Proper Metrological Methodologies to Avoid Severe Systematic Errors When Characterizing High-Power IGBTs on the Installation Field: An Introduction

The exigency of characterizing high-power insulated-gate bipolar transistors (IGBTs) directly on-site increasingly emerges in field installations hosting multilevel converters that employ such devices by the dozen, when not by hundreds, as it is already the reality for modern modular voltage-source HVDC and flexible ac transmission systems (FACTS). Nonetheless, differently from the actualities in permanent laboratories, the typology of the instrumentation rapidly obtainable on the field is often rather restricted by the aspects of logistic, necessary timing of intervention, and ruggedness against the usually harsh environments. Commonly, the only specimens realistically expectable at short notice on a field installation are portable curve tracers (CTs), portable field oscilloscopes with nonisolated channels, passive voltage probes, current probes based on Rogowski coils, portable multimeters, and variable dc power supplies. This paper describes metrological methodologies devised to diagnose and characterize high-power IGBTs by employing solely such a limited set of instruments. The severely erroneous conclusions that possibly descend from the utilization of CTs only are highlighted, both by experiments and by a dynamic model deduced from the nature of the specific metrological context. It is shown how the additional presence of properly connected oscilloscopes and probes becomes mandatory in order to attain a proper characterization. The role of the CTs should be strictly limited to mere stimuli generators only, also because their compensation techniques against the systematic measurement errors occurring with signal and low-power transistors rarely suffice in the case of high-power devices. Additionally, the equivalent differential capacitance between the collector and emitter terminals of a high-power IGBT-diode module is characterized too. It is often insufficiently documented in the datasheets, although its values and evolution are important for the dynamics of- the commutation. Experimental observations from an ABB 5SNA 1200G450300 IGBT module (4.5 kV and 1.2 kA) are discussed throughout the different sections in order to elucidate the details of the presented methodologies.