14  Prakt. Met. Sonderband 50 (2016) 
At roughly half its length, the dampening pin broke into 2 pieces, i.e. approximately 50 mm from as initially sı 
the head end of the part. The fracture surface is almost exactly perpendicular to the longitudinal axis significant ro 
of the failed part. There are small final forced rupture areas (shear lips) that are tilted at almost 
exactly 45° to the longitudinal axis of the bolt. 
The loose-end side of the pin is evenly worn off circumferentially by approximately 0.3 mm where 
it sat in the airfoil. There are no obvious clamping marks or any other signs that would suggest an 
arrest of the loose end in the blade's airfoil which could have caused an arrest of the loose end and a 
cyclic overload of the pin as a result of this. 
It could be verified that the broken pin had been shot peened in the area designated by the drawing, 
i.e. the shrink seat of the sleeve. The characteristic "moonscape" is plainly visible. 
The surface of the entire dampening bolt has a rather rough look to it as it is produced by machining 
operations such as turning. These machining marks are evenly spaced with a distance between 
peaks of about .3 mm. They run circumferentially around the bolt. They are not consistent with 
grinding marks which would probably form irregular circles. It is evident that the original turning 
marks are also visible on the shot peened surface in the form of an underlying superstructure. The 
surface roughness exhibited by the failed bolt is not a condition well suited for subsequent shot 
peening. 
Neither corrosion pits nor other notch-like surface damage like foreign object damage (FOD), that 
may have served as crack initiators were found on the pin surfaces. 
The two broken pieces exhibit very smooth fracture surfaces which run almost exactly perpen- 
dicular to the longitudinal axis of the pin and cover more than 90 % of the fracture surface. The 
small remainder belongs to a shear lip indicating the ductile nature of a forced final rupture. These 
macroscopic fracture characteristics are consistent with fatigue failure and a cyclic tensile load 
normal to the fracture surface. 
The likely site of crack initiation directly opposite the shear lip is visible on both pieces of the 
broken pin. The characteristic "thumb nail", often marking a crack initiation site is plainly visible 
on the fracture surfaces. Also, a somewhat weak evidence of beach marks is detectable. Their ap- 
pearance may have been enhanced by oxidation. A rather large area of the fracture surface is oxi- 
dised, which was probably assisted by the prevailing high temperature. Such oxidisation may both 
obscure or enhance fracture characteristics. On the broken pin, the two fracture surfaces are 
oxidised to a varying degree, partly in different places, i.e. oxidisation patterns do not match. This, 
too, is probably caused because the two pieces of the cracked bolt tended to hammer against each 
other in service. Generally, oxidisation is somewhat weaker the nearer the area is located to the final 
forced rupture. Obviously, by the end of crack propagation, the relative motion increased. 
All fracture characteristics found during the failure investigation suggest the cause of the 
dampening pin fracture is high cycle fatigue. The root cause of the failure is probably a combination 
of the fatigue load normal for usual operating conditions and an increased surface roughness of the 
pin caused by the final machining operation that is, in the case of the notch-sensitive material 
Udimet 720, detrimental to the fatigue limit. The measured surface roughness of the failed part is 
out of specification and traceable to the use of non-OEM components. High temperature corrosion. 
Fig. 3