Another thing about those tight curves and wheel wear. With the rail being new, and I am assuming unground, the contact patch on the rail will tend to be on the gauge corner. Through the turnouts this will be further exasperated as the rails are not inclined. As the wheels track through the tight curves, this very narrow contact patch on the rail will be on the fillet (is that what's it called?) of the wheel between the wheel flange and tyre (flat?). This concentrated pressure would increase the wear rate of the flange. And those iron fillings you see on some curves, is mostly from the wheels and not the rail. Rail steel is a fair bit harder than wheel steel.The rails are inclined inwards at the same angle as the wheel so the flat of the wheel runs on the flat of the rail. ON timber sleepers the incline is provided by the sleeper plates. With concrete sleepers the rail mount itself is inclined.
There was a trial conducted a number of years ago on Cowan Bank to see if switching the lubricators off would reduce the number of train failures due to wheelslip. The trial was halted after the rails started to be eaten quite significantly. And if the rails were wearing fast, the wheels no doubt would have been too.
What happens is this is NOT done on a line with heavy traffic is all the weight is taken as you described on the inside edge of the rail, this stress's the head of the rail where it attachs to the web and one ends with the inside edge of the rail breaking off through fatigue.
For both flanges to drag on the inside edge of the track the track must be AT LEAST 13mm under gauge and I seriously doubt the RRL is that much under gauge. It was widely circulated that the RFR was 12mm under gauge THIS IS NOT CORRECT. I am NOT trying to be a pain in the neck here just trying to provide accurate information.
A senior manager told me the reason why the RFR is 1.5mm undergauge is the sleepers and rail unsulators were made for rail lighter than the 60kg rail used and the 1.5mm under gauging is because the head of the rail is that much wider. I do not know if this is true or not. I have been told on concrete sleepers the rail insulating pads are made interchangeable so different ones are used for rails of different weights so the gauge will be correct (the sleepers used on the NE line are this way) apparently this was not followed on the RFR sleepers.
Yep, both the rail and wheels are inclined at 1 in 20 in Australia, though this won't help much on curves below 800m radius where the self steering ability of rail wheels are lost. Once you get below 800m radius, the outside wheel will ride up against the rail. When you grind rail in a curve, you get the rail head to match up with the shape of the wheel so the rail and wheel match perfectly, reducing the pressure applied to the rail. With unground rail, the match between rail and wheel is not so perfect, creating a more concentrated pressure at the gauge corner. More pressure, more wear.
And if the track is so tight that both wheels are flanging, this is bad. Significant hunting will be the result and wheels tasting dirt and not steel will be the result.
As for the differing size of the rail head between different weight rails, this is true 60kg has a wider head then 53kg or 50kg, or 94lb. I'm sure that you can get differing size biscuits to accommodate differing rail foot widths, though I'm not sure if it changes the centreline of the rail on the rail seat.