Research On Nonlinear Behavior Mechanism And Modeling Method Of Power Inductors

Power inductors are widely used in the smartphone and portable computer industries due to their low DC resistance,high power,and good current resistance characteristics.However,as the process size decreases,the frequency range increases,making it susceptible to parasitic effects such as skinning and proximity effects,making it more difficult to model power inductors.For example,wire-wound power inductors display DC dependency due to fluctuations in the magnetic flux of their cores and the resulting magnetization caused by the non-constant operating current.On the one hand,the operating current of a wire-wound power inductor is not constant,which leads to changes in the flux of the core and consequently magnetization,showing DC dependence;on the other hand,changes in temperature affect the permeability of the core and the inductance changes,showing temperature dependence.In the current research field of inductor modeling,although there exist modeling studies that consider the DC dependence or those that consider the temperature dependence,the effects of DC bias and temperature often exist simultaneously in practical use cases.Therefore,it becomes a very important task to analyze the operating characteristics of nonlinear power inductors,to study the mechanism of their nonlinear behavior and to build a library of high-precision nonlinear inductor models.In this paper,the behavior mechanism,device model and parameter extraction of nonlinear power inductor model are analyzed in depth,and the DC dependence and temperature dependence of nonlinear power inductor model concerns are considered on the basis of static model.The main work of the article is as follows:(1)To begin,we introduce electromagnetism theory,focusing on the relationship between current and magnetic field,in order to lay the groundwork for the subsequent study of the nonlinear power inductor model.(2)Then,in the static equivalent circuit of inductance,self-inductance,mutual inductance,distributed capacitance and loss mechanism of the device are mainly considered.The equivalent circuit model of power inductor is derived by comparing the equivalent circuit model of magnetic bead and RF inductor.Based on the static model,a controlled current source is added to reflect the characteristic changes of electronic components during DC bias.Temperature dependence is also studied in this paper,and the corresponding temperature influence coefficient is obtained by extracting parameters from test data.Thus,the static model,dynamic model and temperature model used in this paper are introduced.(3)The parameters of the two-port network matrix are defined,and the transformation formulas between each network matrix are derived primarily to prepare for parameter extraction.A reasonable de-embedding method,parameter extraction,and modeling process are proposed based on the test data.The model extraction process is further optimized based on the frequency characteristic curve,DC characteristic curve,and temperature characteristic curve to obtain a more accurate nonlinear power inductor model.(4)The static model,dynamic model,and temperature model are all verified separately,and the simulation results are compared to the test data,as well as the frequency characteristic curve,DC characteristic curve,and temperature characteristic curve,to ensure the feasibility and accuracy of the model parameter extraction.Finally,an optimization scheme based on ADS is proposed.