EEE A ET TETE
SEE
——
GERE
RE FETTE
o ——X—
Fig.8 The corresponding profiles of Specimen II, right
side. The specimen was precracked and broken up at
-196? C. The CODi value is smaller than the accuracy of
the measurement. (a) profile 3, (b) profile 4.
The profiles are presented in the following figures and give
a rather good impression, of how the CODi values differ
between the two specimens investigated. The CODi of
specimen I is very small, only of the order of the accuracy
of the analyzing method. The CODi of specimen II is
larger by a factor 4.
Discussion
From the view of fracture mechanics the most interesting
result of this investigation is that the mode of precracking
has only a little effect on the critical stress intensity (as it
should) but a large effect on the critical crack tip opening
displacement. So it seems that the measured COD of
Specimen II was formed already during the precracking. If
the breaking in liquid nitrogen produced this COD, the
CODi of Specimen I would be not so different from the
CODi of Specimen II. But the physical reason of this effect
is not fully understood up to now and should be subject of
future investigations.
The measurement method using stereo images from the
scanning electron microscope is accurate and effective to
compare the fracture surfaces of the specimen halves. This
depends mainly on the possibility of identifying
corresponding points, profiles and structures on the
conjugate surfaces. The material presented in this paper
was very suitable for that task and promises good results
with specimen of similar structured surfaces.
References
[1] S.K. Gosh, Photogrammetric calibration of a scanning
electron microscope. Photogrammetria 31 (1975) 91-
114.
[2] O. Kolednik, A contribution to tereophotogrammetry
with scanning electron microscope. Practical
Metallography 18 (1981) 561-573.
[3] O. Kolednik and H.P. Stiiwe, The stereoscopic
determination of the critical crack tip opening
displacement. Engineering Fractrue Mechanics 21
(1985) 145-155.
