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 >
Limits of accurcacy with TOFD
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

Zawada NDT
Suppliers of instruments to test steel wire ropes in-situ, using magnetic, non-destructive method.

540 views
03:01 Aug-30-2004
Sri
Limits of accurcacy with TOFD

Could anyone let me know the limits on accuracy to be considered when sizing a fatigue crack using TOFD.Can a 0.1" deep crack on a 0.5" thick steel be sized with reasonable accuracy.



 
00:54 Sep-08-2004

Ed Ginzel

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1208
Re: Limits of accurcacy with TOFD Sri:
In answering your question about limitation of TOFD to "size" a fatigue crack, my reply will be like most NDT technicians' responses...a bit vague. Or perhaps more accurately, cautionary.

The condition of a 0.1" (2.5mm) crack in a 0.5" (12.5mm) wall is not too difficult to assess IF the crack is on the opposite wall from the probes, the plate tested is flat and the crack is surface connected. Typically such a crack "tip" can be located within a half wavelength of the frequency used. For 10MHz compression mode in steel the wavelength is about 0.6mm so an approximate "tolerance" of 0.3mm might be the sort of accuracy you might be able to achieve under good conditions.

But if the flaw is on the near surface this will be in the lateral wave ring time and not reasonably sizeable! Also, if the flaw is not surface connected on the opposite wall the ability to resolve the lower tip will limit the accuracy as the ring time from the lower tip could be masked by the backwall ring.

To some extent the resolution will also be limited by the angle used. Better time resolution is had for smaller angles but these may not be the best option for thin wall.

If your flaw is a simple surface connecting crack won the opposite wall and you can use a reasonably high frequency probe with a short ringtime (broad bandwidth) and about a 60 degree nominal refracted angle the extent to which the flaw has propagated from the surface should be reasonably estimated to less than 0.5mm (0.02").

Resolution and error limitation guides are well explained in EN 583-6.

----------- Start Original Message -----------
: Could anyone let me know the limits on accuracy to be considered when sizing a fatigue crack using TOFD.Can a 0.1" deep crack on a 0.5" thick steel be sized with reasonable accuracy.
------------ End Original Message ------------




 
09:00 Sep-08-2004
Srikanth Ramachandran
Thanks for responding: Limits of accurcacy with TOFD Thank you very much.Its of great help in computing possible errors.

Thanks again
Sri


----------- Start Original Message -----------
: Sri:
: In answering your question about limitation of TOFD to "size" a fatigue crack, my reply will be like most NDT technicians' responses...a bit vague. Or perhaps more accurately, cautionary.
: The condition of a 0.1" (2.5mm) crack in a 0.5" (12.5mm) wall is not too difficult to assess IF the crack is on the opposite wall from the probes, the plate tested is flat and the crack is surface connected. Typically such a crack "tip" can be located within a half wavelength of the frequency used. For 10MHz compression mode in steel the wavelength is about 0.6mm so an approximate "tolerance" of 0.3mm might be the sort of accuracy you might be able to achieve under good conditions.
: But if the flaw is on the near surface this will be in the lateral wave ring time and not reasonably sizeable! Also, if the flaw is not surface connected on the opposite wall the ability to resolve the lower tip will limit the accuracy as the ring time from the lower tip could be masked by the backwall ring.
: To some extent the resolution will also be limited by the angle used. Better time resolution is had for smaller angles but these may not be the best option for thin wall.
: If your flaw is a simple surface connecting crack won the opposite wall and you can use a reasonably high frequency probe with a short ringtime (broad bandwidth) and about a 60 degree nominal refracted angle the extent to which the flaw has propagated from the surface should be reasonably estimated to less than 0.5mm (0.02").
: Resolution and error limitation guides are well explained in EN 583-6.
: : Could anyone let me know the limits on accuracy to be considered when sizing a fatigue crack using TOFD.Can a 0.1" deep crack on a 0.5" thick steel be sized with reasonable accuracy.
------------ 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|