simulation schematic simulation schematic and graphic
Transcription
simulation schematic simulation schematic and graphic
SIMULATION SCHEMATIC AND GRAPHIC REPRESENTATION OF HUMAN HUMAN BOBY MODEL ESD Oana Cristina BENIUGĂ Gheorghe Asachi Technical University of Iaşi Oana Maria NEACŞU Gheorghe Asachi Technical University of Iaşi REZUMAT. În această lucrare este descrisă topologia topologia celui mai întâlnit model pentru simularea descărcării electrostatice. Optimizarea circuitului ce reproduce corpul uman este o abordare în plină evoluţie ce implică analiza deteriorărilor echipamentelor electronice. Articolul propune modelarea evenimentului evenimentului de descărcare electrostatică generată de modelul corpului uman utilizând un model realistic ce include toate elementele de circuit ce pot fi implicate în proces. Reprezentarea Reprezentarea schematică a modelului corpului uman ca generator de sarcini electrice nedorite se face utilizând tehnici avansate de instrumentaţie viruală. Simularea în programul SPICE, folosind o arhitectură specifică, permite înţelegerea comportamentului dispozitivelor supuse testării sub acţiunea diverşilor stimuli externi, în speţă corpul corpul uman. Testele au condus la concluzia că descărcarea electrostatică provenită de la corpului uman produce nivele ridicate ale curenţilor de descărcare putând cauza deteriorări serioase ale echipamentelor electronice. Cuvinte cheie: cheie descărcare electrostatică, instrumentaţie virtuală, modelul corpului uman, simulare SPICE, ABSTRACT. In this workpiece is described the topology of the most common model for electrostatic discharge simulation. The HBM model optimization is a evolutionary approach that involves invo lves electronic equipment failure analysis. The present article proposes modeling the HBM ESD event using a realistic model, including all the circuit elements that may be involved in the process. Schematic representation of human body model as electrical charges generator is realized using advanced virtual instrumentation techniques. SPICE simulation, using specific architecture, allows understanding the behavior of devices under test at different external stimulus, in this cause the human body. The test concluded c oncluded that the electrostatic discharge from the human body produces high levels levels of discharge currents being able to cause serious damages of electronic equipments. Keywords: Keywords electrostatic discharge, virtual instrumentation, human body model, SPICE simulation, 1. ISSUES RELATED TO ELECTROSTATIC DISCHARGES Electrostatic discharges (ESD) are a serious reliability issue in electronic environment design and manufacturing. Nowadays, ESD are mainly addressed in standard electrical engineering and the increases in performance and design leads to very small electronic components. In this situation, the thinner regions cannot withstand the variable voltages and discharge currents produced by ESD phenomena. The sources of ESD produce large amounts of charge and the combination between sensitive ESD devices and no protection to charge accumulation increases the incidence of damages and failures. As the electronic technology advances are needed extra precautions and discharge current and fields’ limitation. Those limitations involves circuit protection by improving the self protection of integrate circuits or by adding elements for charge or voltage rerouting. The voltage produced during discharge describes the charge’s forcing conditions and the current produced is described by the speed the charge moves. The ESD event is a two body system, as one charged body comes into contact with an uncharged body and so appears a charge imbalance. The produced current pulse is characterized by two body’s capacitance, the initial voltage differences and the impedance between them during the phenomenon. In order to prevent, limit or estimate the electrostatic discharge, in time were created and standardized different types of event’s equivalent system. _____________________________________________________________________________________________ Buletinul AGIR nr. 4/2011 ● octombrie-decembrie 171 2. GRAPHIC REPRESENTATION OF HUMAN BODY MODEL In the electronic and information & communication technology are standardized three basic models implicated in the ESD event. The models are based on the localization of charge storage and grouped as: a. Human Body Model (HBM); b. Machine Model (MM); c. Charged Device Model (CDM). Under certain conditions the electrical charged human body can transfer his charge to electronic devices by simply handling, operating or assembling them. As the basic immunity test method for personnel electrostatic discharge is IEC 61000-4-2, for the simulation of electrostatic effects – HBM, MM, CDM are adopted the standards EN 61340-3-1, 3-2, 33:2007. The standards covers HBM, MM, CDM ESD waveforms for use in general test methods and for applications to materials or objects, electronic components and other items for ESD withstand test or performance evaluation purposes. The HBM is the most popular ESD model and models the ESD event coming from a person that enters into direct contact with an electronic component or circuit (device under test), represented in figure 1: Fig. 1. Human body – DUT electric discharge The schematic representation of HBM described in EN 61340 -3-1:2007 is illustrated in figure 2: This figure shows the real case HBM ESD event, the static charges being initially stocked in the human body and then transferred to the device under test by finger contact or through a metal tool in contact with the human hand. The accumulation of charges into the test equipment can produce very high voltage or current that can result into irreversible failure or even integral destroy of the device. HBM ESD Generator 100pF/1.5 kΩ switch Device under test R 500 Ω Current transducers Fig. 2. HBM ESD waveform generator The current transducers measure the current resulted from the discharge either through a shorting wire, either through a 500 Ω low inductance resistor, with a tolerance of ± 1% appropriately rated to the voltages that will be used for waveform qualification. The ESD event main failure reasons are related to the discharge current that affects the devices under test functionality. The discharge current is characterized by the rise time, the peak to peak value, the current at 30 ns and at 60 ns. The current transducer introduces a parasitic inductance into the ESD discharge path, which results in slower current rise times. In a ESD event the human body can generate very high voltage levels as 5,000 volts by simply walking across a linoleum floor, 15,000 volts by a carpeted floor, while a human body wearing nylon clothes can supply 21,000 volts. Walking across a carpet with leather shoes and low-humidity can raise the human body capacitance charge up to 25,000 volts. The actual amount of energy in a given ESD event depends on the types of materials involved (wool fabrics generate less than nylon), the humidity (low humidity offers less resistance to the discharge), the amount of physical energy (friction) involved, and how quickly the energy is released. For example, a human body that carries a static charge of 30,000 volts stores an electrical energy about 0,045 joules. In order to model the ESD process are used equations describing the physical process that simulates the response to external stimulus. The mathematical simulation allows understanding the behavior at different stimulus. _____________________________________________________________________________________________ 172 Buletinul AGIR nr. 4/2011 ● octombrie-decembrie 3. HUMAN BODY MODEL SIMULATION APPROACH There are a number of factors that influence the human body model waveform creating a discrepancy between the reality and the standard specifications of this ESD model: - the non uniform ESD environment conditions; - the circumstances of electrostatic discharge that can occur and can not be predicted; - the different configurations of human body (different charges and circuit parameters); - lack of unstandardized procedure or method for capturing the ESD event. Commonly, there are various software solutions that can approximate the ESD spark and the current produced. HBM ESD simulation process takes into account certain considerations. First, the ESD simulation on electrical circuit level must forecast all the entrance data that allows reproducing the real event. Then, the electrical components must be provided with protection design so that those can be unaffected by the electrostatic discharge. The actual ESD standards must deliver opportunities for testing the immunity of electronic components under the action of direct or indirect discharge from human body. Table 1. Approximate values for different sections of human body Fingers holding key Entire hand holding key Forearm Full arm Torso Whole body Diameter (cm) 2 Lenght (cm) 6 C (pF) 2 L (µH) 0.02 7.5 12.5 5 0.02 9 9 30 30 30 60 60 120 10 20 20 60 0.1 0.27 0.13 0.43 The simulation will generate a lot of data that must be graphical estimated and interpreted by specific software and which will help in predicting the levels of break-down, according to the discharge current amplitude. According to the IEEE standard C62.47-1992, the table 1 presents the electrical parameters of human body, calculated based on spherical or cylindrical approximations of the human body segments involved in an electrostatic discharge. The inductances and capacitances of different sections of human body are dependent on their geometry, while their resistance is dependent on various nongeometric factors. For different types of electrostatic discharge, like body/finger or hand/metal, the bulk resistance of body sections is in series with skin resistance, the last one being believed that is influenced not only by humidity but also by skin secretions chemistry. Because the effective resistance can not be accurately estimated, the best way to determine this parameter is to computer model using advanced virtual instrumentation techniques and then to make comparison with real ESD event. In the scientific literature are data that the body/metal electrostatic discharge total resistance is in the range 300 ÷ 1500 Ω and the total capacity is around 100 pF. Typically, the human body model failure modes consists in visible thermal damages or integrate circuit error, as voltage flow which discharges a lot amount of current into the electronic component in a short time period of several hundred nanoseconds. 4. SPICE APPLICATION EXEMPLE The basis for the most circuit’s simulation and modeling today is the program SPICE (Simulation Program with Integrated Circuit Emphasis) and allows connecting linear and nonlinear circuit elements in order to simulate their time or frequency behavior. In the realistic human body model simulation, the resistors will keep their resistance the same, independent of current flowing through them or the voltages across them. The temperature is modeled only to define the ambient temperature at the start of the simulation. A real resistor will have non linear characteristics under high bias conditions, so the results do not model the physical circuit accurately if the current through the resistor causes non linear behavior. A potential way to make the analysis of electrostatic discharge from the human body more accurate is parametric modeling using SPICE software, because this program is well suited to model the electrical circuits. _____________________________________________________________________________________________ Buletinul AGIR nr. 4/2011 ● octombrie-decembrie 173 L2 L3 R2 150nH Arm Torso Legs L4 R3 25 50nH L5 L7 R4 15 40nH Hand – metal object Hand L8 R5 50 50nH R8 50 15nH R1 25 80 3nH 0 1 U1 2 R6 30 R7 V1 L1 150nH L6 15nH 250 C2 30pF C3 30pF C4 5pF C5 4pF C6 5pF Value = 500 1.5pF Load Hand – metal impedance C7 C1 40pF I C9 C8 3pF 10pF 0 0 0 0 0 0 0 0 0 Fig. 3. HBM ESD schematic circuit with VHBM= 25,000 Volts In figure 3 is presented the schematic representation of human body equivalent circuit, modeled in SPICE. There were configured and approximated the RLC (resistance-inductancecapacitance) elements that define the different sections of the human body, holding a metal object. The tests were performed with a very high voltage charge of 25,000 volts. The source simulates a transient pulse generator, as the real ESD event happens. The output load take into account has a value of 500 Ω. Moreover, the analysis allows estimating the current and voltage flow through each circuit element, from source until the output marker. In figure 5 is illustrated the graphical representation of the discharge current in time distribution. As it can be seen, comparing the simulated waveform of the human body discharge from this figure with the waveform presented in the EN 61340-3-1:2007 standard (figure 4), can be noticed the time representation of discharge current waveforms similitude. Fig. 4. EN 61340-3-1-2: 2007 - Current waveform for HBM ESD Fig. 5. Discharge current waveform for HBM ESD, VHBM=8kV The peak value of discharge current I (R9), at the output of human body circuit, before the output load of 500 Ω is approximately 0.75 A in the nanaseconds range. Fig. 6. Discharge current waveform for HBM ESD, VHBM=25kV _____________________________________________________________________________________________ 174 Buletinul AGIR nr. 4/2011 ● octombrie-decembrie Interior shielded wall Exterior shielded wall Connector BNC Voltage source S2 S1 To Coulomb model 6517A, R = 50 kΩ S3 Ceramic rack 10 mm Rubber ground carpet Fig.7. Experimental configuration for HBM indirect measurement If the charge voltage is 25 kV the resulted discharge current peak value is about 12 A. The human body from figure 1, isolated from ground is actually an insulated conductor that has a certain capacitance. This capacitance can be measured through a indirect method. In figure 6 is presented a configuration method for estimating the human body capacitance using an electrometer. The operator is placed inside a Faraday cage (which has the role to remove the disturbing fields), on a ceramic rack, on a rubber carpet. As it is well described in another previous work the indirect method is based on charging with voltage the human capacitor whose value is unknown and then determine his value by placing him in parallel with a precisely know capacitor, the measurements being realized using coulomb function of a programmable electrometer. For charging the human capacitor it was used a build-in programmable source, applying d.c. voltage to the human operator, maintaining the switches S1 and S3 open and the S2 closed. The electrostatic charge was estimated using the formula (1), where UK is the well known voltage from the build-in source. CH = Q UK (1) In a previous paper it was determined that the human capacity is about 300 pF (potential capacity and isolated sphere capacity), different from the standard 100 pF capacitor, illustrated in figure 1. 5. CONCLUSION The present approach enables better evaluation of the electrical discharges from the human body. In this paper was investigated the discharge current generated by the electrostatic human body model, because the electronic components from the electronic technology are very sensitive to this parameter. For human body model simulation was used a virtual instrumentation program that allowed configure and graphic visualizing the discharge current waveform. Moreover, it could be estimated and graphic represented the currents and voltages through each circuit component. The realistic system was test for two different charge voltages, at 8 kΩ and 25 kΩ, the resulting discharge currents being able to cause damages or break-downs to the electronic components which they came in contact. Comparing to other human body model tested circuits, can be concluded that the peak to peak voltage (current) vary with their geometry. The human body capacitance determined through calculations is around 300 pF, and is different from the standard value of 100 pF, which takes into consideration only the capacitance by a completely isolated sphere, neglecting the potential capacity of human body. The high transient current in an ESD event can lead to real reliability problems and the tests using HBM simulation can successfully recreate a real event. _____________________________________________________________________________________________ Buletinul AGIR nr. 4/2011 ● octombrie-decembrie 175 ACKNOWLEDGEMENT This paper was supported by the project PERFORM-ERA "Postdoctoral Performance for Integration in the European Research Area" (ID57649), financed by the European Social Fund and the Romanian Government. REFERENCES [1] E. Franell, S. Drueen, H. Gossner, and D. SchmittLandsiedel, ESD full chip simulation: HBM and CDM requirements and simulation approach Advances in Radio Science, Volume 6, 2008, pp.245-251 [2] Ming-Douker, Jeng-Jie Peng, Hsin-chin Jaing, ESD Test Methods on Integrated Circuits: an Overview, IEEE 2001 ICECS, pg. 1011 - 1014 vol.2 [3] A. Sălceanu, Oana Neacşu, E. Luncă, V.David, Indirect Measurements on the Capacity in the Electrostatic HB Model, 2007, 15-th IMEKO TC 4 International Symposium on Novelties in Electrical Measurements and Instrumentation, Vol. I, pag. 38-41, ISBN 978-973-667-261-3 [4] M.A. Kelly, G.E. Servais and T.V. Pfaffenbach, An Investigation of Human Body Electrostatic Discharge, ISTFA ’93: The 19th International Symposium for Testing & Failure Analysis, Los Angeles, California, USA/15–19 November 1993 [5] EN 61340-3-1:2007 Electrostatics — Part 3-1: Methods for simulation of electrostatic effects — Human body model (HBM) electrostatic discharge test waveforms [6] IEEE Std C62.47-1992 - Guide on Electrostatic Discharge (ESD): Characterization of the ESD Environment _____________________________________________________________________________________________ 176 Buletinul AGIR nr. 4/2011 ● octombrie-decembrie