Pspice仿真常用变压器模型
基于PSpice的升压型开关稳压电源设计和仿真
基于PSpice的升压型开关稳压电源设计与仿真20世纪50年代，美国宇航局以小型化、重量轻为目标，为搭载火箭开发了开关电源。
在半个多世纪的发展过程中，开关电源因具有体积小、重量轻、效率高、发热量低、性能稳定等优点而逐渐取代由传统技术设计制造的连续工作的线性电源，并广泛用于电子、电气设备中。
20世纪80年代，计算机全面实现了开关电源化，率先完成了计算机的电源换代。
20世纪90年代，开关电源在电子、电气设备以及家电领域得到了广泛的应用，开关电源技术进入快速发展期。
Cadence旗下的PSpice是一款电路仿真软件，能够对复杂的模数混合电路进行仿真，而且开关电源也不例外。
1升压变换器拓扑结构升压变换器属于间接能量传输变换器。
供电过程包含能量的存储和释放两方面。
如图1所示，Vclock是脉冲信号源，提供PWM电压，用以功率开关S1的导通与截止。
Rsense为电流取样电阻，Resr为电容的等效串联电阻。
在开关S1导通期间，二极管D1截止，电感储存能量，输出电容单独为负载提供电能。
在开关S1断开期间，二极管D1导通，储存了能量的电感与输入电源串联，为输出提供电能，其中一部分转移到电容C1里。
1.1工作于CCM条件下的升压变换器波形对图1所示电路，借助PSpice进行仿真，获得如图2所示的波形图。
这是典型的电感电流连续导通模式（CCM）。
图1基础升压变压器结构电路图2工作于CCM条件下的Boost变换器波形曲线①代表PWM波形，用于触发功率开关导通或断开。
当开关S1导通时，公共点SW/D电压几乎降到0.相反，当开关S1断开时，公共点SW/D电压增加为输出电压和二极管的正向压降之和，如曲线②所示。
曲线③描述了电感两端电压的变化。
高电平期间，电感左侧电压为Vin,右侧几乎为0,对应功率开关导通；而低电平期间，电感左侧电压仍为Vin,而右侧突变为Vout,因为功率开关截止，同时二极管导通，此时对应电感电压为负值，这就意味着输出电压大于输入电压。
变压器的参数和数学模型
第二节变压器的参数和数学模型⏹双绕组变压器的参数和数学模型⏹三绕组变压器的参数和数学模型⏹自耦变压器的参数和数学模型一.双绕组变压器的参数和数学模型⏹阻抗⏹电阻变压器的电阻是通过变压器的短路损耗，其近似等于额定总铜耗。
我们通过如下公式来求解变压器电阻：（MV A）Rt—电阻（欧）•电抗在电力系统计算中认为，大容量变压器的电抗和阻抗在数值上接近相等，可近似如下求解：Uk —阻抗电压（%），Un —额定电压（kV ），Sn —额定容量（MV A ） Xt —电抗⏹导纳⏹电导 变压器电导对应的是变压器的铁耗，近似等于变压器的空载损耗，因此变压器的电导可如下求解：⏹电纳在变压器中，流经电纳的电流和空载电流在数值上接近相等，其求解如下：二.三绕组变压器的参数和数学模型⏹按三个绕组容量比的不同有三种不同的类型：100/100/100、100/50/100、100/100/50⏹按三个绕组排列方式的不同有两种不同的结构：升压结构：中压内，低压中，高压外降压结构：低压内，中压中，高压外•电阻由于容量的不同，对所提供的短路损耗要做些处理 ⏹⏹对于100/50/100或100/100/50首先，将含有不同容量绕组的短路损耗数据归算为额定电流下的值。
例如：对于100/50/100然后，按照100/100/100计算电阻的公式计算各绕组电阻。
2. 电抗⏹根据变压器排列不同，对所提供的短路电压做些处理：一般来说，所提供的短路电压百分比都是经过归算的三.自耦变压器的参数和数学模型就端点条件而言，自耦变压器可完全等值于普通变压器，但由于三绕组自耦变压器第三绕组的容量总小于变压器的额定容量，因此需要进行归算。
❖对于旧标准：❖对于新标准，也是按最大短路损耗和经过归算的短路电压百分比值进行计算。
第二章 电力系统各元件的特性和数学模型一．电力线路的参数和数学模型二．负荷的参数和数学模型第三节 电力线路的参数和数学模型⏹电力线路结构简述电力线路按结构可分为架空线：导线、避雷线、杆塔、绝缘子和金具等电缆：导线、绝缘层、保护层等架空线路的导线和避雷线导线：主要由铝、钢、铜等材料制成避雷线：一般用钢线1. 架空线路的导线和避雷线❖认识架空线路的标号×××××—×/×钢线部分额定截面积主要载流部分额定截面积J 表示加强型，Q表示轻型J 表示多股线表示材料，其中：L表示铝、G表示钢、T表示铜、HL表示铝合金例如：LGJ—400/50表示载流额定截面积为400、钢线额定截面积为50的普通钢芯铝线。
PSCAD中的变压器模型
1.Introduction to Transformers（引言）EMTDC中使用变压器有两种方法：经典方法和统一的磁等效电路（unified magnetic equivalent circuit (UMEC)）方法。
经典方法用来模拟同一变压器铁芯上的绕组。
也就是说，每一相都是独立的，各单相变压器之间没有相互作用。
而UMEC方法计及了相间的相互作用：由此，可以对3相3臂或3相5臂式变压器构造进行精确的模拟。
每一模型中，铁芯的非线性特征是最基本的不同。
经典模型中的铁芯饱和是通过对选定绕组使用补偿注入电流实现的。
UMEC方法采用完全插值，采用分断线性化的ϕ-I曲线来表征饱和特性。
2.Transformer Models Overview（变压器模型概述）对电力系统进行电磁暂态分析过程中必然会出现变压器。
PSCAD中有两种方法对变压器进行模拟：经典方法和UMEC方法。
经典方法仅限于单相设备，其中不同的绕组处于同一铁芯腿上。
而UMEC方法，考虑到来铁芯的几何外形和相间的相互耦合因素。
除了以上的显著区别外，两种变压器模型之间最基本的区别是对铁芯非线性特性的描述。
在经典模型中，非线性特性采用近似地基于“拐点”、“空心电抗”和额定电压的磁化电流曲线进行模拟。
而UMEC模型则直接采用V-I曲线进行模拟。
与经典模型不同，UMEC模型没有配置在线分接头调整功能。
但是，可以在指定绕组上设置分接头，不过分接头在仿真过程中不能动态调整。
3.1-Phase Auto Transformer（单相自耦变压器）此组件基于经典方法模拟了单相自耦变压器。
用户可以选择采用磁化支路（线性铁芯）或注入电流模拟磁化特性。
理想情况下，可以忽略磁化支路，变压器即为理想模式，仅保留串联的漏抗。
4.3-Phase Star-Star Auto Transformer（三相星形连接的自耦变压器）此组件模拟了由3个单相构成的3相自耦变压器。
用户可以选择采用磁化支路（线性铁芯）或注入电流模拟磁化特性。
变压器隔离推挽式开关电源PSpice仿真设计
变压器隔离推挽式开关电源PSpice仿真设计作者：邵珠雷来源：《科技视界》2013年第01期【摘要】设计了一种采用变压器隔离推挽式输出的开关电源，其输出电压为5V，输出电流为2A，开关频率为25kHZ。
使用仿真软件PSpice进行系统建模，设置相关参数后得出仿真分析结果，由输出电压和输出电流的时域波形图可知，所设计的开关电源输出稳定，符合设计要求。
【关键词】开关电源；推挽式；PSpice仿真0 引言在开关电源的的设计中，输入电源与输出负载之间的公共点常常需要隔离，从而阻隔共模回流和消除地线上的环流。
开关电源在变换器部分的设计中常采用变压器来隔离输入电源与输出负载，并且通过变压器实现电压调节功能。
推挽式开关电源包含两个反相工作的正激变换器，因此推挽式开关电源具有正激式开关电源拓扑的所有优点[1]。
由于电源提供给负载的功率并不在变压器中存储，推挽式开关电源比正激式开关电源能处理更多的功率，且效率更高，控制性能更好。
1 电路拓扑结构及工作原理本设计为采用变压器隔离推挽式输出的开关电源，其输入电压为35V，输出电压为5V，输出电流为2A，纹波为250mV。
为了实现稳定输出，开关电源采用电流型PWM开关电源控制器SG1846构成反馈回路，开关频率为25kHZ。
为缓解变压器磁通不平衡造成的过热失控问题，在电路中采用功率MOSFET作为开关管[2]。
其电路结构如图1所示。
如图所示，开关管M1，M2根据输入电压V1、变压器变比和预期输出电压所决定的占空比每半周期交替导通。
为避免两开关管直通现象的发生，需要在M2开通和M1关断之间设置死区时间td。
二极管D2上的电压为矩形波，幅值为输入电压V1乘以变压器电压比，即v=V■■（1）二级管D2正偏导通，输出电感L1此时间段内充磁。
二级管D2的电流i■（t）与开关管集电极电流i■（t）成比例。
i■（t）=i■（t）■（2）电路的电压调整率可以从输出电感L1的伏秒平衡关系导出，由于每个开关周期输出电感都要充磁两次，开关管开通时间ton的系数为2。
建立PSPICE模型
PSpice库中已有极多模型可用，没有必要自建模型，如果遇到库中没有的器件模型，可以到生产该器件上公司网站上下载，一般大型公司都会提供。
如果一定要自建模型，可以用PSpice中的模型编辑软件实现（“Model Editor”），一般可以用已有的模型作一些修改实现。
可以上网找一些深层次的PSpice书看或是找一些极老版本的PSpice的书看，老版本的书中会较多得提到关于PSpice命令、语言等方面与模型有关的东西。
听过一个高手的培训，PSpice其实就是一个计算器。
只要器件模型对了，就能给出结果。
你调用的是PSpice的模型库中的元器件吗？如果是，基本不会出现因为模型原因而不能仿真的现象！资料你可以上网找，很多的。
先找本简单的看看就行了。
个人认为PSpiec在模拟电路仿真方面是最好的。
关于你出现的问题，是PSpice中常见的，与PSpice的算法有关。
解决的方法是在出现问题的结点处（即提示的node ***）与电路地之间加一个值很大的电阻，这样即不会影响仿真精度，问题也能解决。
值得一提的是在PSpice的电路在必需有一个结点的名称为“0”,一般建议将“地”结点命名为“0”.这与PSpice的算法与电路网表的结构有关，不必深揪！PSpice 如何利用Model Editor 建立仿真用的模型PSpice 提供Model Editor 建立元件的Model，从元件供应商那边拿到该元件的Datasheet，透过描点的方式就可以简单的建立元件的仿真模型，来做电路的模仿真。
PSpice 提供约十多种的元件(Diode、Bipolar Transistor、Magnetic Core、IGBT、JFET、MOSFET、Operational Amplifier、Voltage Regulator、Voltage Comparator、Voltage Reference、Darlington Transistor)来建立元件的模型。
基于PSpice的电力变压器绕组中暂态电压分布的仿真计算
基于PSpice的电力变压器绕组中暂态电压分布的仿真计算高宏;郭颖娜;王世山【摘要】传统对电力变压器绕组中暂态电压分布的仿真计算有多种,但均需编程计算而得,为此提出了一种适用于PSpice仿真的电力变压器绕组中暂态电压分布的电路模型,并对该变压器雷电过电压时的暂态电压分布及正常运行电压分布进行了仿真,结果表明该模型可表达绕组暂态时的表现,且大大简化了计算过程.【期刊名称】《现代电子技术》【年(卷),期】2006(029)013【总页数】3页(P137-139)【关键词】电力变压器;暂态电压;电路模型;PSpice【作者】高宏;郭颖娜;王世山【作者单位】西安石油大学,电子工程学院,陕西,西安,710065;西安石油大学,电子工程学院,陕西,西安,710065;南京航天航空大学,江苏,南京,210016【正文语种】中文【中图分类】TM71 引言对1995～2001年共7年间的常规变压器故障原因统计可得，变压器绕组故障占故障总数的71.2%[1]，且变压器绕组故障大部分是由于绕组本身结构及绝缘不合理所引起的[2]。
因此在处理变压器的绝缘问题时，应该小心对待绝缘裕度，选择恰当的试验电压和完善的试验方法。
变压器在冲击电压作用下的过电压，主要是由线圈内部的自由振荡过程和线圈之间的静电及电磁感应过程引起的。
这两种过程，我们统一称为变压器线圈中的波过程。
当输入电压的波形和幅值一定时，自由振荡电压的幅值主要决定于变压器的最终电压分布和起始电压分布之差。
改善起始电压分布，使他尽量接近于最终电压分布，是降低自由振荡电压的主要方法。
因此，准确计算变压器的起始电压分布对于正确地进行变压器绝缘结构的设计，对于减小变压器的体积、节省材料、运行维修、降低成本以及改善振荡过电压波形都是必须的，研究绕组的波过程具有重要的意义。
起始电压分布，传统的主要以实验的方法测得。
由于受实验条件的限制，多数厂家很难准确测定，而传统的手工计算方法速度慢，又因简化太多使计算精确度大打折扣。
SPICE的器件模型大全(第6章)
SPICE的器件模型大全在介绍SPICE基础知识时介绍了最复杂和重要的电路描述语句，其中就包括元器件描述语句。
许多元器件（如二极管、晶体管等）的描述语句中都有模型关键字，而电阻、电容、电源等的描述语句中也有模型名可选项，这些都要求后面配以.MODEL起始的模型描述语句，对这些特殊器件的参数做详细描述。
电阻、电容、电源等的模型描述语句语句比较简单，也比较容易理解，在SPICE基础中已介绍，就不再重复了；二极管、双极型晶体管的模型虽也做了些介绍，但不够详细，是本文介绍的重点，以便可以自己制作器件模型；场效应管、数字器件的模型过于复杂，太专业，一般用户自己难以制作模型，只做简单介绍。
元器件的模型非常重要，是影响分析精度的重要因素之一。
但模型中涉及太多图表，特别是很多数学公式，都是在WORD下编辑后再转为JEPG图像文件的，很繁琐和耗时，所以只能介绍重点。
一、二极管模型：1.1 理想二极管的I-V特性：1.2 实际硅二极管的I-V特性曲线：折线1.3 DC大信号模型：1.4 电荷存储特性：1.5 大信号模型的电荷存储参数Qd：1.6 温度模型：1.7 二极管模型参数表：二、双极型晶体管BJT模型：2.1 Ebers-Moll静态模型：电流注入模式和传输模式两种2.1.1 电流注入模式：2.1.2 传输模式：2.1.3 在不同的工作区域，极电流Ic Ie的工作范围不同，电流方程也各不相同：2.1.4 Early效应：基区宽度调制效应2.1.5 带Rc、Re、Rb的传输静态模型：正向参数和反向参数是相对的，基极接法不变，而发射极和集电极互换所对应的两种状态，分别称为正向状态和反向状态，与此对应的参数就分别定义为正向参数和反向参数。
2.2 Ebers-Moll大信号模型：2.3 Gummel-Pool静态模型：2.4 Gummel-Pool大信号模型：拓扑结构与Ebers-Moll大信号模型相同，非线性存储元件电压控制电容的方程也相同2.5 BJT晶体管模型总参数表：三、金属氧化物半导体晶体管MOSFET模型：3.1 一级静态模型：Shichman-Hodges模型3.2 二级静态模型（大信号模型）：Meyer模型3.2.1 电荷存储效应：3.2.2 PN结电容：3.3 三级静态模型：3.2 MOSFET模型参数表：一级模型理论上复杂，有效参数少，用于精度不高场合，迅速粗略估计电路二级模型可使用复杂程度不同的模型，计算较多，常常不能收敛三级模型精度与二级模型相同，计算时间和重复次数少，某些参数计算比较复杂四级模型BSIM，适用于短沟道（<3um）的分析，Berkley在1987年提出四、结型场效应晶体管JFET模型：基于Shichman-Hodges模型4.1 N沟道JFET静态模型：4.2 JFET大信号模型：4.3 JFET模型参数表：五、GaAs MESFET模型：分两级模型（肖特基结作栅极）GaAs MESFET模型参数表：六、数字器件模型：6.1 标准门的模型语句：.MODEL <(model)name> UGATE [模型参数] 标准门的延迟参数：6.2 三态门的模型语句：.MODEL <(model)name> UTGATE [模型参数]三态门的延迟参数：6.3 边沿触发器的模型语句：.MODEL <(model)name> UEFF [模型参数]边沿触发器参数：JKFF nff preb,clrb,clkb,j*,k*,g*,gb* JK触发器，后沿触发DFF nff preb,clrb,clk,d*,g*,gb* D触发器，前沿触发边沿触发器时间参数：6.4 钟控触发器的模型语句：.MODEL <(model)name> UGFF [模型参数]钟控触发器参数：SRFF nff preb,clrb,gate,s*,r*,q*,qb* SR触发器，时钟高电平触发DLTCH nff preb,clrb,gate,d*,g*,gb* D触发器，时钟高电平触发钟控触发器时间参数：6.5 可编程逻辑阵列器件的语句：U <name> <pld type> (<#inputs>,<#outputs>) <input_node>* <output_node># +<(timing model)name> <(io_model)name> [FILE=<(file name) text value>] +[DATA=<radix flag>$ <program data>$][MNTYMXDLY=<(delay select)value>]+[IOLEVEL=<(interface model level)value>]其中：<pld type>列表<(file name) text value> JEDEC格式文件的名称，含有阵列特定的编程数据JEDEC文件指定时，DATA语句数据可忽略<radix flag> 是下列字母之一：B 二进制 O 八进制 X 十六进制<program data> 程序数据是一个数据序列，初始都为0PLD时间模型参数：七、数字I/O接口子电路：数字电路与模拟电路连接的界面节点，SPICE自动插入此子电路子电路名（AtoDn和DtoAn）在I/O模型中定义，实现逻辑状态与电压、阻抗之间的转换。
Pspice仿真常用变压器模型
Pspice仿真——常用变压器模型时间：2012-04-12 2176次阅读【网友评论0条我要评论】收藏因为电感元件的参数比较单一，而且在仿真中，主要是仿真元件的电子特性。
所以，这里就不谈电感，而主要讨论一下变压器和耦合电感的问题。
不少朋友在使用pspice仿真的时候，只会使用元件库中的几个理想化的耦合电感和变压器模型，却不会用那种带磁芯参数的耦合电感和变压器。
下面让我们画一张原理图，把常用的理想化的和非理想话的耦合电感及变压器包含进去，进行一个仿真比较，这样才能掌握模型的特点，从而在实际工作中运用。
在这张原理图中，我们一共放置了5个耦合电感和变压器模型。
其中左边的2个是理想化的，右边三个是非理想化，模拟的是带着实际的磁芯的磁性元件，磁芯的规格是3C90材质的ER28L。
有必要先简单说一下耦合电感这个模型，让一些刚入门的朋友便于自己动手尝试。
在图中的K1、K2、K3就是以耦合电感为核心构造的几个变压器。
我们构造这种变压器的时候，需要放置一个耦合电感模型K_Linear 或K_Break或一个带磁芯的耦合电感模型例如K3所用的ER28L_3C90这个模型。
然后需要根据实际的需要放置一个电感模型作为绕组，有几个绕组就放几个电感模型，但对于一个耦合电感模型，绕组不能超过6个。
下面说说这几个模型的设置。
左边两个理想化模型：K1：耦合电感模型为K_Linear，绕组为L1和L2，必须双击K_Linear模型在其参数L1中输入L1，在参数L2中输入L2，才能实现两个绕组的耦合。
耦合系数设定为1，说明是完全耦合。
电感L1和L2的电感量，就代表绕组的电感量。
我们设定L1为250uH，L2为1000uH。
这就意味这初级与次级的匝比为1：2。
因为电感量之比是匝比的平方。
TX1：采用理想变压器模型XFRM_LINEAR，这个模型只有两个绕组，双击模型后设定耦合系数为1，两个绕组的电感量也分别设定为250uH和1000uH。
Pspice仿真类型及不同电源参数
解题步骤略。
1.瞬态分析（Transient Analysis）
瞬态分析即时域分析，可在给定激励信号的情况下，分析电路输出的时间响应，也可在没有激励信号，而有储能元件的情况下，求振荡波形。作瞬态分析可使用5种信号源（皆可通过在属性设置框中设置DC、AC值而作直流、交流分析。具体方法稍后介绍），分别是脉冲信号、分段线性信号、正弦信号、调频信号和指数信号，扫描变量是时间。另外，需要指出的是，数字电路只可作瞬态分析，而不可作直流分析或交流分析。
例：已知电路如图1-25所示，
图1-7
试求各节点电位、各支路电流和电阻消耗的功率。
解题步骤如下：
1．绘图
图1-8
（1）按按钮，点击程序＼OrCAD Family Release 9.2 Lite Edition\Capture Lite Edition，进入Capture电路图编辑界面。
（2）在SOURCE库中调用直流电压源VDC，在ANALOG库中调用电阻R及受控源G1。
按此步骤设置之后，对该电路进行直流扫描分析时，交流源的直流分量即相当于电压值为10V的直流源。
同样的，瞬态信号源包含DC、AC参数项，亦可用于直流或交流分析。设置方法与上类似。
1.直流工作点分析（Bias Point Detail）
直流工作点分析即将电路中的电容视为开路、电感视为短路，并对电路中各个信号源取其直流电平值，然后计算电路的各直流偏置量（例如节点电压、支路电流及元器件功耗等）。仿真结束后，PSice将结果自动存入扩展名为.out的输出文件中。需要指出的是，即使用户未选择进行直流工作点分析，运行PSice程序时，首先也要进行直流工作点分析。
图1-6 Probe窗口的波形显示
一般情况下，直流扫描分析使用直流电源，交流扫描分析使用交流电源，而瞬态分析使用的是瞬态信号源。但需要指出的是，交流源的属性设置框中包含DC参数项，经过设置，交流源也可应用于直流扫描分析电路中。方法是：
变压器隔离推挽式开关电源PSpice仿真设计
Science &Technology Vision 科技视界0引言在开关电源的的设计中,输入电源与输出负载之间的公共点常常需要隔离,从而阻隔共模回流和消除地线上的环流。
开关电源在变换器部分的设计中常采用变压器来隔离输入电源与输出负载,并且通过变压器实现电压调节功能。
推挽式开关电源包含两个反相工作的正激变换器,因此推挽式开关电源具有正激式开关电源拓扑的所有优点[1]。
由于电源提供给负载的功率并不在变压器中存储,推挽式开关电源比正激式开关电源能处理更多的功率,且效率更高,控制性能更好。
1电路拓扑结构及工作原理本设计为采用变压器隔离推挽式输出的开关电源,其输入电压为35V,输出电压为5V,输出电流为2A,纹波为250mV。
为了实现稳定输出,开关电源采用电流型PWM 开关电源控制器SG1846构成反馈回路,开关频率为25kHZ。
为缓解变压器磁通不平衡造成的过热失控问题,在电路中采用功率MOSFET 作为开关管[2]。
其电路结构如图1所示。
如图所示,开关管M 1,M 2根据输入电压V 1、变压器变比和预期输出电压所决定的占空比每半周期交替导通。
为避免两开关管直通现象的发生,需要在M 2开通和M 1关断之间设置死区时间t d 。
二极管D 2上的电压为矩形波,幅值为输入电压V 1乘以变压器电压比,即v=V 1N s N P(1)二级管D 2正偏导通,输出电感L 1此时间段内充磁。
二级管D 2的电流i D 1(t )与开关管集电极电流i c 1(t )成比例。
i D 1(t )=i c 1(t )N PN s(2)电路的电压调整率可以从输出电感L 1的伏秒平衡关系导出,由于每个开关周期输出电感都要充磁两次,开关管开通时间t on 的系数为2。
电压调整率为V o =V 1N s N P 2t onT ()(3)如果考虑开关管与二极管的通态压降,电压调整率为V o =V 1-V cesat ()N s N P -V d []2t onT()(4)式中V cesat 为开关管通态压降,V d 为二极管通态压降,T 为开关周期。
Proteus电路spice器件模型
模拟电路模型常见的有pspice、hspice、通用spice、ti_spice、ISPICE模型。
Prospice（Proteus，采用spice3f5）一、一般元件1、电阻温度特性：。
L1:A, MUTUAL_B=0.5L1:B二、半导体器件1、二极管（19个参数）对于硅二级管，工作电流大概为10uA-100A。
IS为0.01-10uA。
硅整流二极管为5-40uA，稳压二极管为0.1-5uA，发光二极管1-100uA。
对于锗二极管，工作电流100uA-100mA。
IS为10-1000uA。
100mV，ID/IS=45。
200mV，ID/IS=2194。
300m，VID/IS=10^5。
500mV，ID/IS=2.5*10^8。
600mV，ID/IS=1.2*10^10。
700m，VID/IS=5.7*10^11。
1.1 DC模型：Ut=k*T/q=1.3806226e-23*300.15/1.6021918e-19 =0.025864186伏。
推出Ut*ln(10)=0.05955449。
1.2 大信号模型的电荷存储参数Qd：1.3温度模型：φ(T2)=T2/T1*φ(T1)-2*K*T2/q*ln(Is2/Is1);1.4噪声模型1.5IS, RS and CJO are scaled by the area factor。
1.6 直流参数的推导（参数Is、N、Rs）1、Ud=Id*Rs+Ut2*N*ln(Id/IS)方程：设Id=100mA。
① Ud=Id*Rs+Ut2*N*ln(Id/IS)=0.338221② Ud1=Id*e*Rs+Ut2*N+Ut2*N*ln(Id/Is)=0.535911③ Ud2=Id*10*Rs+Ut2*N*ln(10)+Ut2*N*ln(Id/IS)=1.29777②-① Id*(exp(1)-1)*Rs+Ut2*N=0.19769;③-① Id*9*Rs+Ut2*N*log(10)=0.959549;推出Rs=0.9999999，N=0.99990887。
变压器仿真模型
Midcom's Tips for Transformer Modelingby Dave LeVasseur29-Jan-98Transformer characterization is important if you plan to model your circuit in PSPICE or other simulation programs. Unfortunately, not all transformer manufacturers provide circuit models for their products, so you may find you need to determine the equivalent circuit model based on lab measurements. This technical note will help you characterize almost any transformer so you can develop its equivalent circuit model to suit your analysis needs. It is intended to help circuit designers, component engineers understand transformer capabilities and limitations. For further information on transformer behavior and modeling, refer to Midcom Technical Note #69 which may be found on the Midcom web site, .Characterizing a transformer of unknown origin can be tricky, but if you know the application in which it is intended to be used, you can make some educated guesses regarding the working impedances, voltages and bandwidth. Here is the equivalent circuit on which we will base our measurements for the purposes of assigning values to the as-yet unknown elements:Figure 0-1The "complete" transformer equivalent circuitFIRST-ORDER APPROXIMATIONThis technique will give you useful results for most power or audio transformers and is relatively quick to perform. You will need a standard ohmmeter and an LCR bridge capable of displaying series and parallel complex impedances (R ± jX). If you're willing to convert your readings, you may use any means available that will provide complex impedance measurements. If you are patient, you can even do this with a simple VOM using the method described in Midcom Technical Note #35, VOM Measures Complex Impedances.1. DC Resistances (R DC )Measure DC resistances, R Pri and R Sec using a standard ohmmeter. If the ambient temperature is within five degrees of 20°C, your measurements will be within 2% of the 20°C value. (Temperature compensation is covered in a later section)R Pri : ______ R Sec : ______2. Magnetizing Inductance and Core Loss ResistanceSet up the LCR bridge to measure inductance and resistance in the parallel mode: L P , R P . Measure the inductance and resistance at the primary winding at or near the midband of the transformer's frequency range of operation. You can use 55 Hz for 50/60 Hz linear power transformers while 1kHz is a good choice for analog telecommunications magnetics. The values you measure will correspond approximately to the elements L Pri and R core shown in the equivalent circuit.L P : _____________ R P : _______________ Use L P for L Pri , R P for R Core .3. Leakage InductanceMeasure leakage inductance, L Leak , by shorting the secondary and measuring the inductance at the primary. Try 1kHz first, then 10kHz. If you are measuring a transformer with very small turn counts less than about 25 turns on each winding, take a reading at 100kHz. Use whichever value is lowest , but not negative (which would indicate capacitance). The reason you need the lowest reading is explained in section on "Improved Accuracy Modeling". Note: If you insist on using coupling coefficient (k) instead of leakage inductance, please refer to Midcom Technical Note #69 for a means of converting between leakage inductance and coupling coefficient.L Leak : ______________ at ________________ Hz.4. Turns RatioIf you don't know the nominal turns ratio from the transformer's spec sheet, you'll need to measure it. Turns ratio,α, is best measured by using precision decade resistor substitution boxes in a bridge configuration that will be described later. You canmake a rough estimate of turns ratio by applying an ac voltage at the primary, then measuring the voltage at the secondary. Armed with this, you then calculate turns ratio by α =V Pri /V Sec . This method is prone to several kinds of errors, so a better approach is toemploy your LCR bridge once again.Set the LCR bridge to measure L P and Q, then connect the bridge to the primary terminals. Vary the frequency of measurement until the Q is at its highest reading. Laminated sheet steel transformers will typically have Q values of perhaps less than 1 to about 5; maybe 10 if you're lucky. Ferrite-based transformers will have substantially higher Q values than those with steel cores, perhaps as high as 100. The higher your Q value, the more accurate will be your turns ratio estimation. If your Q value is less than 5, you can't count on this method to provide better than a few percent of accuracy.Measure the secondary inductance at the same frequency as you did the primary, noting the Q value.L Pri at ___________Hz, Q = ______________L Sec at ___________Hz, Q = ______________Assuming Q is high enough, turns ratio may be calculated by:1)If this calculation returns a value less than 0.5 or greater than 2, consider adjusting the voltage applied to either winding to account for changes in flux density. The goal here is to keep maintain the following equivalence: .Laminated sheet steel transformers may exhibit significant change in inductance depending on the excitation voltage applied to their windings. Ferrite-core transformers also exhibit this behavior, but to a lesser extent. The reason inductance depends on excitation level is explained in Midcom Technical Note #69.To correct for effects of flux density, reduce the voltage applied to the winding with the lower inductance value (either L Pri or L Sec )by the ratio you first calculated in equation 1. For example, if you calculate α to be 4.0, reduce the measurement voltage of L Sec from 1.0V to 0.25V . Recalculate α if this yields a significantly different (usually lower) value of L sec . Call this L´Sec to differentiate it from the uncorrected flux density.Turns ratio, corrected for flux density, B m :If your uncorrected value for α was less than 0.5, raise V Sec by a factor of measured α, then remeasure and recalculate. Watch out for saturation symptoms such as falling or unstable inductance readings. A single iteration of level adjustment/remeasurement is usually sufficient.5. Distributed Winding CapacitancesThe primary and secondary distributed winding capacitances manifest their effects at the higher frequencies. Most well-designed communications transformers will have negligibly small values of distributed capacitance such that measuring them may be more work than the results are worth. This is less true with switchmode power transformers. To determine if the values of C Pri and C Sec are significant, connect your LCR bridge to the transformer's primary and increase the frequency until the impedance seen at the primary becomes capacitive. The frequency at which the impedance changes from inductive (+jX) to capacitive (-jX) is the transformer's self-resonant frequency. If this frequency is well above the passband of interest, it may not be worthwhile tocontinue. If self-resonance is not too far (within a decade or two) above the midband frequency of the transformer's passband, you can make a quick estimate of distributed capacitance as follows: Raise the frequency of the LCR bridge beyond self-resonance (where the reactance changes from inductive to capacitive) until you find the frequency at which the Q of the capacitance is at its maximum. Use the series (C S ) mode. If the Q of the capacitance is at least 5, preferably 10 or more, you have a useful value of the total distributed capacitance. You can assume that all of this capacitance exists on the primary side and call it C Pri , then assume C Sec is zero. Conversely, you could assume that the entire capacitance exists at the secondary, but to move it over to that side, you must multiply the measured primary capacitance by the square turns ratio, α. You must multiply by α2 instead of dividing (as you would with an impedance) because you are working with capacitance which is proportional to the reciprocal of impedance. This effect allows a transformer with a 0.5:1 turns ratio to effectively multiply by four a capacitance placed on its secondary.That's it. You now have a first-order approximation of the transformer's equivalent circuit model. For an improved accuracy model, keep reading.IMPROVED ACCURACY MODELINGLet's start again with the easy ones: R Pri and R Sec . If you want to get fancy you can use a temperature-compensated ohmmeter with a thermal-sensor matched to the metal comprising the transformer's windings. Copper is the most common metal used in transformer windings and its temperature coefficient of resistance (T ce ) is about +0.4% per degree Celsius. You could also measure resistance with an uncompensated ohmmeter, then consult a nearby thermometer to convert to Standard Temperature of 20°C (not 25°C like the rest of the electronics world). As it turns out, we actually need the uncompensated reading for later use,so record it anyway. Unless you work outside in the Arctic or Death Valley, take your readings with reasonable precision (3 places for xxx, 4 places for 1xxx) on a standard ohmmeter. Use of a four-terminal (kelvin) ohmmeter is recommended for resistances less than about 0.1 ohm. If you don't have a four-terminal ohmmeter, but you do have a stable and accurate current source and a high-accuracy voltmeter, you can use the current source to develop a voltage across the transformer winding. Measure the resulting voltage drop with your high-accuracy voltmeter. This is exactly the same as a four-terminal test, but without the convenience of doing it all with the same meter.WARNING: Applying current to an inductor can result in dangerous and possibly lethal voltages when the source of the current is suddenly removed. (For you calculus buffs, remember that )For safety's sake, use just enough current to provide an adequate reading on the voltmeter. Keep the power dissipated in thewinding low to prevent the winding's resistance from climbing due to a current-induced increase in temperature. Stay below a few tens of milliwatts for transformers larger than your thumb and one or five milliwatts for those that are smaller. A level of 100mW may be appropriate for fist-sized and larger transformers with large wire diameters. Connect the inductor or transformer to the current source with the source turned off. Slowly (over the course of several seconds) raise the current until the voltage read on the voltmeter is stable and comfortably within its accuracy range. If you notice the voltage reading start to climb, you may be heating the winding. Either that or your current source hasn't stabilized.If the winding resistance is less than an ohm, you may have to settle for a voltage less than 100mV to stay below the 100mW power level. Resistances less than 0.1 ohm require even lower voltages, so make sure you have chosen an appropriate voltmeter and current source for the job.After the voltage and current have stabilized, note the voltage and current values, then slowly reduce the current to zero. Physically small inductors and transformers (smaller than your fist) are probably safe enough that several seconds of ramp up and down will keep your hair in place. An inductor the size of a breadbox (does anyone have a breadbox anymore?) is as dangerous when carrying high currents as it would be if someone dropped it your head.INDUCTANCEFor improved accuracy, we need to describe how the transformer's winding inductance changes with varying frequency. We also need to remove effects of R Pri in our attempt to measure L Pri .To begin this process, it is important to know the upper and lower band limits of the transformer's operational frequency range.We also need to know something about the transformer's operational voltage range. If we apply too much voltage at a frequency below the transformer's lower frequency limit, we run the risk of causing core saturation. While saturation does no permanent damage to the core, it won't allow us to accurately model the transformer.To check for saturation, first measure the transformer's inductance. We'll arbitrarily choose the series equivalent at its lowest frequency and highest drive level. Call this operating point f LOW , V HIGH . The inductance obtained under this set of conditions will be the highest inductance value we measure for any subsequent combination of higher frequency and lower voltage. If we find the inductance goes up significantly as we raise the frequency above f LOW , the transformer is operating in an unstable region near saturation. Similarly, if we reduce the voltage below V HIGH and find inductance increases, we have again shown that thetransformer is operating near its saturation point. The following family of curves shows the effect of saturation on inductance:Figure 0-2The equivalent circuit model of a single winding of a transformer is the same as that of an inductor. That equivalent circuit looks like this:Figure 0-3The arrows show the dependency of each element on flux density and frequency. The skin effect is responsible for the frequency dependence of R COIL , but this is beyond the scope of this technical note. Skin effect is covered in greater detail in Midcom Technical Note #69.We found R COIL as our first step (where we called it R DC for simplicity). We now need to account for it in our model. We need the actual resistance, not compensated for temperature, to determine the true core loss resistance and magnetizing inductance values.1.With your LCR bridge in the series equivalent mode, measure L S and R S . If your bridge does not have this capability, or your measurements were recorded using the parallel equivalent, you must convert your measurements to L S and R S . The conversions between bridge parameters are readily available from various sources. Refer to the manual for your LCR bridgeif you can't find them anywhere else. Five common conversions are shown here for reference:2.Subtract the DCR from the measured value of R S . Call this R S ' to differentiate it from the measured reading.3.Calculate the intrinsic inductance and core loss resistance by converting L S and R S ' into L P and R P using the following conversions, noting that core loss, R C , is the parallel element R P in the L P /R P combination.Example:Find L P and R C given measured L S and R S at 1000 Hz are 1.138 henries and 6809 ohms at a frequency of 1 kHz. The DC resistance of the coil is 143.24 ohms:ohmsMAGNETIZING INDUCTANCE AND CORE LOSS RESISTANCE FACTORSMagnetizing inductance and core loss resistance change over frequency. The factors by which they do this were fit to curves by Midcom engineers in the early 1980s. The factors do not include effects due to saturation and are only approximations designed primarily for voiceband applications. The formulae are reproduced here for reference:Inductance change versus frequencywhere:f is the frequency at which we need to determine the inductanceL' is the inductance at a given frequency, fL fR is the inductance at a reference frequency, f R f R is a reference frequency, usually chosen to be near the transformers mid-band frequency f L is a frequency lower than f R , usually about one-fifth the value of f RαL is ratio of inductances at a low frequency, f L to the reference frequency, f R , orCore loss change versus frequencywhere:f is the frequency at which we need to determine the core loss resistanceR Cf is the inductance at a given frequency, f R CfR is the core loss resistance at a reference frequency, f R f R is a reference frequency, usually chosen to be near the transformers mid-band frequencyαRc is the core loss resistance factor, which usually ranges from 0.35 to 0.45We need to be able to calculate the two factors, αL and αRc from measured values. The two equations, solved for their respectivefactors are shown here:Example:A transformer has measured inductances of 1.8 H at 400 Hz and 1.2 H at 1000 Hz. Find its inductance ratio, αL :Example:A transformer has measured core loss resistances of 20k ohms at 500 Hz and 26k ohms at 1000 Hz. Find its core loss resistance ratio, αRc :LEAKAGE INDUCTANCEDirect measurement of leakage inductance can be difficult if the magnetizing inductance is low. The quick-and-dirty approach described earlier will give you useable results in most cases, but you should understand why you raise the frequency to find the most valid reading. An abbreviated equivalent circuit for the leakage inductance case looks like this at frequencies below self-resonance:Figure 0-4Leakage Inductance equivalent circuitWhen L Pri is two or more orders of magnitude larger than L Leak , we can simply raise the measurement frequency until the effects of L Pri do not significantly affect the reading of L Leak . Most communications transformers have ratios of L Pri /L Leak on the order of 200to 1000. Flyback and certain digital telecommunications transformers with gapped cores may have ratios between 20 and 100.Ratios greater than 200 or so order allow direct reading of L Leak with minimal loss of accuracy due to effects of L Pri . The trick is to find the frequency at which the impedance of L Pri is high enough not to interfere with the reading of L Leak . One way to do this is to set your LCR bridge to read L S and sweep the frequency until the inductance reaches it lowest value. A spot-frequency version of this was outlined in the earlier section. Another way is to "back out" the value of L Pri from the Ls and Rs readings, but this ends up being fairly complicated. I worked up a MathCad sheet that calculates 'true' leakage inductance given L Pri , R Pri , R Sec , turns ratio and measured input inductance and resistance. The MathCad sheet works fine, but when the frequency is on the low side of the plateau you must make extremely precise measurements of the input parameters to achieve believable results.The example shown in the MathCad sheet worked fine for 1kHz and 10kHz, but at 100Hz an error of -0.5% in the reading of R Sin gave a result of 165uH (when the correct number was about 14uH). An error of +0.1% yielded an impossible -26uH result. The moral of the story: stay reasonably close to the plateau range shown in figure 5.Figure 0-5Apparent Leakage inductance versus frequency TURNS RATIOTurns ratio errors generally result from poor coupling between the windings being measured. You can ameliorate this by putting the two windings in series then measuring the voltage ratio of one winding to the two in series. If the windings happen to be 1:1you'll get a nice, round number of 2 for turns ratio when you use this method. Common-mode chokes are commonly tested for turns ratio this way since they typically have very poor coupling between the windings. At the "coil" stage of the manufacturing process, transformers may be tested with ungapped cores to reduce the effects of fringing. (The root word is fringe, making the first "g" in fringing a soft sound)Fringing can make turns located near a gap seem to disappear in a magnetic sense. Thus we have turned an electrical conundrum into a philosophical question: Is the turns ratio of a transformer equal to the quotient of each winding's turn count or is it themeasured ratio after the appropriate gapped core is placed on the coil? The answer depends on what you are trying to accomplish.As transformer manufacturers, we need to know the exact turns count our winding machines are applying to each coil. That is why we (usually) test our coils with ungapped cores to reduce the effects of fringing. The fully-assembled product, however, may react differently. Remembering that magnetic flux is actually a gradient, it is possible for any portion of a turn, or more than one turn, to be magnetically absent from the flux path. In a power supply, for example, the fringing effect may lead to lower output voltages than that predicted by the actual turns count applied to the coil. Knowing this, a transformer designer could add physical turns to make up for the apparent loss of turns. In this instance, it is more important that the effective turns ratio be measured since the output voltage is a function of the turns included in the magnetic flux path. This is true regardless of the actual number that are wound on the coil.So the question remains: what is the turns ratio of a coil in the presence of fringing effects? The answer: whatever its effective voltage ratio happens to be. How do we measure this ratio? Any of several ways, but one that works very well in the absence of fringing and reasonably well in the presence of fringing is the resistive bridge method.To use the resistive bridge method, connect the windings as shown at N X and N ref . Choose an oscillator level and voltage consis-tent with the midband frequency and drive level appropriate for the transformer under test. At Midcom, we use 1kHz or 10kHz and occasionally 100kHz, and we set V 1 to 1.0V . The exact values for frequency and voltage aren't critical. The 671-0018 isolates the generator from the circuit, but it won't work very well below 100Hz or above 100kHz. Set R REF to some convenient decade value (10, 100, 1k, etc) ohms. Adjust R X until V 2 is nulled as low as it will go. You should be able to get nulls less than 0.001V 1.If you can't, you may have poor coupling between N X and N REF or the polarity of one winding is reversed. (This method will also verify polarity - an added benefit of the test) The resistive bridge method is Midcom Test Specification 999-2357. Refer to that specification for more details.Figure 0-6Resistive Bridge Turns Ratio Test MethodIf all else fails, you can use the method described in the first section to determine turns ratio using inductance readings. What that method lacks in accuracy is made up in its convenience. If winding resistance is reasonably low, the inductance ratio method will give pretty decent results.DISTRIBUTED CAPACITANCESOnly in the case of very loosely-coupled windings is the result of attempting to determine the individual primary and secondary capacitances worth the effort. Double-tuned RF transformers and some common-mode chokes are the only cases I can think of where it might be beneficial to know the individual values versus the lumped-constant capacitance value. Remember that the lumped-constant equivalent circuit is really only a convenience for us to visualize all of the elements. A distributed-parameter model is more appropriate, but much harder to analyze. In my lab experience, shorting one winding will cause the capacitance one would normally attribute to the shorted winding to "float" (reflect) over to the other side and appear at the measured winding's terminals.As a method of double-checking the capacitance value you found in the first section, you can measure the transformer's self-resonant frequency and then calculate the distributed capacitance by solving the resonance formula for capacitance:Note that you must make some assumptions about L at the resonant frequency. Remembering that inductance is not always constant as frequency is varied, you may need to use the inductance factor to estimate the inductance at the resonant frequency since it cannot be measured directly. By making a series of measurements of inductance below self-resonance, you can see the trend of inductance shift versus frequency, then plug in an appropriate value to the formula shown above. If your impedance bridge is only capable of discrete frequency measurements, none of which is particularly close to the resonant frequency, you can estimate the resonant frequency by noting the two frequencies on either side of resonance that are conjugates of each other and taking the average of them. (You could take the geometric mean, instead if you feel the additional accuracy iswarranted)Midcom Technical Note #82 Page 11 of 11 INTERWINDING CAPACITANCEThis one is so easy I should have put it in the first section. In practice, interwinding capacitance is rarely a concern in telecommunications transformer since it affects only common-mode signals (assuming the transformer is properly designed for good balance). In switchmode power and certain balanced inductors, interwinding and winding-to-core capacitance may result in low self-resonant frequency, reduced rise time and extended ringing.To measure interwinding capacitance, short the primary windings, short the secondary windings, then measure capacitance between the two. If you don't short the windings, you run the risk of error due to a voltage differential being developed across the transformer's windings. If this happens, you will see some inductance inside your capacitance. This effect is most pronounced when you measure from a 'dotted' terminal on the primary side and an 'undotted' terminal on the secondary side. If this happens you will notice a self-resonance at fairly low frequencies. This is true particularly if the interwinding capacitance is fairly high or the winding inductance is high.Figure 0-7FINAL WORDSIf you notice any unusual behavior when you are measuring transformer parameters, send your results to me. As time permits, I will update this technical note with your input, which is welcomed.Best regards,Dave LeVasseurChief Technical OfficerMidcom, Inc.dlevasseur@+1 605.882.0339 (direct)。
Pspice器件模型参数说明
Pspice 器件模型参数说明1、二极管模型及主要参数二极管模型参数如表1所示 名称 符号 SPIC 名称 单位 缺省值 反向饱和电流(Saturation current) I S IS A 10-14 欧姆电阻(Ohmic resistance) R S RS Ω 0 发射系数(Emission coefficient) n N 1 渡越时间(Transit time) τT TT s 0 零偏置电容(Zero-bias junction capacitance) C j0 CJ0 F 0 结电压(Junction potential) V 0 VJ V 1 电容梯度因子(Grading coefficient) m M 0.5 反向击穿电压(Reverse breakdown voltage) V ZK BV V ∞ 反向击穿电流(Current at breakdown voltage) I ZK IBV A 10-10仿真时采用理想二极管，参数不需要设置。
参数说明：I S ：PN 结反向扩散电流，该值远小于PN 结反向（漏）电流，因为它为包括反向空间电荷区产生的电流、表面复合电流、表面沟道电流和表面漏导电流。
n ：一般n =1，测量：正向特性线性区 )/ln(2121D D D D I I V V kT q n −=C j0: CD =C d +C j =m nU U V U C eI U )1()1(0D 0j s TTTD −+−τ0j T T 2)1(TDC e I U nU Us+−≈τV 0：0.7-0.8Vm : 0.3-0.5, 一般为0.332、 稳压管模型及主要参数模型参数如表1所示，参数设置如下： V ZK =U Z I ZK =I Zmin3、 晶体管模型及主要参数模型参数如表2所示名称符号 SPIC 名称 单位 缺省值 传输饱和电流 I S IS A 10-16 正向电流增益 βF BF100 反向电流增益 βR BR 1集电极电阻 R CC’ RC Ω 0 发射极电阻 R EE’ RE Ω 0 基极电阻R BB’ RB Ω 0 理想正向渡越时间τF TFs 0理想反向渡越时间 τR TR s 0 发射结零偏置势垒电容 C je0 CJE F 0 发射结电容梯度因子 m BEJ MJE 0.33 发射结内建电势 V 0e VJE V 0.75 集电结零偏置势垒电容 C jc0 CJC F 0 集电结零偏置势垒电容 m BCJ MJC 0.33 集电结零偏置势垒电容 V 0c VJC V 0.75一般参数设置如下：RB: r bb’RE, RC: 一般设为0 V 0e : =U BE , 一般为0.7V V 0c : 一般为0.75V其它参数说明：0je me0BE 0je je C 2)V U 1(C C ≈−=，此处m BE 约为0.5mc 0CB 0)V U 1(C C +=μμ，此处m BC 约为0.2-0.5 参数设置经验：C je0=0.5C π，C jc0=C μ=C ob4、 MOSFET 模型及主要参数i D 与u GS 、u DS 之间的关系：2GS(th)DO n 2GS(th)GS n 2GS(th)GS n D 2DS DS GS(th)GS n D oxn n n 2GS(th)GS ox n D ox ox oxoxox U I k )U U (k )U U )(L W('k 21i U 21U )U U )[(L W ('k i C 'k )L W()U U )(L W )(C (21i )T (T C =−=−=−−==−==恒流区：可变电阻区：沟道宽长比载流子迁移率，二氧化硅厚度二氧化硅介电常数，μμμεε模型参数设置：KP=k n ’， VT0=阈值电压U GS(th)。
Pspice器件模型参数说明
Pspice 器件模型参数说明1、二极管模型及主要参数二极管模型参数如表1所示 名称 符号 SPIC 名称 单位 缺省值 反向饱和电流(Saturation current) I S IS A 10-14 欧姆电阻(Ohmic resistance) R S RS Ω 0 发射系数(Emission coefficient) n N 1 渡越时间(Transit time) τT TT s 0 零偏置电容(Zero-bias junction capacitance) C j0 CJ0 F 0 结电压(Junction potential) V 0 VJ V 1 电容梯度因子(Grading coefficient) m M 0.5 反向击穿电压(Reverse breakdown voltage) V ZK BV V ∞ 反向击穿电流(Current at breakdown voltage) I ZK IBV A 10-10仿真时采用理想二极管，参数不需要设置。
参数说明：I S ：PN 结反向扩散电流，该值远小于PN 结反向（漏）电流，因为它为包括反向空间电荷区产生的电流、表面复合电流、表面沟道电流和表面漏导电流。
n ：一般n =1，测量：正向特性线性区 )/ln(2121D D D D I I V V kT q n −=C j0: CD =C d +C j =m nU U V U C eI U )1()1(0D 0j s TTTD −+−τ0j T T 2)1(TDC e I U nU Us+−≈τV 0：0.7-0.8Vm : 0.3-0.5, 一般为0.332、 稳压管模型及主要参数模型参数如表1所示，参数设置如下： V ZK =U Z I ZK =I Zmin3、 晶体管模型及主要参数模型参数如表2所示名称符号 SPIC 名称 单位 缺省值 传输饱和电流 I S IS A 10-16 正向电流增益 βF BF100 反向电流增益 βR BR 1集电极电阻 R CC’ RC Ω 0 发射极电阻 R EE’ RE Ω 0 基极电阻R BB’ RB Ω 0 理想正向渡越时间τF TFs 0理想反向渡越时间 τR TR s 0 发射结零偏置势垒电容 C je0 CJE F 0 发射结电容梯度因子 m BEJ MJE 0.33 发射结内建电势 V 0e VJE V 0.75 集电结零偏置势垒电容 C jc0 CJC F 0 集电结零偏置势垒电容 m BCJ MJC 0.33 集电结零偏置势垒电容 V 0c VJC V 0.75一般参数设置如下：RB: r bb’RE, RC: 一般设为0 V 0e : =U BE , 一般为0.7V V 0c : 一般为0.75V其它参数说明：0je me0BE 0je je C 2)V U 1(C C ≈−=，此处m BE 约为0.5mc 0CB 0)V U 1(C C +=μμ，此处m BC 约为0.2-0.5 参数设置经验：C je0=0.5C π，C jc0=C μ=C ob4、 MOSFET 模型及主要参数i D 与u GS 、u DS 之间的关系：2GS(th)DO n 2GS(th)GS n 2GS(th)GS n D 2DS DS GS(th)GS n D oxn n n 2GS(th)GS ox n D ox ox oxoxox U I k )U U (k )U U )(L W('k 21i U 21U )U U )[(L W ('k i C 'k )L W()U U )(L W )(C (21i )T (T C =−=−=−−==−==恒流区：可变电阻区：沟道宽长比载流子迁移率，二氧化硅厚度二氧化硅介电常数，μμμεε模型参数设置：KP=k n ’， VT0=阈值电压U GS(th)。
Pspice仿真常用变压器模型教学提纲
P s p i c e仿真常用变压器模型Pspice仿真——常用变压器模型时间：2012-04-12 2176次阅读【网友评论0条我要评论】收藏因为电感元件的参数比较单一，而且在仿真中，主要是仿真元件的电子特性。
所以，这里就不谈电感，而主要讨论一下变压器和耦合电感的问题。
不少朋友在使用pspice仿真的时候，只会使用元件库中的几个理想化的耦合电感和变压器模型，却不会用那种带磁芯参数的耦合电感和变压器。
下面让我们画一张原理图，把常用的理想化的和非理想话的耦合电感及变压器包含进去，进行一个仿真比较，这样才能掌握模型的特点，从而在实际工作中运用。
在这张原理图中，我们一共放置了5个耦合电感和变压器模型。
其中左边的2个是理想化的，右边三个是非理想化，模拟的是带着实际的磁芯的磁性元件，磁芯的规格是3C90材质的ER28L。
有必要先简单说一下耦合电感这个模型，让一些刚入门的朋友便于自己动手尝试。
在图中的K1、K2、K3就是以耦合电感为核心构造的几个变压器。
我们构造这种变压器的时候，需要放置一个耦合电感模型K_Linear或K_Break或一个带磁芯的耦合电感模型例如K3所用的ER28L_3C90这个模型。
然后需要根据实际的需要放置一个电感模型作为绕组，有几个绕组就放几个电感模型，但对于一个耦合电感模型，绕组不能超过6个。
下面说说这几个模型的设置。
左边两个理想化模型：K1：耦合电感模型为K_Linear，绕组为L1和L2，必须双击K_Linear模型在其参数L1中输入L1，在参数L2中输入L2，才能实现两个绕组的耦合。
耦合系数设定为1，说明是完全耦合。
电感L1和L2的电感量，就代表绕组的电感量。
我们设定L1为250uH，L2为1000uH。
这就意味这初级与次级的匝比为1：2。
因为电感量之比是匝比的平方。
TX1：采用理想变压器模型XFRM_LINEAR，这个模型只有两个绕组，双击模型后设定耦合系数为1，两个绕组的电感量也分别设定为250uH和1000uH。
(整理)PSCAD中的变压器模型.
1.Introduction to Transformers（引言）EMTDC中使用变压器有两种方法：经典方法和统一的磁等效电路（unified magnetic equivalent circuit (UMEC)）方法。
经典方法用来模拟同一变压器铁芯上的绕组。
也就是说，每一相都是独立的，各单相变压器之间没有相互作用。
而UMEC方法计及了相间的相互作用：由此，可以对3相3臂或3相5臂式变压器构造进行精确的模拟。
每一模型中，铁芯的非线性特征是最基本的不同。
经典模型中的铁芯饱和是通过对选定绕组使用补偿注入电流实现的。
UMEC方法采用完全插值，采用分断线性化的ϕ-I曲线来表征饱和特性。
2.Transformer Models Overview（变压器模型概述）对电力系统进行电磁暂态分析过程中必然会出现变压器。
PSCAD中有两种方法对变压器进行模拟：经典方法和UMEC方法。
经典方法仅限于单相设备，其中不同的绕组处于同一铁芯腿上。
而UMEC方法，考虑到来铁芯的几何外形和相间的相互耦合因素。
除了以上的显著区别外，两种变压器模型之间最基本的区别是对铁芯非线性特性的描述。
在经典模型中，非线性特性采用近似地基于“拐点”、“空心电抗”和额定电压的磁化电流曲线进行模拟。
而UMEC模型则直接采用V-I曲线进行模拟。
与经典模型不同，UMEC模型没有配置在线分接头调整功能。
但是，可以在指定绕组上设置分接头，不过分接头在仿真过程中不能动态调整。
3.1-Phase Auto Transformer（单相自耦变压器）此组件基于经典方法模拟了单相自耦变压器。
用户可以选择采用磁化支路（线性铁芯）或注入电流模拟磁化特性。
理想情况下，可以忽略磁化支路，变压器即为理想模式，仅保留串联的漏抗。
4.3-Phase Star-Star Auto Transformer（三相星形连接的自耦变压器）此组件模拟了由3个单相构成的3相自耦变压器。
用户可以选择采用磁化支路（线性铁芯）或注入电流模拟磁化特性。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
Pspice仿真——常常使用变压器模子之青柳念文创作时间：0412 2176次阅读【网友评论0条我要评论】收藏
因为电感元件的参数比较单一，而且在仿真中，主要是仿真元件的电子特性.所以，这里就不谈电感，而主要讨论一下变压器和耦合电感的问题.很多朋友在使用pspice仿真的时候，只会使用元件库中的几个抱负化的耦合电感和变压器模子，却不会用那种带磁芯参数的耦合电感和变压器.下面让我们画一张原理图，把常常使用的抱负化的和非抱负话的耦合电感及变压器包含出来，停止一个仿真比较，这样才干掌握模子的特点，从而在实际工作中运用.
在这张原理图中，我们一共放置了5个耦合电感和变压器模子.其中左边的2个是抱负化的，右边三个是非抱负化，摹拟的是带着实际的磁芯的磁性元件，磁芯的规格是3C90材质的ER28L.
有需要先简单说一下耦合电感这个模子，让一些刚入门的朋友便于自己动手测验测验.在图中的K1、K2、K3就是以耦合电感为核心构造的几个变压器.我们构造这种变压器的时候，需要放置一个耦合电感模子K_Linear或K_Break或一个带磁芯的耦合电感模子例如K3所用的ER28L_3C90这个模子.然后需要根据实际的需要放置一个电感模子作为绕组，有几个绕组就放几个电感模子，但对于一个耦合电感模
子，绕组不克不及超出6个.
下面说说这几个模子的设置.
左边两个抱负化模子：
K1：耦合电感模子为K_Linear，绕组为L1和L2，必须双击K_Linear模子在其参数L1中输入L1，在参数L2中输入
L2，才干实现两个绕组的耦合.耦合系数设定为1，说明是完全耦合.电感L1和L2的电感量，就代表绕组的电感量.我们设定L1为250uH，L2为1000uH.这就意味这初级与次级的匝比为1：2.因为电感量之比是匝比的平方.
TX1：采取抱负变压器模子XFRM_LINEAR，这个模子只有两个绕组，双击模子后设定耦合系数为1，两个绕组的电感量也分别设定为250uH和1000uH.
右边的非抱负化模子：
K2：采取的耦合电感模子为KBreak，同样还需要放置两个电感，这里是L3和L4，双击KBreak的模子，设定耦合系数为1，参数L1为L3，参数L2为L4，把参数Implementation设置为ER28L_3C90.这里要注意了，电感L3和L4的参数分别为10和20.这个数字代表什么意思呢？是电感量吗？不是，千万注意，这里的意思是匝数！凡是采取了带磁芯的模子，就不再用电感量来作为参数了，而是使用匝数.
TX2：采取的非抱负变压器模子XFRM_NONLINEAR，同样这个
模子只有两个绕组.双击模子后，设定耦合系数为1，参数Implementation设置为ER28L_3C90，参数L1_TURNS和
L2_TRUNS分别设置为10和20.很显然，这里也是匝数.
K3：直接采取ER28L_3C90磁芯的耦合电感模子，绕组为L5和L6.双击耦合电感模子，设置耦合系数为1，参数L1为
L5，参数L2为L6.把L5和L6的参数改成10和20.
为什么我这里要把采取ER28L_3C90磁芯的模子的匝比设定为10：20呢，因为这个磁芯的电感系数为250nH/N2，这样刚好使初次级之间的电感量之比
为250uH:1000uH，和抱负模子的参数一致，方便仿真后成果的比较.
现在我们在每一个变压器的初级串一个0.001欧的电阻，次级接上10欧的负载.并放置一个峰值1V，频率10KHz的正弦波电压源给变压器初级提供输入信号，并双击这个信号源，设置初始相位为90度，如图毗连好电路，放置0电位的地.然后点击菜单pspice项，选择new simulation profile建立一个新的仿真.然后选择时域仿真，见下图：设定仿真时间从0秒开端到1ms，最大步长100ns，跳过初始化偏压点计算.然后点击菜单pspice，选择Run，仿真便可以开端了.
待仿真完成后，如最开端的图放置电压探头.我们已经知道这些变压器的变比是1：2，那末实际的电压输出是不是这
样呢？看看吧：
从图中可以看到，输入电压峰值为1V的正弦波，输出为峰值为2V的正弦波.再如下图放置电流探头：
然后点击仿真器的菜单plot，选择Add plot to window，再放置一个如上图中的看输入电压信号的电压探头，可以
同时看到输入电压与输出电流的波形：从波形上可以看出，每一个变压器的输出电流波形几乎是完全一致的.那末有朋友要说了，这么看来，抱负变压器和非抱负变压器模子的表示好像是一样的，没有什么区别呀？下面我们来继续探讨.抱负变压器和非抱负变压器的一个重要的区别就是，抱负变压器不会饱和，而非抱负变压器会饱和.怎么样才干让变压器饱和呢？假如给变压器的初级施加一个直流电压信号，时间长了，励磁电流越来越大，变压器就会饱和.我们来看看是不是这样的.
把输入的交流信号源换成一个0.5V的直流信号源:然后点击工具栏上的这个，再次开端时域扫描（没有改参数，和上次一样）.然后在如图中所示，R1、R5处放置电流探头.在仿真器界面下选择菜单的plot>Add plot to window，再如图在R3、R7、R9处放置电流探头，看看变压器原边电流在长时间施加直流电流会如何变更吧：
从图中可以看到，抱负变压器的初级电流线性上升.而非抱负变压器的初级电流在大约0.76ms的地方开端急剧上升.是
不是变压器在这个地方饱和了呢？我们来计算一下.先看一
下磁芯和材质的参数：根据这些已知参数，依照电磁感应定律：
说明从时间零点开端到0.76毫秒处的磁感应强度增量为
467mT.而根据上表中，3C90材质的饱和点约为470mT，说明采取了ER28L_3C90磁芯的几个非抱负变压器在这个地方的确开端进入饱和状态了！而抱负变压器的初级电流只是线性上升，不会进入饱和！
懂得了这些模子的特性，你便可以依照实际的需要选择合适的模子停止仿真.不管是抱负化的耦合电感模子，还是非抱负的模子，K系列的模子可以支持有6个绕组的耦合电感或变压器.通常足够我们使用了.。