Determination of an Optimal Examination Grid for the Automated Ultrasonic Inspection of Heavy Rotor Forgings

Abstract

Heavy rotor forgings, in particular for the power generation market, are highly stressed components and the ultrasonic inspection is the most important method to guarantee a sufficient material quality throughout the volume. This is why more and more heavy rotor forgings have to be inspected using automated inspection systems guarantying a high probability of detection for possible flaws, good documentation as well as highly repeatable inspection.
In contrast to manual inspection, automated inspection does not allow for an optimization of a flaw reflection by moving the probe, as the probe is continuously moved over the part surface in distinct scan lines, resulting in a distinct pattern of inspection points. To ensure full volume coverage using overlapping ultrasonic beams from neighboring inspection points, a precise definition of an examination grid is required.
To assure that all critical errors are detected, multiple scan directions have to be applied as per VGB-R 504 M [1] to inspect the complete volume, resulting in a high inspection duration. Moreover most of the rotor forgings have a low sound attenuation, resulting in low pulse repetition rates and even longer inspection times. An ideal inspection grid will therefore make sure the full forging volume is covered by the inspection and reduce the inspection duration to a necessary minimum at the same time.
Several standards currently specify an examination grid for manual inspection, which are not simply transferrable to automated inspection.
This paper presents a solution to this problem, developed by the subcommittee “Automated UT” of the national German society for NDE (DGZfP).

Full text

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid
for the Automated Ultrasonic Inspection of
Heavy Rotor Forgings
DGZfP Committee Ultrasonic Testing
Subcommittee Automated UT

ASNT Annual Conference 2013 Dr. Johannes Vrana
Automated Shaft Inspection System
Saarschmiede, Völklingen

ASNT Annual Conference 2013 Dr. Johannes Vrana
Terms
Dx
dx
dy
DyScan direction
Index direction
•dx: Increment in scan
direction
(Defined by pulse repetition
rate and examination
speed)
•dy: Distance between two
adjacent laps in index
direction
•Dx, Dy: Dimensions of the
ultrasonic beam

ASNT Annual Conference 2013 Dr. Johannes Vrana
Automatisierte Disc Inspection System
Schmiedewerke Gröditz

ASNT Annual Conference 2013 Dr. Johannes Vrana
Automated Shaft Inspection System
GE Sensing & Inspection Technologies, Alzenau

ASNT Annual Conference 2013 Dr. Johannes Vrana
Automated Shaft Inspection System
KARL DEUTSCH Prüf- und Messgerätebau GmbH + Co KG, BGH Siegen

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
•Required for heavy rotor
forgings
•Limited optimization
regarding flaw reflection
•Recorded in distinct pattern
•Full volume coverage
required
Automated UT Multiple Scans
•Required by VGB-R 504 M
Low Sound Attenuation
Limited pulse repetition
rates
Limited inspection speed
Cost of ultrasonic inspection depends directly on examination grid
(both in scanning and index direction)
High inspection duration

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
Requirements in Current Standards
EN 10228-3:1998 - Non-destructive testing of steel forgings
Dx
dx
dy
DyScan direction
Index direction
Requirement
•Overlap of at least 10% of
the effective active element
size
•No requirement in scan
direction
Issues for AutoUT
•Apparently assumes a high
pulse-repetition-rate and
slow probe movement.
•Shape and size of the
sound bundle not
considered

ASNT Annual Conference 2013 Dr. Johannes Vrana
Requirements in Current Standards
EN 583-1:1998 - Ultrasonic examination - General principles
Requirement
•Based on size of -6 dB beam
•Requirement in scan and index direction
•The beam of two adjacent -6 dB beams
have to touch
Issues for AutoUT
•Some zones are not inspected with the
required sensitivity
•No formulas provided how to determine an
examination grid
Dx
dx
dy
Dy

ASNT Annual Conference 2013 Dr. Johannes Vrana
Requirements in Current Standards
ASTM E 2375 - 08 - Ultrasonic Testing of Wrought Products
Requirement
•Based on size of -6 dB beam
•Index direction: Overlap of at least 20% of
the effective beam width size
•Scan direction: Scanning speed limited by
detectability of the reference reflectors
Issues for AutoUT
•Some zones are not inspected with the
required sensitivity
•No formulas provided how to determine an
examination grid
Dx
dx
dy
Dy

ASNT Annual Conference 2013 Dr. Johannes Vrana
Requirements in Current Standards
Summary
•Overlap of
at least 10%
of the
effective
active
element
size
EN 10228-3 SEP1923 IIW Handbook
•Overlap of
at least 15%
of the active
element
size
•The beam of two adjacent -
6 dB beams have to touch
EN 583-1
•No requirement in scan direction (or only by
limitation of scanning speed)
•Overlap of the beams –however not
considering the volume to be inspected
•Unclear:
•Effective element size
•Transducer width
•Some zones are not inspected with the
required sensitivity
•No formulas provided how to determine an
examination grid
ASTM E 2375
•Overlap of
at least 20%
of the
effective
beam width
size
ASTM A 418
•Indexing by
75 % of the
transducer
width

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
•Required for heavy rotor
forgings
•Limited optimization
regarding flaw reflection
•Recorded in distinct pattern
•Full volume coverage
required
Automated UT Multiple Scans
•Required by VGB-R 504 M
Low Sound Attenuation
Limited pulse repetition
rates
Limited inspection speed
Cost of ultrasonic inspection depends directly on examination grid
(both in scanning and index direction)
Motivation •Existing standards define examination grids for manual
inspection
Not simply transferable to automated
High inspection duration
Start of development of an Optimal Examination Grid
for the Automated Ultrasonic Inspection

ASNT Annual Conference 2013 Dr. Johannes Vrana
DGZfP Committee Ultrasonic Testing
Subcommittee Automated UT
Peter Archinger, GMH Prüftechnik, Nürnberg
Otto Alfred Barbian, Blieskastel
Dr. (USA) Wolfram Deutsch, Karl Deutsch, Wuppertal
Dr. sc. techn. Peter Kreier, Innotest, Eschlikon/CH
Roland Reimann, AREVA NP, Erlangen
Udo Schlengermann, Erftstadt
Herbert Willems, NDT Syst. & Services, Stutensee
Kay Drewitz, Schmiedewerke, Gröditz
Dr.-Ing. Alexander Zimmer, Saarschmiede, Völklingen
Frank W. Bonitz, Westinghouse, Mannheim
Mathias Böwe, BASF SE, Ludwigshafen
Klaus Conrad, Siemens AG Energy, Mülheim
Dr.-Ing. Werner Heinrich, Siemens AG Energy, Berlin
Dr. Johannes Vrana, Siemens AG Energy, München
Dr. Gerhard Brekow, BAM, Berlin
Wolfgang Kappes, Fraunhofer IZFP, Saarbrücken
Hans Rieder, Fraunhofer ITWM, Kaiserslautern
UT System Manufacturers
Forging Manufacturers (Users)
OEM
Research Institutes

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination
Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Longitudinal section through
adjacent beams
Definition of an Examination Grid
Situation
B: At the end of the near field
C: At 3.5 times near field
Longitudinal Section
A: Directly at the probe
Horizontal Section

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Normalized Grid Rating Rn
Definition of Rn
2
2
2
2
1
y
y
x
x
nD
d
D
d
R
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam
Rn= 1 –Gapless (at least single sampling)
dx
dy
Dx
Dy

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Normalized Grid Rating Rn
Definition of Rn
2
2
2
2
1
y
y
x
x
nD
d
D
d
R
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam
Rn= 2 –At least double sampling
dx
dy
Dx
Dy

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Normalized Grid Rating Rn
Definition of RnRn= 0,5 –Beams touching
2
2
2
2
1
y
y
x
x
nD
d
D
d
R
Rn= 1 - Gapless
Rn= 2 –Double Sampling Rn= 4 –Quadruple Sampling
0 1 2 3 4 5 6 7
Overlap:
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination
Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
4


y
y
x
x
dd
D
d
D
R
Definition of an Examination Grid
Average Grid Rating Rd
Definition of Rd
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam
Area of -6 dB beam
Rd= ----------------------------
Area of ex. grid
Example: Rd= 1
-6 dB beam
dx
dy
Dx
Dy
Examintion
Grid

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Average Grid Rating Rd
Definition of RdRn= 1; Rd≈ 1.81
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam
4


y
y
x
x
dd
D
d
D
R
Rn= 1; Rd≈ 1.57
Optimized
Examination Grid

ASNT Annual Conference 2013 Dr. Johannes Vrana
4


y
y
x
x
dd
D
d
D
R
Definition of an Examination Grid
Average Grid Rating Rd
Definition of Rd
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam
•Optimized examination grid in the case of:
 
n
x
xR
D
d
2
 
n
y
yR
D
d
2
Optimized Examination Grid
and
Optimizing the Examination Grid
durchschnittliche Rastergüte Rd
0,00
1,00
2,00
3,00
4,00
5,00
6,00
7,00
8,00
9,00
10,00
0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 1,00
dx/Dx
durchschnittliche Rastergüte Rd
Rn = 1
Rn = 2
Rn = 3
Rn = 4
Average Grid Rating Rd

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Longitudinal section through
adjacent beams
Definition of an Examination Grid
Situation
B: At the end of the near field
C: At 3.5 times near field
Longitudinal Section
A: Directly at the probe
Horizontal Section
Grid Rating
2
2
2
2
1
y
y
x
x
nD
d
D
d
R
4


y
y
x
x
dd
D
d
D
R

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
How to calculate the sound bundle –Basic Situation
Normal Straight Beam Probe on a Plane Surface
 
2 tanDs

  
6
2
x
D FB
6
2
y
D FL
Dual Element Probe on a Plane Surface
•dx: Increment in scan direction
•dy: Distance between two adjacent laps in index direction
•Dx, Dy: Dimensions of the ultrasonic beam
•s: Soundpath
•FB6, FL6: Focal Width & Length

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Different scans required
Definition of an Examination Grid
Situation
axial
radial radial / axial radial /
tangential radial /
tangential
radial
axial /
tangential
axial/
radial
axial/
radial
Scans

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Situation
Situation
Plane Convex Concave
Normal
Dual Element
Angle

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Probe moved on the surface of the component
•Examination grid dxan dyestablished at the surface
•Beam changes within the part
Definition of an Examination Grid
How to calculate the sound bundle
Angle Probe on a Plane Surface
Dx‘
s
α
φ
φ
For the calculation of the
examination grid the
projection of the beam to
the surface is necessary
   
   




2cos2cos 2sincos2
's
Dx

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Probe moved on the surface of the component
•Examination grid dxan dyestablished at the surface
•Beam changes within the part
Definition of an Examination Grid
How to calculate the sound bundle
Angle Probe on a Plane Surface
Dx‘
s
α
φ
φ
   
   




2cos2cos 2sincos2
's
Dx
 
2 tan
y
Ds

  

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Situation
Situation
Plane Convex Concave
Normal
Dual Element
Angle

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
How to calculate the sound bundle
Normal Straight Beam Probe on Convex Surface
 
2 tanDs

  
6
2
x
D FB
6
2
y
D FL
Dual Element Probe on Convex Surface
11
1
' arcsin sin( )
2 180
xD
DD
Ds




    



  


1
1
'2
x
xDD
DDs


Corrected by:
 
2 tan
y
Ds

  

ASNT Annual Conference 2013 Dr. Johannes Vrana
•E.g. from the outer diameter surface D1
Definition of an Examination Grid
How to calculate the sound bundle
Angle Probe on a Convex Surface
D‘+
D1
D‘-



 
1 1 1
22
11
1
1
1
22
' arcsin sin( ) arcsin sin( ) 2 180 2
with ( 2) 2 ( 2) cos( )
2sin( ) for 0
and 2sin for 0
2sin for 0
xD D D
Drr
r s D s D
D
rD
D

    

  

  

   
        

    
   

     






1
1
' in the case / 2 cos( )
and ' in the case
wit
/ 2 cos( )
hD s D
D s D






 
2 tan
y
Ds

  

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Situation
Situation
Plane Convex Concave
Normal
Dual Element
Angle

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
How to calculate the sound bundle
Normal Straight Beam Probe on Concave Surface
6
2
x
D FB
6
2
y
D FL
Dual Element Probe on Concave Surface
Corrected by:
 
2 tan
y
Ds

  
22
2
' arcsin sin( )
2 180
xD
DD
Ds




    



  


 
2
2
'2
x
xDD
DDs



ASNT Annual Conference 2013 Dr. Johannes Vrana
•E.g. from the inner diameter surface D2
Definition of an Examination Grid
How to calculate the sound bundle
Angle Probe on a Concave Surface
 
2 tan
y
Ds

  
 
2 2 2
22
22
22
' arcsin sin( ) arcsin sin( ) 2 180 2
with ( 2) 2 ( 2) cos( )
xD D D
Drr
r s D s D

    


   
        

   
   

     

ASNT Annual Conference 2013 Dr. Johannes Vrana
Definition of an Examination Grid
Situation
Situation
Plane Convex Concave
Normal
Dual Element
Angle

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the
Examination Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
•For each scan
•Normalized Grid Rating Rn (gapless recommended)
•Examination zone
•Minimum soundpath s1
•Maximum soundpath s2
Determination of the Examination Grid
Necessary Specifications
Determination of Examination Grid
•Calculation of the projection of the sound bundle dimensions both for s1and s2
•s1 : Dx1, Dy1
•s2 : Dx2, Dy2
•Calculation of the optimized examination grid both for s1and s2considering the specified
normalized examination grid rating Rn
•s1 : dx1, dy1
•s2 : dx2, dy2
•Selection of the actually used examination grid dxand dy

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Calculation of the projection of the sound bundle dimensions both for s1and s2
•s1 : Dx1, Dy1
•s2 : Dx2, Dy2
•Calculation of the optimized examination grid both for s1and s2considering the specified
normalized examination grid rating Rn
•s1 : dx1, dy1
•s2 : dx2, dy2
•Selection of the actually used examination grid dxand dy
Determination of the Examination Grid
Determination of Examination Grid
Check of Examination Grid
•OK if both selected values are not bigger than the calculated values
•dxvs. dx1, dx2
•dyvs. dy1, dy2
•Otherwise needs to be tested by calculating Rn using dxand dyfor both Dx1, Dy1 and Dx2, Dy2

ASNT Annual Conference 2013 Dr. Johannes Vrana
Scan s1s2D‘x1 D‘x2 Dy1 Dy2 Rndx1 dx2 dy1 dy2 dxdyRn1 Rn2
(mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm)
Faces, axial, straight 100 300 12.9 38.8 12.9 38.8 1 9.1 27.4 9.1 27.4 9 9 1.03 9.3
Faces, axial/tang., 45° 100 424 13.4 40.1 6.6 19.9 1 9.5 40.1 4.7 19.9 9 4.5 1.09 9.8
OD, radial, straight 120 600 20.4 393 15.5 77.6 1 13.1 280 11.0 55 13 10.5 1.05 52
OD, radial, straight,
dual-element, 5120 514 1 3.6 9.9 3.5 10 1.02
OD, radial/tang., 14° 120 728 21.3 - 15.5 - 2 9.6 - 7.8 - 9.5 7.5 2.1 -
OD, radial/tang., 45° 350 1061 161 141 24 70 1 114 100 16.4 50 100 17 1.09 1.79
D1= 1500 mm, D2= 300 mm, L = 300 mm
°
Determination of the Examination Grid
Example
Disc
Examination Grid

ASNT Annual Conference 2013 Dr. Johannes Vrana
Scan s1s2D‘x1 D‘x2 Dy1 Dy2 Rndx1 dx2 dy1 dy2 dxdyRn1 Rn2
(mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm)
Faces, axial, straight 100 300 12.9 38.8 12.9 38.8 1 9.1 27.4 9.1 27.4 9 9 1.03 9.3
Faces, axial/tang., 45° 100 424 13.4 40.1 6.6 19.9 1 9.5 40.1 4.7 19.9 9 4.5 1.09 9.8
OD, radial, straight 120 600 20.4 393 15.5 77.6 1 13.1 280 11.0 55 13 10.5 1.05 52
OD, radial, straight,
dual-element, 5120 514 1 3.6 9.9 3.5 10 1.02
OD, radial/tang., 14° 120 728 21.3 - 15.5 - 2 9.6 - 7.8 - 9.5 7.5 2.1 -
OD, radial/tang., 45° 120 1061 28 141 10 70 1 19.9 100 7.1 50 20 7 1.0 33
D1= 1500 mm, D2= 300 mm, L = 300 mm
°
Determination of the Examination Grid
Example
Disc
Examination Grid

ASNT Annual Conference 2013 Dr. Johannes Vrana
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Introduction & Motivation
Requirements in Current
Standards
Definition of an Examination Grid
Normalized Grid Rating Rn
Average Grid Rating Rd
Determination of the Ultrasonic
Beam Dimensions
Determination of the Examination
Grid
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
•Current standards
•Not sufficient for the determination of an examination grid for automated UT
•New DGZfP Standard
•Harmonizes the calculation of the examination grid
•Defines
•Normalized Grid Rating
•Average Grid Rating
•How to calculate the UT beam dimensions
•Optimizes the inspection speed
•Can be adopted to other applications
Determination of an Optimal Examination Grid for the Automated
Ultrasonic Inspection of Heavy Rotor Forgings
Summary

ASNT Annual Conference 2013 Dr. Johannes Vrana
Thanks for paying
attention to all the
formulas
 
1 1 1
22
11
1
1
1
22
' arcsin sin( ) arcsin sin( ) 2 180 2
with ( 2) 2 ( 2) cos( )
2sin( ) for 0
and 2sin for 0
2sin for 0
xD D D
Drr
r s D s D
D
rD
D

    

  

  

   
        

    
   

     






1
1
' in the case /2 cos( )
and ' in the case
wit
/ 2 cos( )
hD s D
D s D






2
2
2
2
1
y
y
x
x
nD
d
D
d
R
4


y
y
x
x
dd
D
d
D
R
   
   




2cos2cos 2sincos2
's
Dx
11
1
' arcsin sin( )
2 180
xD
DD
Ds




    



  


1
1
'2
x
xDD
DDs

