| Summary: | In this thesis experimental tests and numerical simulations were carried out on the effects of the grounding system on CFRP test samples for the aerospace industry. The grounding system was assessed as one of the parts that can contribute to increasing the cost effectiveness of the sample and the relative test campaign. A numerical investigation was performed to understand the current distribution within anisotropic CFRP laminates. The influence of the grounding system and the interlaminar impedance on the current distribution were studied using parametric simulations and the results were compared. It was found that the side grounding system produces the best performance between the arrangements investigated. It allows for the best current spreading within the sample and for the lowest voltage drop between the injection and grounding points. Furthermore, it allows for the manufacturing of more cost effective test samples. However, a real implementation of the side grounding system introduces the contact pressure variable between the sample and the grounding electrode which influences the contact resistance. To implement the side grounding system, a test rig was designed and manufactured. This made it possible to control the compression force. A low current DC test campaign was carried out on five test samples with the same number of grounding systems. It was found that the contact resistance for the three side grounding systems is dependent on the compression force applied between the grounding electrode and the side of the sample. The influence was more pronounced for the sample without any metal coating on the side in contact with the electrode. Furthermore, the resistances measured for the two side grounding systems with metalized sides were comparable with the resistances measured for the fastener grounding systems. It was found in the literature that the variation of contact resistance due to compression force, for the side grounding cases, is related to variations in the topography of the surfaces in contact, specifically, the variations in the real contact area. Therefore, an investigation on the effects of the compression force on the surfaces in contact was carried out. It was found that the increase in compression force contributes to the decrease in the roughness mean; the increase in the material ratio curve which is a parameter related to the load bearing area; and the modification of the distribution of the height of the surfaces with the appearance of a negative tail. Further to the low current tests and the characterization of the surfaces of the side grounding samples a high current test campaign was performed. Through the use of typical lightning current waveforms, the thermal and electrical limits of the five grounding systems were investigated. The side grounding systems were tested at the maximum compression force to improve the contact area between the sample and the return electrode and therefore, decrease the risk of the sparking phenomena. The side grounding system paired with the sample with the metal sputtered side showed the best performance in terms of sustaining the highest current peak without appreciably increasing the risk of sparking phenomena at the contact interface between the sample and the IV return electrode. Conversely, the sample without any metal coating on its side showed the poorest performance, showing the sparking phenomena at the lowest current peak. The results of the thermal investigation were used for a visual validation of the numerical model created in the first part of the thesis. A comparison between a numerical model and an analytical model was performed. The dynamics of the temperature when a carbon fibres tow is subjected to a lightning current and thus to Joule effects were studied. The analytical model had the advantage of offering a simpler solution compared to the numerical model and a limited amount of data to input into the problem. The two models were found to produce similar results for the first transient of the temperature. However, the analytical model did not take into account the heat dissipation effects of the carbon fibres in the surrounding environment therefore, after the first transient the analytical model was found not to be as accurate as the numerical model.
|
|---|
