In the framework of the ERRI D 173 programme on fatigue caused by rail
rolling contact, the influence of different rail steel grades on the
occurrence of cracks caused by rolling contact was studied on speciall
y selected curves of four European railway networks. Two lines were op
erated with and two without lubrication by passing locomotives. The ra
ils were made of continuously cast, vacuum-treated steel with no addit
ion of aluminium for deoxidation. The three rail steel grades were 900
A in accordance with UIC 860-V (equivalent to 260 HE hardness in the
as rolled state) and two head-hardened variants with 340 and 370 HE. T
he test sections are inspected at regular intervals in accordance with
a given programme. The results obtained to date after total traffic l
oading of between 60 and 170 MGT are as follows. Wear is reduced subst
antially by the use of harder rails and also by constant lubrication.
Fatigue defects occurred in the form. of head-checks on the high rails
of the curves; the phenomenon was much more pronounced and the crack
distance greater in the case of grade 900 A rails than in that of grad
e 340 and 370 HE. With rails of grade 900 A, spalling occurred in the
case of a lubricated test section with a 350 m (5 degrees) curve, this
damage was so great that the rails had to be removed after 64 MGT. On
samples of grades 900 A (260 HE) and 370 HE taken from this test sect
ion it was found that the grade 900 A material exceeds its ductility l
imit earlier because of pronounced cold deformation so that plastic sh
ear deformations and ultimately cracking and spalling occur. Because o
f their higher yield point, the head-hardened rails prove far more res
istant to fatigue damage. Cold deformation is much lower and so also i
s crack formation. The service life can therefore be expected to be gr
eater than for grade 900 A rails.