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 scanning of pipe-to-fitting welds
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

GRANPECT COMPANY Ltd
GRANPECT, X-ray NDT Expert You Can Rely on

2051 views
07:00 Jun-19-2006

Nigel Armstrong

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1094
TOFD scanning of pipe-to-fitting welds

Prior to pipework and fittings procurement for a materials replacement project I specified a required scanner footprint size of approximately 100 mm minimum for TOFD scanning of welds in 64mm thick steel. I based this on using a 45 degree transmitting wedge with the beam axis to cross the weld centre line at two-thirds of the material thickness, i.e. 44mm. The index-point to centre line distance would be 63 mm. The remaining 37 mm is to allow for the index-point to back of probe distance plus a "margin of comfort". The receiving probe would be positioned equidstant from the weld centre line as the transmitter, i.e. 63 mm.

Now the fittings (elbows) manufacturer is saying that a 100 mm straight section beyond the bevel face at the end of the fitting will be too costly.

I would like to know if anybody has any experience in similar circumstances with offset TOFD. The receiver would be positioned much nearer to the weld centre line than the transmitter. This would then reduce the amount of straight section required on the fitting side. I am thinking of keeping the same probe separation distance so the transmitter woould be moved to an index to centre line distance > the original 63mm. Would this arrangement significantly affect the POD? Or if the transmitter were kept at the same centre line distance thus reducing the probe separation?

Thanks in advance for any comments/info.


 
07:06 Jun-19-2006

Nigel Armstr

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1094
Re: TOFD scanning of pipe-to-fitting welds Sorry, correction to my post. (Monday morning after a busy weekend). The probe centre line distance for focusing a 45 degree wedge at the weld centre line at two-thirds of the material thickness of 64 mm will be 44 mm. and not 63 mm. (that will be the beam path distance).

----------- Start Original Message -----------
: Prior to pipework and fittings procurement for a materials replacement project I specified a required scanner footprint size of approximately 100 mm minimum for TOFD scanning of welds in 64mm thick steel. I based this on using a 45 degree transmitting wedge with the beam axis to cross the weld centre line at two-thirds of the material thickness, i.e. 44mm. The index-point to centre line distance would be 63 mm. The remaining 37 mm is to allow for the index-point to back of probe distance plus a "margin of comfort". The receiving probe would be positioned equidstant from the weld centre line as the transmitter, i.e. 63 mm.
: Now the fittings (elbows) manufacturer is saying that a 100mm straight section beyond the bevel face at the end of the fitting will be too costly.
: I would like to know if anybody has any experience in similar circumstances with offset TOFD. The receiver would be positioned much nearer to the weld centre line than the transmitter. This would then reduce the amount of straight section required on the fitting side. I am thinking of keeping the same probe separation distance so the transmitter woould be moved to an index to centre line distance > the original 63mm. Would this arrangement significantly affect the POD? Or if the transmitter were kept at the same centre line distance thus reducing the probe separation?
: Thanks in advance for any comments/info.
------------ End Original Message ------------




 
01:18 Jun-19-2006

Jan Verkooijen

Director,
Sonovation,
Netherlands,
Joined Nov 1998
29
Re: TOFD scanning of pipe-to-fitting welds Dear Nigel,

Not an uncommon issue in weld inspection.

1) To do the inspection on this thickness with just one setting is slightly over-optimistic. I would say you would need a minimum of two settings to cover the weld adequately. With just one setting on this thickness it is no wonder that some people still go on about dead zones for TOFD. May be you need to look at TS 14751, the European Norm for the set-up of TOFD inspections. You will find guidelines there as to how many settings would be needed for a certain thickness et cetera.
2)The issue of differing thicknesses, tapering to one side of the weld et cetera should not significantly decrease the POD provided your settings were right in the first place and are modified to suit the circumstances. That does include that you will have to shift your probe assembly to cater for the different thickness where your probe will sit on the fitting side.
3)Although your POD is not greatly affected, your depth measurement needs correction (your height measurement is still ~ok), which can be helped greatly by simply drawing it or buying some software help tools to visualise what happens.
4)We have a proven TOFD training course to teach you all the tricks in the book to come to a TOFD inspection that does the method justice and will clearly outperform RT under these circumstances. Alternatively we can help you set up your procedure.

In conclusion: you can inspect the welds connecting pipes to fittings without great problems with the right preparation.

Best regards,

Jan Verkooijen

----------- Start Original Message -----------
: Prior to pipework and fittings procurement for a materials replacement project I specified a required scanner footprint size of approximately 100 mm minimum for TOFD scanning of welds in 64mm thick steel. I based this on using a 45 degree transmitting wedge with the beam axis to cross the weld centre line at two-thirds of the material thickness, i.e. 44mm. The index-point to centre line distance would be 63 mm. The remaining37 mm is to allow for the index-point to back of probe distance plus a "margin of comfort". The receiving probe would be positioned equidstant from the weld centre line as the transmitter, i.e. 63 mm.
: Now the fittings (elbows) manufacturer is saying that a 100 mm straight section beyond the bevel face at the end of the fitting will be too costly.
: I would like to know if anybody has any experience in similar circumstances with offset TOFD. The receiver would be positioned much nearer to the weld centre line than the transmitter. This would then reduce the amount of straight section required on the fitting side. I am thinking of keeping the same probe separation distance so the transmitter woould be moved to an index to centre line distance > the original 63mm. Would this arrangement significantly affect the POD? Or if the transmitter were kept at the same centre line distance thus reducing the probe separation?
: Thanks in advance for any comments/info.
------------ End Original Message ------------




 
04:54 Jun-19-2006

Dick Gray

Engineering,
Canada,
Joined Nov 1999
14
Re: TOFD scanning of pipe-to-fitting welds I suggest you re-visit your original design proposal. Long tangent elbows, both long and short radius, are made by a number of manufacturers e.g. Cunico. In evaluating costs, the savings from automating the welding by effectively having straight portions of pipe butting together should be included. In addition, depending on the tangent length, there is an opportunity to reduce the number of welds in the system.

----------- Start Original Message -----------
: Prior to pipework and fittings procurement for a materials replacement project I specified a required scanner footprint size of approximately 100 mm minimum for TOFD scanning of welds in 64mm thick steel. I based this on using a 45 degree transmitting wedge with the beam axis to cross the weld centre line at two-thirds of the material thickness, i.e. 44mm. The index-point to centre line distance would be 63 mm. The remaining 37 mm is to allow for the index-point to back of probe distance plus a "margin of comfort". The receiving probe would be positioned equidstant from the weld centre line as the transmitter, i.e. 63 mm.
: Now the fittings (elbows) manufacturer is saying that a 100 mm straight section beyond the bevel face at the end of the fitting will be too costly.
: I would like to know if anybody has any experience in similar circumstances with offset TOFD. The receiver would be positioned much nearer to the weld centre line than the transmitter. This would then reduce the amount of straight section required on the fitting side. I am thinking of keeping the same probe separation distance so the transmitter woould be moved to an index to centre line distance > the original 63mm. Would this arrangement significantly affect the POD? Or if the transmitter were kept at the same centre line distance thus reducing the probe separation?
: Thanks in advance for any comments/info.
------------ End Original Message ------------




 
05:45 Jun-20-2006

Nigel Armstr

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1094
Re: TOFD scanning of pipe-to-fitting welds Jan, Dick

Thank you both for your quick responses and information.

Jan, Thanks for your reference to TS 14751, I shall attempt to obtain a copy to check recommended set-ups.

Have you any recommendations for scan visualisation software?

In your 3rd sentence of point 1, you imply that the TOFD near surface dead zone can be eliminated through the correct technique. This is news to me. I was vaguely aware of some potential for offline elimination of the lateral wave with advanced post-data collection signal processing (correct me please if I am wrong). However I have suggested both top fill and root pulse-echo scans due to my (mis)understanding that both these regions were effectively TOFD "blind spots" - more the lateral wave than the back wall, but stipulating the root PE to improve POD of root defects. The top fill PE set-up may be difficult at this thickness, especially as-welded, the choice seems to be either a creeping wave probe or a very long beam path for skipping off the ID. I can foresee calibration problems with both of these. Have there been any field trials focussing on detection capabilities of near-surface and root defects by TOFD?

I have already noted you will be holding a course in November in Oosterhout. I hope to be able to attend.

Dick, Good point regarding cost benefits of welding automisation and reduction of number of welds through long tangent elbows. I shall return to the design team and hope they listen!

Thanks again

Nigel Armstrong


-------- Start Original Message -----------
: Dear Nigel,
: Not an uncommon issue in weld inspection.
: 1) To do the inspection on this thickness with just one setting is slightly over-optimistic. I would say you would need a minimum of two settings to cover the weld adequately. With just one setting on this thickness it is no wonder that some people still go on about dead zones for TOFD. May be you need to look at TS 14751, the European Norm for the set-up of TOFD inspections. You will find guidelines there as to how many settings would be needed for a certain thickness et cetera.
: 2)The issue of differing thicknesses, tapering to one side of the weld et cetera should not significantly decrease the POD provided your settings were right in the first place and are modified to suit the circumstances. That does include that you will have to shift your probe assembly to cater for the different thickness where your probe will sit on the fitting side.
: 3)Although your POD is not greatly affected, your depth measurement needs correction (your height measurement is still ~ok), which can be helped greatly by simply drawing it or buying some software help tools to visualise what happens.
: 4)We have a proven TOFD training course to teach you all the tricks in the book to come to a TOFD inspection that does the method justice and will clearly outperform RT under these circumstances. Alternatively we can help you set up your procedure.
: In conclusion: you can inspect the welds connecting pipes to fittings without great problems with the right preparation.
: Best regards,
: Jan Verkooijen
: : Prior to pipework and fittings procurement for a materials replacement project I specified a required scanner footprint size of approximately 100 mm minimum for TOFD scanning of welds in 64mm thick steel. I based this on using a 45 degree transmitting wedge with the beam axis to cross the weld centre line at two-thirds of the material thickness, i.e. 44mm. The index-point to centre line distance would be 63 mm. The remaining 37 mm is to allow for the index-point to back of probe distance plus a "margin of comfort". The receiving probe would be positioned equidstant from the weld centre line as the transmitter, i.e. 63 mm.
: : Now the fittings (elbows) manufacturer is saying that a 100 mm straight section beyond the bevel face at the end of the fitting will be too costly.
: : I would like to know if anybody has any experience in similar circumstances with offset TOFD. The receiver would be positioned much nearer to the weld centre line than the transmitter. This would then reduce the amount of straight section required on the fitting side. I am thinking of keeping the same probe separation distance so the transmitter woould be moved to an index to centre line distance > the original 63mm. Would this arrangement significantly affect the POD? Or if the transmitter were kept at the same centre line distance thus reducing the probe separation?
: : Thanks in advance for any comments/info.
------------ End Original Message ------------




 
08:31 Aug-23-2006

Nigel Armstrong

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1094
Re: TOFD scanning of pipe-to-fitting welds Thanks to all of you who replied to my original post. And Jan I would love to attend your TOFD course in September, but I have 2 PCN certiificate renewals in that month and I am a lowly contractor here in Kazakhstan, so guess who pays for everything?

Despite months of repetition that fittings (with the exception of 2" OD and less which will be RT)would need to be procured with extended straight sections, procurement turned semi-deaf. They have ordered the extra length for everything over 35mm and 8" OD. All other fittings will be standard. We intend using both TOFD and pulse-echo for final weld acceptance using ASME Code Case 2235-9 (which now also allows phased array).

I appreciate your comments whether we should now insist, and I mean insist not just suggest, on all non-extended length fitting welds should have the cap ground flush to enable the inspection. Thoughts of the wise gentlemen (and ladies) who frequent this board, please.

Nigel

----------- Start Original Message -----------
: Prior to pipework and fittings procurement for a materials replacement project I specified a required scanner footprint size of approximately 100 mm minimum for TOFD scanning of welds in 64mm thick steel. I based this on using a 45 degree transmitting wedge with the beam axis to cross the weld centre line at two-thirds of the material thickness, i.e. 44mm. The index-point to centre line distance would be 63 mm. The remaining 37 mm is to allow for the index-point to back of probe distance plus a "margin of comfort". The receiving probe would be positioned equidstant from the weld centre line as the transmitter, i.e. 63 mm.
: Now the fittings (elbows) manufacturer is saying that a 100 mm straight section beyond the bevel face at the end of the fitting will be too costly.
: I would like to know if anybody has any experience in similar circumstances with offset TOFD. The receiver would be positioned much nearer to the weld centre line than the transmitter. This would then reduce the amount of straight section required on the fitting side. I am thinking of keeping the same probe separation distance so the transmitter woould be moved to an index to centre line distance > the original 63mm. Would this arrangement significantly affect the POD? Or if the transmitter were kept at the same centre line distance thus reducing the probe separation?
: Thanks in advance for any comments/info.
------------ End Original Message ------------




 
02:59 Nov-22-2006
Yadiel
Shane http://fastfish.org/student-financial-aid-services.html


 


© 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|