Checking a Time-of-Flight Sensor Behind Its Production Cover Window

A time-of-flight sensor that ranges correctly on an open evaluation board may behave differently behind a product window. The window, air gap, enclosure walls, and contamination can change the optical path. Treat these features as part of the sensor system before approving a PCB position or production calibration.

Use the correct optical reference

STMicroelectronics’ single-zone time-of-flight cover glass guidance discusses crosstalk associated with the cover assembly and the effect of air gap. It defines that gap relative to the sensor module and cover surface. Use the guidance for the selected device rather than transferring a clearance from another sensor family.

Draw a section through the emitter, receiver, window, and adjacent enclosure features. Label the package reference surface, window thickness, gap, and optical keepout regions. A mechanical clearance measured from the PCB surface may omit package height and therefore fail to describe the optical condition.

Evaluate realistic tolerance combinations

1. Combine board mounting, package placement, window location, and enclosure tolerances.
2. Check that the emitter and receiver paths remain unobstructed.
3. Review any separator or gasket for alignment and compression variation.
4. Include the final window material, surface treatment, and cosmetic printing.
5. Identify assembly residues or protective films that could remain in the path.

For example, reducing the nominal gap does not solve the problem if a tolerance extreme causes mechanical contact with the module. The mechanical and optical reviews should resolve both requirements together. Avoid compensating for an uncontrolled stack solely through a software setting.

Test the assembled optical system

Use representative targets and conditions relevant to the product. Include distances, target properties, ambient lighting, and window conditions that challenge the intended function. Compare results before and after the approved calibration procedure and retain the raw observations.

Keep calibration tied to the final assembly state. If a window is installed after calibration, demonstrate that the process still represents the finished product. A replacement window or changed adhesive may require more than repeating a basic communication test.

The release record should identify sensor variant, optical stack, tolerance analysis, calibration sequence, and measured performance limits. Include a production check capable of detecting an obstructed or incorrectly assembled optical path. Revisit the evidence when cosmetic or mechanical changes affect the window, even if the PCB circuitry is unchanged. The aim is repeatable ranging through the actual product interface, with known limits under the conditions the product will encounter.

Keep window cleaning and handling instructions with the assembly process so qualification samples and production units have comparable optical surfaces.