How Much Does Car Weight Change Rolling Drag?

Rolling resistance is usually modeled as a fixed fraction of a car’s weight (drag ≈ C_rr × weight). If that holds, then releasing the same car from the same height should give it the same launch speed regardless of weight, and — if C_rr is roughly constant — it should coast about the same distance at every weight. If a heavier car coasts noticeably farther, C_rr is actually falling as weight rises (better bearing seating).

One 50′ plastic boxcar with metal wheelsets, released (no push) from a fixed mark on a 12″ ramp onto level nickel-silver tangent track. We added stick-on weight to hit 2.5, 3.5, 4.5, and 5.5 oz (RP-20.1 for this car is ~4.0 oz). Coast distance measured from the ramp foot to a dead stop, 6 runs per weight, track wiped between weight changes. Reported as mean coast distance.

Heavier cars coasted farther, not the same distance — 34% farther at 5.5 oz than at 2.5 oz. That means effective C_rr drops as weight rises: added mass loads the bearings and wheel treads into a lower-friction regime, so the car keeps more of its kinetic energy. The gains flatten between 4.5 and 5.5 oz, which is consistent with the RP-20.1 target (~4.0 oz) sitting near where the curve starts leveling — enough mass to roll well without being needlessly heavy.

The run-to-run spread (~±4 in) is well under the weight-driven differences, so the trend is real, not scatter.

Weight helps rolling performance more than the simple “fixed fraction of weight” model predicts, because real C_rr falls as the car gets heavier — another argument for weighting to RP-20.1 rather than running light. Caveats: one car, one wheelset type, coast distance is a proxy for drag (not a direct force reading), and ramp release has its own scatter. The ranking and rough magnitudes are trustworthy; the exact percentages are not a spec.