structural engineering parametric study of web shear cracking

40
Structural Engineering Concrete Structures Kristian Edekling & Lars Rettne Parametric Study of Web Shear Cracking Lars Rettne Based on the Master’s Project “Improved Design Method for Web Shear Tension Failure in Hollow Core Units” in the International Master’s Programme Structural Engineering

Upload: others

Post on 15-Feb-2022

2 views

Category:

Documents


0 download

TRANSCRIPT

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Parametric Study of Web Shear Cracking

Lars Rettne

Based on the Master’s Project “Improved Design Method for Web Shear Tension Failure in Hollow Core Units” in the International Master’s Programme Structural Engineering

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Web Shear Tension Failure in Hollow Core Units

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Problem Description

”In the present standard the guidance on howto calculate shear tension capacity is not detailed enough and a more precise methodis needed”

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Aim of the Project

• Create finite element (FE) models of single hollow core units

• Perform calculations using Yang’s methodof single hollow core units

• Compare the shear capacity and location of the critical point between the two methods

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Critical Point= The point where web shear cracking starts

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Collaboration

• This project was initiated by Strängbetong, which is the largest hollow core unit producer in Sweden

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Hollow Core Floors

Hollow core floors consist of hollow core units

Hollow core floor Hollow core unit

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Web Shear Tension Failure in Hollow Core Units

Uncracked cross-section σ1 < fctm

σ1

σ1 = Principal stressfctm = Mean tensile strength of concrete

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Web Shear Tension Failure in Hollow Core Units

Web shear cracking σ1 = fctm

Immediately failure (brittle)

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Shear Capacity with Respect to Web Shear Tension Failure

• Failure when the concrete tensile strength is reached, σ1 = fctm

• The principal stresses are dependent of the shear and normal stresses, σ1 = σ1(σx, τ)

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Shear Stress Calculations in Prestressed Members

Yang’s (new)Traditional (present)

P0i P0i P0iP0i

x xσp

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Shear Stress Calculations in Prestressed Members

Yang’s (new)Traditional (present)

)()()(

),(I cpw

cpcpcpcp zbI

xVzSzx =τ )(

)()()(

),( 0

I dxdPf

zbIxVzS

zx i

cpw

cpcpcpcp +=τ

The shape is taken into account

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Shear Capacity Calculations in Prestressed Members

Yang’s (new)Traditional (present)

Assumed location of criticalpoint somewhere along line

Assumed location ofcritical point

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Reference Case – Principal Sketch

Prestressing strand

Cross-section

Self weight and imposed load

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Parametric Study

• Prestressing strand arrangement• Prestressing strand amount• Prestress• Concrete strength class• Concrete strength at strand release• Prestress transfer function• Type of cross-section

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

FE Model

Steel strands: Truss finite elementsStiff steel support plate: Solid finite elements

Concrete part: Solid finite elements

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

FE Model - Boundary Conditions

Roller support

Solid concrete part

Stiff support plate

Line where the boundary conditions were applied

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

FE Model - Boundary Conditions: Symmetry Planes

Boundary conditions along transversal symmetry plane

Boundary conditions along longitudinal symmetry plane

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

FE Model - LoadsSelf weight & Imposed load (pressure)

Initial stress from prestressing (initial condition)

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

FE Model - Mesh

dense mesh (25 mm)

Non-critical region coarse mesh (200 mm)

Critical region

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Results – FE MethodPrincipal Stresses in the Reference case

Critical point

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Results – FE MethodStress Field in the Reference Case

Short term response(28 days age)

Long term response(57 years age)

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Results – FE MethodStress Field in the Reference Case

Short term response(28 days age)

Long term response(57 years age)

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Results – Yang’s Method

Critical point

Principal Stresses in the Reference Case

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Shear CapacitySupport Reaction at Web Shear Cracking

Lin Yang, 35° Critical Path from Mid Support

100

200

300

400

500

600

REFSTIFF

PSArPSAm8PSAm12 PS CC

CSAR

PTFTOCS

TOCS-PSAr

Supp

ort R

eact

ion

[kN

]

t=28 days t=57 years

Support Reaction at Web Shear CrackingFE Analyses

100

200

300

400

500

600

REFSTIFF

PSArPSAm8PSAm12 PS CC

CSAR

PTFTOCS

TOCS-PSAr

Supp

ort R

eact

ion

[kN

]

t=28 days t=57 years

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Shear CapacityRelative Support Reactions at Web Shear Cracking

RLin Yang 35°/RFE Analyses

0,90

0,95

1,00

1,05

1,10

REF

STIFF PSAr

PSAm8

PSAm12 PS CC

CSAR

PTF

TOCSTOCS-PS

Ar

RLi

n Y

ang

35°/R

Aba

qus [

]

t=28 dayst=57 years

Comparison between FEM and Yang’s Method

Yang’s method unconservative

Yang’s method conservative

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Location of Critical PointPosition of Critical Point For the Reference Model (REF)

Comparison Between FE Analyses / Yang

00,020,040,060,08

0,10,120,140,16

0 0,05 0,1 0,15 0,2 0,25

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

-Bo

ttom

Uni

t [m

]

REF 28 days, FEM REF 57 years, FEMREF 28 days, Yang REF 57 years, Yang

Position of Critical Point For the Model with Stiffener (STIFF)Comparison Between FE Analyses / Yang

00,02

0,040,060,08

0,10,12

0,140,16

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Botto

m U

nit [

m]

STIFF 28 days, FEMSTIFF 57 years, FEMSTIFF 28 days, YangSTIFF 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Prestressing Strand Arrangement (PSAr)

Comparison Between FE Analyses / Yang

PSAr 28 days, FEMREF 28 days, FEM

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

] PSAr 28 days, FEMPSAr 57 years, FEMPSAr 28 days, YangPSAr 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Prestressing Strand Amount (PSAm8)

Comparison Between FE Analyses / Yang

PSAm8 28 days, FEMPSAm8 57 years, FEM

0

0,05

0,1

0,15

0,2

0,25

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

PSAm8 28 days, FEMPSAm8 57 years, FEMPSAm8 28 days, YangPSAm8 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Prestressing Strand Amount (PSAm12)

Comparison Between FE Analyses / Yang

PSAm12 28 days, Yang

PSAm12 57 years, Yang

0

0,05

0,1

0,15

0,2

0,25

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

PSAm12 28 days, FEMPSAm12 57 years, FEMPSAm12 28 days, YangPSAm12 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Prestress (PS)Comparison Between FE Analyses / Yang

PS 28 days, FEM

PS 57 years, FEM

REF 28 days, FEM

REF 57 years, FEM

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 0,05 0,1 0,15 0,2 0,25

Horizontal Position from Mid Support [m]V

ertic

al P

ositi

on fr

omB

otto

m U

nit [

m]

PS 28 days, FEMPS 57 years, FEMPS 28 days, YangPS 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Concrete Class (CC)

Comparison Between FE Analyses / Yang

CC 28 days, FEMREF 28 days, FEM

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

CC 28 days, FEMCC 57 years, FEMCC 28 days, YangCC 57 years, YangREF 28 days, FEMREF 57 year, FEM

Position of Critical Point For the Model with Changed Concrete Strength at Strand Release (CSAR)

Comparison Between FE Analyses / Yang

CSAR 57 years, FEM

CSAR 28 days, YangCSAR 57 years, Yang

REF 57 years, FEM

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0 0,05 0,1 0,15 0,2 0,25Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

CSAR 28 days, FEMCSAR 57 years, FEMCSAR 28 days, YangCSAR 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Prestressing Transfer Fuction (PTF)

Comparison Between FE Analyses / Yang

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

PTF 57 years, FEMPTF 57 years, YangREF 57 years, FEM

Position of Critical Point For the Model with Changed Type of Cross-Section (TOCS)

Comparison Between FE Analyses / Yang

TOCS 28 days, YangTOCS 57 years, Yang

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0 0,05 0,1 0,15 0,2 0,25

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

TOCS 28 days, FEMTOCS 57 years, FEMTOCS 28 days, YangTOCS 57 years, YangREF 28 days, FEMREF 57 years, FEM

Position of Critical Point For the Model with Changed Type of Cross-Section and Prestressing Strand Arrangement (TOCS-PSAr)

Comparison Between FE Analyses / Yang

TOCS-PSAr 28 daYang

TOCS-PSAr 57 yeYang0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0 0,05 0,1 0,15 0,2 0,25

Horizontal Position from Mid Support [m]

Ver

tical

Pos

ition

from

Bot

tom

Uni

t [m

]

TOCS-PSAr 2TOCS-PSAr 5TOCS-PSAr 2TOCS-PSAr 5REF 28 days, FREF 57 years,

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Location of Critical PointFE Analyses

Horizontal and Vertical Location of Critical Point Independently of Web, t =28 days, FE Analyses

PSAm8

PSAm12

TOCS

REFPSAr

PSCCCSARSTIFF

TOCS-PSAr

0

0,05

0,1

0,15

0,2

0,25

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Distance from Middle Support [m]

Ver

tical

Dis

tanc

e fr

om

-B

otto

m E

dge

[m]

REF PSAr PSAm8 PSAm12PS CC CSAR PTFSTIFF TOCS TOCS-PSAr

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Location of Critical PointYang’s Method

Horizontal and Vertical Location of Critical Point Independently of Web, t =28 days, Yang's Method

REF

PSAr

PSAm8

PSAm12PS

CCCSAR PTF STIFF

TOCSTOCS-PSAr

00,020,040,060,080,1

0,120,140,160,18

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Distance from Middle Support [m]

Ver

tical

Dist

ance

from

B

otto

m E

dge

[m]

REF PSAr PSAm8 PSAm12PS CC CSAR PTFSTIFF TOCS TOCS-PSAr

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Location of Critical Point:Reference Case

Location of Critical Point For the Reference Case (REF)Comparison Between FE Analyses / Yang's Method

0

0,020,04

0,06

0,08

0,10,12

0,14

0,16

0 0,05 0,1 0,15 0,2 0,25

Horizontal Distance from Middle Support [m]

Ver

tical

Dis

tanc

e fr

om

-B

otto

m E

dge

[m]

REF 28 days, FEM REF 57 years, FEMREF 28 days, Yang REF 57 years, Yang

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Location of Critical Point: Prestressing Strand Amount, PSAm8

Location of Critical Point For the Case with Changed Prestressing Strand Amount (PSAm8)

Comparison between FE Analyses / Yang's Method

PSAm8 28 days, FEMPSAm8 57 years, FEM

0

0,05

0,1

0,15

0,2

0,25

0 0,05 0,1 0,15 0,2 0,25 0,3

Horizontal Distance from Middle Support [m]

Ver

tical

Dist

ance

from

B

otto

m E

dge

[m]

PSAm8 28 days, FEMPSAm8 57 years, FEMPSAm8 28 days, YangPSAm8 57 years, YangREF 28 days, FEMREF 57 years, FEM

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Angle β for Different CasesAngle β** for Different Cases

FE Analyses

15

20

25

30

35

40

45

50

55

60

REF STIFF PSAr PSAm8 PSAm12 PS CC CSAR PTF TOCS TOCS-PSAr

Ang

le [D

egre

es]

t=28 dayst=57 years

β

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Angle β** for Different CasesFE Analyses

15

20

25

30

35

40

45

50

55

60

REF STIFF PSAr PSAm8 PSAm12 PS CC CSAR PTF TOCS TOCS-PSAr

Ang

le [D

egre

es]

t=28 dayst=57 years

Angle β for Different Cases

Recommended β from Yang (1994) β

Comparison between FEM and Yang’s Method

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Angle β for Different Cases

Angle β** for Different CasesFE Analyses

15

20

25

30

35

40

45

50

55

60

REF STIFF PSAr PSAm8 PSAm12 PS CC CSAR PTF TOCS TOCS-PSAr

Ang

le [D

egre

es]

t=28 dayst=57 years

Recommendations of New Angle

β

Suggested β for long term response and long transfer length

Suggested β for short term response and short transfer length

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Relative Shear Capacity

Relative Support Reactions at Web Shear CrackingRYang Ne w Su gge stion s/RFE Analyse s

0,900

0,950

1,000

1,050

1,100

REFSTIFF

PSArPSAm8PSAm12 PS CC

CSAR

PTFTOCS

TOCS-PSAr

RY

ang/R

Aba

qus [

]

t=28 dayst=57 years

New Recommended Angle is Used

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Conclusions

• For long term response and long transfer length, the critical point was found along a path with a smaller inclination than 35°proposed by Yang (1994).

• For long term response and long transfer length, the capacity of web shear tension failure was reduced.

General from FE Analyses

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Conclusions

• Decreased reinforcement amount had a great influence on the position of the critical point; the horizontal distance from the support to the critical point was increased. The capacity in web shear tension failure is reduced.

• Increased prestress from 1000 MPa to 1200 MPadid not affect the location of the critical point, but reduced the shear capacity.

Specific from FE Analyses

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Conclusions

• The agreement between Yang’s method and FEM regarding the location of the critical point was, with some exceptions, good for short term response and short transfer length and less good for long term response and long transfer length.

• The agreement between Yang’s method and FEM regarding the web shear tension capacity was, with some exceptions, good for short term response and short transfer length and less good for long term response and long transfer length.

General from Comparison FEM – Yang’s Method

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Recommendation

• Use β=35° for short term response and short transfer length with Yang’s method.

• Use β=25° for long term response and long transfer length with Yang’s method.

β

Structural Engineering

Concrete Structures Kristian Edekling & Lars Rettne

Thanks for your attention!