PRAECIS · TECHNICAL BRIEF
Praecis unlocks LuphoScan performance.
Praecis unlocks LuphoScan performance that on-board corrections can’t reach.
The LuphoScan is specified to ±50 nm (2σ) form accuracy and <20 nm reproducibility. Taylor Hobson’s MWLI sensor, Invar reference frame, and built-in temperature/pressure compensation do exceptional work at the instrument level — but three physical effects sit outside the instrument’s correction envelope, and they are exactly where measurement time and yield are lost.
Problems built-in algorithms cannot solve.
1. Part soak time. The MWLI sensor measures the surface it sees — not the surface the part will have once it reaches lab temperature. Glass, metal, and mounting hardware all expand at different rates, and a 1 °C gradient across a 100 mm optic produces tens of nanometers of apparent form error. Operators compensate by letting parts sit overnight.
2. Thermal drift during the scan. Areal scans take minutes to hours. The LuphoScan’s four temperature sensors and one pressure sensor correct air refractive index in real time, but they cannot correct mechanical drift in the metrology loop caused by HVAC cycling, operator presence, or adjacent equipment. Published MWLI sensor studies show residual drift of ~15–40 nm from ambient thermal fluctuation and air turbulence alone — a large fraction of the instrument’s 2σ specification.
3. Steep-slope and freeform residuals. Steep aspheres and freeforms are the most thermally sensitive geometries: small frame gradients tilt and bend the measurement loop in ways that show up as low-order Zernike terms — power, astigmatism, coma — that look like real surface features. An independent cross-check against CGH null testing (Xu et al., Light: Advanced Manufacturing, 2024) reported pointwise LuphoScan–CGH deviation of 5.6 nm RMS under well-controlled conditions; in production rooms without active environmental control, the figure is typically several times larger.
What Praecis adds.
Praecis delivers an integrated thermal environment purpose-built for LuphoScan platforms. Rather than competing with the instrument’s algorithms, it eliminates the physical inputs those algorithms cannot model:
• Active part conditioning — brings parts and fixtures to measurement temperature in minutes, not hours, with embedded verification.
• Tight ambient stability — holds the instrument envelope to ±[0.05] °C with gradient control, well below the drift budget that on-board compensation is asked to handle.
• Low-turbulence airflow — laminar, low-velocity exchange around the optical path, directly attacking the turbulence-driven drift documented in MWLI sensor literature.
• Isothermal fixturing — part-mount interfaces that stay gradient-free through the full scan cycle.
Addressing the deficiencies identified in Xu et al. (2024)
Xu, F. et al. (2024) presents the clearest peer-reviewed benchmark of where LuphoScan’s measurements diverge from a reference method. Xu et al. identified that LuphoScan could not accurately resolve surface errors in the 0.16–0.36 mm⁻¹ spatial-frequency band. This is a sensor-sampling and spot-size limitation — it is not thermally driven, and no environmental solution will change it. For that band, CGH null testing, sub-aperture stitching, or white-light interferometry remain the right tools. Praecis improves the accuracy floor of the measurements LuphoScan can make; it does not extend its bandwidth.
Independent studies referenced
• Berger, G. & Petter, J. (2013). Non-contact metrology of aspheric surfaces based on MWLI technology. SPIE 8884, 88840V.
• Wendel, M. et al. (2023). Precision Measurement of Complex Optics Using a Scanning-Point Multiwavelength Interferometer Operating in the Visible Domain. Nanomanufacturing and Metrology 6:23.
• Sharma, S., Eiswirt, P., Petter, J. (2018). Electro-optic sensor for high precision absolute distance measurement using multiwavelength interferometry. Optics Express 26(3): 3443 — documents ~15–40 nm residual drift from thermal and turbulence effects in the MWLI sensor.
• Xu, F. et al. (2024). Accuracy verification methodology for computer-generated hologram used for testing a 3.5-meter mirror based on an equivalent element. Light: Advanced Manufacturing 5: 25. DOI: 10.37188/lam.2024.025 — reports 5.6 nm RMS pointwise agreement between LUPHOScan and CGH under controlled conditions; also documents LUPHOScan high-frequency band limits (0.16–0.36 mm⁻¹).
• Yan, L. et al. (2014). Measurement of air refractive index fluctuation based on a laser synthetic wavelength interferometer. Measurement Science and Technology 25: 095006.