In smartphones, the LCD panel backlight accounts for approximately 40% of the device's total power consumption. Therefore, enabling the backlight brightness to change according to ambient light levels would offer numerous benefits. In relatively dark environments, reducing display brightness can save power, while also reducing eye strain and improving the user experience.
In fact, ambient light sensors (ALS) are widely used in smartphones to provide information about ambient light levels to support the backlight LED power supply circuitry. However, while this application sounds simple, it presents many challenges in practice, as it needs to achieve both significant power savings and comfortable viewing for the user.
The ALS (Advanced Light Detection System) must be placed behind the display screen, where space is extremely limited, and the same component must simultaneously perform proximity detection (turning off the display when the user's face is near) and ambient light measurement. These and other conditions severely restrict design engineers, limiting their freedom to optimize the design.
This article describes the main challenges of implementing ambient light sensing in smartphones, and how to overcome these challenges to achieve higher backlight responsiveness and precise adjustment of backlight brightness based on ambient light.
Bright visual response
The first challenge is that photodiodes react to light differently than the human eye. The human eye is not sensitive to infrared (IR, wavelengths greater than 780 nm) and ultraviolet (UV, wavelengths less than 380 nm) light. On the other hand, standard silicon photodiodes typically sense light with wavelengths between 300 nm and 1,100 nm.
This means the first challenge for designers is how to remove the infrared and ultraviolet components from the sensor output. The ALS (Alternating Light Source) measures the brightness of light incident on a smartphone display (in lux). If this brightness measurement includes ultraviolet, infrared, and visible light, what it presents to the display backlight controller is not what the human eye actually sees. In other words, the sensor's response to ambient light differs from the human eye's "photopic" response. In short, the sensor "senses" a higher ambient light brightness than the human eye perceives.
This is because both natural and artificial light contain infrared components. For example, sunlight (Figure 1) and light from incandescent lamps do. An effective way to remove infrared radiation is to overlay an optical infrared filter onto the sensor . However, in smartphones, the same sensor is often used for proximity detection (accompanied by an infrared LED) to turn off the display and touch controller when the phone is close to the user's face.
Figure 1 shows the spectral power distribution of sunlight, where the high-power infrared component is invisible to the human eye.
Of course, smartphone designers could add a separate infrared photodiode for proximity sensing, but this would be a cumbersome solution: such a design would have to bear the cost of both the optical filter on the ALS and the separate infrared photodiode, which would also take up extra space and require openings in the display surface to allow infrared light to pass through.
ams has proposed a better solution to this problem: a dual-diode module. One photodiode (Channel 0 in Figure 2) is used to sense the full spectrum, while the other (Channel 1 in the figure) is mainly used to sense the infrared portion of the spectrum. Subtracting the output of the infrared photodiode from the output of the full-spectrum sensor yields the visible light measurement result.
Figure 2 shows the spectral sensitivity of the TMD2772 dual diode module from ams Semiconductor. Other products include the TMD27721 and TMD27723 series.
This sensor is not very sensitive to ultraviolet light, and the ultraviolet radiation emitted by common light sources under any circumstances is minimal. In most cases, removing ultraviolet light for ambient light sensing is sufficient simply by using packaging materials that absorb ultraviolet light.
After removing the infrared element from the ALS output, smartphone designers now have to solve a second problem: how to limit the field of view of the ALS/proximity sensor module without affecting its performance. This is about striking a balance between the ALS and the proximity sensor.
For ambient light sensing, the ideal viewing angle is 180 degrees (which is practically impossible) because this is the angle at which ambient light hits the display. However, the opposite is true for proximity sensing: it requires a narrow viewing angle to limit the possibility of crosstalk between the infrared LED and the infrared sensor. Ideally, the infrared sensor should only detect infrared light reflected from the user's face, and the LED should not shine directly onto the sensor, nor should it detect light reflected from the top and bottom of the touch panel. Therefore, a trade-off must be made between the conflicting requirements of the ALS and the infrared sensor.
Through experiments, smartphone designers discovered that a viewing angle of 90-110 degrees provides high-performance proximity detection while also allowing the ambient light sensing system to perform well; narrowing the angle to below 90 degrees significantly impairs the performance of the ALS. Furthermore, if the system operates at a 90-degree viewing angle, the air gap between the bottom of the touchscreen and the top of the sensing module must be extremely small.
Viewing angle isn't the only mechanical design issue affecting ALS performance. To allow light to pass through the screen to the sensor module, designers must create an opening in the display. OEMs want this opening as small as possible to avoid disrupting the touchscreen's smooth, rounded shape. They also add ink to the underside of the screen glass to conceal this opening, darkening it and blending its color with the phone's casing. Both the ink and the opening reduce the intensity of light incident on the sensor module.
Furthermore, OEMs must strictly control ink transmittance variations on the production line. For example, if using ink with 17% transmittance, even a ±1% variation in ink transmittance can result in an additional 5.9% error in ALS output.
The third major challenge in implementing ambient light sensing in smartphones lies in handling light output with a very high dynamic range. Smartphone manufacturers want the brightness of the display backlight to be set appropriately, whether the phone is in near-total darkness (brightness as low as 0.1 lux) or direct sunlight (brightness as high as 220 klux).
This requires the sensor to have high sensitivity over an extremely wide dynamic range, while maintaining extremely low noise floor. Furthermore, the device gain should be controllable to adapt to changes in ambient light.
Implementation of fine-tuning
This article has explained the trade-offs governing ambient light sensing in smartphones, the benefits of dual phototube solutions, and the ALS module characteristics that OEMs need to pay attention to. However, the appearance, mechanical design, and inks of each device are different, which requires various characterization features to develop a tailored brightness equation.
This equation is used to precisely remove the infrared component of ambient light and compensate for limited viewing angles.
To achieve such characterization properties, the smartphone must be exposed to various light sources emitting different proportions of infrared and ultraviolet light. Then, under identical lighting conditions, the ambient illuminance is measured using a high-precision lux meter, simultaneously with an ALS module. The ALS output is then calibrated based on the lux meter's output. The lux meter's metering surface should be covered with a light shield to simulate the limited viewing angle of a light sensor.
To characterize sensor modules, such as the TMD27721 or TMD27723 from Austria Microelectronics, the following equations can be used:
CPL=(ATIME_ms×AGAINx)/20Lux1=(C0DATA-a0×C1DATA)/CPLLux2=(b0×C0DATA-b1×C1DATA)/CPLLux=MAX(Lux1,Lux2,0)<, b>
In the above equations, CPL,a0,b0,b1 are parameters characterizing the properties.
CPL: CountsperLux. C0DATA: Data read from Channel 0. C1DATA: Data read from Channel 1. C0DATA-a0xC1DATA: Light sources with a high weighted infrared ratio. b0xC0DATA-b1xC1DATA: Light sources with a low weighted infrared ratio. MAX: The maximum value among Lux1, Lux2, and 0.
Generally speaking, the more data sets collected under more light sources, the more accurate the characterization properties will be.
By defining appropriate mechanical design, strictly controlling ink transmittance during production, and carefully characterizing the system, the systematic error of ambient light sensing can be limited to no more than ±15%. In some cases, the error can be as small as ±10%. This is sufficient for adjusting backlight brightness to reduce power consumption and improve the user experience.
Of course, OEMs may require more precise ALS for functions beyond display backlight control, which necessitates the use of highly sensitive ambient light sensors (e.g., standalone devices without proximity detection). For such applications, the Austrian Microsystems TSL25911 is an ideal choice.
Summarize
ALS (Advanced Light Filtering) is widely used in smartphones to provide information about ambient light levels to support the backlight LED power supply circuitry. However, while this application sounds simple, it is not easy to implement in practice—because it needs to achieve both significant power savings and comfortable viewing for the user.
The ALS (Advanced Light Detection System) must be placed behind the display screen, where space is extremely limited—the component must be able to perform proximity detection (turning off the display when the user's face is near) and primary ambient light measurement. These and other conditions severely restrict design engineers, limiting their freedom to optimize the design.
