Collimated Backlights for High-Precision Machine Vision Measurement

What a Collimated Backlight Is and When to Use It

A standard diffuse backlight radiates light at all angles. On a thin, flat part that is fine, but on a part with any thickness the camera also collects rays that pass the edge at a shallow angle, so the measured edge blurs and shifts depending on where the part sits in the field — a systematic measurement error. A collimated backlight uses a lens or a structured film to send rays nearly parallel to the optical axis. Only rays travelling straight toward the lens reach the sensor, so the silhouette edge is defined by the part's real profile and stays stable across the field of view. Use a collimated backlight for dimensional gauging of machined parts, thread and bore inspection, glass and transparent-part edges, and any measurement where repeatability at the micron level matters — especially when it is paired with a telecentric lens.

Selection Criteria: Collimation Angle, Wavelength, Footprint

• Collimation half-angle: ±1° to ±5° typical. A tighter angle gives sharper edges on thick parts but a smaller usable area and a tighter mechanical alignment tolerance to the camera axis.
• Wavelength: green 525 nm is the common default — short enough for a sharp edge and where most lenses have their best MTF; blue 460 nm for the finest edges, red 625 nm for compatibility with existing red setups, IR 850 nm to see through some plastics.
• Footprint and depth: the active area must cover the largest part plus margin, and there must be room behind the conveyor for the collimating optic, which is deeper than a flat diffuse panel.
• Telecentric-lens pairing: the backlight's collimation angle should meet or exceed the lens acceptance angle, otherwise the lens re-introduces the edge error the backlight removed.
• Alignment: a collimated source must sit square to the camera axis — pair it with a fine-adjust or swivel bracket so the opposition can be set and locked precisely.