| 摘要: | 本研究以Ti-6Al-4V為母材，TiCuNi為填料(30 µm厚度金屬箔片)，進行真空硬銲後之結構為對象。第一部分的目的為建立最佳製程參數組合，品質參數採用抗拉強度、破壞韌性及微硬度，分別以單目標及多目標最佳化方法進行分析。第二部分的目的為建立疲勞裂縫成長速率資料，探討平均應力及超載負荷之影響，透過等負荷振幅疲勞實驗，得到材料疲勞常數及應力比之影響。利用變動負荷振幅疲勞實驗結果，驗證多種疲勞裂縫成長模式，包括應力比效應、負荷交互效應及裂縫閉合效應等，以建立最佳的疲勞裂縫成長預測模式。
研究結果顯示：以抗拉強度為目標的最佳化參數組合為預熱溫度890℃，預熱時間60分鐘，硬銲溫度975℃，硬銲時間45分鐘。驗證實驗中可以得到抗拉強度1265 MPa與預測值1262 MPa之誤差僅0.24 %。選用抗拉強度、破壞韌性和微硬度作為品質目標，最佳化參數組合為預熱溫度890℃，預熱時間60分鐘；硬銲溫度1005℃，硬銲時間15分鐘。超載負荷比大於2.0時，會造成明顯的疲勞裂縫成長延遲效應。在銲道破斷面觀察方面，在拉伸試片中，抗拉強度較小者，破斷面以小劈裂面為主，抗拉強度較大者，破斷面由韌窩組織及小劈裂面所組成，且強度越高則韌窩組織越多且越緻密。超載負荷會造成裂縫成長速率大幅度的改變，可在破斷面上看出其施加的位置及當時的裂縫前緣。施加的超載負荷越大，該圓弧形的裂縫前緣曲線越明顯。在疲勞裂縫成長模式方面，(1)雨流循環計數法比簡單範圍法佳，(2) 在裂縫閉合效應修正模式中：與Elber模式的預測結果與未修正的Paris模式相近，且接近實驗值，而Schijve模式的預測壽命值則過度小，(3)最接近實驗值的疲勞裂縫成長預測模式為靜態破壞模式，(4) 利用塑性變形區修正的Willenborg模式，其修正結果比未修正者差。次序效應驗證中，對於應力振幅由大到小排列的歷程，在初期產生較大的拉應力，對材料產生超載負荷效應，由此產生的殘留壓應力場，導致後續負荷的有效應力降低，進而降低裂縫擴展速率，增加疲勞壽命。
;This study focuses on the structure after vacuum brazing using Ti-6Al-4V as the base material and TiCuNi as the filler material (metal foils with a thickness of 30 µm). The first part aims to establish the optimal combination of process parameters. Quality parameters, including tensile strength, fracture toughness, and microhardness, are analyzed using both single-objective and multi-objective optimization methods. The second part aims to establish fatigue crack growth rate data, investigating the influence of average stress and overload conditions. Through fatigue experiments under constant load amplitudes, the study obtains data on material fatigue constants and the effect of stress ratio. Utilizing the results from variable load amplitude fatigue experiments, various fatigue crack growth models are validated, including stress ratio effects, load interaction effects, and crack closure effects, to establish the optimal fatigue crack growth prediction model.
The research results indicate that the optimal parameter combination, targeting tensile strength, consists of a preheating temperature of 890°C, a preheating time of 60 minutes, a brazing temperature of 975°C, and a holding time of 45 minutes. Experimental validation yields a tensile strength of 1265 MPa, with a deviation of only 0.24% from the predicted value of 1262 MPa. When utilizing tensile strength, fracture toughness, and microhardness as quality objectives, the optimized parameter combination includes a preheating temperature of 890°C, a preheating time of 60 minutes; a brazing temperature of 1005°C, and a holding time of 15 minutes. Notably, when the overload load ratio exceeds 2.0, a significant fatigue crack growth delay effect is observed. With respect to the fracture surface observation of welded joints, those with lower tensile strength exhibit a fracture surface predominantly characterized by small cleavage facets. On the contrary, specimens with higher tensile strength display a fracture surface composed of ductile dimples and small cleavage facets, with increased density and compactness of the ductile dimples corresponding to higher strength levels. The application of overload stress induces a significant decrease in the crack propagation rate, revealing the applied stress location and the crack front at the time of fracture on the fracture surface. As the applied overload stress increases, the arcuate curve of the crack front becomes more pronounced on the fracture surface. In terms of fatigue crack growth models: (1) The rainflow cycle counting method is superior to the simple range method. (2) In the crack closure effect correction model: The predicted results of the Elber model are similar to the uncorrected Paris model and close to experimental values, while the Schijve model′s predicted life values are excessively small. (3) The fatigue crack growth prediction model closest to experimental values is the static failure mode. (4) The Willenborg model corrected with the plastic deformation zone yields worse results than the uncorrected model. In the verification of sequence effects, the High-to-Low (HtoL) transition, occurred in the early stages, induces significant tensile stresses, leading to an overload effect on the material. The residual stress field generated as a result of overloading diminishes the effective stress for subsequent loading, consequently reducing the crack propagation rate and enhancing the fatigue life. |
