PWM circuit controls sensor's AGC

Dongjie Cheng, Allegro Microsystems, Willow Grove, PA -- EDN, 3/29/2001

All electronic sensors have their limits on working distances and environmental tolerances. Dynamic range defines a sensor's maximum allowable variations in the signal amplitude. ACG (automatic gain control) finds widespread use in systems to extend the dynamic range. Applications using photoelectric or ultrasonic techniques involve both emission and detection energy. In many cases, the emission first establishes a background receiver signal as a reference, and the receiver monitors the signal and detects any changes against this reference. It is often desirable to maintain the background signal at a moderate level so that the sensor works within its limits. A signal that is too weak cannot produce a significant SNR, and a signal that is too strong can disable the sensor by saturation or produce overheating of the sensor. Designers use different techniques to achieve satisfactory signal handling. The simplest method might be adjusting emission power or the receiver's gain by manually configuring jumpers or switches. An example of an AGC solution is to let a µP constantly adjust emission to a suitable level. Figure 1 shows a simple PWM-based technique for maintaining the reference signal at an ideal level.

The concept embodied in Figure 1's technique is that, if the sensor sees a strong background signal, the circuit slowly reduces the emission intensity at the fundamental frequency by reducing the pulse width of the emission drive signal. We verified the idea by using OrCAD/Cadence PSpice modeling, followed by experimental verification. The voltage source, V3, generates a 1-kHz sinusoidal signal named receiver signal. Figure 2 plots this signal for a half-cycle in the third panel down. The model simulates the receiver's filtered, slowly varying background signal. IC_2A functions as an inverter. Its output passes through the lowpass filter R₂/C₂ to generate the Control signal in Figure 2, fourth panel down. IC₂, a 555 timer, acts as a pulse-width modulator. Its output is a pulse train named PWM emission drive (Figure 2, first panel). The frequency of the PWM emission drive follows the 10-kHz trigger signal, modeled with V1 and plotted in Figure 2 in the second panel down. The modeled results in Figure 2 indicate a negative correlation between the width of PWM emission drive and the level of receiver signal. As the receiver signal declines, the control signal widens the pulses of PWM emission drive to boost the sensor's emission power at the fundamental frequency.

The process maintains the width of the PWM signal with the width fluctuating around a stable point. This point depends on the signal strength. If a sensor works near its upper range limit, the signal remains weak despite the increased pulse width, so the stable point shifts toward the maximum pulse width. In this Design Idea, the sensor is sensitive to the 10-kHz fundamental frequency because of a bandpass filter. Therefore, a 50% PWM duty cycle yields the maximum signal amplitude. On the other hand, if a sensor operates near its lower range limit, the pulse width converges to a very narrow width for a medium background signal. The time constant R₁C₁ in Figure 1 also affects the location of the stable point. A longer time constant pushes the PWM stable point to a higher duty cycle and vice versa. In this case, the R₁C₁ time constant is half the trigger period (0.05 msec). You usually need to perform a test to determine R₁, C₁, or both. The test ensures that 50% is the maximum duty cycle from the modulator. The Trigger signal must be a narrow, negative pulse train.

Figure 3 shows a hardware setup to test the PSpice model, using an infrared sensor. IC_1A and IC_1B are 555 timers (the dual TLC556 or LM556). IC_1A is a free-running oscillator that supplies the 10-kHz Trigger pulses for the IC_1B PWM circuit. IC_1B drives the IR LED and Q₁. The 11, 1, and 5.5V voltages come from filtered voltage sources. A transimpedance amplifier first amplifies the 10-kHz photocurrent from the IR photodiode. The signal then goes through a second-stage amplifier and then undergoes filtering and peak detection to generate the quasi-dc receiver signal. IC_2A then inverts the signal to provide the PWM control. The PWM emission drive drives the IR LED in such a way that the IR photodiode receives medium-radiation intensities. Because the receiver's signal amplitude is proportional to the pulse width of PWM emission drive, the sensor constantly tries to reverse any trend of the background signal. In varying the distance between the IR LED and the photodiode, an oscilloscope showed signal behavior as predicted by the model.