IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 55, NO. 1, FEBRUARY2006 257 The New Method of Determining Characteristics of Elements for Overvoltage Protection of Low-Voltage System Predrag Osmokrovic, Boris Loncar, and Srboljub Stankovic Abstract—This paper presents an innovative method for an efficient characterization of relevant characteristics of nonlinear overvoltage protection elements in low-voltage applications. Stan- dard measuring equipment was modified to enable an efficient and repeatable experimental procedure in order to investigate characteristics of overvoltage diodes and varistors, in particular volt–ampere characteristic, volt–ohm characteristic, coefficient of nonlinearity, and dissipating energy. Furthermore, the pseudoem- piric method using the “area law” has been developed to determine a pulse characteristic of a gas-filled surge arresters (GFSA). An innovative aging estimate algorithm was additionally used. The justification of the suggested method on commercial nonlinear overvoltage components was checked. Its protective characteris- tics were examined by both the classical approach and the new sug- gested method. The obtained results from both methods showed minimal discrepancies. Index Terms—Analytical algorithm, F test, overvoltage protec- tion elements, pulse-shape characteristic, U test. I. I NTRODUCTION T HE CONTINUOUS miniaturization trend in the elec- tronic industry inherently causes increased sensitivity of overvoltages and effort in protecting delicate electronic compo- nents. The latest generations of integrated circuits have internal layers of thickness of just several nanometers; thus, relatively small overvoltages may cause damage or even complete de- struction of sensitive electronic components. Additionally, in a microprocessor-based electronic system, overvoltages can cause resetting, temporary disturbance, or internal data cor- ruption. Therefore, sensitive electronic components should be adequately protected against harmful overvoltages (either as a part of initial electronic component design or using an add- on overvoltage protection). By placing adequate overvoltage suppressor at the entrance of the electronic device, the best level of protection is typically achieved [1]. The optimal overvoltage Manuscript received June 2, 2004; revised September 21, 2005. This work was supported by the Ministry of Science and Environmental Protection of the Republic Serbia under Projects 2006 and 2016. P. Osmokrovic is with the Faculty of Electrical Engineering, University of Belgrade, Belgrade 11000, Serbia and Montenegro (e-mail: opredrag@ verat.net). B. Loncar is with the Faculty of Technology and Metallurgy, University of Belgrade, Belgrade 11120, Serbia and Montenegro (e-mail: bloncar@ EUnet.yu). S. Stankovic is with the Institute of Nuclear Sciences “Vinca,” Belgrade 11001, Serbia and Montenegro (e-mail: srbas@rt270.vin.bg.ac.yu). Digital Object Identifier 10.1109/TIM.2005.862023 solution [2], [3] provides an economical solution to the follow- ing requirements: 1) It should limit input voltage to a nondangerous level. 2) It should be well below the current limit for a chosen protective component. 3) It has to be fast enough (response time should also be adequate). 4) It must be located close enough (particularly important for the “fast transients”). The overvoltage protection elements may be divided into two groups: nonlinear [gas-filled surge arresters (GFSA), varistors, and overvoltage diodes] and linear (various types of electrical filters). This paper focuses on representing the development of the methodology oriented toward efficient characterization of nonlinear components only. The essential characteristics of nonlinear components used for overvoltage protection are [4] as follows: 1) volt–ampere (V–A) characteristic; 2) volt–ohm (V–Ω) characteristic; 3) coefficient of nonlinearity α defined by the following equation: α = log I 2 I 1  log U 2 U 1  (1) where U 1 , I 1 , U 2 , and I 2 are coordinates taken from the volt–ampere curve; 4) dissipating energy, which can be calculated from W = n i=1 U i· · I i · Δt (2) where U i value of ith sample of voltage; I i value of ith sample of current; Δt sampling interval of a digital scope. 5) breakdown voltage; 6) volt–second (pulse shape) characteristic, i.e., breakdown voltage of the GFSA as the function of the applied voltage pulse duration; 7) functional aging as a consequence of design and material imperfections. The changes in the characteristics are a function of the number of past activations of current pulse amplitude and duration of the transients. 0018-9456/$20.00 © 2006 IEEE