A simple method for evaluating the transient thermal response of semiconductor devices

In this paper, a simple experimental method for the transient thermal characterisation of semiconductor packages is presented. The method is based upon the assumption that in many cases, device temperature evolution can be accurately described by a few exponential terms, as will be shown to be the case when transient thermal response has widely separated time constants.The proposed method, begins with the experimental determination of transient thermal response followed by numerical extraction of the time constants and amplitudes for the significant exponential terms, and is applied to a number of commercial Smartpack^® modules in order to obtain a dynamic model that can be used in circuit and numerical simulators, in order to predict the dynamic thermal behaviour of the device under any working condition.     

Evaluation of short pulse and short time thermal transient measurements

Thermal transient recording and the time constant spectrum analysis are widely used methods in the thermal testing and qualification of IC packages. A limitation of these methods is that recording of the complete transient response requires long time. This limitation, however, can be overcome by sophisticated procedures. The first method is to apply short power pulse for excitation; the second one is the interruption of the transients long before the thermal equilibrium is reached. The paper offers algorithms to evaluate these short pulse and short time measurements. The presented methods are suitable if the extraction of the little time constants is needed. This is the case if the transient method is used e.g. for die attach quality checking.     

THERMODEL: A Tool for Thermal Model Generation, and Application for MEMS

This paper presents a tool and a method for the generation of reduced order thermal models, in order to assure modeling the effect of the package on the thermal behavior of the packaged device. The method is generic, and can be based either on the simulated or on the measured thermal transient response of the real packages. It is based on the generation of the time constant density spectrum of the thermal response function, from which we automatically generate a reduced order thermal model in the form of an RC ladder network model. Beyond presenting the generic methodology experimental results are also presented, based both on the simulation and measurement of MEMS elements and packages.     

A new approach to the dynamic thermal modelling of semiconductor packages

This paper describes a novel approach to the dynamic thermal characterisation of semiconductor packages. Conceived as an extension of the variable angle model for multilayer structures, the thermal capacitance of each layer is evaluated at the same time and for the same volume used to calculate the thermal resistance, which guarantees the coherence. Because this volume or lump is a function of system geometry, materials and boundary conditions, the model can be said to include 3D aspects. The thermal resistances and capacitances obtained from the model, each pair representing a layer of the structure, are used as input to an electric circuit simulator to obtain the transient response of the package.Moreover, some common assumptions usually made when interpreting the transient thermal response of semiconductor packages are revised and discussed in an attempt to clarify the scene. Particularly, the log-of-time derivative of the step response is shown to give all the information about the dynamic response when the characteristic time constants of the system are widely separated. In that particular case, otherwise very common in practice, a novel analytical method for the obtention of the characteristic time constants and associated amplitudes is presented and their results evaluated for different industrial power packages.     

脉冲方波驱动强化热电制冷的瞬态特性

采用有限差分法对脉冲电压驱动下的瞬态热电效应及其动态特性过程进行了理论分析,探索了非稳态工况下帕尔帖效应、焦耳热效应与傅里叶导热效应之间的耦合关系及其关键制约因素对制冷性能的影响规律,进而探讨了脉冲驱动强化热电制冷性能的作用机理。分析结果得到,在合理电压域值内采用主动控制方法,对热电模块周期性施加数倍于稳态工况理想电压的脉冲突变电压,有益于充分利用帕尔贴制冷效应而推迟出现以焦耳热和傅里叶热耗散形式为主的内部热积聚对热电模块冷端引起的负效应,并能瞬态实现冷端面的制冷强化作用和最大程度实现输入电能的有效转换。该结论不仅为进一步提出脉冲驱动模式的优化控制策略提供了理论依据,也为瞬态热电制冷效应的应用开辟了一条新思路。     

A straightforward analytical method for extraction of semiconductor device transient thermal parameters

Most of the thermal characterization of packaged semiconductor devices is carried out in the time domain by measuring and recording the values of a temperature sensitive electrical parameter (TSEP) as a function of time, during the heating or cooling transient caused by the application of a given power function, generally a step, to the device. Although this time response contains, within measurement resolution, all the relevant information about device thermal behaviour, the “packed” form of this information is not so convenient for its immediate use when we want to obtain a thermal model for further simulation or extract some information on device structure. Real and imaginary parts of device frequency response are a better way of dealing with measurement results because its “unpacked” form leads easily to thermal model and structural information.In this paper we derive a direct analytical calculation of the frequency response from the scalar time domain response as well as an approximate method to obtain an equivalent thermal model.     

On the modeling of the transient thermal behavior of semiconductordevices

A mathematical model of the transient temperature response of integrated devices is presented which takes into account the three-dimensional (3-D) nature of heat flow and the physical structure of the device. Simple analytical relations for the transient thermal impedance and thermal time constants are derived for the first time. The impact of device geometry on the transient thermal response curve is discussed, and simple guidelines for the thermal design of solid-state devices operated in transient or pulsed regime are given     

非制冷红外焦平面阵列器件的热时间常数测试方法

热时间常数是基于微测辐射热计的非制冷红外探测器的关键指标参数,它与探测器的最高有效帧频直接相关,因此准确测量热时间常数对于器件设计和应用都有举足轻重的意义。但目前无论是探测器热时间常数的标称值还是基于单元的热时间常数现有方法的测试值,都无法建立与探测器的频率响应特性的直接定量函数关系,以确定探测器工作的最小帧间时间间隔。直接基于阵列器件测量热时间常数的方法,借助低于1/2帧频的斩波调制,通过变频时域采集,快速傅里叶变换（FFT）等常规测试手段,提取有效的电压响应信号,拟合频响曲线,能快速有效地提取热时间常数。通过实测分析,该测量方法具有准确度高、抗干扰能力强、稳定性高、测试用时短的特点,且均采用通用的测试仪器,无需单独制作测试样品,具有较高的推广价值。     

微测辐射热计的热响应时间测试方法

热响应时间是微测辐射热计的关键参数,它会制约非制冷红外探测器的最高工作帧频。热响应时间的像元级测试能够真实反映传感器的物理热响应时间,为产品设计优化提供及时、有效的数据支持,因此准确测量该参数具有十分重要的意义。但目前像元级测试方法均未能有效补偿微测辐射热计的自热效应,无法精准地测量热响应时间。基于频率响应法测试了微测辐射热计的有效热响应时间。通过用电阻温度系数对自热效应进行补偿,可以精确测量物理热响应时间。通过实验分析了不同偏置电流下测得的物理热响应时间。结果表明,该方法准确度高,稳定性强。     

脉冲工作状态下微波功率管瞬态特性研究

鉴于越来越多的领域要求微波功率管工作于脉冲应用状态,研究管子的瞬态温度特性就显得日益重要。假设热源为一无限大平面,且在芯片表面热源产生的热量只沿垂直于芯片表面一维地传递。从这一简化的理想热传导模型出发,对热传导方程离散化;在热沉端,定义了能够与实际情况相吻合的温度上升与下降时间常数来确定边界条件,从而得到了在脉冲应用情况下,芯片表面瞬态温度随时间变化的计算机模拟结果。通过在连续波工作下与红外显微镜实测温度的比较,验证了模拟程序的准确性,并且给出了最高结温随时间、脉宽、占空比的拟合公式。为微波脉冲功率晶体管的设计提供了育价值的参考。     

Analysis of the temperature rise in PIN diodes caused by microwave pulse dissipation

With the increasing use of PIN diodes in high power, microwave switching components, knowledge of the diode transient thermal response to such pulse dissipation becomes of increasing importance. Thermal models for the microwave dissipation within the diode are proposed for both the forward and reverse biased states. Based on these models, both the maximum and junction temperature step responses and pulse waveforms are determined for both bias states. For the forward bias state, the maximum temperature occurs near the center of the dissipation or I-region and for a step increase in dissipation, initially increases linearly with time then levels off to a constant difference with respect to the junctions. Under the same conditions, the junction temperature (relative to an appropriate heat sink) initially increases linearly also followed by a square root dependence to the steady-state value. The total response is therefore characterized by two fundamental thermal time constants and thermal resistances corresponding to the dissipative and conductive regions. Of major importance is the substantial calculated temperature differences which can occur between the center of the I-region and the junction where the temperature is normally experimentally monitored. For the reverse bias state, the maximum temperature occurs at the junctions because of the more localized dissipation and follows the square root dependence to the steady-state value. With knowledge of the diode thermal parameters, the relations given, and quantitative substantiation of the model proposed, the transient temperature response, together with its imposed component design limitations, are now obtainable.     

Characterization of solar cells by thermal transient testing

The current paper deals with the application of thermal transient testing as a characterization tool for solar modules. Based on the measurement of different samples (concentrator solar cell, single junction silicon solar cell) we prove the applicability of this measurement technique and address some specific issues of the characterization of solar cells by the thermal transient method.From the measurement metrics such as junction-to-base plate thermal resistance and thermal capacitance(s) can be derived and can serve as a basis of a multi domain solar cell model. The used technique also enables us to verify the quality of attachment layers in a solar module allowing fair quality control and reliability analysis of these devices. Finally a method is proposed to regain the data that is covered by the initial electric transient following the power step. This initial electric transient can be high in large surface devices like solar cells, and covers valuable data describing the structure near to the p–n junction. To eliminate this, simulated transients were fitted to the part of the actual measured thermal transient where the electric transient already decayed. This way the part of the thermal transient that was covered by the electric transient can be reconstructed.     

On the determination of minority carrier lifetime from the transient response of an MOS capacitor

Minority carrier lifetime greater than 10^-10seconds can be determined for silicon at room temperature by observing the transient response of the MOS capacitance after the application of a large depleting voltage. The waveform is exponential for heavily doped samples and nonexponential for lightly doped samples. The transition occurs when the oxide capacitance approximately equals the space-charge capacitance. Results are presented for a lightly doped bulk silicon wafer exhibiting an effective minority carrier lifetime of 7.6 microseconds and a thin silicon-on-sapphire film with a lifetime of 4.5 nanoseconds. The ratio of transient time constant to lifetime is typically 10⁵-10⁸at room temperature. Lower lifetime may be determined by cooling the sample. A graphical method is presented to rapidly extract lifetime from the transient response of lightly doped samples when the waveform is nonexponential.     

On the various time constants of wavelength changes of a DFB laserunder direct modulation

The temporal behavior of the optical frequency emitted by several DFB lasers under direct square wave modulation was measured using an all-fiber implementation of a Mach-Zender interferometer with an imbalance of 30 ps. The impulse response of the optical frequency to injection current modulation was found to contain a time constant as short as 10-20 ns, together with a few longer ones. The existence of such a short time constant is consistent with a thermal analysis of a laser structure with finite thermal impedance of the active region and should be taken into consideration in various wide bandwidth applications of direct modulated semiconductor lasers     

In-situ analysis of thermal properties of polymer composites by embedded LED temperature sensor

Real time and accurate measurement of thermal conductivity of polymer composites with thermal conductive fillers challenges researchers in industrial application. Here, we present an in-situ measurement approach by embedding a LED or diode as a combined heat source and temperature sensor into the filled polymer and using the well-established transient measuring method based on forward voltage variation to determine the temperature response of the sensor in polymer. Numerical model fitting is applied to estimate the thermal conductivity of the polymer composites with different filler/polymer ratios. These findings are compared with other thermal conductivity test methods such as the laser flash method and the Modular Differential Scanning Calorimeter (MDSC). The proposed approach provides a quick way of measuring the thermal conductivity in relatively thin polymer composites and agrees well with the MDSC method. Another advantage is that it can work with the real samples made for the application in mind, so its results can be used directly.     

Thermal resistance evaluation of high-speed VCSELs: an isothermaloptical transient technique

A new method for the estimation of the overall thermal resistance of high-speed vertical-cavity surface-emitting lasers (VCSELs) is presented. The method is based on an isothermal optical transient (IOT) approach heavily relying on the following two basic features: 1) the high-speed VCSELs have an optical-electrical transient response much faster than their thermal response, and 2) each operating point, defined by the driving current, the emitted optical power, and the internal temperature of the VCSEL is unique. As a proof of concept, the IOT procedure has been simulated using accurate VCSEL models. These simulations have clearly demonstrated the potential of the IOT method to achieve accurate estimates of the overall thermal resistance of VCSELs     

Static and Dynamic Error Vector Magnitude Behavior of 2.4-GHz Power Amplifier

This paper presents error vector magnitude (EVM) behavior under two different modes of a reference voltage (V_ref); one is a static EVM with a constant V_ref and the other is a dynamic EVM with a pulsed Vref. A 2.4-GHz two-stage power amplifier (PA) is implemented by GaAs HBT technology to investigate EVM behavior for IEEE 802.11 g wireless local area network application. A square wave pulse with the duty cycle of 50% is applied as V_ref for dynamic operation mode. In this paper, it is shown that a dynamic EVM is worse than a static EVM because a transient response of PA output distorts preamble and signal in orthogonal frequency division multiplexing signal bursts. This transient response is explained by a two-pole thermal network in this study, and transient response experiments with a resistor-capacitor (R-C) network at a power supply were performed for verification     

Application of the TLM method to transient thermal simulation ofmicrowave power transistors

The transmission line matrix (TLM) explicit method of numerical simulation has been used to model the transient thermal properties of various microwave heterojunction bipolar transistor (HBT's) power structures, used in a pulsed mode. Control of the time step during the simulation is of paramount importance and the paper outlines some of the problems encountered using time step control methods currently published and describes an improved algorithm. This improved time step control method has been implemented in a general purpose 3D TLM transient thermal simulator. Some simulation results are described for a variety HBT transistor structures with very different thermal time constants