Martensitic stainless steels have represented since few years a material of choice for the manufacture of mechanical parts such as hydroelectric turbines. The development of the alloy has led to grades with very low amount of carbon giving them a good weldability.
The assembly of these parts, made by autogenous welding, requires the use of materials with low transformation temperature (LTT) such as 410NiMo. These filler metals are also used for assembly by heterogeneous welding of steel parts susceptible to cold cracking. The transformation of austenite to martensite occurring at low temperature, residual stresses from single-pass welding operation are different from those normally found and reduce the risk of cracking. By cons, industrial experience shows that in situation of multipass welding, the risks of cold cracking are still present.
This project aimed to determine a cracking test for assessing susceptibility to cold cracking of 13%Cr-4%Ni stainless steel according to the welding procedure, in autogenous welding situation.
Literature contains much information about cold cracking phenomena. That phenomena occurs under three conditions. These conditions are: a high diffusible hydrogen level, significant residual stresses and a brittle microstructure. It seems that despite the low mass ratio of carbon (0.022% C) and the low diffusible hydrogen level (< 3 ml/100g) risks of cold cracking remain present during multipass deposits.
Use of cracking tests was necessary to assess the sensitivity to cracking of the martensitic stainless steel. Before the work preliminary tests have been made or tested Tekken GBOP and testing to determine that to obtain the most representative of the industrial reality results. Then they have been modified to reverse the compression stress in the seam test to tension by the addition of a second weld. This inversion occurs in multipass welding and has been targeted as an important factor in the occurrence of cold cracking phenomenon. The results of these tests show that Tekken test is not suitable for LTT testing. It was also demonstrated that GBOP test with two juxtaposed seams configuration gave results consistent with the industrial observations.
The second stage of the project was to study the cracking test selected. Acoustic emission tests were done during welding and cooling of GBOP test. These tests were conducted in order to detect when the cracking of the test occurred and to validate the method of inspection. This inspection is done after separation of the specimen, by observation of the fracture surface. Usually, cliveage zone on the fracture surface can be associated with cold cracking and dimple zones can be associated with the specimen separation. Through these tests, it was possible to validate this assertion. Then the relevance of the addition of a second weld has been validated by studying the residual stress by the contour method. It was possible to observe an area of the first bead in tension, promoting cracking of the test.
Finally, some test runs were made with various filler metals in order to confirm that the utilization of the modified GBOP test for 13%Cr-4%Ni was adequate. A fractographic study of some sample was also made.
| Date | 24 Feb 2014 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Philippe Bocher (Supervisor) & Yves Verreman (Co-supervisor) |
|---|
Paquin, M. (Author),
Bocher (Supervisor) & Verreman (Co-supervisor),
24 Feb 2014Student thesis: Master's thesis › Master in Engineering: Mechanical Engineering