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 >
Re: Automated UT inspection of steel mill work rolls
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

02:54 May-01-1997

Tom Carodiskey

Engineering
GE Inspection Technologies,
USA,
Joined Nov 1998
1
Re: Automated UT inspection of steel mill work rolls
: We currently use eddy current flaw detection techniques in our roll grinding shop. I am interested in enhancing the process by
: combining this with automated surface wave and through body UT inspection.
: Can the panel indicate the level of crack resolution I am likely to achieve and suggest the best type of phased array transducer
: for this purpose.
: The following URL will take you to our site if you are interested in knowing more about Dofasco http://www.dofasco.ca

Scanning with a conventional surface wave transducer would be more cost effective than a phased array system. The depth of penetration of a surface wave is approximately 0.5 wavelength so higher frequencies would give good reflections from most cracks while lower frequecies would reflect well only from deeper cracks. However, a very smooth surface finish is required for surface wave inspection.
We understand that some rework stations mount UT equipment on the lathe and inspect the roller as it is machined.
Also, more traditional angle beam or creeping wave techniques should be considered.
Tom Carodiskey


 
03:42 May-02-1997

Ed Ginzel (Waterloo, Ontario)

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1199
Re: Automated UT inspection of steel mill work rolls : We currently use eddy current flaw detection techniques in our roll grinding shop. I am interested in enhancing the process by
: combining this with automated surface wave and through body UT inspection.
: Can the panel indicate the level of crack resolution I am likely to achieve and suggest the best type of phased array transducer
: for this purpose.
: The following URL will take you to our site if you are interested in knowing more about Dofasco http://www.dofasco.ca

More information would be needed to adequately assess your needs. How thick are the sections now being tested by eddy current (I assume remote field)?
Are the probable flaw orientations all in the same direction or can they be randomly oriented? The response by Tom Carodiskey correctly assessed the
problems with surface waves and perhaps it is a linear array not a phased array you would be considering. Several TOFD probes in some sort of array
might provide the detection capability you need for the volumetric examination. TOFD has the advantage that it has a reduced dependence on flaw
orientation compared to pulse-echo techniques. A recent paper by Jan de Raad et al indicated a Dutch programme is being carried out to apply TOFD
independent of other techniques as they find it an effective tool for both detection and sizing.




 
00:56 May-06-1997
Trevor Easton
Re: Automated UT inspection of steel mill work rolls : : We currently use eddy current flaw detection techniques in our roll grinding shop. I am interested in enhancing the process by
: : combining this with automated surface wave and through body UT inspection.
: : Can the panel indicate the level of crack resolution I am likely to achieve and suggest the best type of phased array transducer
: : for this purpose.
: : The following URL will take you to our site if you are interested in knowing more about Dofasco http://www.dofasco.ca

:
: I don't completely understand the problem. Well, it's a
: fact, that EC testing in ferritic material is restricted.
: I don't know, if there is a static magnetic field for
: premagnetization used, to enhance the penetration of
: EC. If not, the penetration is very poor, but it should
: inprove, if cracks are open to the surface and I think,
: this should be the normal case. Remote field eddy current
: shouldn't work in the case of a solid roll.

: But in all cases, the sizing of the defect depth with
: EC is still a problem. Will you use UT for sizing the
: defect depth? Do you want to use phased arrays for im-
: proving the accuracy of sizing by using several angles
: of incidence or is the reason the arbitrary orientation
: of the cracks?

: We have experience with both. For example, we tested a
: PA-probe for lateral skewing to +- 45 degrees. In this
: case, the angle of incidence should carefully be
: selected, because by skewing the angle of incidence
: will change.

: If the question is the voluminous scanning of your rolls,
: PA probes are an adequate tool for doing this. With
: selected angles of incidence you can inspect the whole
: volume of your rolls and you can visualize the results
: with tomografic methods. The capture of flaws perpenticular
: orientated to the surface will be enhanced by using
: wave conversion methods.

We currently use the EC to detect fire cracks (perpendicular to rolling direction) and bar cracks (parallel to rolling direction) on the surface of the rolls. The system was installed originally when we we using iron rolls. We are now using tool steel rolls and the information we have been given indicates that UT may be superior to EC for tool steel probably because of the magnetic permeability of the material as you mention.
Our hope is to develop a system combining the two technologies in a synergystic relationship to give the optimum flaw detection and recording. The object of using a phased array was, as you suspect, to accomodate random orientation of flaws.
By displaying the data on a C scan we want to be enable the operator to easily interpret the roll condition and also to detect changes in flaws from rolling campaign to campaign.


 
01:49 May-06-1997
P.W. van Andel
Re: Automated UT inspection of steel mill work rolls : We currently use eddy current flaw detection techniques in our roll grinding shop. I am interested in enhancing the process by
: combining this with automated surface wave and through body UT inspection.
: Can the panel indicate the level of crack resolution I am likely to achieve and suggest the best type of phased array transducer
: for this purpose.
: The following URL will take you to our site if you are interested in knowing more about Dofasco http://www.dofasco.ca

Lismar Engineering in the Netherlands recently developed an automated roll
inspection system that uses multiple surface waves. Their objective is to find cracks
with arbitrary orientation, and with a depth of at least 0.1mm.


 
03:29 May-07-1997

Rainer Meier

R & D
retired from intelligeNDT Systems & Services,
Germany,
Joined Nov 1998
15
Re: Automated UT inspection of steel mill work rolls : : : We currently use eddy current flaw detection techniques in our roll grinding shop. I am interested in enhancing the process by
: : : combining this with automated surface wave and through body UT inspection.
: : : Can the panel indicate the level of crack resolution I am likely to achieve and suggest the best type of phased array transducer
: : : for this purpose.
: : : The following URL will take you to our site if you are interested in knowing more about Dofasco http://www.dofasco.ca

: :
: : I don't completely understand the problem. Well, it's a
: : fact, that EC testing in ferritic material is restricted.
: : I don't know, if there is a static magnetic field for
: : premagnetization used, to enhance the penetration of
: : EC. If not, the penetration is very poor, but it should
: : inprove, if cracks are open to the surface and I think,
: : this should be the normal case. Remote field eddy current
: : shouldn't work in the case of a solid roll.

: : But in all cases, the sizing of the defect depth with
: : EC is still a problem. Will you use UT for sizing the
: : defect depth? Do you want to use phased arrays for im-
: : proving the accuracy of sizing by using several angles
: : of incidence or is the reason the arbitrary orientation
: : of the cracks?

: : We have experience with both. For example, we tested a
: : PA-probe for lateral skewing to +- 45 degrees. In this
: : case, the angle of incidence should carefully be
: : selected, because by skewing the angle of incidence
: : will change.

: : If the question is the voluminous scanning of your rolls,
: : PA probes are an adequate tool for doing this. With
: : selected angles of incidence you can inspect the whole
: : volume of your rolls and you can visualize the results
: : with tomografic methods. The capture of flaws perpenticular
: : orientated to the surface will be enhanced by using
: : wave conversion methods.

: Weused an EC system on ferromagnetic materials and detected and evaluated quite precise
: (with precision comparable to twice the roughness of the surface - typical for EC - in our case
: bout 0.3 - 0.5 mm) simulated (sawed) surface cracks with depths up to 3 mm.

: How deep do You want to go with EC, especially for EC? For testing steel mill work rolls,
: the only purpose of using EC, IMHO, is the testing of the near-surface zone
: in a very sensitive way - cracks in this zone are the most dangerous - because
: traditional US testing approaches do not give good results in the near-surface
: zone (about what I know).

: Besides, as I do not know too much about automated US testing, tell please,
: what's a phased array? (Please don't laugh too loud.)

: Jurcovici Florin

One difficulty for EC inspection of ferromagnetic material is changing permeability
from one surface point to another, which could lead to misinterpretation of the
resulting signal. One other point is the accuracy of depth sizing, especially
for non voluminous cracks. You have experience with depth sizing of small notches
in ferromagnetic material. Is this experience transferable to steel material
with cracks?

A phased array probe consists of a number of small piezo elements. The width of
each element is so small (in the range of the half wavelength), that a very broad
sound beam results. The individuel elements will be excited separately. Also the
received signals from the individual array elements will be amplified separately.
By firing the array elements time delayed, the resulting sound field can be formed.
This means,that as well the beam angle as also the focus can be controlled
by electrical steering of the time delays. Normally the received individual
signals will also be delayed in the same way, before summarizing to the resulting
signal.
Most of the phased arrays are linear arrangements of 8 to 32 elements. In the
future, with decreasing size and price of electronics, 2-D-arrays should become
popular.
Examples for linear phased arrays you will find in the "Ultrasonic Testing
Encyclopedia" under URL http://www.ultrasonic.de/article/ut_az/ut_p/ut_p.htm or in the
NDT-exhibition under URL http://www.NDT.net/exhibit/cust_sh/sie_sh1.htm.

Rainer Meier



 


© 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|