Table 1

Material properties of UD HexPly  T700-M21GC。

The  damage  profiles  vs。  the  normalized  cycles  (N/Nf, where

Nf = 5000 cycles) are shown in Fig。 3。 When conventional   machin-

Young’s modulus (GPa)

Shear modulus (GPa)

Poisson’s ratio

Fiber content (%)

Ply thick (mm)

ing is used, the damage is less than 5% for loads below 8 kN (47% of UTS)。 When the load reaches 9 kN (53% UTS) the cumulative dam- age does not exceed 10%。 At 10 kN (59% UTS) the specimens rup- tured。 For the specimens drilled with abrasive water-jet, the cumulative damage was about 10% for 9 kN load。 The damage reaches 30% at 10 kN and the specimens ruptured at 11 kN (65% UTS)。

fixed on a tripod and placed at a distance of 100 cm from the car-

bon epoxy composite plate in order to capture stable images with- out any vibration。 Thermographic detection was conducted using an infrared system with a precision of 320 × 256 pixels – providing extremely high temperature sensitivity (a little under 20 mK at    30 °C) at high-speed frame rates (1–150 Hz)。 For our tests, a fre- quency of 5 Hz has been used to record thermal dissipation。 The   IR camera was equipped with an integrated motorized lens and could be easily focused using the software。 The rectangular plate

specimen  of  carbon/epoxy  composite  has  emissivity  e = 0。9。

3。Results and discussion

3。1。Static analysis

Fig。 2 shows the stress–strain curves for CM and AWJM speci- mens under static loading。 From Fig。 2, it was  observed that, for  all specimens, the stress–strain curves pass through the following phases: The first phase, which is linear characterized by an elastic modulus E0; in the second phase, the slope of the curve decreased reflecting the beginning of the viscoelastic and viscoplastic behav- ior; and in the last phase, the stress decreases until the final break。 For the CM and AWJM specimens respectively, the average failure stress is 202 MPa and 209 MPa, and the average strain is 8。79% and 8。70%。 A sudden fracture in the vicinity of the hole was observed for all specimens。 From these static test results it is evident that the machining processes have no effect on load bearing perfor- mance of carbon–epoxy  plates。

3。2。Fatigue analysis

3。2。1。Damage analysis

The change in stiffness during cyclic loading is commonly used to quantify the damage in the specimen。 The damage accumulation

(D) is related to change in the ratio of dynamic stiffness (Ei) to sta- tic stiffness (Eo) by the following   equation:

Ei

Fig。 4 plots the histories of damage progression (damage vs。 normalized cycles) for cyclic loading amplitude of 10 kN (59% UTS)。 In tests conducted at 10 kN (59% UTS) shown in Fig。 4, the CM specimens ruptured after less than 3900 cycles (N/Nf = 0。8), accu- mulating damage levels between 50% and 66%。 Under the same loading, after the same number of cycles, the AWJM specimens continued to withstand cyclic stresses and accumulated damage levels of only 20% (a difference of 30% points when compared    to

CM)。 The AWJM specimens lasted all 5000 cycles without ruptur- ing, accumulating a maximum damage level of only 33%。 These re- sults indicate that for any given load level, AWJM specimens display higher resistance to fatigue damage than CM    specimens。

This difference in the damage accumulation between conven- tionally and non-conventionally drilled specimens is mainly attrib- uted to the machining process。 This is supported by the surface topography (roughness) tests as shown in Fig。 5。 It can be seen that specimens machined by abrasive water-jet are characterized by streaks damage in the same the direction of the displacement of