Explosion of high-temperature pipes
Thread Content
A few years ago, for a certain project involving pipes exposed to temperatures of 200 degrees Celsius, the pressure testing was successful; no non-destructive testing was carried out, and the pipes were in operation for some time as well. Yet leaks occurred at the welds, which led to an explosion. Why?
You yourself said it – no non-destructive testing was done. In some areas, the welds contain pores or slag inclusions. During the pressure test, it withstood the stress, but during operation, internal corrosion, pipe vibration, and other factors can affect the welds. I haven’t seen it, but I’ve heard that welds can have pores; during pressure tests, water seeps out slowly, and when struck with a small hammer, the water sprays out.
For large pipelines, 3091 welded steel pipes were chosen as the material for steel plate coiled pipes. Those who are aware of the differences between the two, as well as the considerations for engineering design and construction, please share your insights.
There is a leak in the weld; it is not specified whether it refers to the circumferential weld of the pipe during installation or the welds on the body of the welded steel pipe. Additionally, details such as the welding process and heat treatment conditions for the 3091 steel pipes purchased initially are unknown. Passing the pressure test does not mean that there are no issues with the welds; it simply means that those problems were not revealed during the pressure test. Defects such as slag inclusions, pores, and delamination still need to be detected using non-destructive testing methods.
Post-weld heat treatment: For carbon steel pipes, in addition to considering the wall thickness, the medium in use also needs to be taken into account; for example, with wet H2S, heat treatment is necessary regardless of the wall thickness.
Ordinary hydrogen and nitrogen pipelines, free of ammonia and hydrogen sulfide.
Hydrogen-nitrogen mixtures are also commonly encountered in ammonia synthesis plants. When using carbon steel, heat treatment is required, as residual stresses and high hardness in the welds resulting from lack of post-weld heat treatment can make them more susceptible to hydrogen corrosion. Most cases of hydrogen corrosion occur in the welds and heat-affected zones, rather than in the base material.
For high-pressure hydrogen-nitrogen mixtures used in ammonia synthesis, where the wall thickness exceeds 19 mm, post-weld heat treatment is required in accordance with standards such as 50235. I don’t have a hydrogen-ready condition here; the pipeline is large, with a DN1200 diameter, and it’s short and straight.
This post was last edited by Yimi on 2015-12-24 09:36. Heat treatment is also required for synthesis gas compression pipelines (with wall thickness less than 19 mm). Hydrogen corrosion does not necessarily occur only under critical conditions; it can be quite dangerous when the weld quality is poor and residual stresses are present. Short and straight pipes are also prone to stress concentration. Regarding the accident you mentioned, I think it’s necessary to determine the exact cause and conduct a thorough accident analysis. And just talking about the material choice, for a hydrogen-nitrogen mixture, I personally wouldn’t use 3091 – hydrogen molecules have too high permeability. Another point: if by \"pipe\" you mean resistively welded pipes, their use at temperatures above 200°C is not recommended, as specified in clause 7.1.8 of SH/T3059.
The accident analysis result was: poor welding quality. The construction party bears primary responsibility. Design bears secondary responsibility.
According to TSG D0001-2009 Code for Safety Inspection of Pressure Piping – Industrial Piping, Article 86… For pipe weld joints that have not undergone non-destructive testing, the installation unit shall also be responsible for their quality. The direct cause is lack of penetration. Of course, even if they all followed the design requirements, it doesn’t mean that nothing will go wrong. However, as a result, the blame for the issues fell mainly on the installation unit, and the design also suffered a great deal.