The Influence of The Degree of the Rivet Hole Sizing on the Fatigue Life

The paper presents the results of fatigue tests of specimens with sized rivet holes. Samples for tests were made of 0.05” (1.27 mm) thick non-clad plates of aluminium grade 2024-T3. Rivet holes were prepared assuming that they shall be used for 3 mm nominal diameter snap head rivets for aviation-related purposes. Different sizing degrees were achieved by drilling holes of different diameters in the samples followed by the sizing process using a sizing mandrel of the same diameter of 3.15 mm. Holes in the test samples were drilled using special device ensuring appropriate quality and repeatability of the holes. Five different sizing degrees were achieved be means of five drills of different diameters.

Samples with holes of varying degrees of sizing were tested under constant amplitude sinusoidal loading conditions (cycle asymmetry coefficient R = 0) at the load frequency of 5 Hz. The study was conducted at three levels of maximum stress in the cycle. Fatigue life test results were presented in the form of fatigue diagrams (regression lines) determined in the bi-logarithmic coordinate system log N, log Smax. The results of the fatigue life tests received for the drilled as well as drilled and reamed holes were also presentedfor comparison. The results obtained lead to the conclusion that that the lowest fatigue life characterises samples with drilled holes and drilled and reamed holes. Fatigue life of specimens with holes for rivets improved (by 50% to 74%, depending on load level), even as a result of the hole surface polishing only (minimum sizing degree), whilst the two-fold growth of fatigue life was achieved for holes of a slight sizing degree. A further significant increase in fatigue life was achieved by the cold work of the hole's surface. Based on the location of the fatigue diagrams and the fatigue life tests results, it may be concluded that the higher a sizing degree, the higher the fatigue life growth. The growth is also proportional to the specimen load level: the lower the load level, the higher the fatigue life growth.

Fatigue diagrams obtained from tests were divided into three groups: diagrams for drilled holes and drilled and reamed holes, diagrams for holes with a low degree of sizing and diagrams for holes with a high degree of sizing. This division was confirmed by statistical tests of regression lines parallelism by the “peer-to-peer” method.