Abstract

This paper shows a numerical polynomial approximation to the issue of how junction transistors "BJT" and field effect transistors "FET" can work in safe or unsafe conditions in explosive atmospheres.  We have analyzed the most widely used transistors with thermal imaging, working in a controlled environment, to characterize their thermal behavior. With this characterization, we can predict their possibility to ignite a classified location (explosive atmosphere) fundamentally by mean of thermal-conduction and reaching the minimum activation energy of the combustible vapor, dust or flying. We have bring these transistors to their nominal rated values specified by their currents and working voltages, and we found that the effect of heat dissipation on the base-emitter junction of a transistor BJT is really non linear and much greater than on the threshold voltage of polarization of the FET due to the inherent loses to obtain the nominal voltage of the diode that forms the base-emitter junction. We have found experimentally that it is very easy to obtain a thermal difference of more than 200 °C in overheating of a common BJT compared to a MOSFET with a similar load in fixed polarization. We found temperatures over 300ºC in BJTs in common operating ranges, when the accepted “safe” temperature is supposed to be no more than 200ºC in any case. This issue is addressed with performance-based analysis, focused on temperature, and it suggest that equipment with BJT technologies should not be implemented in some areas of hazardous or explosive locations; so MOSFET technologies are preferable.

Abstract

The thermal-conductive models for the calculation of the electrical conductors and their energy dissipations have been able to solve doubts about the power conductors, which has been permanently unsuspected. Before thinking about thermally characterizing an electrical conductor, it was enough with the application of simple Joule-type equations, with constant resistive intrinsic properties in the conductor, as well as invariable considerations in the conditions that surround the conductor. In fact, just as is exposed, this has been published on the top of the best known electrical standards, such as those issued by the National Fire Protection Association (NFPA), as well as those edited by the International Electrotechnical Commission (IEC). It has been demonstrated that electric cables with higher electrical intensities, move away from the behavior predicted by the Joule type models, sometimes with energy dissipations 50% higher than those predicted by Joule. This is due to the non-linearities of the intrinsic properties of the cable, and the inconsistency of the environmental conditions surrounding the cable at the time it is going to heat. The novel thermal-conductive models with just ten years of development, are today a very complete solution with unparalleled accuracy (errors less than 0.01% in energy and temperature) in the thermal predictions of conductors; but due to its high degree of mathematical complexity requires in the most of cases to use numerical methods to obtain practical solutions. In this paper a new general algorithm is shown, which stands out for its simplicity compared to the first thermoconductive models, solving the models in thermal predictions of power electrical cables, without the need of huge databases, with simplicity, maneuverability and excellent practical results.