where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
Flaw sizing with UT contact transducer in 2" steel
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

IDEKO S.Coop
We provide Precision, Quality, Reliability and Productivity in manufacturing processes among our knowledge of Inspection and measurement technologies

1346 views
07:35 Apr-13-2005

Jeff Medford

NDT Inspector
Nucor Steel,
USA,
Joined Apr 2005
6
Flaw sizing with UT contact transducer in 2" steel

I am with a company that had a claim against us. They were using 1/2" 2.25Mhz transducer inspecting through 2" steel plate Long. Wave. They said they were finding defects that were sized 2 to 3 mm. Now my question is can you use UT to legally measure a discontinutiy?
I have used this all the time but I was taught that it was not acurate by any means.

I do thank anyone for the time to answer my posting

Jeff Medford



 
08:14 Apr-14-2005

S.V.Swamy

Engineering, - Material Testing Inspection & Quality Control
Retired from Nuclear Fuel Complex ,
India,
Joined Feb 2001
782
Re: Flaw sizing with UT contact transducer in 2" steel ----------- Start Original Message -----------
: I am with a company that had a claim against us. They were using 1/2" 2.25Mhz transducer inspecting through 2" steel plate Long. Wave. They said they were finding defects that were sized 2 to 3 mm. Now my question is can you use UT to legally measure a discontinutiy?
: I have used this all the time but I was taught that it was not acurate by any means.
: I do thank anyone for the time to answer my posting
: Jeff Medford
------------ End Original Message ------------

Dear Jeff,

While sizing of defects (strictly speaking they should be called discontinuities) by UT or any other NDT method is a little approximate for the reason that almost all NDT methods are indirect, a close approximation can be obtained. For example, DGS scales do provide for size estimation. Even without that, suitable Flat Bottomed Holes drilled into the plate will enable a good estimate of the size. The question which needs to be asked and answered is: What size discontinuities were cause for rejection? The answer depends on the type of discontinuities. In Steel plates, for example, it is not uncommon to find small laminar discontinuities. Different codes and specifications allow discontinuities of various sizes.

It is almost impossible to produce steel plates or forgings with zero discontinities. But the type of discontinuities (round, planar, orientation etc.) is important.

Swamy


 
00:27 Apr-14-2005
Ed T.
Re: Flaw sizing with UT contact transducer in 2" steel ----------- Start Original Message -----------
: I am with a company that had a claim against us. They were using 1/2" 2.25Mhz transducer inspecting through 2" steel plate Long. Wave. They said they were finding defects that were sized 2 to 3 mm. Now my question is can you use UT to legally measure a discontinutiy?
: I have used this all the time but I was taught that it was not acurate by any means.
: I do thank anyone for the time to answer my posting
: Jeff Medford
------------ End Original Message ------------

UT can be used and is used all the time to size flaws. However using a 1/2" 2.25 MHz. transducer and sizing 2-3mm defects is a challenge unless you compensate for your beam spread.
I would recommend measuring your beam spread in a 2" thick plate with flat bottomed holes approximately 2-3mm in diameter drilled to the same depth of the defects that were found.
You may even want to try using focused probes, but still using the flat bottomed holes.
Then measure the defects.
If you don't know your beam spread at the depth of the indications, then the measurements could be and probably are unreliable.


 
01:48 Apr-14-2005

Doug Breeze

Director, Consultant; R&D
Inspection Software Limited,
United Kingdom,
Joined May 2000
20
Re: Flaw sizing with UT contact transducer in 2" steel ----------- Start Original Message -----------
: I am with a company that had a claim against us. They were using 1/2" 2.25Mhz transducer inspecting through 2" steel plate Long. Wave. They said they were finding defects that were sized 2 to 3 mm. Now my question is can you use UT to legally measure a discontinutiy?
: I have used this all the time but I was taught that it was not acurate by any means.
: I do thank anyone for the time to answer my posting
: Jeff Medford
------------ End Original Message ------------
Jeff,

1. Contractul Issues. Not sure who's sueing who here, but contractually (if not legally), both parties need to assure themselves that they were working in accordance with their pre-agreed contracts, or with recognised codes of practice, if they weren't agreed beforehand.

2. Choice of Transducer. The 0.5", 2.25 MHz transducer possesses a huge beam-spread, low sensitivity to small defects and unreliable defect dimensioning especially at longer path lenghts. If it's a single crystal, the near surface capabilities are likely to be poor. If it's dual-crystal the beam spread (and sensitivity) could vary depending on its configuration, and the path length to the defects. In either case it's probably a poor choice of transducer for 2" steel. A larger, 20mm 5MHz single-crytal transducer would possess a more definable beam spread, or a 20mm 5MHz dual-crystal focussed transducer would optimise the chance of defect detection, and more accurate measurement. Conversely it may increase inspection time because of the need for more scans with the narrower beam width.

3. Dimensioning: You must differentiate between defect detection and defect measurement, although both are compromised by the choice of transducer in this case.
Defects much smaller than 2-3mm can be easily detected with an appropriate transducers. But in the 'industrial' arena, measurement of small defects (repeatability, accuracy, precison etc.,) gets murky... in manual UT, too many variables are at play to make such measurements (estimates, is a better term) reliable. The DGS method mentioned by SV Swamy has much merit, but it's reliable only if the defects are planer in nature, parallel to the surface, and with similar reflectivity to a machined flat-bottomed-hole. And the result is defined by area, not X-Y dimensions.

4. Alternatives. If there's a legal issue to be resolved or perhaps just the technical issue, you could undertake an automated very high-resolution UT map of a representative area of the plate. Careful selection of transducers, and good surface prepartion should yield results at least an order of magnitude greater than manual UT.

Or for a fraction of the cost, just section some plate to verify what defects are been found.



 
03:44 Apr-14-2005

Ed Ginzel

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1185
Re: Flaw sizing with UT contact transducer in 2" steel ----------- Start Original Message -----------
: : I am with a company that had a claim against us. They were using 1/2" 2.25Mhz transducer inspecting through 2" steel plate Long. Wave. They said they were finding defects that were sized 2 to 3 mm. Now my question is can you use UT to legally measure a discontinutiy?
: : I have used this all the time but I was taught that it was not acurate by any means.
: : I do thank anyone for the time to answer my posting
: : Jeff Medford
: Dear Jeff,
: While sizing of defects (strictly speaking they should be called discontinuities) by UT or any other NDT method is a little approximate for the reason that almost all NDT methods are indirect, a close approximation can be obtained. For example, DGS scales do provide for size estimation. Even without that, suitable Flat Bottomed Holes drilled into the plate will enable a good estimate of the size. The question which needs to be asked and answered is: What size discontinuities were cause for rejection? The answer depends on the type of discontinuities. In Steel plates, for example, it is not uncommon to find small laminar discontinuities. Different codes and specifications allow discontinuities of various sizes.
: It is almost impossible to produce steel plates or forgings with zero discontinities. But the type of discontinuities (round, planar, orientation etc.) is important.
: Swamy
------------ End Original Message ------------

Swamy:
First a semantic item: Instead of "discontinuities" which could include ANY sort of indications I prefer flaws. Once determined critical to service they become defects.

As to sizing them, I think all the explanations provided are valid considerations. Another consideration is which dimensions are sized. The AVG (DGS in English) system is based on ideal reflectors of an assumed uniform shape. Ermolov derived equations to make equivalent relationships but there too were ideal reflectors.

At the stated 2.25MHz, even in shear mode, the temporal resolution of a single cycle pulse will be hard pressed to resolve the 3mm (vertical height) using back diffraction!

Length assessments are typically made using the 6dB drop but as others have stated, that is an approximation highly dependent on the flaw shape. For flaws having lengths less than the beam width divergence effects oversize using this method. Sizing "policies" built into the test procedure could at least ensure a consistent treatment of sizing.

All measurments are prone to error. Even when you think it is easy. Give 30 people a small block of steel and an old mechanical Vernier caliper and ask them to provide the 3 dimensions. Probability states that you will get a gaussian distribution of sizes of each dimension.

This scatter in sizing estimates is a now common study in NDT. So I agree with Swamy, flaw sizing is not much more than a reasonable approximation when using UT.

Ed



 
02:32 Apr-15-2005

N.Kuppusamy

Consultant, Level-III
United Testing Co. Pte Ltd,
Singapore,
Joined Jan 2003
13
Re: Flaw sizing with UT contact transducer in 2" steel ----------- Start Original Message -----------
: : I am with a company that had a claim against us. They were using 1/2" 2.25Mhz transducer inspecting through 2" steel plate Long. Wave. They said they were finding defects that were sized 2 to 3 mm. Now my question is can you use UT to legally measure a discontinutiy?
: : I have used this all the time but I was taught that it was not acurate by any means.
: : I do thank anyone for the time to answer my posting
: : Jeff Medford
: Dear Jeff,
: While sizing of defects (strictly speaking they should be called discontinuities) by UT or any other NDT method is a little approximate for the reason that almost all NDT methods are indirect, a close approximation can be obtained. For example, DGS scales do provide for size estimation. Even without that, suitable Flat Bottomed Holes drilled into the plate will enable a good estimate of the size. The question which needs to be asked and answered is: What size discontinuities were cause for rejection? The answer depends on the type of discontinuities. In Steel plates, for example, it is not uncommon to find small laminar discontinuities. Different codes and specifications allow discontinuities of various sizes.
: It is almost impossible to produce steel plates or forgings with zero discontinities. But the type of discontinuities (round, planar, orientation etc.) is important.
: Swamy
------------ End Original Message ------------

Deal All,

I agree with others. Defect sizing by manual method is not accurate due to many parameters already discussed in this thread. DGS may give equivalent defect size. Since UT and most other NDT methods are qualitative, we can not come out with precise sizing with reliablility/repeatability.


What is the test specification/procedure requirement? Is it says that all discontinuties greater than 2mm equivalent flaw size are to be reported or rejected. Is your test procedure is capable of finding 2mm equivalent defect size? Is it proven by any other means?

If the acceptance criteria rejects 2mm equivalent defect size and you accept it (or missed it), then only the question of claim araises.

In case if 2mm equivalent defect size is reportable but not rejectable, then you may give explanation for not reporting. Because in many places, enough time is not given to UT technicians to report all their findings.


Regards,

N.Kuppusamy


 
07:39 Nov-23-2005
singita
singita test. ingore it. (ceevbuf)


 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|