Current Monitoring in Switching Converters via Midpoint Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing current monitoring systems in switching converters, particularly in skip mode operation, suffer from noise-induced inaccuracies due to switching transitions, limiting their effectiveness in both continuous conduction mode (CCM) and discontinuous conduction mode (DCM).
Innovation Solution
The implementation of an acquisition circuit comprising a sample and hold circuit and a low-pass filter, along with a phase frequency detector and control logic, generates a narrower sampling window aligned with the midpoint of the turn-on signal, reducing noise and enabling accurate current monitoring by sampling midpoints far from switching transitions. Additionally, a low-pass filter network averages the voltage signal to generate a noise-free DC value, and in DCM, the sensed DC value is scaled to account for dead times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current monitoring is performed by sensing current on power FETs and filtering the signal, then current measurement capability is provided, but noise from switching transitions degrades measurement accuracy
Solution Approach 1:
The patent applies preliminary action by generating a sampling pulse that is synchronized with the switching cycle and positioned to sample the inductor current at the optimal moment (when the current waveform is at its peak or crossing zero in DCM). This pre-planned sampling timing avoids the noise from switching transitions by capturing the signal at a predetermined safe point in the cycle, before the harmful switching events occur.
Solution Approach 2:
The patent uses feedback by employing a phase-frequency detector (PFD) that compares the sampling pulse timing with the actual switching transitions. The PFD generates error signals that are fed back to adjust the timing and width of the sampling window, ensuring it remains optimally positioned relative to the switching events. This closed-loop feedback mechanism dynamically adapts to maintain accurate sampling despite variations in switching timing.
2Object-affected harmful factors
If additional blanking circuitry is added to remove noise spikes, then noise rejection is improved, but the duty cycle range of the switching regulator is limited
Solution Approach 1:
Instead of using blanking circuitry that removes portions of the signal, the patent employs preliminary action by strategically timing the sampling operation to occur at the optimal point in the switching cycle. The sampling pulse is generated in advance and synchronized to capture the current waveform at its peak or zero-crossing point, avoiding the need for post-sampling noise removal that would limit duty cycle range. This approach maintains full adaptability across different duty cycles.
Solution Approach 2:
The patent extracts only the useful portion of the current waveform by using a narrow sampling window that captures only the relevant current information at the optimal timing point. By taking out only the necessary signal portion and discarding the rest (including the noisy switching transitions), the system achieves noise rejection without requiring blanking circuitry that would constrain the duty cycle range.
3Quantity of substance
If a wider sampling window is used to capture more signal, then signal capture is improved, but noise from transitions is included in the measurement
Solution Approach 1:
The patent applies local quality by concentrating the sampling effort precisely at the most informative point in the current waveform (the peak or zero-crossing point) rather than attempting to capture the entire waveform. The sampling window is localized to a narrow time interval around this critical point, ensuring high signal capture quality while excluding the noisy switching transitions that occur at other times in the cycle.
Solution Approach 2:
The patent uses partial action by sampling only the essential portion of the current waveform needed for accurate measurement, rather than capturing the entire waveform. The sampling window is deliberately kept narrow and positioned to capture only the critical current information, providing sufficient signal for accurate measurement without including the excessive noisy portions of the waveform.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach provides accurate current monitoring in both CCM and DCM modes by minimizing noise interference and ensuring precise current measurement across a wide range of operating conditions, enhancing the reliability of switching converters.
Implementation Method 1
a low-pass filter network samples and averages aligned voltage signal to generate a DC value that represents the sensed current
Implementation Method 2
an acquisition circuit that comprises a sample and hold circuit and a low-pass filter to control the sampling of at least one section of a waveform of a voltage
Data Source
AI summary
Various embodiments of the invention increase current monitoring accuracy in switching converters. In particular, certain embodiments of the invention allow reduce noise associated with transients that are typically generated at transitions when power FETs are turn on and off and allow to accurately sense inductor DC current of switching converters, thereby, increase current monitoring accuracy without requiring any blanking circuitry. In certain embodiments of the invention, this is accomplished by an acquisition circuit that dynamically monitors current in various operating modes. A phase frequency detector (PFD) and control circuit in the acquisition circuit automatically align a narrow sampling window and the midpoint of a turn-on signal. Certain embodiments utilize an analog multiplier circuit to sense current in skip mode operation.