RELACIÓN DE PUESTOS DE TRABAJO 2018 I. FUNCIONARIOS DE CARRERA

After investigating the effectivity of the electrical resistance sensing system to measure the severity of damage and discussions in the areas that are needed to be researched to improve the accuracy of the current sensing system. Another challenging subject was the investigation of the sensor to enable damage location within the panel. The main advantage of this technique was the ease with which the damage can be located. This technique has an advantage over C- scan in that it located the damaged area within few seconds while C-scan is a point-test and it requires an extensive work to locate internal damage such as delamination. However, a four- probe electrical resistance technique cannot map the damage while C-scan can do that in 2D and 3D view. The ease with which the damage can be located was attributed to innovative sensing mats, four-probe electrical resistance technique, and a simple data processing proce- dure.

The data acquired from the DAQ system was presented Table 7-2. The sensing mats divided the CFRP laminate panels into segments as shown in Figure 7-37, at which the electrical re- sistances were measured before and after each impact events. the four-probe electrical re- sistance technique offered a precise way to measure the resistance by eliminating electrical contact resistances. The electrical resistance at each segment was measured and then the electrical resistance changes at each segment were measured again at various impact ener- gies. The electrical resistance readings were analysed by plotting the change in electrical re- sistance (∆R) using Equation 7 – 10 against the sensing electrodes that were used to measure them.

∆R = Ri – Ro 7-10

Where ∆R is the change in resistance at a segment, Ri is the electrical resistance after an impact

Figure 7-37: Representation of damage identification technique using electrical resistance change method.

The damage was located by significant local variations in the electrical resistances occurred in a panel as shown in Figure 7-38 to Figure 7-43. Those figures were generated in the following method: changes in electrical resistance in the horizontal direction were (∆RA1, ∆RA2, ∆RA3, ∆RA4)

and changes in electrical resistance in the vertical direction were (∆RB1, ∆RB2, ∆RB3, ∆RB4);

changes in horizontal and vertical directions were multiplied to generate a 4 x 4 matrix and then a cubic interpolation function was used to expand the matrix from 4 x 4 to 7 x 7 matrix. The purpose interpolation function was that to make the transition between damage and undam- aged regions smooth and then the contour was plotted using “contourf” function in MATLAB. It can be seen from the figures that when impact energy was low, the changes in electrical resistance were low, therefore, it was hard to locate damage precisely since the local variations in electrical resistances were high in comparison with undamaged areas in a CFRP laminate panel. This can be seen clearly in Figure 7-38a, however when the impact energy increased damage became more defined hence damage was located accurately.

Figure 7-38: Damage location in AB panel that was impacted at a) 0.406 J, b) 0.607 J, c) 1.426 J, d) 3.333 J, and e) 4.96 J. The arrows on C – scan image at the top left and right corners represent the current flow direction.

It can also be seen that C-scan images helped to determine damage profiles. The smaller the damage the more spread the contours were. The arrows around C-scan images represent the direction of electrical current applied into the CFRP laminate panels. All panels were impacted in Segment 2 (Figure 7-37), therefore the highest changes in electrical resistance were occurred in the region between A2, A3, B2, and B3. In spite of the fact that the sensing electrode in this region experienced the highest electrical resistance changes, however the changes in electrical resistance were varied from a sensing electrode to the others. This in turn was shifted damaged areas on the contours from Segment 2 to other segments slightly. This challenge was attributed to amount of electrical contacts made with the carbon fibres in CFRP laminate panels during attaching the sensing mats to CFRP laminate panels. It is important to state that this type of error was inherited in the electrical resistance sensing technique and it required an advanced

signal processing technique to overcome it. In this thesis a simple and straightforward relation- ship was used to determine the damage location as give in Equation 7-10.

When impact energy increased to 4.96 J, all types of damage occurred (matrix crack, fibre breakage and delamination), therefore the damaged area was big as shown in Figure 7-38e. When the thickness of the CFRP laminate panel increased to 1.63 mm (Figure 7-39), the changes in electrical resistance were less than CFRP laminate (Figure 7-38). When the thick- ness increased to 2.54 mm the changes in electrical resistance were as low as 11 x 10-4 _{Ω when}

the panel impacted at 4.96 J (Figure 7-40). However, damage was located, and damage was more localised when the CFRP panel was impacted at 9.996 J.

Figure 7-39: Damage location in AC panels that were impacted at a) 1.426 J, b) 3.333 J, and c) 4.96 J. The C-scan images at the top left and right corners show the damage profile, the arrows around C-scan images show the current flow direction.

Figure 7-40: Damage location in AD panels that were impacted at a) 4.96 J, and b) 9.996 J. The C-scan images at the top left and right corners show the damage profile, the arrows around C-scan images show the current flow direction.

4.96 J and 9.996 J impacted energies caused barely visible impact damage in 3.5 mm CFRP laminate panel; the changes in electrical resistance was as low as 3.5 x 10-4 _{and 2.5 x 10}-3 _Ω

respectively. It can be concluded according to Equation 7 – 1 that fibre volume fraction has higher impact on the electrical resistance sensing system than panel thicknesses.

Figure 7-41: Damage location in AE panels that were impacted at a) 4.96 J, and b) 9.996 J. The C-scan images at the top left and right corners show the damage profile, the arrows around C-scan images show the current flow direction.

Damage location in two CFRP laminate panels, that were made using a VARTM technique, was investigated. It was found that VB panel in Figure 7-42 followed the same pattern as AB panel in Figure 7-38, however the changes in electrical resistance were higher which was attributed to the manufacturing technique, where fibre volume fraction was lower as discussed in Section 6.2.4. Also fibre-fibre contact network was lower than AB panel, therefore disruptions caused to fibre-fibre contact network caused due to damage had a higher impact on the electrical re- sistance of the panels. The CFRP laminate panels made using a VARTM technique had bigger resin rich areas and void contents than their equivalent made in an autoclave as discussed in Section 6.2.1 and Section 6.2.4. All those factors played an important role in damage detection and location.

Figure 7-42: Damage location in VB panels that were impacted at a) 1.426 J, b) 3.33 J, and c) 4.96 J. The C-scan images at the top left and right corners show the damage profile, the arrows around C-scan images show the current flow direction.

Impact energy of 4.96 J caused visible damage in VB panel (full perforation was occurred) and that caused damage not only to the CFRP laminate panel but also to sensing electrodes, there- fore damage was spread over a large area. Although VC panel had higher fibre volume fraction than VB panel (Table 6-8) but it also had a bigger resin rich area, therefore that was reduced fibre-fibre contact network in through-thickness direction as show in Figure 6-8. Hence the changes in electrical resistance due to BVID was lower than VB panel as shown in Figure 7-43. Given these results the electrical resistance sensing system showed a great advantage to lo- cate damage in a short time (few seconds) over conventional non-destructive techniques, such as C-scan.

Figure 7-43: Damage location in VC panels that were impacted at a) 1.426 J, b) 3.33 J, and c) 4.96 J. The C-scan images at the top left and right corners show the damage profile, the arrows around C-scan images show the current flow direction