Measure a run you can compare
A useful speed test starts with a measured section and a repeatable timing method. This guide explains the calculation used by Rail Speed Log and what its results can tell you.
Set up the measured section
- Choose a section where the locomotive can reach a steady speed before the first marker. Keep the test conditions consistent.
- Measure the distance along the track between two visible markers. On a curve, measure the travelled path, not the straight chord between its ends.
- Choose the model’s actual scale ratio. HO is preset to 1:87.1; British N and US N use different presets. For another ratio, use Custom. Track gauge alone does not uniquely determine scale.
- Start and stop timing when the same point on the locomotive crosses each marker. A longer section makes reaction time a smaller fraction of the total time.
You can enter seconds from your own stopwatch or use the built-in timer. Keep the browser tab visible while timing. A browser timer is not a calibrated sensor, and the displayed hundredths do not establish hundredth-second measurement accuracy.
The formula
For the conventional model railway equivalent speed, multiply measured model speed by the scale denominator. This is a linear length-scale conversion, not the square-root relationship used in some physical-model similarity calculations.
Scale km/h = distance in metres ÷ time in seconds × scale denominator × 3.6
Scale mph = scale km/h ÷ 1.609344
For example, a 1:48 O-scale model travelling 12 inches in 3 seconds moves at 4 inches per second. Multiplying by 48 and converting units gives 10.91 scale mph, or 17.56 scale km/h. One foot is exactly 0.3048 metres.
The NMRA Train Speeds data sheet, page 5 describes gauging model speeds from measured distance and elapsed time. Its quick-reference table uses rounded approximations; this calculator uses the explicit ratio and unit conversion above. The historical railway speed limits elsewhere in that document are not current operating advice.
Compare like with like
Name each locomotive and enter a consistent throttle label, such as “step 32 / 128”. A label of “32” alone is less useful if different controllers use different ranges. Forward and reverse runs are kept separate.
Repeat each measurement under the same conditions. The comparison groups identical locomotive names, scale ratios, throttle labels and directions. It shows the arithmetic mean of individual run speeds and the observed minimum–maximum range. The dots represent actual saved runs on a shared scale. A single run has no measured variation yet; a narrow range is not a confidence interval.
Small differences can come from reaction time, measured distance, locomotive temperature, track conditions or load. Similar averages do not by themselves prove that locomotives can be coupled safely. This tool does not program decoders or recommend specific CV settings. The North County Model Railroad Society’s speed-matching article illustrates the broader process of recording throttle and direction settings using dedicated equipment.
Keep your evidence
Saved runs stay in this browser’s local storage on this device. They are not uploaded or synced. Clearing browser/site data can erase them. Download CSV after a session; the export contains only the locomotive filter’s shown runs, including their original distance, seconds, ratio, direction and calculated speeds.
Each row is a measurement, not a decoder configuration backup. CSV download requests do not prove that your browser saved the file; open the downloaded file and check it before deleting anything.