending stres
41、s measured from misalignment should typically be within 5% of the applied axial stress. 4.3.3 If the testing program includes tension compression cycling (as opposed to tension-tension cycling), it may be necessary to grip the specimen at a shorter length, (i.e., in the transition region between the
42、 specimen gage section and grip end) to avoid buckling during the compression part of the loading. For example, gripping the specimen geometry shown in Figure 2(a) at a width of about 15 mm, which is the average of the gage section and grip section widths, has proven successful without inducing grip
43、 end failures. As mentioned earlier, Copyright General Motors Company Provided by IHS under license with General Motors CompanyNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-GM WORLDWIDE ENGINEERING STANDARDS GMW16970 Copyright 2013 General Motors Company All Righ
44、ts Reserved July 2013 Page 4 of 13 however, positioning the specimen precisely and consistently in the grips is essential for good alignment and validity of the test results. 4.3.4 Prior to testing, the temperature and humidity should be recorded. For other than room temperature testing, the specime
45、n should be stabilized at the testing temperature for a pre-determined duration in the environmental chamber prior to testing to ensure the specimen temperature is the same as the temperature in the chamber. 4.3.5 The loading to be used can be specified by stress amplitude Pa and R ratio (ratio of m
46、inimum load Pmin to maximum load Pmax) or mean load Pm, or by specifying the maximum and minimum loads. The following relations exist between these quantities: R = Pmin/Pmax Equation (1) Pmax = Pm + Pa = Pa (2/(1-R) Equation (2) Pmin = Pm - Pa Equation (3) Pa = (Pmax - Pmin)/2 Equation (4) Pm = (Pma
47、x + Pmin)/2 Equation (5) The R ratio of -1 indicates fully-reversed loading and a zero R ratio indicates pulsating tension loading. The most commonly used R ratio for fatigue testing of polymers is 0.1. The load amplitude has a primary effect on fatigue life, while the mean load has a secondary but
48、still significant effect. Stress values corresponding to the aforementioned load quantities are calculated by dividing the loads by the initial measured cross sectional area. The required R ratio value for the fatigue testing will be provided by GM Material Engineer. The testing machine can be progr
49、ammed to follow a variety of dynamic load waveforms (sinusoidal, triangular, square, etc.) Unless otherwise specified, it is recommended to use a sinusoidal waveform. Test data may be recorded at specified intervals during each fatigue test using an automated data acquisition system. This data may include cycle number, load amplitude, mean load and, if desired, deformation amplitude and mean. An appropriate data recording scheme should be used to record the data at selected cycles. E
