Analysis of distributed coupling along nonparallel traces using PEEC with phase term expansions

Michael A. Cracraft, James L. Drewniak
2008 2008 IEEE International Symposium on Electromagnetic Compatibility  
When evaluating electromagnetic modeling software, there is usually a significant focus on the "accuracy" of the software. Differences between the results generated by the software and the "correct" solution are the result of several potential sources of error. Of those potential error sources, differences between what the modeler wants to analyze and what the software is actually modeling is usually the greatest source of error even when the analysis is being done by experienced modelers. In
more » ... is paper, three full-wave time-domain EM modeling codes are evaluated by analyzing three simple canonical problems. These codes employ the three most common time-domain modeling techniques: Transmission Line Matrix method (TLM), the Finite Integration Technique (FIT) and the Finite Difference Time Domain technique (FDTD). The three canonical problems are a center-driven dipole, a circuit board power-bus structure, and a power-bus structure with a cable attached. Electrically large problems require full-wave calculations that can be handled using PEEC using either static phase approximations, phase term expansions, or the dynamic Green's function directly. When distributed coupling is significant between traces, parallel or nonparallel, a quasi-static approximation will return incorrect results in the coupling terms. Nonparallel coupled lines pose a problem that is not easily solved by transmission lines, but may in some cases be analyzed by lumped element models or quasi-static models when near-end coupling dominates. The statistics of the fields scattered by nonlinear equipment in a reverberation chamber are investigated using both a Monte-Carlo simulation of a simple statistical model and measurements of real equipment. The probability distributions of the scattered fields of the second and third harmonic are predicted and compared to measurement. The implications of the results for the immunity assessment of digital hardware and harmonic cross-section measurements in reverberation chambers are discussed. A multiconductor transmission line (MTL) model for the time domain considering losses, incident fields and skineffects for the modeling language VHDL-AMS was developed. The model is based on the FDTD method. It can be extended by various VHDL-AMS circuit files and used for linear and non-linear time domain simulations. Modern standardized modeling languages like VHDL-AMS have the important advantage that growing model libraries permit fast creation of large simulation models. Exchange of models and extension of models is easily possible. After a short introduction to VHDL-AMS and presentation of the MTL and FDTD theory used for modeling, examples are shown. The developed model is compared to calculation results gained with other simulation methods. The validity of the implemented VHDL-AMS model is proved. A hybrid method is proposed for radio frequency interference (RFI) prediction of a metal enclosure with an aperture on the top wall. The structure is divided into several segments. While the fields in rectangular segments are described by cavity model, the segments with apertures are modeled by the commercial finite element solver (HFSS). Tangential field continuities along the common boundaries of different segments are enforced by the voltages and currents of boundary ports. Good agreement has been achieved between the hybrid method and full wave simulation. The Discontinuous Galerkin Finite-Element Time-Domain method is presented. The method is based on a high-order finite element discretization of Maxwell's time-dependent curl equations. The mesh is decomposed into contiguous sub-domains of finite-elements with independent function expansions. The fields are coupled across the sub-domain boundaries by enforcing the tangential field continuity. This leads to a locally implicit, globally explicit difference operator that provides an efficient high-order accurate time-dependent solution. An efficient implementation of the perfectly matched layer media boundary truncation is also presented that allows general tetrahedral meshing through the PML region. Correlation between radiated emission from chassis with PCB and the junction current which is the current flowing through screws connecting PCB GND to chassis GND was investigated, and new technique to reduce the junction current was proposed. The measurement results of the junction current frequency spectra showed strong correlation with the emission from chassis with PCB. Also investigation using meshed LCR network SPICE model of PCB and chassis was performed. The calculation results of the junction current showed good correlation with measurement results for frequency spectrum and it suggested that the closer to screw bypass capacitor is placed, the lesser current flows through junction. That new concept was validated by actual measurement and the results of radiated emission showed 19 dB improvements at 320 MHz. Effective decoupling is crucial for the optimum performance of the power distribution network in an electronic system. As component packaging technologies evolve enabling tighter integration and faster operation of electronic systems, it is important to develop better decoupling strategies. This paper describes several new or proposed packaging structures and evaluates the connection inductance associated with possible decoupling capacitor locations. As expected, connections made on the chip tend to have a lower inductance than connections made on the package; and connections made on the package tend to have a lower inductance than connections made to the board. This illustrates the importance of providing decoupling capacitance as close to the chip as possible in order to maximize the effective bandwidth of the power distribution network. This paper proposes a technique for the fast evaluation of antenna-induced noise at the terminations of PCB traces on compact mobile phones platforms. The main approach is based on the classical theory of transmission lines with external field excitation. We show that a single full-wave electromagnetic simulation is needed to compute the coupling coefficients for an arbitrary location of the trace on the phone PCB. Therefore, the approach is ideally suited for the automated incorporation of EMI constraints within routing and placement algorithms and for parametric and what-if analyses. A model to estimate the minimum voltage between a heatsink and a PCB that is required to generate a given radiated electromagnetic field strength from a PCB with attached cables is introduced. The model is based on a previously introduced technique for calculating the maximum possible radiated field due to a known voltage driving a heatsink against a circuit board with attached cables. A closed-form expression is derived that can be used to determine if a measured voltage on a heat sink is capable of generating a field strong enough to exceed an FCC or CISPR radiated emissions limit. A common-mode antenna model designed specifically for estimating the emissions from PCBs very quickly is composed of an excitation source and an antenna. This paper provides the experimental validation of an imbalance difference model of common-mode excitation. In the model, the excitation source is defined by the product of the difference of a current division factor related to the cross-sectional structure of the transmission line, with the voltage between signal line and return plane. Through an evaluation of radiation suppression due to a guard trace placed to narrow return plane, imbalance difference model is found to be generalized to the multi-conductor system and provide helpful information for noise suppression in PCB design. The WG2 of CISPR/SC-I is discussing a reference test method for measurement of conducted disturbance voltage at mains port. A statistical approach seems to provide reasonable evaluation results to determine that multiple test methods are alternative methods. In the study reported here, we obtained data from measurements of the conducted disturbance voltage at mains port, taken with two different test arrangements, The data recently gathered by VLAC (Voluntary EMC Laboratory Accreditation Center, Inc.) in two proficiency testing (PT) programs; we processed the data statistically and evaluated the results. From the results, we are able to conclude that measurements taken with two test arrangements could be used as alternative test methods. In situ testing of large machines: alternative methods for conducted emission measurement J. A. Catrysse, Khbo, Oostende, Belgium The new European EMC Directive 2004/108 allows for alternative validation methods for large machinery. Following the related harmonized European standard EN50370-1, subassembly testing combined with a final monitoring of the machine may show proof of compliance. A related European Research Project TEMCA2 was developping new, alternative test methods that can easily be used for this final control of large machines, by in-situ testing. This paper is dealing with the description and validation of the methods for conducted emission. In situ testing of large machines: alternative method for radiated emission measurement J. A. Catrysse, Khbo, Oostende, Belgium The new European EMC Directive 2004/108 allows for alternative validation methods for large machinery. Following the related harmonized European standard EN50370-1, subassembly testing combined with a final monitoring of the machine may show proof of compliance. A related European Research Project TEMCA2 was developing new, alternative test methods, that can easily be used for this final control of large machines, by in-situ testing. This paper is dealing with the description and validation of the methods for radiated emission. This paper discusses the procedure to convert total-radiated power measurements data to equivalent electrical field data, which might be get on a standardized test site. The algorithm in use is based on two parameters, which are the geometrical factor gmax and the directivity of equipment-under-test (EuT) Dmax. Both parameters are unknown for unintentional radiators. The aim of this paper is to collect the published information to predict these parameters and to discuss the influencing factors to get a useful approximation for gmax and Dmax. Especially the prediction of Dmax depends on the effective size of the EuT described by the radius of a sphere containing the EuT. In radiated emission measurements an error is introduced by the directive receive antenna. The Monte Carlo Method was used to calculate this error where measured antenna pattern had been taken into account. Although there are large differences between the classical test set-up and the two improvements antenna tilting and antenna bore sighting the impact to the measurement uncertainty is low. For a test distance of 3 m the measurement uncertainty can be reduced from 5.62 dB to 5.43 dB with antenna tilting. With a bore sight antenna tower a reduction to 5.31 dB is feasible. The bias and uncertainty given by CISPR 16-4-2 are not adequate. New radiated RF immunity/susceptibility test method using RF-pulsed rotating-EM field A new radiated radio-frequency (RF) immunity/susceptibility test method using an RF-pulsed rotatingelectromagnetic (EM) field is proposed. The RF-pulsed signal that has an arbitrary bandwidth can be generated by controlling the pulse width. Controlling the pulse width can generate the RF-pulsed signal having an arbitrary bandwidth. By using such RF-pulsed rotating-EM field as an incident-EM field of the immunity/susceptibility test, it is possible to easily investigate the immunity-weak points of electronic equipments. In this paper, the basic characteristics of the RF-pulsed rotating-EM field are clarified. Moreover, some susceptibility characteristics of a cavity with a slit measured by using the RF-pulsed rotating-EM field are shown as an example. Analyzing interference problems in vehicle wiring harnesses requires fast and accurate methods of approximating crosstalk. Worst-case approximations using lumped element models are fast and easy to use, but run the risk of overestimating problems. Statistical methods that account for the random variation of wire position help prevent overdesign, but are often difficult and time-consuming to apply and lack a clear link between problems and their cause. Here we investigate the use of simple lumped-element models to predict the statistical variation of crosstalk in wire harness bundles. Models are based on lumped-element approximations, where inductance and capacitance values are calculated for a single bundle cross-section, and only the circuit position is varied. Accuracy was evaluated by comparing results to numerical simulations. The method does a good job of quickly predicting the reasonable worstcase values of crosstalk due to inductive or capacitive coupling. One of the big challenges for the automobile and vehicle antenna manufacturers remains designing the vehicle active antennas for weak signal performance along with meeting the requirements of vehicle radiated immunity tests according to ISO 11451-2 and ISO 11452-2. The past vehicle testing have shown that some vehicle active antenna systems included during those tests had performance issues caused by the vehicle body sizes. Component radiated immunity tests of ISO 11452-2 challenges active antenna system designs. In order to reach effectively the vehicle performance goal, a deep knowledge about the design of the vehicle active antennas is required. There are three major topics this paper addresses: 1) Clarifying the technical parameters of the vehicle antenna designs; 2) Characterizing parameters useful for the "Electromagnetic Compatibility" (EMC) in order to pass the radiated immunity tests; and finally 3) Identifying the causes of the antenna amplifier distortion In this paper a methodology for taking into account the frequency dependence of a possible referencing (grounding) techniques in LTCC boards for satellite applications is presented. The aim of this work is to analyze and compare the input impedance and S-parameters values for different strategies of inter connection of the reference planes among them. The numerical analysis presented has proven the importance of the lateral metallization set up in the LTCC board working into a metal housing. For board used stand alone the lateral metallization can be replaced by an adequate density of vias stitching among the planes. The modeling agrees well with real grounding solution. A planar screen made of a periodic array of small resonant particles showing a strong magnetic resonance is studied in detail. Two types of inclusions are considered: the double split-ring resonator (SRR) and the spiral resonator (SR). The homogenized screen is shown to behave as a homogeneous slab with an effective magnetic permeability having large positive or negative values around the magnetic resonance of the particle, giving rise to selective shielding properties. The shielding performance of the screen are investigated, and the effects of several geometric parameters are carefully evaluated through an equivalent-circuit representation of the single particle and full-wave simulations. Abstract With the increasing numbers of transistors in today's die as well as incorporating mixed technology (i.e. analog, RF, digital, and power), large scale devices with their large switching currents create internal voltage drops due to the package design (i.e. wire bond or flip-chip) that can be a source of common mode noise. The normal lumped components, the printed circuit boards, the harnessing, and the packaging are well known contributors and there are various measurement techniques to identify these noise sources. However, what is lacking are some measurement techniques applicable specifically for determining if a large scale integrated circuit type devices could be a noise source. This paper will look at a couple of proposed techniques specifically for the chip level. The IEC 61967-2 TEM cell standard allows for variations in test parameters which may cause variations in the measured emissions from integrated circuits (ICs). To test the impact of these parameters, two printed circuit boards were designed within the IEC standard using "poor" PCB design strategies and using "good" design strategies. Emissions from three pin-for-pin compatible 8051 microcontrollers were tested. Emissions were measured using both PCBs, changing the PCB configuration, and changing test parameters like the program running on the IC, the rise time of the input clock, and I/O switching. Emissions from the "poor" PCB were about 3-8 dB higher than emissions from the "good" PCB. A change in the program run by the IC, the clock rise-time, and I/O caused a 4-15 dB change in emissions. Emissions differed considerably among the ICs. Possible causes for variations in emissions with the test parameters are discussed. This paper details the simulation and measurement of low frequency (LF) magnetic field shielding by open surface structures. The approach of this paper is to develop equivalent electric sheet impedance boundary conditions for use by LF surface based electric field integral equation (EFIE) solvers. From these sheet impedance forms, a figure of merit is developed to determine the boundary condition applicability. Next, transmitting and receiving coil designs are analyzed and modifications are made to eliminate shield currents in measurement. Finally, simulated and measured magnetic field shielding is compared for electrically small open surface material disks. The paper focuses the attention on the reverberation chamber method for shielding properties evaluation of equipment-level enclosures. The whole testing situation, chamber plus enclosure under test, is numerically modeled by a in-house FDTD code able to predict the field inside the enclosure and the voltage captured by probes placed inside it. After the code validation, the effect of the probe position, on the enclosure shielding effectiveness, is examined. In particular a field normalization is introduced in the shielding effectiveness definition. Finally, the effect of a frequency stirring application is also investigated. An equivalent circuit method is used to define ports at the slots of enclosures to separate the external field domain from the internal over a broad frequency range. The aspect ratios of internal to external structures of electronic devices are usually difficult to handle in joint domain simulations. Our method provides the opportunity to simulate the domains separately and to connect the two models afterwards. We present an example for a cubical enclosure with a slot. We use a cavity model and an impedance network at the slot to consider the radiation loss. The results are compared with HFSS simulations.
doi:10.1109/isemc.2008.4652106 fatcat:hua5e7gzlvhq5ng6hrt46h4t2q