Home › Guide › Accuracy
Ranging accuracy and the error budget that actually matters
A datasheet accuracy of ±2 m sounds precise until you put it next to a 0.3 mrad pointing error at 5 km. Here is how the terms stack up, and which ones you can buy your way out of.
What the module contributes
Two shapes dominate the catalogue. Fixed: ±2 m, the same at 200 m and at 20 km — characteristic of the long-range 1535 nm family. Proportional: ±0.3 m under 100 m plus a distance term such as D × 0.7 ‰ above it, common on the integrated 905 nm modules. At 1.2 km that proportional term is almost a metre on its own.
What the platform contributes
- Pointing error. 0.3 mrad at 5 km = 1.5 m lateral. This is usually the largest term, and it is not in the rangefinder datasheet.
- Boresight drift. A mount aligned on a bench at +20 °C moves through a −40 °C to +70 °C swing. Align at the far field, and re-check hot and cold.
- Time sync. If the range is stamped on receipt and the platform moves at 30 m/s, every 100 ms of latency is 3 m of along-track error.
- Atmosphere. Not an accuracy term in the datasheet sense, but it biases long ranges systematically in haze.
Adding them up
| Source | 5 km contribution | Buyable? |
|---|---|---|
| Module range error | ±2 m (fixed) or ±3.8 m (0.7 ‰ term) | Yes, by model |
| Pointing error, 0.3 mrad | ±1.5 m lateral | Yes, by stabilisation |
| Time sync, 100 ms at 30 m/s | ±3 m along track | Yes, by trigger design |
| Boresight drift, 0.2 mrad | ±1 m lateral | Yes, by mount design |
Roughly, the root-sum-square of the lateral terms at 5 km is 2–4 m regardless of whether you bought the ±1 m module or the ±2 m one. Pick the module on range and mass, then spend the effort on the mount.
When fixed accuracy is worth paying for
Short-range metrology, docking and industrial positioning, where the target is close and the platform is static. There, a ±0.3 m 905 nm sensor and a good mount get you centimetres, and the proportional term has not yet eaten the budget.