Factors affecting PD detection in GIS using a carbon nanotube gas sensor

Weidong Ding, Kohei Ochi, Junya Suehiro, Kiminobu Imasaka, Ryota Hayashi, Masanori Hara

Research output: Contribution to journalArticlepeer-review

23 Citations (Scopus)


The authors have previously demonstrated that a carbon nanotube (CNT) gas sensor can detect SF6 decomposition byproducts generated by partial discharges (PDs). There are several factors to be considered, however, before applying the CNT gas sensor to practical diagnosis of a gas-insulated switchgear (GIS). In this paper, three major factors, namely, the effects of operating temperature and installation location of the CNT gas sensor and SF 6/N2 gas mixture, were investigated. The stability of sensor conductance and the sensor response to PD were improved by controlling the sensor operating temperature. The maximum sensor response was obtained at about 70 °C. The CNT gas sensor, which was installed in an external pipe connected to the GIS tank, could detect PD generated in the tank, although the sensor response became lower and slower as the sensor was located further away from the PD source. It was found that the sensor response to PD showed a clear dependence on the mixing ratio of the SFS/N2 gas mixture. The maximum sensor response was achieved for SF6/N2 (20%/80%) gas mixture. This result was discussed based on Fourier transformation infrared (FTIR) analysis of PD decomposition byproducts adsorbed on the CNT surface.

Original languageEnglish
Pages (from-to)718-725
Number of pages8
JournalIEEE Transactions on Dielectrics and Electrical Insulation
Issue number3
Publication statusPublished - Jun 2007

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering


Dive into the research topics of 'Factors affecting PD detection in GIS using a carbon nanotube gas sensor'. Together they form a unique fingerprint.

Cite this