Track Cross-Level Deviation and Warp Against a Class Limit

ScopeA rigid truck bridging a section that rises on one rail and falls on the other has one wheel carrying much less than its share, and a lightly loaded wheel on a curve with lateral force is the wheel that climbs. That is the mechanism, and it is why warp limits tighten faster with class of track than most other parameters do. The measurement detail that decides what the reading means is the REFERENCE. FRA warp (49 CFR 213.63(a)) is the raw difference in cross level between two points less than 62 ft apart. On a curve of constant elevation the superelevation cancels out of that difference: readings of 4.6 in and 3.2 in 62 ft apart on a curve designed for 4 in are plus 0.6 in and minus 0.8 in of deviation, and the warp is 1.4 in either way. On a spiral it does not cancel, and it should not -- the wheel feels the designed runoff as twist too, and 213.59(b) holds spiral runoff to the same limits. Readings of 4.6 in and 2.56 in against a design of 4.0 in and 2.76 in are only 0.80 in off the design, but the warp is 2.04 in, over a 1.5 in Class 5 limit. So the verdict judges the raw difference, and the split shows how much of it is designed runoff (fixed only by a longer spiral or a lower speed) and how much is deviation from design (fixed by surfacing). A reading exactly at the limit passes, since the rule says may not be more than. The same twist is also reported scaled to a 31 ft base, because the two measurement lengths carry different limits. This does not ship the FRA limit tables. They are set by class and by parameter in 49 CFR 213 and the adopted regulation governs, so the applicable limit has to be entered. It does not evaluate the other geometry parameters -- gauge, alignment, surface, and the combinations of them -- each of which has its own limits and any of which can independently restrict speed, and it does not handle the special limits that apply within a specified distance of a joint, on a bridge, or through a turnout. It does not address the qualification and frequency requirements for track inspection, the recording of exceptions, or the remedial action a defect requires, all of which are regulatory obligations rather than calculations. Track geometry defects are a derailment hazard: 49 CFR 213, the qualified track inspector, and the track owner govern.

Formula or tablewarp = |cross level at A - cross level at B| (49 CFR 213.63(a), the raw readings); split into designed change = designed A - designed B and deviation twist = (measured A - designed A) - (measured B - designed B); pass when warp <= the entered limit; margin = entered limit - warp; the same twist scaled to a 31 ft base = absolute warp x 31 / distance.

Source and edition49 CFR 213.63(a) (as amended at 91 FR 22735, April 28, 2026): warp is the difference in cross level between any two points less than 62 ft apart; 213.59(b) holds spiral runoff to the same limits. 49 CFR 213 is named as the source of the limits by class of track. The limit tables are not reproduced; the applicable limit is entered. The qualified track inspector and the track owner govern.

Free access49 CFR 213 is US federal regulation and public domain; this is subtraction on the user's own level-board readings against a limit they supply.

What governsEstimate. AHJ and licensed professional govern.

API field namesmeasured_a_in, designed_a_in, measured_b_in, designed_b_in, distance_ft, warp_limit_in, deviation_a_in, deviation_b_in, warp_magnitude_in, margin_in, warp_per_31ft_in

• Warp is the raw cross-level difference the designed runoff on a spiral counts toward the limit; constant superelevation cancels out49 CFR 213.63(a), 213.59(b)
• Measurement length matters 31 ft and 62 ft warp carry different limits; the length must be stated with the reading49 CFR 213
• Limits are entered, not shipped the class limits are set by the adopted regulation and must come from it49 CFR 213