De-embedding Balun Effects in Differential Signal Acquisition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Test and measurement systems face challenges in de-embedding combiners, such as baluns, that do not go to zero frequency, leading to noise issues and inefficiencies in acquiring differential signals, as existing software applications are not effective for such devices and may introduce aliasing errors.

Innovation Solution

A test and measurement instrument with processors that generate and apply de-embed filters to remove the effects of combiners, using calibration procedures involving differential pulse generators, PRBS generators, and vector network analyzers to create filters that reduce noise and allow single-channel acquisition of differential signals.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a balun or combiner that does not go to zero frequency is used to combine differential signals, then only one channel of the test instrument is needed to acquire the signal, but the inverse of the passband results in a large de-embed boost in the DC area and introduces noise

Engineering Contradiction:
Improvenumber of channels usedVSAvoidsignal accuracy
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent applies preliminary action by performing calibration measurements before actual signal acquisition. The system measures the combiner's frequency response in advance, stores the magnitude and phase data, and uses this pre-characterized information to compute de-embed corrections during signal processing. This preliminary characterization enables accurate removal of combiner effects without introducing noise or requiring the combiner to extend to DC.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If software applications use S-parameter models to de-embed devices, then de-embedding can be performed, but it only works reliably when devices go to zero frequency and may introduce aliasing errors

Engineering Contradiction:
Improvede-embedding capabilityVSAvoidde-embedding accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by transitioning from traditional S-parameter model-based de-embedding to a frequency response-based approach. Instead of relying on S-parameter models that require DC extension, the system uses measured magnitude and phase parameters across the actual passband of the combiner. This parameter transformation enables accurate de-embedding for combiners that do not extend to zero frequency, improving measurement precision while maintaining ease of operation.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10895588B1Apparatus and method for de-embedding a combiner from a balanced signal
Publication Date: 2021.01.19 TEKTRONIX INC
  • US10895588B1 patent drawing
  • US10895588B1 patent drawing
  • US10895588B1 patent drawing

AI summary

A test and measurement system including a plurality of channels and one or more processors. The one or more processors are configured to cause the test and measurement system to receive, via a first channel of the plurality of channels, a positive side of a reference differential signal pair, receive, via a second channel of the plurality of channels, a negative side of the reference differential signal pair, and produce a reference signal based the reference differential signal pair. A combined signal is received, from a combiner, that is a balanced signal produced from the reference differential signal pair. A de-embed filter is generated based on the reference signal and the combined signal and an additional signal is received from the combiner and an effect of the combiner is removed from the additional signal by applying the de-embed filter to the additional signal.