Multiplexed Signal Conditioning for Shared A/D Sensor Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing signal conditioning methods for AC sensor signals, such as LVDTs and resolvers, are costly, large, and power-intensive due to separate anti-aliasing filters and demodulators for each signal, and introduce errors from inherent variations between circuits.
Innovation Solution
A signal conditioning circuit that time-shares anti-aliasing filters and A/D converters using multiplexers, allowing multiple sensors to share a single A/D converter, reducing size, cost, and power consumption, and mitigating errors through synchronized sampling and digital processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate anti-aliasing filters and demodulators are used for each AC sensor signal, then signal processing accuracy is maintained, but circuit size, cost and power consumption increase
Solution Approach 1:
The patent merges multiple separate anti-aliasing filters into a single shared filter that processes multiple sensor signals sequentially. The filter is time-shared among multiple signals, allowing one filter circuit to perform the function of what would traditionally require multiple separate filter circuits, thereby reducing overall circuit size while maintaining processing accuracy
Solution Approach 2:
The anti-aliasing filter is designed to be universal, capable of processing multiple different sensor signals (e.g., from LVDTs and resolvers) through time-division multiplexing. This single filter structure serves multiple functions that would traditionally require separate dedicated filters for each sensor type
2Measurement precision
If separate anti-aliasing filters and demodulators are used for each AC sensor signal, then signal processing accuracy is maintained, but power consumption increases
Solution Approach 1:
Multiple filter circuits are merged into a single time-shared filter, reducing the total power consumption of the system. Since only one filter is active at any given moment processing one signal, the power consumption of having multiple simultaneous filters is eliminated, while still maintaining the ability to process multiple signals with high accuracy
3Device complexity
If separate circuits are used for each sensor signal, then signal processing is simplified, but inherent variations between circuits introduce errors
Solution Approach 1:
The patent merges multiple sensor signal paths into a single shared filter circuit, ensuring that all signals undergo identical processing. This eliminates the variations between separate circuits that would otherwise introduce errors in angle calculations, as the same physical hardware processes all signals
4Device complexity
If multiple sensors share a single A/D converter, then circuit size and cost are reduced, but sampling speed requirements increase
Solution Approach 1:
The system uses periodic time-division multiplexing to share the A/D converter among multiple sensors. Each sensor signal is sequentially routed to the converter in regular time slots, allowing a single converter to service multiple sensors at high speed through rapid periodic switching, thereby reducing circuit size while meeting sampling requirements
Data Source
AI summary
A signal conditioning circuit time share multiplexes anti-aliasing filters and an A/D converter. A plurality of first tier multiplexers each time share multiplex one of a plurality of antialiasing filters between a plurality of AC or baseband input signals from a plurality of sensors. A second tier multiplexer selects its inputs from the outputs of the first tier multiplexers. The output of the second tier multiplexer feeds a high speed A/D converter. Thus, the A/D converter is time share multiplexed by the second tier multiplexer. In this manner, a plurality of sensors can share a single A/D converter. After allowing a settling time for the multiplexers and antialiasing filters, a plurality of samples of the input signals are taken, such as for one period. The samples of each AC input signal are multiplied by a sine vector and a cosine vector. The product vectors are then each averaged and the root mean square of the two averages yields the magnitude of the input signal. Mechanical angle of the input signal can be determined based upon the sign of the sine and cosine product vector averages.