Multipath Inductor Cross-Over Architecture for Skin Effect Loss Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In high-frequency applications, existing on-chip inductor designs face challenges with reduced quality factor (Q-factor) and increased capacitance, leading to skin and proximity effect losses, which complicate the design of efficient transformers and inductors.

Innovation Solution

A series stacked, solenoidally wound, multipath inductor structure with cross-over architecture is employed, where spiral turns on multiple layers have varying segment lengths to equalize current flow and reduce interwinding capacitance, skin, and proximity effect losses, utilizing a cross-over architecture to connect segments between layers and maintain equal path lengths.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional on-chip inductor designs are used, then the inductor can be integrated into CMOS circuits, but the quality factor decreases and capacitance increases at high frequencies

Engineering Contradiction:
Improvequality factorVSAvoidskin and proximity effect losses
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The inductor coil is divided into multiple segments along its length, with each segment having a dedicated shielding layer. This segmentation allows independent optimization of each section and reduces the overall skin and proximity effects by breaking up the continuous current path into discrete segments with controlled electromagnetic coupling.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from planar 2D inductor designs to three-dimensional stacked configurations with multiple metal layers. By utilizing the vertical dimension with stacked inductor coils and interleaved shielding layers, the design achieves reduced capacitance and improved Q-factor while maintaining compact footprint.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Speed

If the inductor operates at high frequencies, then the application requirements are met, but skin and proximity effect losses increase

Engineering Contradiction:
Improveoperating frequencyVSAvoidskin and proximity effect losses
Core Design Contradiction:
SpeedVSLoss of energy

Solution Approach 1:

Shielding layers are introduced as intermediary structures between adjacent inductor segments and turns. These shielding layers act as electromagnetic mediators that redirect and contain the magnetic flux, reducing the proximity effects between adjacent conductors and minimizing skin effect losses at high operating frequencies.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent optimizes geometric parameters including segment length ratios, shielding layer positions, and metal layer thicknesses to minimize skin and proximity effects. By carefully controlling these parameters, the inductor maintains low loss characteristics across the target frequency range while achieving the required inductance values.

Inventive Principle:
Principle #35Parameter changes

3Quantity of substance

If multi-layer stacked structure is used, then inductance density increases, but device complexity increases

Engineering Contradiction:
Improveinductance densityVSAvoidstructure complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The shielding layers serve multiple functions simultaneously: they reduce skin effects, control magnetic flux distribution, provide electrical isolation between segments, and contribute to the overall mechanical structure. This multi-functionality reduces the need for additional dedicated components and simplifies the overall design despite the multi-layer configuration.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The inductor design employs nested structures where shielding layers are interleaved between inductor segments, and multiple inductor coils are stacked within a compact volume. This nesting approach maximizes the use of available space and achieves high inductance density while maintaining a manageable structural complexity through systematic layering.

Inventive Principle:
Principle #7Nested doll (Nesting)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This design achieves higher inductance density, reduced capacitance, and improved quality factor, enabling superior performance at higher frequencies while minimizing skin and proximity effect losses, thus enhancing the figure of merit compared to conventional structures.

Implementation Method 1

A series stacked, solenoidally wound, multipath inductor includes a plurality of turns disposed about a center region on two layers

Methodology Applied
Scientific EffectElectromagnetic induction: Electromagnetic Induction

Implementation Method 2

Skin effect is the tendency for high-frequency currents to flow on the surface of a conductor

Methodology Applied
Scientific EffectSkin effect: Skin Effect

Implementation Method 3

Proximity effect is the tendency for current to flow in other undesirable patterns, e.g., loops or concentrated distributions, due to the presence of magnetic fields generated by nearby conductors

Methodology Applied
Scientific EffectProximity effect:

Data Source

PatentUS9865392B2Solenoidal series stacked multipath inductor
Publication Date: 2018.01.09 GLOBALFOUNDRIES US INC
  • US9865392B2 patent drawing
  • US9865392B2 patent drawing
  • US9865392B2 patent drawing

AI summary

A series stacked, solenoidally wound, multipath inductor includes a plurality of turns disposed about a center region on two layers. The turns on the two layers have corresponding geometry therebetween. Each of the plurality of turns includes two or more segments that extend length-wise along the turns. The segments have positions that vary from an innermost position relative to the center region and an outermost position relative to the center region. A cross-over architecture is configured to couple the segments of a turn on one layer with the segments on a turn on another layer to form segment paths that have a substantially same length for all segment paths in a segment path grouping between the two layers.