Bisect De-Embedding Using Open and Short Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current de-embedding techniques for high-frequency on-wafer measurements are limited by lumped element assumptions and require precise fabrication, making them difficult to implement and apply effectively, especially at higher frequencies where probe-probe coupling becomes significant.
Innovation Solution
The bisect de-embedding method uses open and short structures to improve the accuracy of ABCD parameter determination for a substantially symmetric THRU, allowing for the mathematical splitting of a THRU structure into mirrored halves, which can then be removed from a two-port measurement structure, thereby isolating the device under test.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Y and Z subtraction methods are used for de-embedding, then the procedure is simple and widely used, but the accuracy deteriorates at higher frequencies due to lumped element assumptions and probe-probe coupling effects
Solution Approach 1:
The patent segments the THRU structure into two mirrored halves by introducing a virtual mid-plane, allowing separate characterization of each half. This segmentation enables the use of open and short structures at the mid-plane to independently determine series and shunt parasitics, thereby improving accuracy at high frequencies while maintaining procedural simplicity through systematic measurement steps
Solution Approach 2:
The patent introduces open and short structures as intermediary elements placed at the virtual mid-plane of the THRU structure. These intermediaries serve as reference standards that enable precise determination of parasitic elements through ABCD parameter optimization, resolving the accuracy issue without complicating the overall de-embedding workflow
2Reliability
If probe-probe spacing is reduced to mitigate lumped element limitations, then the frequency range extends higher, but probe-probe coupling increases significantly negatively impacting measurements
Solution Approach 1:
The patent extracts and separately characterizes the series and shunt parasitic elements by using open and short structures at the mid-plane. This extraction approach allows the measurement system to account for coupling effects mathematically through ABCD parameter optimization, enabling high-frequency measurements without requiring reduced probe spacing that would exacerbate coupling
Solution Approach 2:
The patent changes the measurement parameters by introducing additional measurement configurations (open and short circuits at the mid-plane) and using ABCD parameter optimization instead of direct S-parameter subtraction. This parameter transformation enables the system to compensate for probe-probe coupling effects and extend the reliable frequency range without physical modification to probe spacing
3Measurement precision
If advanced de-embedding techniques are used to avoid lumped element constraints, then measurement accuracy improves, but the device complexity and fabrication precision requirements increase
Solution Approach 1:
The patent creates a universal de-embedding methodology that works across a broad frequency range by using standard open and short structures that can be implemented with existing fabrication capabilities. The method universally applies ABCD parameter optimization to various THRU configurations, achieving high accuracy without requiring specialized complex structures or ultra-precise fabrication
Solution Approach 2:
The patent uses idealized open and short structures as reference copies that represent extreme cases of impedance boundaries. These copied reference structures serve as mathematical anchors for the optimization algorithm, enabling accurate parasitic extraction without requiring complex physical reference structures, thus reducing fabrication complexity while maintaining measurement precision
Data Source
AI summary
Application of open and short structures may result in improved accuracy in determination of ABCD parameters of a substantially symmetric two-port network for purposes of bisect de-embedding. Either one or both of the open and/or short techniques may be used to improve results of an ABCD optimization algorithm. Bisect de-embedding may then be performed to determine the ABCD parameters of a device under test based on the ABCD parameters of the substantially symmetric two-port network and measured s-parameters of the substantially symmetric two-port network and the embedded device under test.