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 >
TOFD notches
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

LECOEUR ELECTRONIQUE
The Specialist of electronics for ultrasonic testing.

1114 views
06:53 Sep-23-2006

Ed Ginzel

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1208
TOFD notches

I am trying to determine the function of the reference notches as described in the various TOFD documents.
BS-7706 indicates the notch as one of three options (grain noise and side drilled holes being the other options). BS-7706 suggests a near sided sharp pointed notch machined (EDM) to half the wall thickness. The sensitivity based on the diffracted signal being set to a set value (e.g. 80%FSH).
EN583-6 indicates only a sensitivity setting based on 5% grain noise. It then lists the diffracting notches in the Annex but does not state how they are used, simply that:
"The sensitivity setting can now be checked making use of representative flaws or diffracting artificial defects in a reference block as described in Annex A. The results can be used to justify
reducing the gain setting or give warning that the signal-to-noise ratio is insufficient."
The new EN Specification 121377 uses the Lateral wave set between 40-80% for single zones, the backwall set between 18-30dB over FSH where the lateral wave is not practical or present, and grain noise set at 5-10% for all other conditions. Reference blocks that include the 60° pointed notches are described in the Informative Annex A. Yet there seems to be no described method of using these targets. There are references in the document to using these for "verifying" sensitivity, for determining "adquacey of the testing (e.g. coverage and sensitivity)". But nowhere does it explain how these are to be done.

I think there are a few good reasons to have TOFD "notches", but the sensitivity verification items seem to be a nonissue when the lateral wave, backwall and grain structure noise are all valid limits indicating we should use maximum sensitivity possible.
The notches provide a method of verifying adequate beam coverage on the far side to carry out a single centred pass for a D-scan or if we are to use scans offset from the centreline because the beam divergence is not adequate for HAZ coverage.
The notches also provide a reasonable method of determining the near and far-surface resolutions (deadzones) so the operator can decide if a higher frequency may be required.

Can anyone suggest why the notches are required for sensitivity and why the specific tip shape is imposed (i.e. 60° point)?

Thanks
Ed


 
04:19 Sep-25-2006

Nigel Armstrong

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1094
Re: TOFD notches Hello Ed

I hope you dont mind some clarification comments on the document status first. (I had to look this up, cos I was not sure)

CEN upgraded the CEN/TC 121 committee document 377 to Technical Specification CEN/TS 14571:2004 "Welding - Use of time-of-flight-diffraction technique (TOFD) for examination of welds" on 11th July 2004 for "provisional application" with an initial limited validity of 3 years.

BSI along with 27 other European countries are required to announce the existence of the document in the same way as for an European Standard and to make it available promptly at the national level in "an appropriate form". BSI has done this as a Draft for Development - DD CEN/TS 14751:2004.

BSI's National foreword emphasises in bold print "This publication is not to be regarded as a British Standard". It expatiates that "..the time-of-flight technique has not been widely used for these applications (weld testing). It should be applied on this provisional basis, so that information and experienceof its practical application may be obtained".

After 2 years CEN members are requested to submit their comments, particulalrly on the question whether it can be converted into an European Standard. If so agreed then all conflicting National Standards must be withdrawn upon its full issue as an EN.

My input to your question now, Ed. Table 1, Section 5 of the DD (if I may call it that) specifies Examination Levels of which there are 4, viz. A thru D of increasing reliability. Column 4 of the table is "reference block for sensisitivity settings (see 10.1.4)" and stipulates there is a requirement for such for examination levels B thru D. All examination levels are applicable for pre-service inspections (yer pays yer money, yer takes yer choice!). Level D is for in-service inspection.

Section 10 "Range and sensitivity settings" and 10.1.4. "Sensitivity settings" specifies the hierarchy of lateral wave, backwall, grain noise as you stated in your post. It then continues that for reference levels B thru D "the sensitivity shall be VERIFIED by use of.." blocks containing either real flaws or machined discontinuities. Section 10.3.4. "Reference reflectors" gives the number required, i.e t = 6 - 25 mm minimum of 3, t > 25mm minimum of 5 - and Fig A4 shows a notched reference block and Fig A5 with side-drilled holes, all at various depths. Which is used seems to be a matter of agreement or perhaps referencing code.

10.2. "Checking of the settings" requires a check at least every 4 hours and sfter com[pletion of the test. and refers to Table 3 which requires that, for a sensitivity deviation from the reflector > 6dB, then "Settings shall be corrected and all examinations carried out since the last valid check shall be repeated".

So my understanding is that once the test sensitivity has been set by use of lateral, backwall or grain noise then the maximum screen height of each notch's analogue signal should be recorded along with the original system gain setting. If at the original gain setting any of the analogue responses diminish by > 6 dB then settings shall be corrected and all examinations carried out since the last valid check shall be repeated. There is a similar clause for the range for a deviation > 0,5mm or 2% of depth-range (whichever is greater). So the location of the notches will need to be initially measured for depth accuracy and recorded and checked at least every 4 hours.

I haven't tried it yet - our TOFD equipment is spanking new and not been used in anger yet - but I'll certainly post expereiences once we build up experience.

As for the 60 degree tip shape (could be difficult to machine and to verify), I dont know!

Cheers

Nigel

----------- Start Original Message -----------
: I am trying to determine the function of the reference notches as described in the various TOFD documents.
: BS-7706 indicates the notch as one of three options (grain noise and side drilled holes being the other options). BS-7706 suggests a near sided sharp pointed notch machined (EDM) to half the wall thickness. The sensitivity based on the diffracted signal being set to a set value (e.g. 80%FSH).
: EN583-6 indicates only a sensitivity setting based on 5% grain noise. It then lists the diffracting notches in the Annex but does not state how they are used, simply that:
: "The sensitivity setting can now be checked making use of representative flaws or diffracting artificial defects in a reference block as described in Annex A. The results can be used to justify
: reducing the gain setting or give warning that the signal-to-noise ratio is insufficient."
: The new EN Specification 121377 uses the Lateral wave set between 40-80% for single zones, the backwall set between 18-30dB over FSH where the lateral wave is not practical or present, and grain noise set at 5-10% for all other conditions. Reference blocks that include the 60° pointed notches are described in the Informative Annex A. Yet there seems to be no described method of using these targets. There are references in the document to using these for "verifying" sensitivity, for determining "adquacey of the testing (e.g. coverage and sensitivity)". But nowhere does it explain how these are to be done.
: I think there are a few good reasons to have TOFD "notches", but the sensitivity verification items seem to be a nonissue when the lateral wave, backwall and grain structure noise are all valid limits indicating we should use maximum sensitivity possible.
: The notches provide a method of verifying adequate beam coverage on the far side to carry out a single centred pass for a D-scan or if we are to use scans offset from the centreline because the beam divergence is not adequate for HAZ coverage.
: The notches also provide a reasonable method of determining the near and far-surface resolutions (deadzones) so the operator can decide if a higher frequency may be required.
: Can anyone suggest why the notches are required for sensitivity and why the specific tip shape is imposed (i.e. 60° point)?
: Thanks
: Ed
------------ End Original Message ------------




 
04:25 Sep-26-2006

Ed Ginzel

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1208
Re: TOFD notches Nigel:
Thanks for the information on the EN Codes. Following the dozens of EN and ISO documents and the various letters and numbers used to identify them in their various stages of development is more challenging that understanding wave mechanics.
I found a few PowerPoint presentations online, especially from Jan Verkooijen at Sonovation, that were particularly helpful in trying to trace all this TOFD Code development!

But the issues of the notches were not addressed in these PowerPoints.

When using the Technical Specification with the 4 Levels examination it is clear that we can perform the inspection without a Reference block.
583-6 on the other hand does not address the issue as specifically. In 7.4 it states “The sensitivity setting can now be checked making use of representative flaws or diffracting artificial defects in a reference block as described in Annex A”.
I read the “can” as nonmandatory, but Annex A is a “Normative” (mandatory) annex.(?) Here in the Annex the standard makes note that the targets are used for setting sensitivity and volume coverage (the potential use of offset scans is specifically implied).
BS7706 is a “Guide” but makes excellent observations and recommendations. Concerning reference reflectors it states they have only 3 functions:
a) Verification of the distribution of energy within the specimen
b) A means of reproducing inspection sensitivities
c) Demonstrating inspection resolution.
That same paragraph (6.5.4) also states the obvious:
For TOFD there is generally no direct correlation between the amplitude of the signal and the severity of the flaw.
As most people are aware, this also applies for all UT including pulse-echo. So the function of “checking sensitivity” via the 60° tip notches is moot when the Procedure has used the lateral wave set to 80% as the sensitivity level. Even in "verifying" the sensitivity, you would set the probes on the specimen where you originally set the lateral wave sensitivity and verify that the lateral wave amplitude is still as it was set at the outset of the examination.

The function of far surface notches for validating the Technique for volume coverage and surface resolution is still valid. But is this not possible with any small width notch of the same depth? Why 60° points.
Ed


----------- Start Original Message -----------
: Hello Ed
: I hope you dont mind some clarification comments on the document status first. (I had to look this up, cos I was not sure)
: CEN upgraded the CEN/TC 121 committee document 377 to Technical Specification CEN/TS 14571:2004 "Welding - Use of time-of-flight-diffraction technique (TOFD) for examination of welds" on 11th July 2004 for "provisional application" with an initial limited validity of 3 years.
: BSI along with 27 other European countries are required to announce the existence of the document in the same way as for an European Standard and to make it available promptly at the national level in "an appropriate form". BSI has done this as a Draft for Development - DD CEN/TS 14751:2004.
: BSI's National foreword emphasises in bold print "This publication is not to be regarded as a British Standard". It expatiates that "..the time-of-flight technique has not been widely used for these applications (weld testing). It should be applied on this provisional basis, so that information and experience of its practical application may be obtained".
: After 2 years CEN members are requested to submit their comments, particulalrly on the question whether it can be converted into an European Standard. If so agreed then all conflicting National Standards must be withdrawn upon its full issue as an EN.
: My input to your question now, Ed. Table 1, Section 5 of the DD (if I may call it that) specifies Examination Levels of which there are 4, viz. A thru D of increasing reliability. Column 4 of the table is "reference block for sensisitivity settings (see 10.1.4)" and stipulates there is a requirement for such for examination levels B thru D. All examination levels are applicable for pre-service inspections (yer pays yer money, yer takes yer choice!). Level D is for in-service inspection.
: Section 10 "Range and sensitivity settings" and 10.1.4. "Sensitivity settings" specifies the hierarchy of lateral wave, backwall, grain noise as you stated in your post. It then continues that for reference levels B thru D "the sensitivity shall be VERIFIED by use of.." blocks containing either real flaws or machined discontinuities. Section 10.3.4. "Reference reflectors" gives the number required, i.e t = 6 - 25 mm minimum of 3, t > 25mm minimum of 5 - and Fig A4 shows a notched reference block and Fig A5 with side-drilled holes, all at various depths. Which is used seems to be a matter of agreement or perhaps referencing code.
: 10.2. "Checking of the settings" requires a check at least every 4 hours and sfter com[pletion of the test. and refers to Table 3 which requires that, for a sensitivity deviation from the reflector > 6dB, then "Settings shall be corrected and all examinations carried out since the last valid check shall be repeated".
: So my understanding is that once the test sensitivity has been set by use of lateral, backwall or grain noise then the maximum screen height of each notch's analogue signal should be recorded along with the original system gain setting. If at the original gain setting any of the analogue responses diminish by > 6 dB then settings shall be corrected and all examinations carried out since the last valid check shall be repeated. There is a similar clause for the range for a deviation > 0,5mm or 2% of depth-range (whichever is greater). So the location of the notches will need to be initially measured for depth accuracy and recorded and checked at least every 4 hours.
: I haven't tried it yet - our TOFD equipment is spanking new and not been used in anger yet - but I'll certainly post expereiences once we build up experience.
: As for the 60 degree tip shape (could be difficult to machine and to verify), I dont know!
: Cheers
: Nigel
: : I am trying to determine the function of the reference notches as described in the various TOFD documents.
: : BS-7706 indicates the notch as one of three options (grain noise and side drilled holes being the other options). BS-7706 suggests a near sided sharp pointed notch machined (EDM) to half the wall thickness. The sensitivity based on the diffracted signal being set to a set value (e.g. 80%FSH).
: : EN583-6 indicates only a sensitivity setting based on 5% grain noise. It then lists the diffracting notches in the Annex but does not state how they are used, simply that:
: : "The sensitivity setting can now be checked making use of representative flaws or diffracting artificial defects in a reference block as described in Annex A. The results can be used to justify
: : reducing the gain setting or give warning that the signal-to-noise ratio is insufficient."
: : The new EN Specification 121377 uses the Lateral wave set between 40-80% for single zones, the backwall set between 18-30dB over FSH where the lateral wave is not practical or present, and grain noise set at 5-10% for all other conditions. Reference blocks that include the 60° pointed notches are described in the Informative Annex A. Yet there seems to be no described method of using these targets. There are references in the document to using these for "verifying" sensitivity, for determining "adquacey of the testing (e.g. coverage and sensitivity)". But nowhere does it explain how these are to be done.
: : I think there are a few good reasons to have TOFD "notches", but the sensitivity verification items seem to be a nonissue when the lateral wave, backwall and grain structure noise are all valid limits indicating we should use maximum sensitivity possible.
: : The notches provide a method of verifying adequate beam coverage on the far side to carry out a single centred pass for a D-scan or if we are to use scans offset from the centreline because the beam divergence is not adequate for HAZ coverage.
: : The notches also provide a reasonable method of determining the near and far-surface resolutions (deadzones) so the operator can decide if a higher frequency may be required.
: : Can anyone suggest why the notches are required for sensitivity and why the specific tip shape is imposed (i.e. 60° point)?
: : Thanks
: : Ed
------------ End Original Message ------------




 


© 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|