The optical interface is physically located on the same side of the device pins as the electrical interface for the PicoStar? variant and facing away from the pins for the SOT-5X3 variant, as shown in Figure 9-2 and Figure 9-3

In case of the PicoStar? variant systems, light that illuminates the sensor must come through the FPCB. Typically, the best method is to create a cutout area in the FPCB. Other methods are possible, but with associated design tradeoffs. This cutout must be carefully designed because the dimensions and tolerances impact the net-system, optical field-of-view performance. The design of this cutout is discussed more in the Section 9.5.2.

Physical components, such as a plastic housing and a window that allows light from outside of the design to illuminate the sensor (see Figure 9-4), can help protect the device and neighboring circuitry. Sometimes, a dark or opaque window is used to further enhance the visual appeal of the design by hiding the sensor from view. This window material is typically transparent plastic or glass.

Generally for both package variants, any physical component that affects the light that illuminates the sensing area of a light sensor also affects the performance of that light sensor. Therefore, for the best performance, make sure to understand and control the effect of these components. Design a window width and height to permit light from a sufficient field of view to illuminate the sensor. For best performance, use a field of view of at least ±35°, or preferably ±45° or more. Understanding and designing the field of view is discussed further in application report OPT3001: Ambient Light Sensor Application Guide (SBEA002).

The visible-spectrum transmission for dark windows typically ranges between 5% to 30%, but can be less than 1%. Specify a visible-spectrum transmission as low as, but no more than, necessary to achieve sufficient visual appeal because decreased transmission decreases the available light for the sensor to measure. The windows are made dark by either applying an ink to a transparent window material, or including a dye or other optical substance within the window material itself. This attenuating transmission in the visible spectrum of the window creates a ratio between the light on the outside of the design and the light that is measured by the device. To accurately measure the light outside of the design, compensate the device measurement for this ratio.

Although the inks and dyes of dark windows serve their primary purpose of being minimally transmissive to visible light, some inks and dyes can also be very transmissive to infrared light. The use of these inks and dyes further decreases the ratio of visible to infrared light, and thus decreases sensor measurement accuracy. However, because of the excellent red and infrared rejection of the device, this effect is minimized, and good results are achieved under a dark window with similar spectral responses.

For best accuracy, avoid grill-like window structures, unless the designer understands the optical effects sufficiently. These grill-like window structures create a nonuniform illumination pattern at the sensor that make light measurement results vary with placement tolerances and angle of incidence of the light. If a grill-like structure is desired, then the device is an excellent sensor choice because the device is minimally sensitive to illumination uniformity issues disrupting the measurement process.

Light pipes can appear attractive for aiding in the optomechanical design that brings light to the sensor; however, do not use light pipes with any light sensor unless the system designer fully understands the ramifications of the optical physics of light pipes within the full context of his design and objectives.