ASME b31.3 and AWS 1-1

[Naif1981]

Hi gents
let us say that i am preparing a contractual documents, and the list of scope of work. one of the list Items is welding
I wrote: Welding of all pipes connection as per ASME B31.3 and AWS 1-1.
my question is:
1 st: Is the contractor must do the welding including NDE either (UT, RT)

(I will not mention any NDE or NDT in the scope of work)

2nd if NDE not included, if the welding done as per AWS or ASME B31.3 that means will be no welding points will be defected? correct me if wrong

what I am looking for is the procedure of welding on ASME B31.3 or AWS can ensure there will be not failure in welding points after NDE?

the material of the pipes is A53 grade B size (2" to 8")

 

[BigInch]

They will ONLY have to do the NDE/T that is specified in your referenced codes (prob 10%-20%, but you read your design code). No more. It can be beneficial to specify more than what the code requires in many circumstances, or in some locations, such as within dense cities, water crossings and sensitive environmental areas.

It can be VERY beneficial to NOT permit the welding contractor to also do the inspection or testing.

You are wrong. The number of bad welds detected is an inverse function of welding quality and a direct (but not linear) function of the number of welds inspected. Even if you inspect all welds, you still may get failures, because not all defects are detected. If you do the minimum inspection, you will detect the very minimum amount of defects. Although a 20% inspection rate is usually enough to get a high confidence level of 99.5% passing on a long pipeline, but you still might have 1/2 percent of tested joints being false positives; bad welds passing the inspection. On a short pipeline with the number of joints below 20,000 or so, the sample size percentage will have to be higher for the same confidence level.





**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Naif1981]

Thanks Biginch,
What i understand is even the contractor follow procedure of welding as per the codes i will get a defected weld but within 5% of total welded points?

What about 80% failure I think if I got a 80% failure ? i am sure the contractor not follow the codes right?

 

[BigInch]

The codes do not have welding procedures.

With testing, you will find some bad welds, but not all.

With an appropriate welding procedure, you will still get bad welds, maybe 3 to 5%. Most of those bad welds will be found in the testing process and will be repaired, or cut out, retested and eventually passed.

If you have 100,000 welds to make, you should find most all defects 99.5%, if you actually test 20%, but you will never find about 1/2%. That would be about 500 bad welds probably not found.

If you get 80% failure, you have the wrong welding procedure and/or very, very bad welders. With a testing procedure testing only 20%, but with 80% of all welds failing and with 100,000 total welds, you will only find 16,000 bad welds, but you obviously expect to have 80,000 bad welds. No testing procedure will work for that high a failure rate, unless you test 100%.

In that case, EVEN IF the contractor FOLLOWED the code and if the code only required test 10%, you would test 10,000 welds and find 8,000 bad welds and repair them, but you would NOT FIND 72,000 bad welds. RESULT: YOU FOLLOWED THE CODE, BUT MADE 72,000 BAD WELDS THAT WERE NEVER DISCOVERED.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[mrsongs]

We always do a pyramiding method for weld tests. We test approximately 10%, or whatever % we decide on for that area. We pay for our own NDQ. Then, for any failure the company must repair it at their expense, retest at their expense and shoot one more weld at their expense. That way for every failure you shoot two. If quality control is poor then they will will be found and we will find more as they fail more.

 

[BigInch]

If they have a bunch of failures you'll be doing 2x2x2x2x2x2x2x2x2x2x2x2 .... well... let's just day you'll probably still be doing NDT well after the job is finished, packed up and the lights turned off.

**********************
"The problem isn't working out the equation,
its finding the answer to the real question." BigInch

http://virtualpipeline.spaces.live.com/

 

[rneill]

Keep in mind that code requires "progressive examination" which means that as you find bad welds you have to shoot additional welds and in effect your examination rate goes up. If you read the sections on progressive examination, you'll find that it doesn't take to many bad welds before the code actually mandates 100% radiography (for the designed lot anyway).

You might want to get a copy of Pipe Fabrication Institute ES48 as I believe it provides some additional clarity around the requirements for random examination.

I have a flowchart on progressive examination somewhere and if I can find it, I will try to post it.

 

[stanweld]

Naif1981,
First off, you should not have referenced AWS D1.1 if the pipeing is designed, fabricated and inspected per ASME B31.3. ASME B31.3 requires welding procedure qualification per ASME IX, subject to the additional limitations included within B31.3. Second, you need to specify "normal service" or "Cyclic service" to define required NDE frequency. AWS D1.1 does not apply to piping unless that piping is fabricated to make a steel structure, not to provide transportation of fluids.