Noise Suppression Effect Evaluation of Low Ripple High Voltage Power Supply in Precision Measuring Instruments
Precision measuring instruments demand extremely stable and clean power supplies to achieve their specified accuracy. Low ripple high voltage power supplies are essential for instruments where voltage noise directly affects the measurement accuracy. Evaluating the noise suppression effectiveness of these power supplies requires understanding the noise coupling mechanisms and the sensitivity of the measurement to power supply variations.
Precision instruments that use high voltage include mass spectrometers, electron microscopes, photomultiplier detectors, and various analytical instruments. In these instruments, the high voltage biases detectors, accelerates particles, or establishes electric fields for deflection or focusing. Noise on the high voltage causes variations in the measurement that can limit the resolution or accuracy.
Ripple refers to the periodic variation in the output voltage, typically at the switching frequency of the power supply or its harmonics. The ripple amplitude is usually specified as peak to peak or root mean square values, often in parts per million of the output voltage. Low ripple supplies have ripple levels of a few parts per million or less.
The effect of ripple on the measurement depends on the instrument type and the measurement principle. In mass spectrometers, voltage ripple causes variations in the ion energy and trajectory, affecting the mass resolution and peak shape. In electron microscopes, voltage ripple causes beam deflection that degrades the image resolution. In photomultipliers, voltage ripple causes gain variations that appear as noise in the output signal.
The sensitivity of the measurement to voltage variations can be characterized through analysis or experiment. Analytical models relate the voltage variation to the measurement variation based on the instrument physics. Experimental characterization measures the output noise as a function of applied voltage ripple, establishing the transfer function from power supply to measurement output.
Noise coupling mechanisms determine how power supply ripple affects the measurement. Direct coupling occurs when the voltage variation directly affects the physical quantity being measured, such as the ion energy in a mass spectrometer. Indirect coupling occurs through parasitic paths, such as capacitive coupling to sensitive circuits or magnetic coupling through ground loops.
Low ripple power supply design addresses the ripple at its source. Switching power supplies generate ripple through the periodic switching of the power semiconductors. Higher switching frequencies enable smaller filter components but may increase high frequency noise. The filter design must attenuate the switching ripple to acceptable levels. Linear post regulators provide additional ripple rejection.
Output filter design involves trade-offs between ripple attenuation, transient response, and size. Larger inductors and capacitors provide better ripple attenuation but increase the filter size and cost. The stored energy in large filters affects the response to load changes. The filter resonance can interact with the control loop, affecting stability. Optimal filter design balances these considerations.
Ripple measurement requires sensitive instrumentation and proper technique. Oscilloscopes with high vertical resolution can measure ripple directly at the power supply output. Spectrum analyzers provide frequency domain analysis of the ripple components. The measurement bandwidth must be sufficient to capture the relevant frequency components. The measurement setup must not introduce noise that contaminates the measurement.
System level noise evaluation measures the instrument output noise with the power supply operating. This evaluation captures the combined effect of all noise sources, including the power supply ripple, detector noise, and electronic noise. Comparing measurements with different power supplies or with battery power can isolate the power supply contribution.
Correlation analysis relates the power supply ripple to the measurement noise. Simultaneous measurement of the power supply output and the instrument output enables calculation of the cross correlation. A strong correlation at the ripple frequency indicates that the power supply ripple is contributing to the measurement noise. This analysis can identify the dominant noise sources and guide improvement efforts.
Specification of acceptable ripple levels depends on the instrument requirements and the noise coupling. Starting from the required measurement accuracy, the allowable measurement noise can be derived. Through the noise coupling analysis, the allowable power supply ripple can be determined. This system level approach ensures that the power supply ripple specification is appropriate for the application.
