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 >
Realtime detection of porosity and lack of fusion in titanium weld deposits
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

1870 views
03:48 Dec-10-1998
Dave Abbott
Realtime detection of porosity and lack of fusion in titanium weld deposits

I am interested in obtaining more information (or any info) on in-process detection of pores (1-20 mils) and lack of fusion between layers for a laser cladding process. Some have recommended I look into laser UT and/or EMATS. Does anybody have any experience with either? Any other suggestions?

Dave Abbott
AeroMet Corporation
Eden Prairie, MN
612-974-1808


 
01:37 Dec-11-1998

Rolf Diderichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
598
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits : I am interested in obtaining more information (or any info) on in-process detection of pores (1-20 mils) and lack of fusion between layers for a laser cladding process. Some have recommended I look into laser UT and/or EMATS. Does anybody have any experience with either? Any other suggestions?

: Dave Abbott
: AeroMet Corporation
: Eden Prairie, MN
: 612-974-1808
---------------------

You may find information in the following topic of the transducer workshop in Sep 1997.
'Smallest detectable flaw size, especially for low frequency test (f= 0.5 MHz) and in general' at http://www.ndt.net/wshop/wshop_tr/messages/72.htm. Especially the reply from Yosi Bar-Cohen offers some more general aspects.

I tried our new Internt crawler, in a minute I could identify a page of EWI
'Nondestructive evalution integral to proper joining'
http://bushwood.ewi.org/insights/old/mar98/art1.html
On this page you may find some ideas or at least the contact to EWI's NDT experts.

The sizes of 1-20mils seems for me to small to be detected reliable with UT.
Also Laser or even EMATs may not help a lot since both have a poor S/N ratio.
With conventional ultrasonic testing you may not have the excellent resolution of laser UT, but on the other hand you can use more amplification which is an important issue if you apply high frequency UT for small defects.

Rolf



 
02:02 Dec-11-1998
Shaun Lawson
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits : : I am interested in obtaining more information (or any info) on in-process detection of pores (1-20 mils) and lack of fusion between layers for a laser cladding process. Some have recommended I look into laser UT and/or EMATS. Does anybody have any experience with either? Any other suggestions?

: : Dave Abbott
: : AeroMet Corporation
: : Eden Prairie, MN
: : 612-974-1808
: ---------------------

: You may find information in the following topic of the transducer workshop in Sep 1997.
: 'Smallest detectable flaw size, especially for low frequency test (f= 0.5 MHz) and in general' at http://www.ndt.net/wshop/wshop_tr/messages/72.htm. Especially the reply from Yosi Bar-Cohen offers some more general aspects.

: I tried our new Internt crawler, in a minute I could identify a page of EWI
: 'Nondestructive evalution integral to proper joining'
: http://bushwood.ewi.org/insights/old/mar98/art1.html
: On this page you may find some ideas or at least the contact to EWI's NDT experts.

: The sizes of 1-20 mils seems for me to small to be detected reliable with UT.
: Also Laser or even EMATs may not help a lot since both have a poor S/N ratio.
: With conventional ultrasonic testing you may not have the excellent resolution of laser UT, but on the other hand you can use more amplification which is an important issue if you apply high frequency UT for small defects.

I would agree that most UT is unlikely to detect defects down to 1mm in an online
situation (with noise, elec/EM intereference etc). Our recent
BRITE-EURAM project looked at using TOFD for online detection of flaws including
porosity and LOF (see link below) but the minimum defect size used was about 20mm -
though various 'unintentional defects' - most likely small pores - were deetected
quite regularly - though we haveno 'real' data to say we could detect defects of
this size.

Shaun.



 
06:01 Dec-11-1998

Tom Nelligan

Engineering,
retired,
USA,
Joined Nov 1998
390
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits

: I am interested in obtaining more information (or any : info) on in-process detection of pores (1-20 mils) : and lack of fusion between layers for a laser : cladding process. Some have recommended I look into : laser UT and/or EMATS. Does anybody have any : experience with either? Any other suggestions?

Under good conditions, it is possible to detect porosity down to a 1 or 2 mil (25 to 50 micron) minimum void size with very high frequency, sharply focussed piezoelectric immersion transducers in the 50-100 MHz range. This can often be done in relatively thin (less than a few millimeters), smooth-surfaced, grain-free metals and ceramics. However, because of the very sharp focus required, the transducer needs to be indexed at very small increments, on the same order as the minimum flaw size, so this type of test may take too long for in-process use. My experience has been that this technique is generally used in an off-line mode where inspection time is less critical.

If you would like to post further details about the part you need to inspect -- material, thickness, surface condition, geometry, and available inspection time -- I may be able to comment further.

Tom Nelligan
Senior Applications Engineer, Panametrics, Inc.
http://www.panametrics.com



 
02:27 Dec-11-1998
Dave Abbott
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits I think I need to clarify the size of prosity we want to detect. The maximum flaw size for the wall thicknesses of the parts we are making is on the order of 0.006" (0.15 mm). Is this below the lower limit for UT? The wall thicknesses themselves are on the order of 0.1" (2.5 mm) to 0.3" (8 mm). IF not UT, what else should I consider?

If you would like to see some of the parts we're currently making, please visit our Web site @ http://www.aerometcorp.com.

Thanks,

Dave Abbott
AeroMet Corporation
612-974-1808

: I am interested in obtaining more information (or any info) on in-process detection of pores (1-20 mils) and lack of fusion between layers for a laser cladding process. Some have recommended I look into laser UT and/or EMATS. Does anybody have any experience with either? Any other suggestions?

: Dave Abbott
: AeroMet Corporation
: Eden Prairie, MN
: 612-974-1808




 
02:41 Dec-11-1998
Dave Abbott
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits ded I look into : laser UT and/or EMATS. Does anybody have any : experience with either? Any other suggestions?

: Under good conditions, it is possible to detect porosity down to a 1 or 2 mil (25 to 50 micron) minimum void size with very high frequency, sharply focussed piezoelectric immersion transducers in the 50-100 MHz range. This can often be done in relatively thin (less than a few millimeters), smooth-surfaced, grain-free metals and ceramics. However, because of the very sharp focus required, the transducer needs to be indexed at very small increments, on the same order as the minimum flaw size, so this type of test may take too long for in-process use. My experience has been that this technique is generally used in an off-line mode where inspection time is less critical.

: If you would like to post further details about the part you need to inspect -- material, thickness, surface condition, geometry, and available inspection time -- I may be able to comment further.

: Tom Nelligan
: Senior Applications Engineer, Panametrics, Inc.
: http://www.panametrics.com





 
06:45 Dec-12-1998
Wieslaw Bicz
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits : I am interested in obtaining more information (or any info) on in-process detection of pores (1-20 mils) and lack of fusion between layers for a laser cladding process. Some have recommended I look into laser UT and/or EMATS. Does anybody have any experience with either? Any other suggestions?

: Dave Abbott
: AeroMet Corporation
: Eden Prairie, MN
: 612-974-1808

I think, you can detect your flaws, if you use the technique, similiar
to our techniques. As you can find on our page (http://www.optel.com.pl), we are using acoustical
holography for recognising the fingerprints.

I think, it could be possible to detect your flaws, using transducers
coupled on the side of your part, that is not laserformed and detecting
the scattering of ultrasound in many directions in the similiar way, we
are doing it with the fingerprints. We can detect a structures smaller
as 0.1 mm (single points can be even smaller)

It would be interesting for me, to develop the version of this technique
that could be used in your machines. But it must developed for your
needs, and I dont think, it will be easy.

Wieslaw Bicz


 
02:51 Dec-14-1998

Tom Nelligan

Engineering,
retired,
USA,
Joined Nov 1998
390
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits
: I think I need to clarify the size of prosity we want to detect. The maximum flaw size for the wall thicknesses of the parts we are making is on the order of 0.006" (0.15 mm). Is this below the lower limit for UT? The wall thicknesses themselves are on the order of 0.1" (2.5 mm) to 0.3" (8 mm). IF not UT, what else should I consider?

: If you would like to see some of the parts we're currently making, please visit our Web site @ http://www.aerometcorp.com.

Thanks for the additional information. The minimum flaw size that you specify is certainly not in itself below the minimum that can be detected by UT, using high frequency (20 MHz and up) focussed immersion techniques, but from the rest of your description of the process and the picture of the sample part on the web site, I would have to reconsider whether an in-process test of the sort I mentioned in my last posting would be practical in your application. Any test of that type would involve squirting water at the part for acoustic coupling,which given your description of the welding process probably isn't going to be desireable. The inspection surface would have to be very smooth. It would also be necessary to follow potentially complex contours and make a number of passes with a very sharply focussed beam, which could be both expensive and relatively slow.

Assuming that the surface of the part is smooth, I still think this could potentially be done as an off-line test using a high frequency focussed transducer and a scanning mechanism. However, to fully scan a part like the one pictured at the necessary resolution would take many minutes, which I realize probably doesn't meet your needs.

--Tom Nelligan
Senior Applications Engineer, Panametrics, Inc.
http://www.panametrics.com




 
05:05 Dec-17-1998
Rick Finlayson
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits Dave,
Are you the Dave Abbot that was at Ohio State when I was there. I was doing my MSc with Dr. Rohklin in the Welding and Eng. Department and specifically NDE.
Rick
: Dave Abbott
: AeroMet Corporation
: Eden Prairie, MN
: 612-974-1808




 
05:57 Jan-11-1999
Bob Lasser
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits


 
05:45 Jul-03-2000
Steve Adams
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits : I am interested in obtaining more information (or any info) on in-process detection of pores (1-20 mils) and lack of fusion between layers for a laser cladding process. Some have recommended I look into laser UT and/or EMATS. Does anybody have any experience with either? Any other suggestions?

: Dave Abbott
: AeroMet Corporation
: Eden Prairie, MN
: 612-974-1808

Given the environment and inability to couple the part to a transducer, it would seem that the high frequency ultrasonics are not really going to get you where you want to go. I would assume that what you are after is the capability for feedback to allow the removal of gaseous inclusions in the weldment as you build it up and before you cover it up with the next layer - tricky, but extremely useful form a quality standpoint if you can pull it off. The ony feasible method that comes to mind for your application would be a real-time through-transmission X-ray. It would not be cheap, and a complete assembly to do exactly what you wantprobably doesn't exist, but certainly all the peices do, and it wouldn't be rocket science to get from here to there. You probably couldn't get away with a horizontal transmitting system which loked through the deposited metal from the side because there are too many geometries that you couldn't get a look at, (and crashing your transmitter into the part would get old) but you could do it from above. The source could be mounted directly behind the laser (and shielded from the environment with a transparent barrier), and the reciever would have to be below the part but following the motion of the ransmitter/laser. You would be transmitting through the full thickness of the layup at any given time, but through digital subtraction of the previous pass image, you could "filter" down to what you just deposited if you wanted to. You could even get clever analyze the image automatically, and with the use of a second small follower laser, release the trapped porosity as it was detected without stopping the powder-injected laser. Probably requires redesign of your chamber and some sreious capital, but it is doable.




 
01:55 Jan-02-2001
jxhcvytkzufy
Re: Realtime detection of porosity and lack of fusion in titanium weld deposits :
: : I think I need to clarify the size of prosity we want to detect. The maximum flaw size for the wall thicknesses of the parts we are making is on the order of 0.006" (0.15 mm). Is this below the lower limit for UT? The wall thicknesses themselves are on the order of 0.1" (2.5 mm) to 0.3" (8 mm). IF not UT, what else should I consider?

: : If you would like to see some of the parts we're currently making, please visit our Web site @ http://www.aerometcorp.com.

: Thanks for the additional information. The minimum flaw size that you specify is certainly not in itself below the minimum that can be detected by UT, using high frequency (20 MHz and up) focussed immersion techniques, but from the rest of your description of the process and the picture of the sample part on the web site, I would have to reconsider whether an in-process test of the sort I mentioned in my last posting would be practical in your application. Any test of that type would involve squirting water at the part for acoustic coupling, which given your description of the welding process probably isn't going to be desireable. The inspection surface would have to be very smooth. It would also be necessary to follow potentially complex contours and make a number of passes with a very sharply focussed beam, which could be both expensive and relatively slow.

: Assuming that the surface of the part is smooth, I still think this could potentially be done as an off-line test using a high frequency focussed transducer and a scanning mechanism. However, to fully scan a part like the one pictured at the necessary resolution would take many minutes, which I realize probably doesn't meet your needs.

: --Tom Nelligan
: Senior Applications Engineer, Panametrics, Inc.
: http://www.panametrics.com




 


© 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|