Investigation on Feature Guided WavesInvestigation on Feature Guided Waves David Fan, Mike Lowe...

Investigation on Feature Guided Waves

David Fan, Mike Lowe

Non-Destructive Testing Group

David Fan, Mike Lowe

Department of Mechanical Engineering

Imperial College London

SW7 2AZ

United Kingdom

1/18Feature Guided Wave NDT Laboratory

Motivation

• Large area inspection by guided wave


Discovery of feature guiding

• Inspection of plate with weld

Weld guided waveWeld guided wave

J S t (BAE t ) 2006


J.Sargent (BAE systems), 2006

Finite element modelling

Semi-Analytical Finite Element (SAFE) method:

3x2x 12 mm

1x

Ab bi iAb bi i S l PlSt l Pl t W ld

2 mm

6 mm

Absorbing regionAbsorbing region Steel PlateSteel Plate Weld

400 mm800 mm 800 mm


SAFE results – compression mode

6000Phase velocity 150 kHz

m/s Weld guided mode

6mm plate S0

10 l t S0

loci

ty in

m

SH0 in a steel plate

Weld guided mode10mm plate S0200 kHz

Phas

e ve

l

Leaky Non-300 kHLeaky

leaky 300 kHz

50 150 250 350 4500

Frequency in kHz


Discovery of a new mode – shear mode

3500Phase velocity

100 kHz3500

m/s

100 kHz

ocity

in m

SH0 in a steel plate 150 kHz

Phas

e ve

lo

Weld guided mode

P

250 kHz

50 100 150 200 2503000

Frequency in kHzFan & Lowe

Proc. R. Soc. A, 2009


,

Why the waves are trapped?

3x2x

1x

Weld Pl t

1

Weld Plate

SAFE method Analytical method



L dPh l it

• Condition 1: Similar mode shape Weld cap Plate

S0 dL mode6000

Phase velocity

Longitudinal mode

S0 mode

F mode 1in m

/s

SH0 mode

velo

city

i

Flexural mode 1

Torsional modeSH0 mode

Phas

e

Flexural mode 2F mode 2 A0 mode

250500

F i kH


25050 Frequency in kHz


• Condition 2: Lower velocity in the weld

plate(faster) weld

(slower)



• Condition 2: Lower velocity in the weldT d d

6000Phase velocity

S0 L mode

Trapped modes:

S0 mode

n m

/s

Longitudinal mode

velo

city

in

F mode 1

Torsional mode SH0

F mode 1 SH0 mode

Phas

e v F mode 1

F mode 2

A0

0 Exciting mode for inspection


25050 Frequency in kHzc t g ode o spect o

Typical results from FE (crack parallel to the weld)( p )

Absorbing regionMonitor 1

Monitor 2


Defect scattering study

Laser interferometer

Monitor

Shear transducer


Typical results from Experiment

30mm slot parallel to the weld

0.14

30mm slot parallel to the weld

0.1

0.12

oeffi

cien

t

0.06

0.08

ectio

n C

o40 50 60 70 80 90 100 110 120 130 1400.02

0.04Ref

l40 50 60 70 80 90 100 110 120 130

Frequency (kHz)


Discussion - crack parallel to the weld

Different length of the crack:

ent

30 mm24 mm16 mm

0.12

0.14SH0 wave interaction with axial crack on a plate

n C

oeffi

cie

0.08

0.1

Ref

lect

io

0 04

0.06

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.30.02

0.04

Length of crack to wavelength ratio


Discussion - crack parallel to the weld

FE results Experimental results

0.12

0.14

cien

t 0.12

0.1430 mm24 mm16 mm

0.08

0.1

ion

Coe

ffic

0.08

0.1

0.04

0.06

Ref

lect

i

0.04

0.06

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.30.02

0.04

L h f k l h i

0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.30.02

0.04

L h f k l h iLength of crack to wavelength ratio Length of crack to wavelength ratio

Possible reason: difference between the milled slot (experiment) and the FE crack


milled slot (experiment) and the FE crack

Discussion - crack normal to the weld

E i l l

0.24

0.26

t 0 35

0.4FE results Experimental results

0 18

0.2

0.2240 mm

Coe

ffici

ent

30 mm0.3

0.35

40 mm

0.14

0.16

0.18

20 mm

Ref

lect

ion

C

0 2

0.25

20 mm

70 80 90 100 110 120 130 1400.1

0.12

R

70 80 90 100 110 120 130 140

0.2 20 mm

Frequency (kHz) Frequency (kHz)

Possible reason: un-uniform of the weld geometry


the weld geometry

Discussion: Flat-bottom hole

Different depth of the hole:

0.2

0 12

0.14100%

FE results Experimental results

0.16

coef

ficie

nt

0.1

0.12

84 %

0 08

0.12

Ref

lect

ion

c

0.06

0.08

49 %

65 %

0.04

0.08R

0.02

0.04

17 %

33 %

0.3 0.40.2

Length of diameter to wavelength ratio0.2 0.25 0.3 0.35 0.4

0

Length of diameter to wavelength ratio


Summary

The feature guided wave has been studied by the SAFE h d B h h i d d hSAFE method. Both the compression mode and the shear mode have been investigated and compared.

The reason for the trapping effect has been di d d i ddiscussed and summarized.

Experiments have been taken to validate the existence of the trapping mode and its sensitivity to certain defects


Investigation on Feature Guided WavesInvestigation on Feature Guided Waves David Fan, Mike Lowe...

Documents

Transcript of Investigation on Feature Guided WavesInvestigation on Feature Guided Waves David Fan, Mike Lowe...