A Novel Seal Packaging Structure Applied in Sub Miniature Push-On Connector Based on Ni47Ti44Nb9 Shape Memory Alloy Seal Ring

At present, the tin-lead vacuum soldering process is commonly used to complete the seal packaging of Sub miniature push-on (SMP) connectors. However, when the number of SMP connectors exceeds thousands, it is difficult to guarantee the consistency of soldering, which greatly restricts the large-scale application of SMP connectors. Fortunately, Shape Memory Alloy (SMA) materials have the good low-temperature phase change characteristics and could produce a big fasten force, thus can be used for seal packaging of SMP connectors. Therefore, this paper proposes a novel seal packaging structure for SMP connectors based on the Ni47Ti44Nb9 SMA seal ring to solve the problem of seal packaging of SMP connectors. In view of the SMA’s good performance, three seal packaging structure forms for SMP connectors based on the SMA seal ring are proposed, which are named as structure A, structure B and structure C, respectively. The characteristics of the seal packaging structures of each form is briefly described. The comparative study is carried out on the three structure forms with the seal packaging leakage rate as the target. The test results demonstrate that structure A can provide the best sealing performance while taking up the least space, which meets the requirement of the normal application of electronic communications equipment. Then, the further research on structure A is conducted, a prediction model of seal packaging leakage rate is established with the response surface method (RSM), by which the relationship between the structural parameters including the wall thickness of seal ring, the roughness, the thin wall thickness of cavity and the seal packaging leakage rate is investigated. Finally, to reduce the seal packaging leakage rate, the structural parameters of the seal ring are further optimized with an improved Adaptive Genetic Algorithm (IAGA), and the efficiency of the IAGA is verified through comparison with the conventional Adaptive Genetic Algorithm (AGA).