Prakt. Met. Sonderband 46 (2014) 319
effect. Both 2. MINI CASE STUDIES
1 increasing
mechanism. 2.1 CRACKED COMPRESSOR DISK RIM
ooh oss. During a scheduled outage, a compressor disk was found cracked at the rim. The crack
jical system initiated from a disk slot, adjacent to the turbine side disk rim face, and propagated
ncing these outward towards the disk rim's outer diameter. No fragments of the disk rim became
peening or liberated yet. The crack was open and somewhat gaping at a width of ca. 0.1 mm and
nce surface visually detectable with the unaided eye (Fig. 2).
ties of the 2.2 CRACKED LOCKING PINS
ded with all
ld partly or Locking pins used to arrest moving blades in a gas turbine compressor were found crack
on between after a testing campaign. They were made of the soft-martensitic, precipitation-
ult in fretting hardenable high-chromium stainless steel 17-4 PH (AISI 630). In this case, fretting was
might vary only a contributing factor. The primary cause of failure was a machining error that caused
2pending on some fillet radii to become much too small, causing a notch effect not accounted for by
in contact design (Fig. 3).
1) disruption
act with the 2.3 FRACTURED SEAL PLATE
ing and the , : : ,
ant oxide..." While the first two cases described above dealt with failures that occurred at lower or
tting oxides moderate temperatures, this case study is about a component that failed at temperature.
lly there are The seal plate fractured in a high-cycle fatigue mode. However, the crack was induced in
etimes very a fretted zone where the seal plate's edge was in contact with a groove in the turbine disk
etallic wear of a heavv-duty stationary gas turbine and suffered relative movement there.
termined by 2.4 FRACTURED COMPRESSOR BEARING SUPPORT
behavior of
to note that In this last mini case study, a fracture of a forged compressor bearing support of a large
ers for their stationary gas turbine is shown that cracked in service. The primary cause of failure was
/ vulnerable, unrelated to fretting. However, it is interesting to note that once the fatigue crack, that was
hat category induced at a machining fault and was unrelated to fretting, fractured one side of the U-
rding to the shaped support, the clamping force lessened sufficiently to allow the relative motion of the
d by fretting mating partners of this tribological system that is necessary for fretting to occur. By this,
notch effect mating parts of a multi-component system that are normally at rest became somewhat
rse, this only mobile by the loosened assembly (Fig. 4).
r due to the
jory existed, 3. REFERENCES
nding on the .
cracking can [1] Waterhouse, R.B. (ed.): Fretting Fatigue. Applied Science Publishers, London
> features of 1981.
and, in steel, } N } i
; at an angle [2] Waterhouse, R.B.; Lindley, T.C. (ed.): Fretting Fatigue. Mechanical Engineering
on the com- Publications Limited, London 1994. (Papers presented at the International
n previously Symposium on Fretting Fatigue held at the University of Sheffield).
ks were not
y of the area [3] Nowell, D.; Dini, D.; Hills, D.A.: Recent developments in the understanding of
undamaged fretting fatigue. Engineering Fracture Mechanics 73 (2006) 207 - 222.
corners and
) fatigue can [4] Christiner, T.; Reiser, J.; Godor |.; Eichlseder, W.; Trieb, F.; Stuhlinger, R.: The
ading on the fatigue endurance limit of a high strength Cr-Ni steel in a fretting dominated
regime. Tribology International 59 (2013) 97 - 103.