"The scanner is accurate" is a claim; validation is proof. If you never validate, you are shipping unknown error to the mill.
How to validate
- Scan a calibrated artefact (known spheres / bars) 10×.
- Measure in CAD; compute mean deviation and RMS.
- Repeat monthly; a drift >20 µm means recalibrate.
Where error hides
| Source | Effect |
|---|---|
| Dry mouth / glare | Tracking loss, holes |
| Fast passes | Edge blur |
| Uncalibrated tip | Systematic offset |
One unvalidated scanner can quietly add 100 µm to every margin in the practice.
The Scanning Series includes a calibration artefact and open export for independent checks.
The error budget
| Source | Typical error |
|---|---|
| Scanner noise | 10–20 µm |
| Moisture | 30–50 µm |
| Movement | Up to 100 µm |
| Stitching span | 20 µm per 10 mm |
How to validate
- Scan a reference model and overlay in inspection software; report RMS.
- Repeat after each calibration; a drift chart is your QA log.
- Keep ambient light low; strong light floods the sensor.