HomeEducationDoctorate (PhD & PDEng)Current candidatesPhD infoUpcoming Public DefencesPhD Defence Jesper Lansink Rotgerink | Crosstalk Analysis in Aerospace Environments

PhD Defence Jesper Lansink Rotgerink | Crosstalk Analysis in Aerospace Environments

Crosstalk Analysis in Aerospace Environments

The PhD defence of Jesper Lansink Rotgerink will take place (partly) online.

The PhD defence can be followed by a live stream.

Jesper Lansink Rotgerink is a PhD student in the research group Power Electronics (PE). Supervisor is prof.dr.ir. F.B.J. Leferink from the Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS).

Aviation industry is highly focused on the development of emission-free aircraft. One of the pillars towards achieving this is the More Electric Aircraft (MEA) concept, which should ideally include full electric propulsion in the upcoming decades. Innovations have been driven towards replacement of conventional systems by electrical alternatives, resulting in hundreds of kilometres of on-board wiring and as a consequence an increased risk of malfunction due to undesired electromagnetic coupling - crosstalk. More than ever, this calls for optimisations in the weight of an Electrical Wiring Interconnection System (EWIS), while keeping compliance with safety and Electromagnetic Compatibility (EMC) regulations.

Multiconductor Transmission Line (MTL) models, including all its extensions, or even commercially available full-wave solvers based on for instance Method of Moments (MoM), are suitable for making predictions of crosstalk in complicated cabling systems. However, their significantly high computation times prevent them from applications in which a high variety of realisations has to be computed, such as EWIS optimisations and early risk assessments. Therefore, this thesis focuses on the derivation of methods able to provide efficient estimations of crosstalk that will be of use in early design stages of a EWIS.

The thesis starts with the introduction of a method for low-frequency approximations to the MTL equations. This method is used to derive closed-form expressions for crosstalk between wire pairs with and without a nearby perfectly conducting ground plane. Such expressions clearly relate all designable parameters to crosstalk, which can be used as design rules to make early decisions about routing and segregation of low-risk signals, or early identification of high risks. Moreover, the low-frequency approximations are further applicable to greatly reduce computation times for simulations that involve more complicated nonuniform transmission lines.

One of the measures to protect cabling from crosstalk or external field effects is to apply shielding. However, excessive shielding will result in an unnecessary increase in weight of a EWIS. Consequently, shielding provides opportunity for trade-off and optimisation in terms of EMC protection versus weight. Therefore, the third chapter of this thesis derives generic prediction of crosstalk in cases that involve shielded cables. Clear distinctions are made between regions of various frequency behaviour, and the transition frequencies as well as crosstalk levels of each region are again related to all designable parameters.

On top of carrying many electr(on)ic systems, modern aircraft also exhibit a huge change in the materials used. Nowadays, roughly 50% of aircraft comprise composite materials, such as Carbon-fibre Reinforced Plastics (CFRP), which have a much lower conductivity than the conventional aluminium. Therefore, this thesis provides two methods to incorporate the effects of lossy ground planes into the MTL equations. These methods are compared against full-wave simulation results as well as measured crosstalk for the cases of crosstalk between wire pairs above a single lossy ground plane and in between two lossy ground planes. The latter can be regarded as a step towards transmission line modelling of embedded or integrated wiring.

Finally, exact representation of a complex EWIS in aircraft seems utterly impossible. Uncertainties in for instance input data and cabling geometry prevents us from making exact predictions of crosstalk. Design rules will help in making decisions in the early stages of EMC risk assessment. Moreover, sensitivity analyses can provide insight in the parameters that are most crucial to crosstalk levels, as well as insight into the measures to be taken when further optimisations of the wiring systems in later stages of the design process are required. A final step in this thesis showcases the application of the developed transmission line models to perform efficient sensitivity analyses in cases of crosstalk between twisted pairs in cable bundles, as well as the transfer impedance of braided shields.