B31.3 RT Requirement

By pipewelder_1999 (****)

Date 03-12-2003 02:07

Start with Para 340 and you will come to the subject paragraph and travel through many other useful paragraphs.

G Austin

By NDTIII

Date 03-12-2003 13:08

I'd like to elaborate on Mr. Austins reply if I may. Don't feel alone. There's always alot of confusion when it comes to this issue. So here is how I interpret B31.3.

If you look at Paragraph 341.4.1(b) states that "not less than 5% of circumferential butt and miter groove welds shall be examined fully by random radiography or by random ultrasonics. They shall be selected to to ensure that the work product of eack welder or welder operator doing the production welding is included. They shall also be selected to maximize coverage of intersecting long seams. When there is an intersecting long seam, 38mm or 1 1/2 in. shall be included in the examination".

The actual extent of NDE should be included in at least one of the project documents. Whether it be the purchase order, the contract, the design package, the project specs., a letter from the client or even the minutes of the kick-off meeting. But the actual extent should be agreed upon by all involved parties "prior" to the commencement of work. This is very important!

B31.3 is ambiguous as to what they mean by 5% of production welds. Is that "ALL" production welds. It should be agreed beforehand as to the percentage of what. % of Isometrics? % of systems? % of daily production? What percentage? 5%? 10%? You also have to consider the difference between field welds and shop welds. Carbon steel systems compared to stainless steel systems and any other exotic materials. You may want to increase NDE on these materials.

Of course it would be ludicrous to examine 100% of all welds or even 10% of each welder. That is more NDE than I care to think about.
You also have to consider the progressive sampling requirements when you have a repair weld. This is clearly spelled out in paragraph 341.3.4.

Another thing you want to look at is Table 341.3.2.
This table addresses the flaws that ASME is concerned with. The first page tells you which flaws are relevant for each method of NDE. These are the defects and the terminology you must use.

For example, B31.3 does not address burn through. So if you see it on a radiograph, what do you do? Ignore it? My opinion is that you evaluate it as root concavity unless your approved RT procedure does address it. That is possible.

You should get a copy of B31.3 and get thouroughly familiar with it. The NDE coverage can get quite confusing on a large project. But yes, par. 340 is a good place to start.
Hope that helps. Feel free to ask for more info if you need it.

By swift

Date 03-12-2003 23:29

I re-read the customer requirements today and was pleasantly surprised to read in their specification, "shall be visually inspected". 4 word that can make a world of differance.
Swift

By NDTIII

Date 03-15-2003 08:22

Yes all welds get visually inspected, but that does not relieve you from the NDE requirements.

By insp76

Date 03-15-2003 16:29

Swift, if your contracted to weld and fabricate according to B31.3 then you must satisfy the minimum requirements of that code, or if your repairing or revising piping that was built to B31.3 this also applies.The owner/client specs shall meet or exceed these minimum requirements.If in the case the owner/client of the equipment wants to wave some of the minimum requirements ,the owners engineer must sign off before you procede with the job. If this is not done the welding/fabricating contractor will be working out of compliance with the code and will be held liable for any failure of the piping.The statement "shall be visually inspected" means all fabrication and welding shall be visually inspected.In this statement they are in fact exceeding B31.3 because this code states only 5% is required under normal fuid sevice to be visually inspected unless it is classified to be severe cyclic or high pressure in that case 100% visual is required along with other stringent nde requirements.I suggest you get real familiar with the code book and most of all talk and ask questions to exprienced inspectors before you react to a situation,confusion on your part could cause a problem on the job.Good luck

By joe valdez (*)

Date 03-15-2003 17:26

Hi Swift,
In my opinion most commentators were correct ,and let me share with you what we encounter on our refinery construction project here in FT Mc Murray ,Alberta Canada.Halliburton Kellogg Brown & Root is the main contractors operating under EXXon Mobil stringent QA/QC regulations
on the Syncrude /UE-1 Tar sands project ($ 5.7 Billion0, the largest capital cost project in the world as of today.
All piping are classified according to line classes( Bechtel System) irrespective if they are materials in Group1 class 1 such as SA 53B or SA 106B or S1 group 1 materials such as API 5LB. Line classes are used in accordance with what type commodity it transport,wich then brings back to your question regarding minimum % you should flag for RT.
First let us assume welder XYZ starts welding for you today on XYZ line class,we flag and shoot his first 2 welds irrespective of line classes, thereafter we are obligated to then RT a minimum of 5% per line class he/she welds (randomly) thereafter.
Should He/she fail an RT we then require that welder repair weld (subject to re shoot)and then we will shoot 2 following welds as tracers.
If welder than fails any tracer the process re peats itself once.Should welder fail twice we remove the individual and recommend retraining.
As you can see 5 welds and one rejection is not a good passing rate (20 % failure) for XYZ welder,it then makes us wonder how qualiified this individual is..
Regards
Joe Valdez