Charge Pump Error Cancellation for Fractional-N Synthesizer Linearity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional delta-sigma fractional-N frequency synthesizers suffer from noise leakage due to non-linearity, degrading in-band phase noise and spur performance, with existing linearization techniques either worsening reference spur or in-band noise performance.

Innovation Solution

A charge pump linearization technique using an error canceling circuit with matched pairs of charging and discharging current sources, along with an error cancellation circuit that replicates currents to eliminate the non-linear term, ensuring the charge pump operates without increasing in-band noise or adversely affecting reference spur performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional charge pump is used in delta-sigma fractional-N synthesizer, then the synthesizer can operate in fractional mode, but non-linearity causes noise leakage that degrades in-band phase noise and spur performance

Engineering Contradiction:
Improvefractional mode operationVSAvoidlinearity
Core Design Contradiction:
Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The charge pump current is segmented into multiple paths: a main charge pump path and a separate error cancellation path. The error cancellation path contains matched current sources that replicate and cancel the non-linear components, separating the linear and non-linear current components for independent handling.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The non-linear current components, which normally cause noise leakage and performance degradation, are converted into a beneficial cancellation signal. By replicating these error currents through the error cancellation path and injecting them with proper phase and polarity, the harmful non-linearities are transformed into useful correction signals that eliminate the noise leakage.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

2Manufacturing precision

If existing linearization techniques are applied, then linearity is improved, but reference spur or in-band noise performance worsens

Engineering Contradiction:
ImprovelinearityVSAvoidreference spur and in-band noise
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The error cancellation current sources are designed to dynamically adjust their current magnitude based on the operating conditions. By changing the current parameter in the error cancellation path to match the non-linear error components, the system achieves adaptive cancellation that maintains linearity across different operating points without introducing fixed spurs or noise.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The error cancellation mechanism operates as a feedback system where the non-linear error currents are continuously replicated and cancelled. The matched current sources in the error cancellation path provide real-time compensation for the non-linearities, creating a closed-loop effect that maintains linearity while preventing the generation of reference spurs and in-band noise.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS7969247B2Charge pump linearization technique for delta-sigma fractional-N synthesizers
Publication Date: 2011.06.28 RENESAS ELECTRONICS AMERICA INC
  • US7969247B2 patent drawing
  • US7969247B2 patent drawing
  • US7969247B2 patent drawing

AI summary

A delta-sigma fractional-N frequency synthesizer having a charge pump with error canceling circuitry eliminates a non-linear term from the charge pump transfer function. The charge pump includes a matched pair of charging current sources, each supplying a first current IP1 to a common node, when enabled. The charge pump also includes a matched pair of discharging current sources, each sinking a second current IN1 from the common node, when enabled. The error canceling circuitry includes a charging current source, which supplies a current equal to the second current IN1 to the common node, when enabled. The error canceling circuitry also includes a discharging current source, which sinks a current equal to the first current IP1 from the common node, when enabled. The charging and discharging current sources of the error canceling circuitry are both enabled when either one of the matched pairs of charging and discharging current sources is enabled.