An engineering consultancy in York has played an integral part in a European project assessing the way electromagnetic compatibility (EMC) of railway vehicles is measured.
This work is important because rolling stock can produce EM emissions that interfere with railway signalling and communications systems and public broadcast services.
York EMC Services (YES) was a partner in the TREND project (Test of Rolling stock ElectromagNetic compatibility for cross-Domain interoperability), which was funded under the EU’s FP7 programme.
Train manufacturers have to make sure their products comply with the EN50121 European standards. One element of the project was to design test set-ups that would help harmonise approval testing against these standards, while another work package analysed the standards themselves, with a view to informing future updates.
At the recent Infrarail exhibition in London, Eva Karadimou, an EMC engineer at YES, explained more about the project. Potential sources of EM interference from rail vehicles range from power electronics to ‘hotel’ equipment or even devices brought on board by passengers, but the major concern is arcing arising from the current collection, as pantographs bounce and shoegear crosses gaps.
For this project, team members in Sweden created arcing effects by deliberately icing a catenary wire and measuring the resultant emissions from passing trains. YES’s role was to replicate these effects in a reverberation chamber and show that this would be a reliable test method.
What the York engineers did sounds surprisingly crude, but it worked. Within the screened chamber they set up a rotating bicycle wheel close to a brass rod to represent the contact wire and pantograph respectively. Then they generated sparks at the end of the rod to create arcing.
To measure the effects, an antenna within the room was connected to a spectrum analyser outside, and the radiated power was measured at particular frequencies for various parameters: wheel speed, gap length and resolution bandwidth.
“That let us see how the power radiated from the pantograph would affect [services at] different frequencies,” Karadimou told E&T. “We looked at 900MHz because that’s used for GSM, and also at 1.5GHz. The standard covers up to 1GHz, and we wanted to see if there’s a need to go above that. We’ve shown that there is.”
The TREND project was completed in April 2014. Among its conclusions were that a reverberation chamber is a representative environment to test the set-up, that standards bodies should reconsider the limits and methods used for measuring transient phenomena, and that measurements and limits above 1GHz should be considered within the standard, “given the broadband nature of arcing events”.