Differential Signal Converter With Stable Zero-Crossing Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional single-ended to differential signal converters often suffer from low linearity and unstable zero-crossing properties, which degrade the performance of Time to Digital Converters (TDCs) due to non-ideal delay compensation and asymmetrical signal paths.
Innovation Solution
A signal converter device comprising a frequency multiplier and a fully symmetrical differential frequency divider, which multiplies and divides the single-ended signal by two to generate a differential signal with highly stable zero-crossing properties, eliminating the need for signal inversion and delay compensation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional inversion-based methods are used to generate complementary phases, then the differential signal can be produced, but the zero-crossing properties become unstable and linearity degrades due to non-ideal delay compensation
Solution Approach 1:
The patent extracts and eliminates the problematic inversion stage and delay compensation circuits from the signal path. By using a fully differential ring oscillator that directly generates complementary phases without inversion, the design removes the source of zero-crossing instability and eliminates the need for complex delay compensation mechanisms.
Solution Approach 2:
Instead of inverting the input signal to generate complementary phases (conventional approach), the patent inverts the approach by using a fully differential ring oscillator that naturally generates equal and opposite phases through its symmetrical structure. This reverses the causal relationship: rather than creating complementary signals from a single-ended input and then compensating, the system generates them simultaneously in a balanced manner.
2Measurement precision
If asymmetrical signal paths are used in the converter, then the circuit can be simpler, but the linearity and zero-crossing stability are degraded
Solution Approach 1:
The patent applies asymmetry in a controlled manner by introducing deliberate mismatch elements (such as different sized transistors or adjustable resistors) in specific locations within the symmetrical ring oscillator structure. This allows fine-tuning of the signal paths to achieve perfect balance and zero-crossing alignment, transforming the inherent asymmetry of manufacturing variations into a tool for achieving symmetry.
Solution Approach 2:
The patent implements local quality adjustments by applying different component values or configurations at specific locations within the differential structure. For example, certain transistors may have different widths or lengths, or specific resistors may be adjusted, to locally compensate for process variations and achieve global symmetry in the signal paths.
3Reliability
If additional resynchronization inverters and pass-gates are added to force paths to work in phase opposition, then zero-crossing stability improves, but device complexity and manufacturing difficulty increase
Solution Approach 1:
The patent makes the ring oscillator circuit perform multiple functions simultaneously: it generates the differential signal, establishes phase opposition, and self-synchronizes all paths without requiring additional dedicated components. The core oscillator structure inherently provides all these functions through its design, eliminating the need for separate resynchronization inverters and pass-gates.
Data Source
Figure 1a~3b
Figure 4a~4b
Figure 5a~5b
AI summary
It is described a signal converter device (100) for converting a single-ended signal to a differential signal, the device (100) comprising: i) a multiplier device (110), configured to receive a single-ended incoming signal (105), and multiply the incoming signal (105) to provide a multiplied signal (115); and ii) a divider device (120), configured to receive the multiplied signal (115), and divide the multiplied signal (115) to provide a differential signal (125a, 125b). Further, a corresponding signal conversion method is described.