What acceptance criteria decide whether a grinding job passed?

Question

Accepted Answer

What acceptance criteria decide whether a grinding job passed?

Five criteria, all of them measured rather than looked at: profile tolerance against the target template, surface roughness, absence of burn, residual corrugation amplitude, and depth still inside the envelope. The practical difficulty is not the list — it is that every one of the five is a comparison, so without a value recorded before the possession none of them can be judged. An acceptance sheet with thresholds but no baseline will pass a bad job and fail a good one with equal ease.

Rail grinding machine working along a railway line — Acceptance is a comparison between two sets of measurements at the same positions, taken far enough apart in time to include the work.

The five criteria, and how each one is judged

Criterion	How it is judged	Threshold	What failure looks like
1. Profile tolerance	Head shape measured at the same fixed cross-sections used for the baseline, then evaluated zone by zone against the target template the operator set	Zone 1, the rail top centre, within +0.2 /−0.4 mm; zones 2, 3 and 4 within ±0.2 mm. Area quality index of 85 or more is rated good, 70 or more acceptable; a section held at 85 or above with a standard deviation of 8 or less is the strong result	In tolerance along the running band but still out at the gauge corner, which is exactly the zone that was under-covered
2. Surface roughness	Roughness measured with a portable tester at the same points as the baseline, plus a 1 m straightedge laid on the running surface	Roughness 10 µm or finer and no step greater than 0.3 mm under the straightedge, with the specified ramp at the ends of the cut	A surface that looks bright but is torn; metal picks up on it within a few thousand tonnes and the roughness climbs again
3. No burn	Every metre of the ground band walked and looked at while the surface is still as the machine left it. Colour is a temperature record: below 471 °C leaves none, 471–600 °C leaves pale yellow, 600–735 °C leaves blue	No blue, no yellow streaking. Workpiece temperature under 150 °C after grinding, and no fracture, crack or abnormal vibration during the work	Blue marks clustered where the machine slowed or dwelt — a hard, brittle layer that cracks and drives new fatigue
4. Residual corrugation	Re-measured at exactly the baseline positions and the same chord, so the before and after describe the same steel	Amplitude brought below the level at which the line triggers corrective action — on conventional lines, 0.3 mm of corrugation depth or lipping width	The peak has shifted rather than shrunk: same wavelength, lower amplitude on paper, the crest now sitting where the trough used to be
5. Depth within limit	Metal removed at the controlling zone compared with the envelope for that rail section and curve class, and with the acceptance cap for a single job	Inside the envelope, and no deeper than the 0.5 mm acceptance cap — a ceiling, never a target	A profile that measures perfectly and has spent two or three future cycles' worth of section to get there

Note what the thresholds are not. The 0.5 mm figure is the most commonly misused number in the set: it caps what a job may take off, and it is not the depth a job should aim for. Planned depths are smaller, and they differ by job class — the separation between a planned depth, a trigger line and an acceptance cap is set out in what grinding depth is safe. Likewise the 0.3 mm corrugation figure is where corrective action is triggered, so a job that leaves 0.29 mm has technically passed and practically achieved very little.

Four ways a failing job produces a passing sheet

What happened on site	What the sheet says	What is actually true
The baseline was taken from an older survey, not from the same possession sequence	Profile within tolerance	The comparison spans a traffic interval as well as a grinding job. Tolerance can be met by a rail that was already in tolerance before the machine arrived
The after-measurement moved to convenient positions	Corrugation reduced by a stated percentage	The two data sets describe different steel. This is the single most common reason a good job and a bad job produce similar-looking reports
Burn judged after the line reopened	No discolouration	Traffic has already polished the surface. Burn has to be read on the band as the machine left it, which is why that inspection belongs to the possession, not to the following week
Depth checked only at the rail top centre	Depth within limit	Removal is not uniform around the head: on one recorded trial the outer zone gave up 0.165–0.326 mm while the inner zone gave up 0.142–0.241 mm. Checking the shallowest zone certifies the whole head on the strength of its best area

Two further checks belong to acceptance even though they are effect measurements rather than workmanship ones. On turnouts, the contact band after grinding should sit at the width the line calls for — 25–30 mm on high-speed turnouts, 35–40 mm on conventional and heavy-haul. And where the reason for grinding was dynamic, the before-and-after ride measurement is part of the evidence: turnout grinding has been recorded taking lateral body acceleration from about 0.20 m/s² down to 0.04–0.08 m/s². Neither is a substitute for the five criteria, but a job that passes all five and moves neither number has not fixed the problem it was ordered for. The turnout-specific version of this work is scoped in rail turnout grinding solutions.

Rail grinding unit on track in a maintenance possession — The band a machine leaves is read against the band that was there before it, zone by zone rather than as one figure.

When a criterion fails, what it usually means

Failed criterion	Most likely cause	First thing to change
Profile still out at the gauge corner	The angle pattern does not reach that zone, or the passes do not overlap enough to cover the band between them	The pattern. Adding passes at unchanged angles cuts the same zone deeper and leaves the same zone untouched. The profile side of this is in rail profile grinding techniques and standards
Roughness above the limit	Working speed too high for the depth being taken, so the wheel scrubs instead of cutting	Speed and depth together, verified on a short trial length before the whole section is signed off
Blue patches	Dwell, or too much pressure concentrated in one spot, or a single pass too deep for the cooling that exists	Pressure and pass depth, and the time the head spends on one position
Corrugation back within a year	The peaks were cut but the contact geometry that generates them was left as it was	The target profile and the interval, not the number of passes. Background in restoring the rail head contact surface

There is a hard limit to what acceptance can tell you. It says the metal that was removed was removed to specification; it cannot say the right metal was chosen. A job that passes all five criteria on a section that should have been ground deeper — or renewed instead — is a correctly executed wrong decision, and the acceptance sheet will not catch it. That judgement sits upstream, in grind or replace, and it is also why the acceptance data has to feed the next plan rather than close the file. A supplier's acceptance record is one of the few hard things you can compare when choosing one at all, as set out in choosing a rail grinding company.

Railway grinding machine on a main line during maintenance — A sign-off is only as strong as the position records behind it, because every criterion above is a before-and-after comparison.

Where the whole loop is documented — target setting, acceptance thresholds and the standards they come from — the material is in grinding acceptance standards, with the measurement side in how track geometry is measured. What the machine is actually doing to the head while those numbers are being generated is covered in how a grinding train removes rail defects. The consumable is one input to the result rather than the result itself, which is why Molaton delivers wheels against a stated removal and roughness target instead of a dimensional catalogue alone; the programme-level view is in rail grinding solutions.

Why trust this answer — RailwayCare (product brand Molaton) has manufactured rail grinding wheels since 2004 — the first dedicated producer in China, born from the friction-materials laboratory of Wuhan University of Technology. We drafted the industry standard JB/T 11431 for rail grinding wheels, are certified to ISO 9001 / ISO 45001 / CRCC, and every claim below is backed by on-track tests on high-speed, heavy-haul and metro networks.

Will your next job survive its own acceptance?

Send the target profile, the before-grinding survey and the thresholds you intend to sign against. We will tell you which of the five criteria your current data can actually evidence — and where a Molaton wheel set has to be specified to reach them.

Ask About Acceptance Criteria

Professional Rail Track Maintenance Solutions Provider

Safe Tracks Proven by China

What acceptance criteria decide whether a grinding job passed?