Astm stp 637 1977

CYCLIC STRESS-STRAIN AND PLASTIC DEFORMATION ASPECTS OF FATIGUE CRACK GROWTH A symposium sponsored by ASTM Committee E-9 on Fatigue AMERICAN SOCIETY FOR TESTING AND MATERIALS St Louis, Mo., 2-8 May 1976 ASTM SPECIAL TECHNICAL PUBLICATION 637 L F Impellizzeri, symposium chairman List price $25.00 04-637000-30 F^,:0„,„=F^, Stess - S t r a i n -Curve - FIG I—Simulation of local stress-strain behavior Unit S - N D a t a Required Const Amplitude Ref of Damage Test Hysteresis Strain Control Loop: Rt =-1.0 Representation Consideration Accu- of Mean Stress mulation hree Line1 EQ Om Linear, s -N-Curve AE' Initiation " ' E (Unnotched) âã(D N, Np Damage per Strain Control Polygon Rt =-1.0 Element (Unnotched) Stress Cycle L Stress Control S-N-Curve Stress Cycle Stress Control Determination of Linear 0^-2£„ = f(N,) Constant Life (Notched or Unnotched) ®»(3) Propagation 0^„,-2E^ Cliange ofS-NCurve Slope with Change Diagram S-N-Curve Linear of Residual Stress RAE-Method Linear (Notched) (K,'Sa)s =f(N) FIG 2—Damage calculation in local stress fatigue analysis methods Copyright by ASTM Int'l (all rights reserved); Mon Dec 21 11:24:29 EST 2015 Downloaded/printed by University of Washington (University of Washington) pursuant to License Agreement No further reproductions a SCHUTZ AND GERHARZ ON FATIGUE LIFE EVALUATION METHODS 213 accumulated? Some research work has been undertaken to answer this particular question, but it was not intended to develop a procedure of general applicability or to compete with the available local stress fatigue analysis methods To achieve this goal, the stress-strain history at a notch root was determined experimentally, thus avoiding the uncertainties inherent in the theoretical simulation of the elastoplastic stress-strain behavior at the notch root Procedure Specimen Load Program Notched and unnotched specimens shown in Fig were manufactured from both the aluminium alloys Al-Cu-2Mg (sheet) and AZ 74/72 (extrusion) equivalent to 2024-T and 7075-T respectively The center notched specimens had an 8-mm-diameter hole which, for the geometry adopted, corresponded to a stress concentration factor of Ki = 2.5 Two program load sequences (Fig 4) were chosen to represent the load history at the wing lower surface of a transport'airplane The first stress sequence (Fig a) contained a ground-air-ground (GAG) cycle and was called the standard program This standard program was modified by omitting the ground load, and was then called program without GAG cycle (Fig h) Both programs included one large air-load cycle occurring once -270135- t^ dimensions in mm FIG 3—Specimens, notched and unnotched Copyright by ASTM Int'l (all rights reserved); Mon Dec 21 11:24:29 EST 2015 Downloaded/printed by University of Washington (University of Washington) pursuant to License Agreement No further reproductions authorized 214 FATIGUE CRACK GROWTH Reference Area Blocked Cycles per Flight in Every 200 th Flight Flighl- (b) Program without (Q ) standard Program G-A-G Load Cycle FIG 4—Loading programs in every 200 flights which was first applied in the 100th flight Excluding the GAG cycles and the large air-load which occurred every 200 flights, each flight contained 131 load cycles More detailed information on the toad program is given in Ref Experimental Procedures Notched and unnotched specimens were loaded with the flight-by-flight nominal stress sequences and also under constant-amplitude loading conditions in a tape-controlled servohydraulic testing machine (Schenck-Hydropuls) by a 60-kN cylinder The deformation of the hole was recorded continuously during fatigue testing as described in Ref to determine the life-to-crack initiation At that stage, the crack was still very small with a projected surface of 0.3 mm^ It is the authors' opinion that, for this size of crack and severity of stress concentration, the crack development is still governed by the stress concentration Local strain at the hole was measured continuously and recorded by strain gages installed in the hole \S\ Because of both poor fatigue strength of the strain gages and the zero-point drift, the measured values can be relied on only over a relatively short portion of the lifetime Accordingly, a measuring procedure was developed which was able to distinguish between the zero-point drifts of the strain gages arising in the course of long recording times and the actual strain redistributions that occurred at the notch root Copyright by ASTM Int'l (all rights reserved); Mon Dec 21 11:24:29 EST 2015 Downloaded/printed by University of Washington (University of Washington) pursuant to License Agreement No further reproductions authorized I SCHUTZ AND GERHARZ ON FATIGUE LIFE EVALUATION METHODS 215 The loads needed to force an unnotched specimen to follow the recorded strain sequence were also measured and recorded continuously This method, called the companion specimen method, generally is applied [9-11] for determining the stress behavior at the notch root Life Predictions From the continuous records of the local stress and strain sequences, the mean strains and mean stresses were determined and plotted over the elapsed lifetime (number of flights) Figure shows (for the Al-Cu-2 Mg material) the mean strain and mean stress sequences for the air loads between the large air load occurring in every 200th flight It was found that the mean strains and stresses tended to stabilize within to 20 percent of the life-to-crack initiation The average mean stress approached by the stabiUzation process may be designated the "effective mean stress." After stabihzation, the residual stresses (a«) can be defined by OK = Tim - K,Sm, where a = effective aon L—J>—JL—- aoio- h QD09- > - aooe Material: AlCuMg2 (202i-T3) Frequency 17 Hz 0.006- aoo5aoo40 IX 3M 500 700 900 1100 1300 Number of Flights Effective Notch-Root Mean Stress •12.i daN/nnm^ rk I f FIG 5a—Measured notch root mean stress and mean strain sequence: standard load program Copyright by ASTM Int'l (all rights reserved); Mon Dec 21 11:24:29 EST 2015 Downloaded/printed by University of Washington (University of Washington) pursuant to License Agreement No further reproductions authoriz 216 FATIGUE CRACK GROWTH 0011 a ^ 0009 (/> c • am z g aooT- Material: AICuMg2 (202i-T3) I Frequency: 17 Hz aoos (1005- 1100 > « ;.E 24- VI 16- « (/) c 12- :E e- O o ft) 1900 2300 2700 Number of Flights • w• Effective Notch-Root Mean Stress *S.S doN/mm' - * r f Q: 4- FIG 5b—Measured notch root mean stress and mean strain sequence: load program without GA G cvcle mean stress, and 5m = nominal mean stress The effective mean stresses derived from flight-by-flight and constant-amplitude loading are summarized in Table Fatigue life-to-crack initiation was predicted using three methods for defining the stresses to be used in the damage calculations Within each method, the partial damage was linearly accumulated, and the phenomenon of rapid local mean stress stabilization was utilized by Methods A and B Method A employed the local mean stresses (Table 1) and the cyclic stressstrain curves together with plots of "quasi-elastic" notch root stress versus measured notch root strain as shown in Fig to determine the local alternating stresses Unnotched specimen S-N data reported in Ref 12 for AlCu-2 Mg (2024-T 3) and in Ref 13 for AZ 74/72 (7075-T6) were used as a basis for these damage calculations For Method B, the virgin open-hole specimens were preloaded by one load cycle (Ref 8) to induce stresses and strains of the same magnitude as those of the effective mean stresses (Table 1) and mean strains measured I Copyright by ASTM Int'l (all rights reserved); Mon Dec 21 11:24:29 EST 2015 Downloaded/printed by University of Washington (University of Washington) pursuant to License Agreement No further reproductions authorized SCHUTZ AND GERHARZ ON FATIGUE LIFE EVALUATION METHODS p »n o ,m p » n p i / ^ — — ',t^ — «N '(N a -J w ^ i n p v ^ p o • —