RAILPROF measurement data are transferred automatically to a smart phone, which calculates the QI (Quality Index), a measure for the vertical force, directly related to the first derivative of the recorded vertical rail geometry. The force-based standards were developed by TU Delft and this concept received the Best Paper Award at the 7th World Congress on Railway Research in Montreal, Canada, in 2006.
The admissible QI in relation to the line speed is presented in the table, which values are presently the official standards used by ProRail in The Netherlands and the tram and metro companies in Holland.
The vertical wheel rail contact force is approximately proportional to the square of the speed. In this way the standards are speed dependent.
Furthermore the contact force is linearly related to the inclination (first derivative dy/dx of the geometry). The inclination is determined from the versines measured with the RAILPROF on a base of 1 m, sampled at an interval of 5 mm. During data processing the samples are averaged (low-pass filtered) via a moving window of 5 samples. This means that the shortest wavelength in the resulting signal is 25 mm. The calculated QI values are based on this signal. See PDA screen below.
Quality Index (QI) on Smart Phone screen
Blue line represents the vertical geometry, orange line is QI. The person doing the finish grinding can immediately see where the QI is exceeding the norm (by definition QI = 1) - in this case 6 % - and which spots should be ground. In the figure it is clearly seen that the slope right of the middle is too large and so the center part should be ground to decrease the inclination and thus the maximum QI.
Gradient rather than versine
From the right-hand graphs it is obvious that the contact force does not correlate (9 %) with the versine (geometry), whereas the inclination (QI) very well correlates (91 %) with the force. This means, amongst others, that a steel straightedge, as still in use on a wide scale, is absolutely inadequate.
According to recent ProRail figures the claimed track maintenance cost reductions of 10 - 20 % are too conservative. Present practice reveals that these figures are more in the order of 20 - 40 %!