Research on behaviour of butt-welded joints with imperfectionseng-scoop.org/papers2014/IWME/1.ViktorStojmanovski.pdf · Research on behaviour of butt-welded joints with imperfections

Research on behaviour of butt-welded joints with

imperfections Static and dynamic behaviour

Vladimir Stojmanovski, MSc

Centre for Research, Development and Continuous Education – CIRKO, LLC

Inspection Body for Pressure Equipment, Metal Structures and Cableways

Skopje, Republic of Macedonia

[email protected]

Prof. Zoran Bogatinovski, PhD / Prof. Viktor Stojmanovski, PhD

Ss Cyril and Methodius University in Skopje

Faculty of Mechanical Engineering - Skopje

Skopje, Republic of Macedonia

[email protected]

[email protected]

Abstract—Due to discontinuities from various imperfections

found on the outer contour of the welded joint, there is irregular

stress distribution in the joint with elevated peaks. The influence

of these peaks cannot be precisely estimated during the

calculation of the joint. In practice, the solution of the problem,

in order to prevent the existence of such imperfections, lays in

establishment of rigorous criteria prescribed by the regulation.

Sometimes there is a question whether these rigorous criteria are

reasonable due to the fact that they directly influent the costs of

the welded structure. On the other side, in some separate cases

as far as there is discontinuity, it is very likely that during the

reparation the situation might worsen, particularly if there is a

location where the reparation is hard to be made. Considering

these facts, in some cases, it is necessary to make judgment

whether there is need to make reparation on the discontinuities

found on the outer contour during the examination. The purpose

of this paper is to endorse the influence of the discontinuities on

the capacity of the welded joint in order to make appropriate

judgment.

Keywords—welded joint, butt-weld, quality assessment,

imperfection, material testing, static testing, dynamic testing, FEA.

I. INTRODUCTION

The behaviour of butt welds with discontinuity of the outer contour caused by various imperfections

1 is subject of analysis

in this paper. Various test-samples (plates) from materials S355J2G3 (material 1) and S235JR (material 2) are welded for the testing. The selected materials are commonly used for the production of welded structures. For the test, various imperfections are simulated in the welds of the plates. Standard test-samples for mechanical examination are formed from the plates. Static load tests of the basic material and the welded joints are performed for all considered cases.

1 see Table 3

For some characteristic cases dynamic test was performed too. Stress distribution with real dimensions of the models and defects was analysed with Finite Element Method using ALGOR software.

II. BASIC MATERIAL

Chemical composition and mechanical properties of the used materials gained from the material tests are presented in Table 1 and Table 2.

From the test results it may be concluded that chemical composition and mechanical properties meet the requirements of the standard EN 10025 for both examined materials.

TABLE I. CHEMICAL COMPOSITION TABLE II. MECHANICAL PROPERTIES

Scientific Cooperations International Workshops on Engineering Branches 8-9 August 2014, Koc University, ISTANBUL/TURKEY

261

mailto:[email protected]



According to the tests of the material it may be concluded that chemical composition and mechanical properties meet the requirements of the standard EN 10025 for both examined materials.

III. WELDING TECHNOLOGY

Welding of the plates was performed with TIG welding procedure (141) for the root weld and ARC welding procedure for filling and finish. ARC welding was done with basic electrode type E424 32 X5 according to EN499 and E7018 according to AWSA 5.2

Fig. 1. Welding technology [1-root TIG (141), 2-filling ARC (111), 3-finish ARC (111)]

The macrostructure of the joint is presented on Figure 2.

Fig. 2. Macrostructure of the welded joint (x2,5)

Microstructure in some typical locations is presented on Figure 3.

Location 1

Basic Material (x100)

Location 2

Weld (x100)

Location 3

Transition zone (x100)

Location 4

Root of the weld (x100)

Location 5

Normalized zone (x100)

Fig. 3. Microstructure in characteristic locations

IV. TEST-SAMPLES

The test samples are produced by welding the plates made of material 1 and material 2. Characteristic imperfections of the weld are simulated and the corresponding marks are signed to the samples according to Table 3.

A. Permitted dimension deviations of the imperfections

The permitted sizes of the characteristic imperfections (defined by ISO6520), depending of the level of quality of the weld (B, C and D), are defined by ISO 5817. The dimensions of the analysed imperfections, for each particular case, are presented in Table 3. In most of the cases the measured dimensions of the analysed imperfections excess the maximum values permitted by ISO 5817. This conclusion is based on the visual examination of the welded joints. The results from the visual examination are provided in Table 3.

TABLE III. RESULTS FROM VISUAL INSPECTION AND DIMENSION

CONTROL FOR VARIOUS IMPERFECTIONS

Scheme Dimensions from the visual examination

S355J2G3 S235 JR

Grinded face and

grinded root of the

weld

/ /

Sample 1.1 Sample 2.1

Regular type of weld


Weld with excess

metal on the face

and the root


Weld with shallow

root


Incomplete root

penetration


Sagging,

incompletely filled

groove


Continuous undercut



262

B. Radiographic control of the joints

The welded test samples were radiography tested. Radiograms and the records of the analysed welded joints are presented in Table 4

TABLE IV. RECORDS FROM THE RADIOGRAPHY

S355J2G3 S235JR





V. TEST OF THE WELDED PLATES

Results from the tests of the basic material S355J2G3 and related welded plates with various discontinuities (marked with 1.1 to 1.7) are presented in Table 5.

For assessment of the capacity of the joint, most relevant element is the breaking force Fm. For both materials the breaking force is presented on Fig. 4 and Fig. 5 respectively.

TABLE V. TEST RESULTS FOR MATERIAL S355J2G3

Test

sample

A0

(mm2)

Rp0,2

(N/mm2)

Fm

(N)

RM

(N/mm2)

Location of

fracture

1. 232,75 522 147140 632 Δ 5=22%

1.1 227,24 500 138100 607 Basic material

1.2 232,75 507 141700 608 Basic material

1.3 229,90 532 146430 636 Basic material

1.4 236,55 482 144760 610 Weld

1.5 197,60 442 106320 538 Weld 1.6 196,00 553 136120 697 Weld 1.7 196,00 542 139490 681 ZTI

Fig. 4. Breaking force for material S355J2G3

Results from the tests of the basic material S235JR and the appropriate welded plates with various discontinuities (signed with 2.1 to 2.7) are provided in Table 6

TABLE VI. TEST RESULTS FOR MATERIAL S235JR

Test

sample

A0

(mm2)

Rp0,2

(N/mm2)

Fm

(N)

RM

(N/mm2)

Location of

fracture

2. 233,70 366 120980 517 Δ5=31,5%

2.1 220,80 386 123290 558 Basic material

2.2 229,90 368 123200 535 Basic material 2.3 217,80 388 122240 561 Basic material 2.4 227,85 342 126280 554 Basic material 2.5 188,76 381 91760 486 Weld

2.6 212,96 393 126580 594 Basic material

2.7 215,60 355 119520 554 ZTI

Fig. 5. Breaking force for material S235JR


263

VI. FINITE ELEMENT ANALYSIS OF THE SAMPLES

Stress distribution for the specific cases made of material

S355J2G3 (related to Table 3), is presented in Table 7

TABLE VII. STRESS DISTRIBUTION AND IMPERFECTIONS / MATERIAL

S355J2G3

1.1

1.2

1.4

1.7

Stress distribution for the characteristic cases made of material S235JR (related to Table 3), is presented in Table 8

At sufficient distance from the welded joint, due to the force Fsr (see Figure 6), there is steady stress condition for each separate case. Analysing the stress distribution could be concluded that at the location where the discontinuity ends -

the stress peak occurs. These peaks are close to the yielding stress of the used material.

TABLE VIII. STRESS DISTRIBUTION AND IMPERFECTIONS / MATERIAL

S235JR

2.1

2.2

2.4

2.7

From the results of the Finite Element Analysis can be

summarized:

The imperfections (defects) have significant influence on the stress distribution,

Near the imperfections (defects) the stresses achieve values that are close or equal to the yielding stress of the material,

Verification of good FEA modelling is based on the fact that at sufficient distance from the welded joint (defect) there is steady stress condition that represents the stress calculated


264

when the force Fsr is divided by the area of the cross section of the particular test sample.

VII. DYNAMIC TEST

Determination of the dynamic strength D with the discontinuities considered in this paper, for each particular case is far extensive, enduringly and expensive. Consequently, only the welded samples 1.1, 1.2, 1.5 from material S355J2G2 and 2.1, 2.2 and 1.5 from material S235JR are considered in the test.

The tests were performed on the device type “Alfred Amsler & Co Tchaffhausen Schweitz 131/63”.

The samples were dynamically analysed with tensional pre-stressed one directional cyclic loading. The test cycle is presented on Figure 6.

Fg - maximum loading

Fd - minimum loading

Fa - amplitude

Fsr - mean value loading

Fig. 6. The test cycle

The parameters of the tests are presented in Table 9

TABLE IX. DYNAMIC TEST PARAMETERS

S355Ј2G3 S235JR

Test samples 1.1, 1.2 and 1.5 Test samples2.1, 2.2. and 2.5

Fmax = 115 KN Fmin= 45 KN

Fa=35 KN

Fsr=80 KN f=250 cycles per minute

Fmax = 95 КN Fmin = 35 КN

Fa= 30 КN

Fsr= 65 КN f=250 cycles per minute

The number of cycles when the fracture occurred, for each analysed case, is presented on Table 10

TABLE X. NUMBER OF CYCLES UNTIL FRACTURE

Material Sample Number of cycles

S355J2G3

1. 130000

1.1 33500

1.2 24500

1.5 1300

S235JR

2. 150000

2.1 76700

2.2 69500

2.5 1800

The appearance of the samples before and after the test and the appearance of the fracture are presented in Table 11 and Table 12 for each case respectively.

TABLE XI. THE APPEARANCE OF THE SAMPLES FROM MATERIAL

S355Ј2G3

S355Ј2G3

Sample Appearance of the sample and the fracture

1.

Before

After

1.1

Before

After

1.2

Before

After

1.5

Before

After

TABLE XII. THE APPEARANCE OF THE SAMPLES FROM MATERIAL S235JR

S235JR

Sample Appearance of the sample and the fracture

2.

Before

After

2.1

Before

After

2.2

Before

After

2.5

Before

After


265

VIII. COMPARATIVE ANALYSIS

A. Analysis of the results from the test of the welded plates

The performed static examination of both tested materials S355J2G3 and S235JR and the related welded joints with various discontinuities are done according to the existing standards for each testing.

1) Comparative analysis of the materials S355J2G3 and

S235JR (sample 1. and sample 2.) – case without weld

Comparing the test results for both materials it can be

summarized:

The material S355J2G3 has significantly better mechanical properties than the material S235JR,

Observing the deformation properties, the situation is contrary. The material S235JR has better deformational characteristics than the material S355J2G3.

Both materials have good weld-ability. For larger dimensions (thickness), better weld-ability goes for material S235JR. This is due to the fact that S355J2G3 contains more carbon, causing more Cekv which influents the weld-ability.

Material S355J2G3 has better strength properties due to its finer granular microstructure caused by addition of microelements in the fusion.

2) Comparative analysis of the separate test-samples

a) Analysis of the results from the tests of the weld with

grinded face and grinded root (samples 1.1 and 1.2)

The tension test of welded joint with grinded face and grinded root (sample 1.1) determined breaking force Fm=138100N. The area of the cross section is 227,24mm

2. If

the breaking force is reduced to the same cross section of the basic material without the weld there is:

The decrease of the capacity of the welded joint compared to the capacity of the basic material is about 6%. This is due to the thermal influence during the welding process. It is important to highlight that the fracture in the test of the sample 1.1 occurred in the basic material.

The tension test of the welded joint with grinded face and grinded root (sample 2.1) determined breaking force Fm

(2.1)=123290N. The area of the cross section is 227,24mm

2.

This force is close to the breaking force of the basic material Fmom=120890N. The deviation is tolerable and the breaking occurred in the basic material

b) Analysis of the results from the tests of the regular

type of weld (samples 1.2 and 2.2)

The nominal thickness of the plate1 and plate 2 is 10mm and the actual measured thickness is t=9,5mm. From the amount of the excess metal on the weld and the root it can be concluded that the welded plates of the models 1.2 and 2.2

meet the requirements of the quality level B according to ISO 5817

The tension test of the welded joint with the regular weld 1.2 determined breaking force Fm

(1.2)=141700N. By comparing

the results of the basic material 1 and sample 1.1 it can be summarized:

Regular welded joint form the sample 1.2 has insignificantly inferior capacity compared with the non-welded plate 1 (basic material). The difference is around 3,5%, which is in tolerable limits with reference to precision of the measurement of the sample dimensions. It might be concluded that the welded joint with regular dimensions 1.2 has the same capacity as the non-welded plate (basic material).

Stress concentration due to discontinuity of the outer contour (shown on table 7) doesn’t have significant influence on the capacity of the joint (difference inferior by about 3,5%), which is confirmed by the fracture of the sample in the basic material.

Welded joint of the regular weld 1.2, due to static load, has superior capacity compared to the joint with grinded surface 1.1. This is due to the fact that by grinding of the surface the active area of the cross section is decreased.

The tension test of the welded joint with the regular weld 2.2 determined breaking force Fm

(2.2)=123200N. By comparing

the results of the basic material 2 and model 2.2 may be summarized:

Regular welded joint form the sample 2.2 has slightly superior capacity compared with the non welded plate 2 (basic material) (Fm

(2.2)=123200N>Fm

(2)=120980N).

The excess metal and the absence of influence of stress concentration due to discontinuity of the outer surface due to the increased active cross section cause this.

Regular type of weld from the sample 2.2 has the same capacity as the grinded weld from sample 2.1 (Fm

(2.2)=123200 N≈Fm

(2.1)=123290 N).

Stress concentration due to discontinuity of the outer contour (shown on table 8) doesn’t have any influence on the capacity of the joint. Due to superior deformation characteristics (superior ductility) of the material S235JR, the influence of the stress concentration on the joint 2.2 is inferior compared to the joint from the sample 1.2.

The fracture of the test sample occurred at the location of the basic material that confirms above presented findings.

c) Analysis of the results from tests of the weld with

incomplete root penetration (samples 1.5 and 2.5)

From the visual examination and the dimension control was concluded that test samples 1.5 and 2.5 don’t meet the requirements of the criteria B, C and D according to ISO 5817.


266

Tension test of the welded joint with incomplete root penetration 1.5 determined breaking force Fm

(1.5)=106320 N.

By comparing the results of the basic material 1 and samples 1.1 and 1.2 it may be summarized:

The cross section area of the sample 1 is A0

(1)=232,75mm

2. The cross section area of the sample

1.5 is A0(1.5)=

197.60 =mm2 and A0

(1.5)=0,846A0

(1). After

the reduction Fm(1.5)

=125673N. The reduced force (capacity) Fm

(1.5) is inferior compared to the breaking

forces form the samples 1.1 and 1.2

The fracture occurs at the weld.

The stress concentration due to discontinuity of the outer contour of the weld (presented in Table 7) has significant influence of the joint capacity in static loading.

Tension test of the welded joint with incomplete root penetration 2.5 discovered breaking force Fm

(2.5)=91760N. By

comparing the results of the basic material 2 and samples 2.1 and 2.2 it can be summarized:

The cross section area of the sample 2 is A0(2)=

233,70 mm

2. The cross section area of the sample 2.5 is

A0(2.5)=

188,76 mm2. Therefore A0

(2.5)=0,81A0

(2). After

the reduction Fm(2.5)

=113284N. The reduced force (capacity) Fm

(2.5) is inferior compared to the breaking

forces form the samples 2.1 and 2.2

The fracture occurs at the weld.

The stress concentration due to discontinuity of the outer contour (presented in Table 8) has significant influence of the joint capacity due to static loading. This influence, compared to the sample 1.5 is inferior due to increased ductility of the material S235JR compared to material S355J2G3

B. Analysis of the results from the dynamic test

Observing the results from dynamic test can be summarized:

For the dynamic test it is peculiar to precisely define the maximum force Fmax, the minimum force Fmin and the frequency f (cycles per minute) for previously given (presumed) lifetime.

Discontinuities of the outer contour on the face and the root of the weld influence the dynamic strength of the joint. These discontinuities produce stress concentrations (peaks).

If the discontinuity is removed by grinding, then the better strength is going to be achieved.

In order to eliminate the discontinuity of the outer contour by grinding the plates, the inferior dynamic strength will be achieved compared to non-welded plate (basic material). This is due to the structural changes in the weld, the zone of temperature influence and the residual stresses from the welding process.

Incomplete root penetration is large stress concentrator. It is confirmed by significantly inferior dynamic strength of the welded joints with incomplete root penetration.

Analysing the fracture can be summarized that, at all times, breaking of the samples occurred at the locations of maximum stress concentration (stress peaks).

The failure of the dynamic strength for all cases of tested samples (1,1.1,1,1.5 and 2.2.2,2.5) is higher in material S355J2G3 (material 1) compared to S235JR (material 2). This confirms that stronger materials in static are more sensitive on stress concentration in dynamics. Therefore, the advantages of the stronger materials in static loading are vanishing in dynamics. This is verified by analysing the results from static and dynamic tests.

IX. CONCLUSIONS

From the results of the tests and the comparative analysis it can be concluded:

Materials S355J2G3 and S235JR, considered in this work, completely fulfil the requirements of the standard EN10025

The material S235JR is less sensitive on stress concentration than S355J2G3. That causes better dynamic behaviour and small lowering of dynamic strength of material S235JR compared to S355J2G3.

Welded joint with regular weld face from the sample 1.2 has the same capacity as plate without weld (basic material). The fracture of the sample occurred in the basic material. Stress concentration due to the static loading has no influence on the capacity of the welded joint.

Welded joint with regular weld face (from the sample 2.2) has the same capacity as the joint with grinded weld face and grinded root (from the sample 2.1) and it has superior capacity compared to the capacity of basic material. This confirms the fact that stress concentration in the situation of static loading doesn’t affect the capacity.

Welded joint with excess weld metal 1.3 has superior capacity compared to 1.1 and 1.2. This is due to the fact that excess metal increases the active cross section and stress concentration in the situation of static loading doesn’t affect the capacity.

Welded joint with excess weld metal 2.3 has superior capacity than to 2.1 and 2.2.

In the case of incomplete root penetration 1.4, if the root penetration is superior, than it influences the capacity of the welded joint. In the present case 1.4 the root penetration is minor and does not significantly influence the capacity. This is due to the fact that the basic material 1 (S355J2G3) is more sensitive on stress concentration compared to basic material 2 (S235JR).


267

Significantly incomplete root penetration (samples 1.5 and 2.5) where the depth of the incomplete penetration is higher compared to the allowable level, influences the capacity even in the static load condition. That is caused by decreased cross sectional area and the effect of stress concentration.

Incompletely filled groove on the one side of the weld, (samples 1.6 and 2.6), depending of the penetration, can influent the capacity in static and dynamic loading both. This influence is higher in the welds made of material S355J2G3.

Continuous undercut of both sides of the weld (samples 1.7 and 2.7), depending of the penetration, has influence on the capacity. The influence is higher in the dynamic conditions and in the welds made of material S355J2G3

In the situation of static loading, during the quality assessment of the welded joints (in term of imperfections that interrupt the outer contour of the weld) may be permitted certain violation of the dimension of imperfections associated to limit values prescribed in ISO 6520.

In the situations of dynamic loading, where the level of quality B is prescribed, in all cases, there is no need for grinding the face and the root. Certain violation of the dimensions of outer contour of the weld may be tolerated.

During the quality assessment of the welded joints the level of stress in the weld, the form of the stress and the form of the load of the structure must be considered.

During the quality assessment of the welded joints, in the case of dynamic loading, the material of the structure must be considered. This is due to the fact that sensitivity of the stress concentration for variety of materials is different.

Based on the results from this work, in some cases, the weld can be judged positively even if there are certain imperfections that are caused by discontinuity in the outer contour. For bringing such judgment the person must have good understanding of materials, welding, design etc.

REFERENCES

[1] Werner Mewes; Kleine Schweibkunde fur Maschinenbauer, 2 Auflage, VDI-Verlag GmbH, Du basic material sseldorf 1992.

[2] Георгиевски В.: Испитување и контрола на заварени врски и конструкции, Универзитет Св. Кирил и Методиј, Скопје, 1982

[3] Г.А. Николов. Сварние конструкции, Машгиз, Москва, 1982

[4] Zienkiewicz O.C., Taylor R.Z.: The Finite Element Method, Vol 1, Vol. 2, McGraw-Hill, London, 1991.

[5] Leung A. Y. T.: Dynamic Analysis of Thin-Walled Structures, Journal of Thin-Walled Structures, Volume 14, Issue 3, pp.209-222, Elsevier Science Ltd.,1992.

[6] Vinson J.R.: The Behavior of Thin Walled Structures: Beams, Plates and Shells, Kluwer Academic publishers, Doredecht, 1989.

[7] Neumann A. ,,Schweibtechnisnes Handbuch fur Konstrukterre” – Teil 1, DVS – Verlag GmbH, Dusseldorf, 1996.

[8] Neumann A. ,,Schweibtechnisnes Handbuch fur Konstrukterre” – Teil 3, DVS – Verlag GmbH, Dusseldorf, 1996.

[9] Neumann A. ,,Kompendium der Schweibtechnik”, Band 4: Berechung und Gestaltung von schweib konstruktionen, DVS – Verlag, Dusseldorf, 1997.


268

Documents

Research on behaviour of butt-welded joints with imperfectionseng-scoop.org/papers2014/IWME/1.ViktorStojmanovski.pdf · Research on behaviour of butt-welded joints with imperfections