The chevron notch specimen is especially useful for measuring the plane strain fracture toughness K I_ of brittle materials. The specimen's unique advantages are: (i) ~ sharp natural crack is produced during the early stage of test loading so that no pre-cracking is required, and (2) the test load passes through a maximum at a constant, material-independent crack length/width ratio for a specific geometry, and (3) no post-test crack length measurement is required. For the chevron notch specimen the fracture toughness may be expressed aswhere Pma is the maximum test load, and A is dependent on the specimer geometry ~nd the manner of specimen loading.The chevron notch specimen has recently been studied in three con. figurations: short bar [1,2], short rod [5,4], and four point bend [5] The factor A has been obtained for the short bar specimen by the superposition of ligament-dependent and ligament-independent solutions for straight through crack, referred to as the straight through crack assu~ tion (STCA), and also from experimental compliance calibrations [6]. The factor A has recently been obtained for the short rod specimen frol experimental compliance measurements [7]. The Klc-Pm-x relation for the four point bend specimen has been developed using t~e straight through crack assumption and the slice model of Bluhm [5]. For both the straight through crack assumption and the slice model, complex relations for the calculation of A have resulted.In this paper, a closed form relationship between the factor A and specific geometries of the four point bend specimen is presented.The chevron notch four point bend specimen is characterized by the dimensional variables shown in Fig. i. The crack length and notch parameters are those expressed in dimensionless form as ~ = a/W, ~o = a_/W, and e. = al/W. The relation between K._ and the applied load duty ing cra~k extension, assuming a flat c r a~ growth resistance curve, is then given by [6]: P y, = P 1 dC;r ~i-~o]1/2 KIc = ~ ~ [7 d= ~-~ (2 o where C~ = E'BCtr, a dimensionless compliance expressed in terms of the compliance Ctr of a specimen whose crack front is trapezoidal, as Int Journ of Fracture 16 (1980

Fracture toughness of a sintered A12O3 was determined with four‐point‐bend specimens having either straight‐through or chevron notches. For the straight‐through notched specimens, measured KIc decreased with decreasing notch width. For the smallest notch width (66 μm) KIc= 3.42±0.13 MN m−¾. For specimens with chevron notches, a crack initiates and extends from the tip of the notch under increasing load. KIc is calculated from the maximum load without measuring crack length, under the assumption that the derivative of the compliance is the same as that for a specimen with a straight‐through crack. A refined calculation accounts for the truncated chevron crack shape at maximum load using Bluhm's slice model. For the chevronnotch configuration, a value of KIc= 3.49±0.11 MN m−¾ was measured, which appears to be independent of the initial notch length a0 (distance from the crack mouth to the tip of the triangular notch). An effect of a1 (length of the chevron notch at the surface) on KIc was observed, independent of whether the calculation of KIc was based on the straight‐through crack assumption or on the slice model.

For the determination of fracture toughness especially with brittle materials, a short bar specimen with rectangular cross section and chevron notch can be used. As the crack propagates from the tip of the triangular notch, the load increases to a maximum then decreases. To obtain the relation between the fracture toughness K~ and maximum load P~ax, calculations of Srawley and Gross for specimens with a straight-through crack were applied to the specimens with chevron notches. For the specimens with a straight-through crack, an analytical expression was obtained. This expression was used for the calculation of the Kfc -P m a x relation under the assumption that the change of the compliance with crack length for the specimen with a chevron notch is the same as for a specimen with a straight-through crack.Comparative compliance calibrations with specimens of different geometries agreed very closely with the analytical results for the K~c -P m a x relation. For the first part of crack extension before reaching maximum load, the dimensionless quantity Y* = KIcB X/-W/P where B and W are the specimen thickness and width, and P the applied load, is greater for the analytical approach than that obtained from the experimental results. This difference can be explained by applying the slice model proposed by Bluhm.

The short rod chevron-notch specimen has the advantages of (1) crack development at the chevron tip during the early stage of test loading and (2) convenient calculation of Ktc from the maximum test load and a calibration factor which depends only on the specimen geometry and manner of loading. For generalized application, calibration of the specimen over a range of specimen proportions and chevron-notch configurations is necessary. Such was the objective of this investigation, wherein calibration of the short rod specimen was made by means of experimental compliance measurements converted into dimensionless stress intensity factor coefficients. List of symbolsa crack length a0 initial crack length (distance from line of load application to tip of chevron) al length of chevron notch at the surface (distance from line of load application to point of chevron emergence at specimen surface) a,~ crack length at minimum of Y* A constant of proportionality between Kit and Pmx used by Barker and Guest [4] C specimen compliance (load line displacement divided by load) C' = E' DC Ct, compliance of specimen with trapezoidal crack C~, = E'DC,, D specimen diameter E Young's modulus E' = E for plane stress, = E/(I -v 2) for plane strain KI mode I stress intensity factor Ktc plane-strain fracture toughness P load Pmax maximum load Y dimensionless stress intensity factor coefficient for a straight-through crack, = KDX/-W/P stress intensity factor for a trapezoidal crack, = KDX/-W/P,= y(~)l,2--y* dimensionless Y* minimum of Y* a = a/W ao = ao/ W al = adW am = am/W v Poisson's ratio

No abstract