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 >
Able to convert AWS defect rating to DAC
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

10:17 Jun-13-2009

Dao Huu Xuan

Joined Feb 2007
Able to convert AWS defect rating to DAC

Hi all,

This discussion refers to AWS D1.1, table 6.2 and 6.3 - UT Acceptance-Rejection Criteria.

These Criteria has been established based on the calculation of the attenuation of sound pressure at far field. Thus, the attenuation by 2 decibels for each increased inch in sound path. (This is included beam spread and absorption, where the first factor is decisive)

Today, many type of commerce digital UT-sets have had build-in AWS software. And this software also used a/b/c/d formula with the fixed the attenuation by 2 decibels for each increased inch in sound path.

The section of the Standard is allows angled transducer dimensions as: [15 to 25 mm] in width and [15 to 20 mm] in height and the maximum width to height ratio shall be 1.2 to 1.0, and the minimum width-to-height ratio shall be 1.0 to 1.0.

This means that the formula of attenuation by 2 decibels is applied for probe sizes from 15x15 to 24x20. But, actually the beam spread of these difference size of probes are difference.

By using IOW DAC block (SDH 1.5mm); the probe 15x17 (2.25 MHz) show the attenuation is about 2.8 decibels per each increased inch in sound path. And the probe 22x20 (2 MHz) show the attenuation is about 1.9 decibels per each increased inch in sound path.

All SDH in IOW block should have the same defect rating (d). But probe 15x17 (2.25 MHz) show the defect rating is difference, especially for the SDH with sound path further 75mm. By table 6.2, the defect rating (d) between Discontinuity Classes A, B, C and D are different only 1 dB.

By above reason, I think that DAC is better and more accurate than a/b/c/d formula in determination the defect rating. So, Table 6.2 and 6.3 needs to convert to related DAC. And, an example of conversion corresponding to material thickness 8-20mm as below:
Table 6.2 (8-20mm):
Discontinuity Classes Response Pulse Amplitude
A >= 25% DAC (DAC-12dB)
B >= 22.4% and <25% DAC
C >= 20% and <22.4% DAC
D < 20% DAC (DAC-14dB)

Table 6.3 (8-20mm):
Discontinuity Classes Response Pulse Amplitude
A >= 45% DAC (DAC-7dB)
B >= 40% and <45% DAC
C >= 36% and <40% DAC
D < 36% DAC (DAC-9dB)

Finally, any one who have experience on this please give me an advice.
Thank you,

16:33 Jun-13-2009

Ed Ginzel

R & D, -
Materials Research Institute,
Joined Nov 1998
Re: Able to convert AWS defect rating to DAC In Reply to Dao Huu Xuan at 10:17 Jun-13-2009 (Opening).

Huu Xuan
I suspect the origins of why AWS D1.1 decided to use a simple "fudge-factor" instead of actual observed attenuation as on a DAC curve, are now lost in history. Echo-transmittance curves (seen in the Appendices of the old Krautkramer book) show that the returned pressure changes with angle. Your observations show that variations in available pressure vary with probe parameters (frequency and dimensions). Some steels (even plain carbon) have anisotropic characteristics, so properties (including attenuation) vary with the angle as well.
Clearly, the D1.1 method to use a simple fixed value does not consider any of these. D1.1 therefore is only a means of arbitrary assessment. Since it does not even make consideration for probe variations it is not even a useful technique for ensuring consistency from one operator to the next with both using "approved" equipment.
AWS D1.1 makes provision in more recent editions to use a DAC as described in the Appendices. Unfortunately, this requires Engineer Approval and submission of an Acceptance criteria to replace the Discontinuity Class Levels (also requiring Enigineer approval).
The wording in D1.1 was perhaps useful in the 1960s. However, now that the equipment and techniques have advanced in the past 50 years it no longer seems to fit today.

19:09 Jun-13-2009

Roger Duwe

NDT Inspector, API-510, 570, 653
Joined Jan 2009
Re: Able to convert AWS defect rating to DAC In Reply to Ed Ginzel at 16:33 Jun-13-2009 .

Concur, Ed. Theoretically, doing a 'fudge factor' analysis of the return from a flaw, in your head, would be easier than working to a DAC curve hand-drawn on the face of an analog scope. Didn't work that way for me, DAC is easier [and more accurate].

On a good [Kraut or Panametrics] digital scope, I find that DAC is still easier than using the onboard D1.1 software. And there is the accuracy of flaw evaluation.

The origin of D1.1 UT criteria may have been something that could be performed with only one calibration block. Seems like a poor reason to sacrifice accuracy. Maybe someone on the D1 committee will take notice of these discussions and observations.

00:41 Jun-20-2009

Neil Breslow

Teacher, - -
NDT Supply.com. Inc.,
Joined May 2009
Re: Able to convert AWS defect rating to DAC In Reply to Dao Huu Xuan at 10:17 Jun-13-2009 (Opening).

Please let me supply some historic background on the origin of the AWS Code. This code was adapted from the Bureau of Public Roads (USA) specification which as adapted from an in-house procedure of American Bridge and Iron and dates back to the late 1960's. As I understand it the orginal draft used the Krautkramer DGS System which was rejected because of its origin and that the US companies Branson and Automation Industries did not want the German company Krautkramer to become the authority on flaw sizing. So, the writer of the specification attempted to "Americanize" the system by simulating the decay curve for a 2.25 MHz rectagular probes in a size range similar to the Krautkramer WB series (20x22mm). Please note that at that time the concern was only for flat work on structural grade steel.

Much has happened to the development of steels and ultrasonic instruments and testing techniques over the past 40 years. But there has been little to no developments to the Ultrasonic Testing Procedures in the AWS Code over these years.

Also, there was an article puplished in the late 60's or early 70's in Materials Evaluation that compares the the AWS Code to to the ASME Pressure Vessel Code. If anyone is interested let me have your e-mail address and I will find it and send you a scanned copy.


© 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|