8/3/2019 Analisis Kerusakan Dalam Unit Komponen Logam Menggunakan Teknik Termoelastik_Marine Transport_N_Sathon

http://slidepdf.com/reader/full/analisis-kerusakan-dalam-unit-komponen-logam-menggunakan-teknik-termoelastikmarine 1/1

• Around a defect region, there exists a high stress gradient and hence a thermalgradient. The irreversible process of heat diffusion occurs and leads to attenuationof thermoelastic response

• The thermoelastic equation accounting for the heat conduction can be expressedas a generalised heat conduction equation:

• Thermal response in terms of magnitude and phase change at any point and timeon the object can be obtained by solving the above equation

Thermoelastic Stress Analysis (TSA)

N. Sathon and J M Barton

School of Engineering Sciences, University of Southampton, UK 

Damage analysis in metallic components using thermoelastic techniques

Fig. 1: TSA equipmen t arrangement 

Experimental TSA Setup:

1. A highly sensitive infrared detecting system is used to acquire thermal response onthe specimen surface due to thermoelastic effect

2. Virtually adiabatic condition for a small stress gradient on a general object can beachieved by applying periodic loading to the object

Damage analysisackground

novation

heory

 AcknowledgementsThe Royal Thai Navy for financial support and

Dr Janice Barton

Specimens

• Three types of materials (Al, Steel, PMMA) are chosen• Each set of specimens made from the same material consists of three

specimens of different damage severity as shown in Fig.4

• The crack-like damage was represented by a through minute slotmanufactured by EDM

• Each specimen was tested at various loading frequency to observe thebehaviour of the non-adiabatic effect

-200

-100

0

100

200

300

400

500

600

700

800

1 11 21 31 41 51 61 71 81 91 101 1 11

Distance (pixel)

 U n c a

  l  i  b r a t e

  d

 U n

  i t

  ( S x

 a n

  d

 S y

  )

6Hz_Sx

12Hz_Sx

18Hz_Sx

24Hz_Sx

6Hz_Sy

12Hz_Sy

18Hz_Sy

24Hz_Sy

Out-of-phase data

In-phase data

6 Hz

12 Hz

18 Hz

24 Hz

In-phase image Out-of-phase image

a /t =0 .7 5 a /t =0 .5 a /t =0 .2 5 a /t =0 .7 5 a /t =0 .5 a /t =0 .2 5

a/t=0.75 a/t=0.5 a/t=0.25

Fig. 6: Example results Al-all oy specimens (load freq 2 and 5Hz), the line  plot was taken from a location of t he (white) arrow shown in Fig.5 

• 2-D FE simulations of thermoelastic effect were carried out to study the effect of damage severity and material properties on the thermal response

• There is a good agreement between FE results and thermoelastic data. Anydiscrepancy may be caused by the generic material properties used in the FEA.

• The more damage in the specimen, the larger phase difference around the damage

site• In PMMA specimens, the adiabatic condition is achieved at very low load frequency

because the material has low thermal diffusivity

• On the opposite face of the crack front, this region suffers from complex stresses andthe thermal response becomes more complex and can be predicted by FEA.

Finite Element Analysis

• TSA can be used to detect sub-surface flaw by assessing phase information from thethermoelastic data close to the damage site

• Validity of the proposed technique has demonstrated by conducting FE simulationsbased on the theory of thermoelastic effect

• FEA results have shown that damage severity can be established from the phase of ththermal response

Damage evaluation is essential for structural assessment to prevent a catastrophicfailure

Thermoelastic Stress Analysis (TSA) is a non-contacting technique for stress analysis

TSA has been used successfully in LEFM for determination of damage severity in termsof stress intensity factors from a crack tip stress field. However, a practical usage of TSA to determine the severity of an internal part-through defect does not exist

A new approach for stress and damage assessment of sub-surface flaws using phaseinformation from thermoelastic techniques is proposed

Experimental evidences show that it is possible to use this approach to locate andevaluate the severity of the sub-surface damage

Conclusions

Fluid Structure InteractionsResearch Group,

School of EngineeringSciences

t T 

t 

T C T k  kk 

t  p∂

∂+

∂

∂⋅=∇

σ α  ρ 

0

2

kk 

 p

t 

cT T  σ 

 ρ 

α ∆−=∆

0

Under adiabatic condition

Preliminary testing

Fig. 2: A l plate specimen with artificial crack like flaw damag e has been tested to observe the hermoelastic effect from the un-damaged face 

Fig. 3: Thermoelastic signal at different load frequency and damage severity 

-30

-20

-10

0

10

20

30

0 5 10 15 20

Distance from the notch (mm)

   P   h  a  s  e

   (   d  e  g   )

FEA_025

FEA_050

FEA_075-30

-20

-10

0

10

20

30

0 5 10 15 20

Distancefrom thenotch (mm)

   P   h   a   s   e

    (   d   e   g   )

TSA _0 25 TSA _0 50 TSA _07 5

-30

-20

-10

0

10

20

30

0 5 10 15 20

Distance from notch (mm)

   P   h  a  s  e

   (   d  e  g   )

F EA _0 25 F EA _0 50 F EA _0 75

-30

-20

-10

0

10

20

30

0 5 10 15 20

Distance from notch (mm)

   P   h  a  s  e

   (   d  e  g   )

T SA _0 25 T SA _0 50 T SA _0 75

Fig. 4: Defect geometry , the drawing only show s the middle section of the specimen in Fig. 5 

Fig. 5 Typical TSA data on the plan e x-y