From your input it appears that the failure could be due to cyclic loading of PSA operation. Please check the support details of the particular line.Since the line operating temperature is less you can carry out radiography of remaining weld joints in the circuit to detect the presence of fatigue cracks.Weld harness will also give some clue about the reasons for failure.If you have failed sample photograph kindly send to me for further opinion. sriramchandra@rediffmail.com regards,

MVN Sastry PSA outlet in hydrogen service - with cyclic loading and swings - can fail due to : fatigue and also due to low temp hydrogen embrittlement. For fatigue failure (low cycle fatigue), the :"T" uneven size - 8" x 4" will see uneven stresses - even if well supported by piping design standard. Since you have experienced sudden cracking, the most probable reason is low temp hydrogen embrittlement - which can happen at ambient temperature if the original fitting is given with high hardenss. Action plan can include : 1. In-situ metallography of the "T" joint . If it is due to HIC (which is quite possible) you could expect predominant mode to be inter-granular crackng. 2. NDT : WFMPT of the other joints - (outlined by Debasis) would be faster and reliable for surface crack evaluation ( to assess other welds and joints with likely restraint as in this ) 3. RT will be time taking but is a good alternative for evaluating other areas in the same product out let pipe (suggested by Krish) and also for future record and monitoring. 4. Since this is 4" line, UFDT needs a careful prior calibrated sample to evaluate. Normal 4" Sch 80 thickness (generally used for this pressure and temperature in such pipelines) will give erroneous interpretative results. Before attempting UFD, on such low thickness joints, use a prior calibrated sample with a low thickness piece of the same material. 5. Repair of plain carbon steel is not so complicated once you have evaluated by in-situ metallography and with NDT of other joints in the pipeline. Trust this will help you. C.V.Srinivasan Nishi Engineers Pvt Ltd India Aug 2,2007 E-mail: nishi@vsnl.com >Hai, > >In our Hydrogen unit, recently a PSA outlet product pipe >line (8" header with 4" branches) developed a >circumferential weld crack over approx. 75% weld length on >4" pipe in the HAZ at the Tee weld joint( Uneaqual Tee 8" >x 4"). The pipe line service is Hydrogen at ambient temp. > >This line is subjected to a cyclic loading with pressure >swings from 0-320psig in 6minutes. The pipe is API5L GrB . >The service life is about 7 years so far. Though the >probable reason appears to be due to fatigue >cracking/original weld defects/HIC, I wish to know further >on this failure, in case anybody experienced similar failure >in the past. > >I am looking for the following things to be clarfied: >1.Most Probable reasons for such sudden crack of weld >2.Any advanced NDT techniques available to ascertain and >ensure mechanical integrity of other welds onstream (while >in service). >We proposed for Radiography and UFDT. > >Regards, >Sastry