where expertise comes together - since 1996 -

# The Largest Open Access Portal of Nondestructive Testing (NDT)

Conference Proceedings, Articles, News, Exhibition, Forum, Network and more

where expertise comes together
- since 1996 -
 2549 views
Technical Discussions
Stan
NDT Inspector,

Stan

NDT Inspector,
Joined Jan 2009
31
22:02 Feb-25-2010
Code Case 2235 question

Can someone help me out.
When using Table 2 in CC2235-9, what is the method on handling indications with a measured aspect ratio of greater than 0.5? Do you use the a/t for an aspect ratio of 0.5, is it am automatic reject, or is there another way of dealing with these indications?
Stan

Ed Ginzel
R & D, -
Materials Research Institute, Canada, Joined Nov 1998, 1286

Ed Ginzel

R & D, -
Materials Research Institute,
Joined Nov 1998
1286
14:49 Feb-26-2010
Re: Code Case 2235 question
In Reply to Stan at 22:02 Feb-25-2010 (Opening).

Stan;
The table for CC2235 first has you assess "a" as a percentage of the wall thickness. The aspect ratio applies to the allowed length "l" after "a" has been determined. This means that the maximum flaw vertical extent determines the allowed length. Because the table plots "a"/"l" it indicates that lesser lengths make greater aspect ratios. Clearly a flaw that has the maximum allowed vertical extent will have a larger value of aspect ratio if the flaw length is less than the maximum allowed; so indeed, the aspect ratio of 0.6 is possible...but it is acceptable because the length does not exeed the maximum allowed.

emil shavakis
emil shavakis
16:21 Feb-26-2010
Re: Code Case 2235 question
In Reply to Ed Ginzel at 14:49 Feb-26-2010 .

So I found this flaw using UT. Its at the end of a repair in a 1.1 T vessel. Im pretty sure its slag running vertically up the end of the repair groove. Its a " is pretty close to 1 but its "l" seems to be only .1 - a tall but skinny flaw. So my aspect ratio is 9.5. And since I dont come anywhere near the 4T length, I can accept a 95% thru-wall flaw  right?

Ed Ginzel
R & D, -
Materials Research Institute, Canada, Joined Nov 1998, 1286

Ed Ginzel

R & D, -
Materials Research Institute,
Joined Nov 1998
1286
18:06 Feb-26-2010
Re: Code Case 2235 question
In Reply to emil shavakis at 16:21 Feb-26-2010 .

Emil: It is unfortunate that NDT people are being expected to carry out engineering functions. CC2235 is filled with inadequacies and I am assured by committee members that these will be addressed. However, until then, I think that the traditional fracture mechanics protocols should be observed, hence the proviso in my explanation that we calculated the maximum allowed flaw height as the first step. Maximum "a" in the tables is had for the item in the last row where the aspect ratio is indicated as 0.5 (the maximum length of 4T that you mentioned is only at the other end of the table where the flaw has the smallest height). Flaws greater than that height are (in traditional fracture mechanics) not allowed. In codes like API 1104 and CSA Z662 a maximum flaw height (typically 50% wall thickness for buried defects) so you will see the ECA curves with a straight line drawn back to 0mm length from the minimum allowed length for the maximum allowed height. CC2235 uses a more conservative maximum height than the pipeline industry but the concepts are built on the same principles.
I personnaly would not feel confident accepting a 95% through wall flaw as an NDT technician. This is my "understanding" of how the CC2235 was to be used, it is clearly not a "Ruling".

Do you think that a short 95% through wall flaw would be acceptable?

There is a process in ASME to answer such questions and our discussions here are not the place to get formal rulings.
ASME has a process whereby such items can be ruled on and you can find the instructions in each of the Sections under the Mandatory Appendix "Submittal of Technical Inquiries to the Boiler and Pressure Vessel Committee".

Your question as to the acceptability of the 95% through wall flaw would be published in the Interpretations publications of ASME. Will you be submitting the Inquiry?

emil shavakis
emil shavakis
00:20 Feb-27-2010
Re: Code Case 2235 question
In Reply to Ed Ginzel at 18:06 Feb-26-2010 .

Ed,
Thanks for your response. I come from many years of the "nuclear" Section XI side of ASME and have long worked with these criteria and all this has been well vetted in that arena. In my experience the maximum intended subsurface flaw, considering 2a, is at the highest a/l of .5 and could be as much as 28.6% thru-wall and accepted. Longer flaws with a lower a/l are subject to rejection at the lower assigned depth, according to as-measured a/l of up to 4T. If a flaw failed at that level we turned to LEFM analysis to IWX 3600 and let the engineers treat the flaw accordingly, subject to operational restrictions and reexamination. We would also bring any flaw we were accepting outright forward to the ANII and regulator for address. It is my personal opinion that ASME construction Codes have stopped short of analytical acceptance since unlike nuclear there is no inservice control. An ASME vessel (or pipe) needs to have a certain quality for unfettered operation after stamping.

I am troubled by the reading of current Code as an "and" statement as opposed to the more accepted "or" - a flaw is rejectable if either the thru-wall for the established aspect ratio were violated the flaw is rejected OR the length exceeds 4T. Unlike the 2235-9 Code case there was no flaw length limit in service.

I have recent experience with persons with a similar philosophical look at CC Table 1, and ASME has ruled (interpreted) that a flaw is rejected if it exceeds the Table stated "a" OR "l".

Ed, you are a very highly respected NDT expert an I thoroughly enjoy your research, and generally agree with your opinions, but know that many people take your word as official and "Code". As you advised me - please leave Code interpretation to the Code people as a body.

I do agree that is a giant leap for your average level II (or III) to adapt to the CC criteria. The Code needs to be more clear and someone, Code or otherwise, needs to provide purposeful training on the subject.

Emil

Nigel Armstrong
Engineering, - Specialist services
United Kingdom, Joined Oct 2000, 1096

Nigel Armstrong

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1096
07:20 Feb-27-2010
Re: Code Case 2235 question
In Reply to emil shavakis at 00:20 Feb-27-2010 .

I have a copy of an ASME paper - "Technical basis for ASME Section VIII Code Case 2235 on ultrasonic examination of welds in lieu of radiography - Rana, Cowfer, Hedden and Boyce" -2000?

Flaw Evaluation Examples

At the time this paper was written, l max was 0,18" rather than the current 0,25". Allowable a/t values were the same as present.

Surface flaw in 0,75" thick vessel. Table 1

Rana postulates a surface flaw, l = 0,15", a = 0,102" (How that was determined!!!)

Using Table 1 for surface flaw material 0,5" to 1,0"

1. a/t = 0,102/0,75 = 0,136
2. l = 0,15"

Flaw is unacceptable since a/t is greater than allowable limit of 0,087

From this I read that Table 1 is an OR statement when a/t exceeds the limit of 0,087, i.e irrespective of length, l

However how does the OR apply where a/t < 0,087 BUT l > 0,15". it cannot apply unilaterally since 2235-9, (i) (4) Flaw evaluation and acceptance criteria, (c) Subsurface flaws states "Flaw length l shall not exceed 4t." Perhpas it does apply unilaterally to surface flaws as these are not covered elsewhere in 2235.

Certainly I agree with both Ed and Emil that this is really pushing the limits of knowledge not only for Level II or III technicians but also to all others interested in proper Code application of accept/reject criteria - and most have no formal training for CC 2235 application. Most people flip its pages, see the defect interaction rules diagrams and decide its not for them!

emil shavakis
emil shavakis
21:14 Mar-01-2010
Re: Code Case 2235 question
In Reply to Nigel Armstrong at 07:20 Feb-27-2010 .

My opinion -

Long flaws - I don't think that hiding the 4T statement in the text is wise but it is in there. 4T takes over when a long (low aspect ratio) flaw is found. E.g. a 7% undercut in a 1.1" vessel that is measured at 6" long. Though acceptable to the table, it exceeds 4T, and becomes rejectable.

Ed Ginzel
R & D, -
Materials Research Institute, Canada, Joined Nov 1998, 1286

Ed Ginzel

R & D, -
Materials Research Institute,
Joined Nov 1998
1286
00:20 Mar-02-2010
Re: Code Case 2235 question
In Reply to emil shavakis at 21:14 Mar-01-2010 .

The lack of illustration of the use of the traditional curves used in fracture mechanics has made no end of difficulties for people trying to use CC2235. The application of fitness-for-service criteria has been around for a long time (the standard BS PD6493 goes back to the early 1980s). But readers of NDT.net have seen them regularly in several papers such as one by Jan van der Ent (http://www.ndt.net/article/apcndt2006/papers/71.pdf). I uploaded Figure 6 from this paper. There you see the normal treatment of flaw sizes using the same principles as are described in CC2235 (in fact Rana et al used PD 6493 as their reference for computations). The green line has a maximum allowed flaw height (dashed green horizontal line in the upper left side). Any flaw higher is allowed NO length. There is also a maximum allowed length (dashed green vertical line) indicating no flaw of any height may exceed that length. In pipeline applications the values are calculated for every project but ASME CC2235 provides a generic table so is much more conservative since it must assume the weakest (least tough) material is used.
The curve makes for easy flaw assessment. Inside the "box" made by the green line the flaws are acceptable. Flaws above or right of the green box are unacceptable.

emil shavakis
emil shavakis
01:56 Mar-02-2010
Re: Code Case 2235 question
In Reply to Ed Ginzel at 00:20 Mar-02-2010 .

As I undertsand ASME, at least Sections I & VIII D1, fitness-for service is not the standard for construction. Fitness-for-service and the analysis that goes with it is not even mentioned in Section VIII. Code rules are minimum "rules".

ASME VIII makes vessels for many applications - kitchen kettles through refinery cokers, air tanks though ammonia storage, and the vessel is assumed to be as good as stamped on its nameplate and as reported in the Manufacturer's Data Report and nothing less. Until the Code Cases the acceptance criteria was quaitive and primarily based on workmanship and images in a film. There is no room for ECA to supercede the Code UNLESS there is agreement that the vessel is a SPECIAL case acceptable to the owner-user and the jurisdictional authority (state, province or country).

The curve shown by Ed is a fairly common shape curve of a common shape and is similar to those used in ASME preservice and inservice nuclear (Section XI (IWB3600)) justification but only when the basic TABLES fail, but not as a new construction basis. The curve typically prevents catastrpihc collapse, but not leaks - it is not normally appropriate for a long-lived vessel with no inservcie inspection mandate.

The acceptance criteria of basic Section XI, as passed through to CC2235/181 have margin to anything expected from and ECA. These are the 1st tier of acceptance. To pull out an ECA for a new ASME construction is not typical - save that for later or when the Code provides for a mandatory ECA.

Emil

Product Spotlight

#### TraiNDE UT

TraiNDE UT is a virtual tool associated with a signal database which simulates real inspection con
...
ditions for numerous applications (Type A/V1 block, DAC block, welds and plates).
>

#### Semi-Automated Phased Array Immersion System for Small Composite Parts

Turn-key semi-automated system as an improved and affordable solution for inspection of small comp
...
osite parts. Includes support table, immersion tank, scanner, PA instrument, PC, Analysis software, database, wedge management and other options.
>

#### Compact NDT inspection-heads for measurements with active thermography

The compact inspection head is suitable for thermographic ndt tasks. The uncooled infrared camera
...
is specially developed for NDI-tasks and offers a thermal sensitivity until now known only from thermal imagers with cooled detector. All required components and functions are integrated into the inspection-head. You will only need an ethernet cable to connect the sensor with the evaluation system.
>

#### NDT Master Lecturer

In the program both university professors and practitioners will give lectures, which guarantees the
...
oretical depth and practical inside. Academic Director: Prof. Dr. Christian Boller
The following lecturers are not complete: Prof. Tadeusz Stepinski, Prof. Wieslaw Staszewski, Prof. Frank Walther, Prof. Giovanni Bruno, Prof. Gerd Dobmann, Prof. Philippe Guy
>

Share...
We use technical and analytics cookies to ensure that we will give you the best experience of our website - More Info
Accept
top
this is debug window