B31.3 Random RT

Date 03-21-2005 04:14

The welds you are performing 5% RT on would only require you to RT 45 welds. However, you must include the work of each welder in that random sampling. Even if it is only 1 weld.
Different codes, different requirements.
If you were to RT 100% of the welds that fall under B31.1 and try to use them to cover the B31.3 welds, you would actually be excluding the welds that are covered by B31.3 from the random sampling that is required by B31.3.
If the 100 welds fall under B31.1 and the other 900 fall under B31.3, each code would be applied seperately. Same thing as PWHT. If you have a trunkline coming into a plant and it is 25mm thick, once you reach the isolation valve, the weld on the B31.3 side of the valve would require PWHT, but the weld on the pipeline side of the valve would not.
But of course that would depend on what your client would accept.

By jon20013 (*****)

Date 03-21-2005 15:03

Shane; I used to be quite proficient with both B31.1. and B31.3 but admit it's been a number of years since I've had to deal with either on a routine basis. I seem to recall a few things however need a bit more information (and need to try to locate my current version of both codes!). First, you didn't mention an operating temperature or pressure for the B31.1 works, from my memory there are certain categories which only require VT, I am taking your statement as meaning you have already evaluated and determined that RT is in fact required for these welds but you may want to double check it. Second, you mention Category NFS & D for the B31.3 welds; I seem to recall that Category D permits an inservice leak test ~ have you looked into that? Finally, if hydro is not possible, what about pneumatic testing? I used pneumatic testing extensively while on contract in South Korea as the systems we were working were all fuel systems and didn't want the possibility of any remaining water in these systems since they were intended for US Military aircraft. Just a few thoughts.

By Jim Hughes (***)

Date 03-21-2005 18:31

Hello Shane, the answer to your question is no. You have to apply the NDE requirments to each code. If B31.1 requires 100% x-ray, that has no affect on the 5% required for B31.3 Also good job on your reject rate. Those are great numbers.
Jim

By chall

Date 03-21-2005 19:16

The answers provided are quite insightful.

One thing that comes to my mind would be a simple request to the client. Point out to them that you are using the same welders for all projects. The intent of 5% random RT is to make sure you don't have a welder who is not performing up to standard. If you point out that 100% of that welder's welds are subject to RT on the other system, and point out the success rate, they may be agreeable to relaxing the 5% random requirement in some instances. If it was my decision to make, and my money, I would approve the request based on your performance. If it was your money to be saved, I may decide otherwise. This may have an impact on the client's answer, so if there is a $ savings, maybe you should offer to share some in a cost reimbursment.

There is obviously no guarantee, but it never hurts to ask.

Charles

By Shane Feder (****)

Date 03-21-2005 21:33

Gentlemen,
Thank you all for your response.
I had a meeting with the clients rep after putting this posting on the forum yesterday morning.
We looked at how many welds had been performed,how many had been radiographed and we discussed our interpretation of the intent of the code regarding random NDT.
At the present time all 22 of my welders (10 was just easier for my hypothetical question) have a ratio of welds performed to welds radiographed of over 10% (it ranges from 17 -25%).
The clients rep has agreed that if the percentage stays above 5% for all welders (this includes the 100% systems) for the duration of the project they will accept my proposal.
Once again,thanks for the input,
Regards,
Shane

By BamaDave (**)

Date 03-22-2005 14:22

Shane,
Another important thing to consider when determining the amount and selection of your random sampling it that the percentages to be radiographed cannot be considered for welds that don’t exist. Meaning, you cannot RT welds and count them towards a batch or lot for welds that haven’t been completed. Many projects that I have worked on in the past this has been an issue. Far to often the contractor is attempting to safe guard the project by sampling in this manor. Almost like the fox guarding the hen house. On the front end of the project it is very important to have a pre-project meeting to discuss this, as this can often be an issue. Sometimes it’s hard to explain that 5% of one weld is one weld and once the radiographs have been taken the lot that it is sampled from is closed and new lots from each welder will be developed from that point on. Often times 5% random sampling can actually be much more than 5%.

Regards,

D Payne
ASNT Level III, AWS-CWI,
API 510/570/653/QUTE

By Jim Hughes (***)

Date 03-22-2005 19:32

Shane, I think you could be missing the point of the code. I think you need to re-examin B31.3 341.4.1 to 344.5.2. I could be wrong, and maybe someone could show this to me, but I don't see anywhere in B31.3 that says you can take examinations from other work under a different code (B31.1) and apply it to B31.3 work. Maybe I missed the point of your original question. Let me know if I'm reading this wrong.
Jim

By Shane Feder (****)

Date 03-23-2005 01:42

Jim,
I have spent the majority of my career as the clients CWI, this is my first project as the contractors CWI.I have seen every "short-cut" imaginable and I can assure you I am not looking for an easy option.
Just trying to learn by communicating with fellow inspectors.
This is also the first time I have used the welding codes for power piping and pressure piping on the same project.
My interpretation of B31.3 is that random examination is used in an attempt to ensure the required standard of the welders or welding operators work is maintained.As noted in 344.1.3 (3) " Random or spot examination will not ensure a fabrication product of a prescribed quality level throughout."
B31.1 appears to have a more stringent acceptance criteria than B31.3 NFS so I think it is safe to assume that a welder who is consistently performing work that complies with B31.1 could also be considered to be welding to the standard required by B31.3.
My welders may perform a B31.3 butt in the morning and a B31.1 butt in the afternoon,they have no idea of the different codes they are working to.They are trained to treat every weld as if it is going to be radiographed.
Hypothetically, one welder does 100 butts to B31.1 and they are all (100%) radiographed and comply.He then does 1 weld on a B31.3 line(5%).
Does that mean that I have to radiograph that one weld in order to comply with the B31.3 code.?
As Charles mentioned previously is it up to the client to decide what they will accept.?
As I said before I am not looking for an easy option, purely looking to better understand the code/s.
Regards,
Shane

By chall

Date 03-23-2005 13:43

This is a well reasoned response and anyone who works for a contracting company in similar circumstances may benefit by having a good dialog with the client.

I attended a negotiating seminar that was pretty informative. The best advice passed on is to simply; "ask the question".

To my understanding, with the exception of Jurisdictional (or local) laws, all the requirements of the construction codes may be applied at the discretion of the client's engineering representative. I'm not advocating trying to obtain waivers for inspection/testing or engineering requirements, etc,. just to save money. However, when circumstances allow for some latitude, it's worth pursuing to save schedule impact and possibly some cash.

Charles

By Jim Hughes (***)

Date 03-23-2005 15:12

Thanks for the feed back. I have been working to B31.1 and B31.3 since 1991 and I have never seen the codes applied this way. So thanks for the learning curve. Although I'm still concered about how B31.3 344.6 fits into this. 344.6 says: The welds to be examined shall be selected to ensure that the work product of each welder or welding operator doing the production welding is included.
Jim

By BamaDave (**)

Date 03-23-2005 17:33

Shane,
I totally disagree with the ability to cross apply the percentages when it comes to meeting the referenced governing codes. As far as my interpretation of the requirements are concerned this type of sampling is a code violation. The client or the owner/used typically will not approve violations to the code due to liability issues. Realistically, how can you cross apply percentages from different code governed welds when they should not be considered as part of each other’s specific lots? There is no grey area here so saving money is not something that should be considered when it comes to meeting the code requirements. I hope you obtained some documentation that approving this method of sampling, as it is clearly violation.

Regards,

D Payne
ASNT Level III, AWS-CWI,
API 510/570/653/QUTE

By jon20013 (*****)

Date 03-23-2005 18:49

BamaDave; maybe my words might be a bit harsh here but the first question that comes to my mind is your statement: "As far as my interpretation of the requirements are concerned this type of sampling is a code violation." Interpretations are made by committee so unless you are the B31 Committee you are offering nothing more than an opinion, just like the rest of us. I have been on many, many B31.1 and B31.3 jobs and there are always questions with regard to how the random RT SHOULD be applied. The bottom line for the B31.3 portion of work is this; the work of each welder is to be represented by X% RT for the work that they have performed, furthermore welds which were NOT subjected to RT are considered acceptable if they pass pressure testing ~ (this is from reading the DuPont History of the B31.3 Code ~ for those who do not know, DuPont was the entity that first requested and did all of the original studies for justifying random RT). Since Shane didn't provide us with pressure and temperature details for the B31.1 work it is hard to answer him on that issue. Shane was simply asking the members of this message board opinions, not stating that his way was the only way or the even necessarily the correct way so I think some of the verbiage used in responding to his inquiries has been a little bit unjustified. Have a nice day.

By Shane Feder (****)

Date 03-23-2005 23:16

Hello guys,
Jon,
I have low pressure steam systems that are nominated for 5% RT(B31.1)
Maximum Design Pressure - 180 kpa / Maximum Design Temperature - 190° C with maximum wall thickness of 9.53 mm.
I have water lines that are nominated for 5% RT (B31.3 Cat D)
Maximum Design Pressure - 410 kpa / Maximum Design Temperature - 50° C with maximum wall thickness of 9.53 mm.
I have numerous other process lines (cooling water, fire water etc) that are B31.3 Cat D but require 5% RT.
It is my understanding that all of these examples noted above have only VT as a minimum requirement.RT could be considered surplus to code requirements.
I am also struggling with all the codes I have to comply with on the one project.
I have B31.1,B31.3,AS1210 (Australian Boiler & Pressure Vessel Code)AS 4041 (Australian Pressure Piping Code) and EN 25817 that are all interconnected and all have different percentages of NDT required.
These are the main reasons I approached the client with a request for clarification on random examination.
My primary objective is to assist in the supply of a quality product to the client and to head to the next project with my reputation (hopefully) enhanced.
Best regards,
Shane

By NDTIII

Date 03-24-2005 04:15

It all depends on what the owner is willing to accept. This would include what his insurance company is willing to accept. It's all about liability.
You should apply each code seperatley, but if the same welders are working on both parts of the job and the owner and the authorized inspector feel they have a resonably good product, they can accept whatever they want.
B31.1 requires RT depending on the pressure and temperature. When a system falls under the pressure and temp requirements of B31.1, it requires 100% RT. The 5% comes from B31.3. The category "D" welds are included in the 5% required by code, however most places I've been require the 5% or 10% whatever the case may be for different lots. Maybe of each line class or maybe each system or maybe even each welder. It depends on you contract.

By jon20013 (*****)

Date 03-24-2005 11:11

Shane, thanks or sharing that information. I agree with NDTIII. Often times customers and inspectors struggle with the term "lot" and then come to some conclusion that it applies to both the welder and some particular system. Were you saying that on your project that ASME, AS and EN Standards share jurisdiction within the same system??? That would be a tough one to sort out! Based on what you've mentioned regarding the B31.1 portion regarding pressure and temperature, if it were my project I would cease RT on that part (unless required by AS / EN) and finish it out with VT. I think virtually everyone has agreed your accept/reject percentage is excellent.

Back to your original question towards 100% RT vs Pressure testing, have you gotten any further with that? Although I am recalling from memory and don't have specific paragraph reference in front of me I believe you may find some relief in the Code to do an in-service leak testing for Category D portions. It also sounds like your approach of getting customer concurrence is the best approach under the circumstances as it sounds like your project specifications exceed code requirements (which isn't unusual!).

All in all, it sounds like you're doing a hell of a good job!

By tab_1999 (**)

Date 05-12-2005 13:39

Hi Shane

It appears from the notes below that it would be much simpler to manage the 3 contracts as they are. Separately.

Good Luck!