“The determination of light intensity can be crucial, for instance, in designing room lighting or preparing for photography. In the era of the Internet of Things (IoT), determining light intensity also plays a vital role in what is known as smart agriculture. In such cases, a key task is to monitor and control essential plant parameters to facilitate optimal growth and accelerate photosynthesis.
Therefore, light is one of the most critical factors. Most plants typically absorb light in the visible spectrum, including wavelengths of red, orange, blue, and purple light. Green and yellow light wavelengths in the spectrum are generally reflected and contribute minimally to plant growth. By controlling partial spectra and light irradiance during different growth stages, growth can be maximized, ultimately leading to increased yield.
Figure 1 depicts a circuit design used to measure light intensity within the visible spectrum for experiments related to plant photosynthesis. Three different colors of photodiodes (green, red, and blue) are used here, each responding to different wavelengths. The measured light intensity from the photodiodes can now be used to control light sources according to specific plant requirements.
Figure 1. Circuit design for measuring light intensity
The circuit shown comprises three precision transimpedance amplifiers, one for each color (green, red, and blue). The outputs of the transimpedance amplifiers are connected to the differential inputs of a Σ-Δ analog-to-digital converter (ADC), providing measured values as digital data to a microcontroller for further processing.
Conversion of Light Intensity to Current Depending on light intensity, varying amounts of current flow through the photodiodes. The relationship between current and light intensity is approximately linear, as shown in Figure 2. The graph displays the characteristic curves of output current versus light intensity for red (CLS15-22C/L213R/TR8), green (CLS15-22C/L213G/TR8), and blue (CLS15-22C/L213B/TR8) photodiodes.
Figure 2. Characteristic curves of output current versus light intensity for red, green, and blue photodiodes
However, the relative sensitivities of the red, green, and blue photodiodes differ, requiring individual gain adjustments via feedback resistor (RFB). To achieve this, the short-circuit current (ISC) for each diode must be obtained from datasheets, and sensitivity (S, pA/lux) should be determined at the operating point derived from it. RFB can be calculated as follows:
Where VFS,P-P is the desired full-scale output voltage range (peak-to-peak); INTMAX represents the maximum light intensity, which is 120,000 lux for direct sunlight.
Current-to-Voltage Conversion High-quality current-to-voltage conversion requires operational amplifiers with minimal bias current since the output current from photodiodes is in the picoampere range. A larger bias current can lead to significant errors. The offset voltage should also be small. The AD8500 from ADI is an ideal choice for such applications, with typical bias current around 1 pA and maximum offset voltage of 1 mV.
Analog-to-Digital Conversion For further processing, the converted photodiode currents must be digitized for a microcontroller to handle. This can be done using an ADC with multiple differential inputs, such as the 16-bit ADC AD7798. Therefore, the output code for the measured voltage is:
Where AIN is the input voltage, N is the number of bits, GAIN is the gain factor of the internal amplifier, and VREF is the external reference voltage. To reduce noise further, each differential input of the ADC employs common-mode and differential filters.
All components mentioned are highly power-efficient, making this circuit suitable for battery-powered portable field applications.
Conclusion Sources of errors such as device bias current and offset voltage must be considered. Additionally, the amplification factor inside the ADC affects signal quality (the offset voltage of the transimpedance amplifier is multiplied by the gain inside the ADC, amplifying the error), which subsequently affects the final sampled results. Employing the circuit illustrated in Figure 1 can relatively easily convert light intensity into electrical quantities for further data processing.”