Current Step Generation and Measurement with Rise-time in the Range of Nanoseconds

Abstract

A current step generator based on a charged coaxial cable is designed and tested for characterizing impulse current shunts. This thesis has developed a traceable calibration infrastructure for fast shunts and other current sensors, defined measurement techniques for a current step and improved the test procedure and measurement capabilities. For calibration of shunts, current coil sensors are used in the measurement circuits.

Since no calibration services are currently available for impulse current measuring systems, a best circuit combination is proposed for current step generation with a rise time of less than 5 ns, along with a proposed reference shunt that aims to provide the best and most stable measurement results with negligible noise, oscillations, and droop in the measured current step.

Based on techniques found in the literature, current steps are generated, and different sensors were used to measure the generated steep front current steps. The generation system consists of a 110-m long, 50-Ω coaxial cable and a spark gap. Various spark gap switches, including the SF6 spark gap, are used for generating current steps. With the coaxial cable charged from one end, a current step is generated after reflecting back from the open end with a step length of twice the cable transmission delay. The cable is than discharged to the shunt (or coil) through the spark gap.

The measurement system consists of shunts and coil current sensors, 5:1 and 6.6:1 attenuators based on the requirement of the sensors. The recording instrument is a 1-GHz, 8-bit, 1-GS/s digitizer. The proposed step generator can produce current steps with a stable current of up to 100 A. The rise time of the step varies from 1.6 ns to 15 ns, depending on the spark gap used for switching. The produced current is constant within 0.5% for a step length of 960 ns generated with a coaxial cable 110 m in length.

To improve the test procedure and measurement capabilities, the thesis also analyzed factors affecting current step measurement, such as the type of coaxial cable, type of connection, extra shielding, clearances, interference sources, media of the spark gap, and the spark gap electrode distance (arc length). It is found that the measurement system and the rise time of current step is affected by many factors, including the coaxiality of the connection, impedance mismatch, interference, clearances, stray capacitances, and stray inductances. These results will enable future standardization of impulse current sensors.