Which grinding wheel removes small radius curve corrugation best?

Why curve corrugation is a different problem from tangent-track corrugation

• The damage comes as a package. Where the curve radius is tight, the dominant defects are not a single wave pattern: side wear and corrugation appear together, along with fatigue cracking at the gauge corner. The damage mix is radius-driven — on urban and metro lines with radius below about 1,000 m, side wear and corrugation lead; on heavy-haul lines, crushing and flange flow lead; on high-speed lines, contact fatigue leads. A wheel chosen from the defect name alone will miss the two defects travelling with it.
• The surface is the symptom, not the cause. Once the rail head profile has moved away from the design shape, contact concentrates on the gauge corner. That raises the local slip and the local stress, and the corrugation regenerates from the same mechanism that created it. This is why a wheel that removes the wave without restoring the profile changes the calendar, not the outcome.
• The crack window is narrow. Rolling-contact fatigue builds through ratcheting of the surface layer: micro-cracks nucleate, then initiate, then propagate, and the first two stages take up the great majority of the fatigue life. When crack depth is still within about 0.2 mm, a normal grinding pass removes the cracked layer completely. Past that depth the surface can be cleaned while the crack root survives — and it will come back. This is the technical reason for grinding curves on a short cycle rather than on condition.
• Heat is harder to shed. Curve passes often run at lower speed and higher unit pressure, which is exactly the combination that produces burning and white layer. Ground-surface finishing temperature has to stay below 150 °C, and the white layer generated at the surface is roughly 870 HV0.3 against about 340 HV0.3 for the parent metal. That brittle layer fragments under traffic and feeds new cracking, so a wheel that burns to gain cutting depth is a false economy.

What the wheel has to deliver on a curve

One caution on hardness, because it is the most common selection error. A harder wheel is not a more durable cutting tool on a curve. In our own series tests, increasing the bond content raised compressive strength but reduced the amount of material removed — the range tested from GS-10 to GS-15 fell from 1.0 g to 0.2 g of stock removed while the grinding ratio rose from 12.8 to 24.1. Higher strength means the grain stays covered and slides instead of cutting. The middle of the range is the optimum, and an over-hard wheel on a curve produces exactly the glazed, burnished surface that operators mistake for a good result.

Selection by machine

Whichever machine you use, the same pass-depth arithmetic applies. Main-line contract limits cap base-material removal at 0.5 mm per pass, and the typical active-grinding removal per pass is about 0.2 mm. That is the budget you are working with: the wave has to be removed inside a small number of passes, which is why a coarse, sharp wheel and a properly corrected profile matter more on a tight curve than on plain line.

The grinding cycle is part of the wheel specification

Wheel grade and cycle length are one decision, not two. Chinese network practice sets preventive cycles by radius and by traffic:

Two consequences follow for the wheel you buy. First, a curve wheel runs more pass-kilometres per year than a tangent wheel, so its cost per pass-kilometre dominates the curve budget — which is why wear rate, not unit price, is the number to compare. Second, a short cycle only works if each pass cuts shallowly and consistently; a wheel that forces you to take deep, damaging passes to compensate destroys the very profile you are trying to restore.

What our own curve and heavy-haul data shows

Wear figures are duty-dependent, so compare them only against lines with similar conditions. Ours, on measured programmes:

• Liuzhou, GMC-96X, 2022 — 18.72 pass-km completed inside a single 2-hour night possession, with wheel wear of 20-30 mm for the composite design against 43.5-59 mm for the imported reference wheel under the same conditions.
• Yiyang line — 5.7 pass-km per mm of wheel wear, averaging 228 pass-km per wheel.
• Shuohuang heavy-haul — 4.98 pass-km per mm, averaging 199 pass-km per wheel; a single-trolley comparison in 2023 recorded 5.96 against 4.99 for the wheel in use, an improvement of about 19.4%, with a best single wheel reaching 332.3 pass-km.
• Hewu high-speed line, PGM-96C, 2020 — 4.28 pass-km per mm against 3.27 for the wheel then in use, or about 1.31 times, with an average of 214.22 pass-km per wheel and no continuous burning.
• Bench comparison at 7 km/h and 8 kW — composite wheels reached 100-150 pass-km against 50-60 pass-km for ordinary and imported reference wheels, and cleared defects in 2-3 passes against 4-6. The pass-count difference is the one that matters on a curve, because it is what lets you finish inside a short possession.

None of those numbers replaces a trial on your own track. Run a small set against your incumbent wheel on the same machine, in the same windows, and measure wear in pass-kilometres rather than in calendar time.

How to tell the corrugation is actually gone

Verification is where curve jobs most often get signed off too early. Check the wave as well as the finish:

1. Wave removal — measure the longitudinal profile before and after; a smoother surface with the wave periodicity intact is a failed pass, not a partial success.
2. Profile conformance — GQI of 85 or better is the good band, with zone tolerances of rail-head centre +0.2 / −0.4 mm and ±0.2 mm across zones 2, 3 and 4.
3. Finish and heat damage — Ra of 10 µm or better, no continuous blue band, and a straightness reading within 0.3 mm under a 1 m straightedge with a ramp-off steeper than 1‰.
4. The next cycle — the real verdict arrives one cycle later. If the corrugation reappears early, the profile correction was incomplete, not the wheel.

For the measurement side, rail wear measurement sets out how to record it; for the acceptance framework, see grinding acceptance standards. The corrugation mechanism itself is covered in this analysis of causes and grinding solutions and in how to remove corrugation and verify the result. If your corrugation is on tangent track rather than tight curves, the selection logic and the wheel recommendations are different — see how to choose a wheel to remove rail corrugation. Curve side wear and the accompanying profile work are covered in managing side wear on curved track, and for the machine side of the purchase, how to choose a rail grinding company sets out the evaluation criteria.