Welding on top of rail, to avoid noise in curves

[kingnero]

For light rail (metro, tramways, ...) in cities, in tight curves the top of the rail is sometimes welded with a 307 filler material (austenitic, work-hardenable layer with 18-8-6 Cr-Ni-Mn chemistry). The goal of this layer is (supposed to be) the mitigation of noise due to the wheel slip.
It is generally assumed in the railway world that this "squeal noise" is caused by sideways slip (and not wheel slip due to fixed axes/difference in longituninal distance covered).

I'm involved in this as the welding engineer (with years of experience in railway welding, but not in acoustics). Welding pearlitic and bainitic steel often poses a whole new set of difficulties. Also, it makes rails more vulnerable for defects, especially so in areas that are already very prone to damage (mostly corrugation, also due to wheel slip, but squats and head checks as well because of the hertzian contact stresses ...).
This welding layer is specified by the railway authority, with about as much knowledge and experience in this as the average forklift driver in fluid mechanics.
Is there any literature about this phenomenon (the application of an anti-squeal noise weld layer)? What is the goal (I suppose lowering the coefficient of friction), is it effective, are there alternatives, why choosing an austenitic filler for this application, ...
I'm specifically trying to learn about the acoustics side of things, I think we've got the welding part quite covered.
Thanks for your insights!

 

[GregLocock]

I can help once the rail starts vibrating, but the excitation mechanism is new to me. This paper seems relevant but I can't download it

Control of wheel/rail noise and vibration​

NTL Rosa-P (.gov)
https://rosap.ntl.bts.gov


PDF

by PJ Remington · 1982

 

[IRstuff]

I had no problem downloading the the paper; it downloaded into my download manager just from clicking the link, but it is from 1984. File is attached herein. They seem to indicate that the radius of curvature seems to be the critical parameter.

 

[kingnero]

Thanks for the publication! Under § 5.1.1, they do mention hardfacing (which is typically very different from a 307 alloy!).
Using a Etecta 5 Spezial filler metal, in Germany... I didn't find anything, but I've forwarded this question to one of my contacts at voestalpine, one of the major German (filler) metal producers. Hopefully they can find something in their archives.





This sure seems more effective in maintaining the contact area between wheel and rail, as to what is specified (and applied) currently:
A soft, workhardenable filler in a machined groove, grinded flat after welding. After some tonnage (we're talking weeks, not even months), the filler is coldformed and compacted into the groove, and the wheel's contact area shift to the sides, thus contacting the hard, pearlitic steel again...
In the below image, disregard the weldment on the right hand side of the rail head, this is the anti-wear layer in curves.