Validation of the Weighted Peak Method in the time domain for EMF assessment in manual arc welding application: Compliance with IEC 62822-2:2016 and 2013/35/EU

In manual arc welding a workplace assessment for human exposure to electromagnetic fields is required. This can be done via calculation with parameters of the setup or direct measurement of the arising fields using probes. The limits to determine are frequency dependent, and therefore an appropriate weighting of the magnetic field is needed for correct exposure assessment. One of the recommended methods is known as the weighted peak method in the time domain. In this research a comparison between analogue and digital weighting procedure is made while drawbacks of digital weighting are exposed. In addition, different measurement methods were discussed in order to highlight the problems and pitfalls of measuring uniform and non-uniform fields. The results showed that for uniform magnetic fields, the deviation between the calculated and measured exposure values is below 5 %. For non-uniform fields (realistically occurring along the welding leads), the digital method overestimated the exposure index, yielding the error to increase with distance between probe and welding cable. It was also found that for very high sample rates, the exposure index is drastically overestimated by several magnitudes for uniform and non-uniform fields, caused by numerical instabilities related to common signal processing libraries and the usage of common weighting functions. Therefore, a different weighting function that resulted in a numerically stable signal weighting is recommended from our study.