As a result of x-ray monitoring of the seams, samples of tubes with from 6 to 20% nonfusion were selected for the tests. The actual dimensions of the nonfusion zones were determined after fracture. The criterion for evaluation of nonfusion was the ratio of the total area of the defects in the welded seam to the rated area of the tube. The static pressure in the samples was set up at normal temperature. The value of the pressure of the pressurization was selected from the condition of the attainment of plastic deformations in the welded joint in a nonfusion zone, and the assurance of the relaxation of the residual welding stresses. The plastic deformations in the welded joint were monitored by the strain-gage method using PKB-5-100 strain-gage pickups. Under these circumstances, the deformation in the axial direction after pressurization was 0.2-0.3%. For tubes with a diameter of 103 3 ram, the pressurization pressure was 132 kg/cm 2, and, for tubes with a diameter of 38 x 3 ram, 412 kg/cm 2. The holding time of the samples under the load did not exceed 2 min. For the testing of tubes with cyclic loading by bending and elongation, special attachments were built for mass-produced testing machines. Tests of tubes with a diameter of 38 x 3 mm with simple cyclic bending were made in a VU 70/100 vibration test stand. At cryogenic temperatures, liquid nitrogen was fed into the cavity of the sample during the tests. The frequency of the cyclic loads with testing in the stand was 16-40 Hz. Tests of tubes with a diameter of 103 3 mm with axisymmetric loading were made in a PDM-100Pu pulsation machine, equipped with a cryogenic chamber for low-temperature tests. The frequency of the loading (elongation) of the samples in these tests was 10 Hz. The adopted criterion of fracture of the samples was the moment of the appearance of a through fatigue crack with a length of 10-15 ram. As the tests showed, with the appearance of a through microcrack, it grows considerably. With tests in a VU 70-100 test stand, this moment is characterized by a change in the frequency of the vibrations of the sample-vibration test-stand table system. The change in the frequency of the vibrations was recorded by a special vibration pickup, connected into the control circuit and shutting down the test stand. With tests in a TsDM-100Pu vibration machine, the moment of the appearance of a through fatigue crack coincided with a lowering of the reading of the load according to a dynamic manometer, which was recorded by an operator. With both kinds of cyclic loading with a frequency up to 16-20 Hz, the temperature of the samples rose only very slightly. With the testing of samples with a high frequency of the loading, there was a considerable increase in the temperature. The fatigue limits of the welded joints of the tubes were determined using the method of accelerated tests [1-3] of ten samples, and were verified experimentally on five samples, on a base of 5 x 106 cycles.
Read full abstract