Capacity check Theoretical background

 

The capacity check for fibre reinforced concrete is based on the calculation of the resistance of cross-section with the current fibre dosage and standard reinforcement and comparison of it with acting load to get equilibrium of forces.

The calculation precognitions for finding plane of equilibrium of steel fibre concrete with reinforcement are taken from Figure R.4 - Determining stresses and strains for steel fibre reinforced concrete from [1] DAfStb Guideline "Steel fibre reinforced concrete".

The following conditions of equilibrium must be fulfilled.

Fc+Fct+Fs++Fs-+NEd =0

Fc. zc + Fct.zct + Fs+. zs++Fs-.zs- + MEd =0

Following procedure is used for calculation of the Capacity check for fibre reinforced concrete.

The calculation of the height of the compression zone

Calculation of the reinforcement material parameters

The reinforcement material parameters are calculated independently for upper and lower surface, they can be generally different.

fyd+(-) = fyk+(-) / γs

Es+(-)

where

fyk+(-)

Characteristic yield strength of reinforcement, different according Type of calculated reinforcement:

  • Nearest - The strength of reinforcement, which is the nearest to the upper (lower) surface
  • Average - The average strength of all reinforcement which are on upper (lower) surface according to formula

fyk+(-) = ∑(As+(-),i ∙ fyk+(-),i) / ∑(As+(-),i)

where

As+(-),i

The area of i-th reinforcement bar on the upper (lower) surface

fyk+(-),i

Characteristic yield strength of reinforcement of i-th bar on the upper (lower) surface

γs

Partial factor for reinforcing steel, national dependent value in EN 1992-1-1, §2.4.2.4(1), for details see "National annexes theoretical background".

Es+(-)

Modulus of elasticity of reinforcement, different according Type of calculated reinforcement:

  • Nearest - The modulus of elasticity of reinforcement, which is the nearest to the upper (lower) surface
  • Average - The average modulus of elasticity of all reinforcement which are on upper (lower) surface according to formula

Es+(-) = ∑(As+(-),i ∙ E+(-),i) / ∑(As+(-),i)

where

As+(-),i

The area of i-th reinforcement bar on the upper (lower) surface

E+(-),i

Modulus of elasticity of reinforcement of i-th bar on the upper (lower) surface

Calculation of the concrete material parameters

Calculation of the design value of compressive concrete strength

fcd = αcc ∙ fck / γC

where

αcc

The coefficient taking account of long term effects on the compressive strength and of unfavourable effects resulting from the way the load is applied, national dependent value in EN 1992-1-1, §3.1.6(1)P, for details see "National annexes theoretical background".

fck

Characteristic compressive cylinder strength of concrete at 28 days

γC Partial factor for concrete, national dependent value in EN 1992-1-1, §2.4.2.4(1), for details see "National annexes theoretical background".

Calculation of the strains for concrete stress-strain diagram

Proper strains are used according selected type of stress-strain diagram

Calculation of the extra rotation point

The extra rotation point is used in the calculation

HeightRatio = 1 - (εc2(3) / εcu2(3))

StrainRatio = (εc2(3) / εcu2(3)

Calculation of the strength reduction factor for strength of concrete and position of concrete compressive force

For bilinear stress-strain diagram

η = 1 - (εc3 / εcu3) / 2

β = 1 - (εcu32 / 2 - εc32 / 6) / (εcu32 - εcu3 ∙ εc3 / 2)

For parabolic stress-strain diagram

η = 1 - {εc2 / [(n + 1) ∙ εcu2]}

β = 1 - {εcu22 / 2 - εc22 / [(n + 1) ∙ (n + 2)]} / [εcu22 - εcu2 ∙ εc2 / (n + 1)]

where

εc2(3)

The strain at the reaching the maximum strength for parabolic (bilinear) stress-strain diagram according to Table 3.1 from EN 1992-1-1

εcu2(3)

The ultimate strain for parabolic (bilinear) stress-strain diagram according to Table 3.1 from EN 1992-1-1

n The exponent for parabolic stress-strain diagram according to Table 3.1 from EN 1992-1-1

Calculation of the area of reinforcement

The area of reinforcement is calculated as the sum of all reinforcement for upper and lower surface on the 2D element

As+(-) = ∑As+(-),i

Calculation of the diameter of reinforcement

The diameters of reinforcement bars are calculated independently for upper and lower surface, they can be different.

Different procedure is used according Type of calculated reinforcement:

ds+(-) = ∑(As+(-),i ∙ ds+(-),i) / ∑(As+(-),i)

where

As+(-),i

The area of i-th reinforcement bar on the upper (lower) surface

ds+(-),i

The diameter of reinforcement of i-th bar on the upper (lower) surface

Calculation of the cover of reinforcement

The cover of reinforcement is calculated independently for upper and lower surface, they can be different.

Different procedure is used according Type of calculated reinforcement:

c- = 0.5 ∙ h + Y(z)-

c+ = 0.5 ∙ h - Y(z)+

where

h

Thickness of 2D element

Y(z)-

Y-coordinate (vertical) of the reinforcement which is the nearest to the lower surface

Y(z)+

Y-coordinate (vertical) of the reinforcement which is the nearest to the upper surface

c-(+) = 0.5 ∙ h - zs-(+)

where

h

Thickness of 2D element

zs-(+)

The lever arm of forces in reinforcement from the centre of gravity

Calculation of the lever arm of forces in reinforcement from the centre of gravity

The lever arms of forces in reinforcement from the centre of gravity are calculated independently for upper and lower surface, they can be different.

Different procedure is used according Type of calculated reinforcement:

zs- = -0.5 ∙ h + c- + 0.5 ∙ ds-

zs+ = 0.5 ∙ h - c+ - 0.5 ∙ ds+

where

h

Thickness of 2D element

c-(+)

The cover to the lower (upper) reinforcement

ds-(+)

The diameter of lower (upper) reinforcement

zs+(-) = ∑(As+(-),i ∙ zs+(-),i) / ∑(As+(-),i)

where

As+(-),i

The area of i-th reinforcement bar on the upper (lower) surface

zs+(-),i

The Y-coordinate (vertical) of i-th bar of reinforcement on the upper (lower) surface

Calculation of the effective depth of the cross-section

The effective depth of cross-section is calculated from cover and diameter of reinforcement

d = h - c-(+) - 0.5 ∙ ds-(+)

where

h

Thickness of 2D element

c-(+)

The cover to the lower (upper) reinforcement

ds-(+)

The diameter of lower (upper) reinforcement

The decision if lower or upper reinforcement characteristics are used is based on the value of bending moment (positive → lower reinforcement, negative → upper reinforcement.

For fibre concrete without longitudinal reinforcement (As,sum = 0 mm2) is d = h.

Calculation of the centre to centre distance from more compressive edge

The centre to centre distance from more compressive edge is calculated from cover and diameter of reinforcement

as = c-(+) + 0.5 ∙ ds-(+)

where

c-(+)

The cover to the lower (upper) reinforcement

ds-(+)

The diameter of lower (upper) reinforcement

The decision if lower or upper reinforcement characteristics are used is based on the value of bending moment (positive → lower reinforcement, negative → upper reinforcement.

For fibre concrete without longitudinal reinforcement (As,sum = 0 mm2) is as = 0 mm.

Calculation of the balanced parameters of the cross-section

The balanced parameters ξbal for balanced depth of compression zone and depth of the compression zones xbal are calculated as follows

The decision if εcu2 or εcu3 are used is based on the type of stress-strain diagram (εcu2→ parabolic, εcu3→ bilinear).

The decision if lower or upper reinforcement characteristics are used is based on the value of bending moment. For more tensile edge: positive → lower reinforcement, negative → upper reinforcement. For less tensile edge: positive → upper reinforcement, negative → lower reinforcement.

ξbal1,t = εcu2(3) / [εcu2(3) + (fyd-(+) / Es-(+))]

ξbal1,c = εcu2(3) / [εcu2(3) - (fyd-(+) / Es-(+))]

ξbal2,t = εcu2(3) / [εcu2(3) - (fyd+(-) / Es+(-))]

ξbal2,c = εcu2(3) / [εcu2(3) + (fyd+(-) / Es+(-))]

The balanced depth of compression zone is than calculated as follows

xbal1 = ξbal1,t(c) ∙ d

xbal2 = ξbal2,t(c) ∙ d

Calculation of the fibre concrete material parameters in tension

The fibre concrete material parameters are calculated according the recommendations from the Entwurf DAfStb Guideline for Steel fibre reinforced concrete.

The design value of tensile concrete strength

For calculation procedure of ffctd,L1 and ffctd,L2 see the chapter "Design values" in "Materials theoretical background".

The balanced depth of compression zone for fibre concrete

ξfbal,t = εcu2(3) / (εcu2(3) + εfct_L2)

where

εfct_L2

The strain in the fibre reinforced concrete at reaching residual tensile strength in the performance class 2

The tensile strain and stress in fibre concrete

The calculation of stress and strain in fibre concrete in divided to two branches according the ratio of xc / d with ξfbal,t.

εfct = εcu2(3) ∙ (d - xc) / xc

εfcc = εcu2(3)

IF ffctd,L2 < ffctd,L1

THEN ffctd = ffctd,L2 + (ffctd,L1 - ffctd,L2) ∙ (εfct_L2 - εfct) / (εfct_L2 - εfct_L12)

ELSE ffctd = ffctd,L1 + (ffctd,L2 - ffctd,L1) ∙ (εfct - εfct_L12) / (εfct_L2 - εfct_L12)

εfct = εfct_L2

εfcc = εfct_L2 ∙ xc/ (d - xc)

where

εfct_L12

The strain in the fibre reinforced concrete at starting descending branch of residual tensile strength in performance class 1

The reduction factor for tensile strength of concrete and position of concrete tensile force

Two support values are calculated

Afct = 0.5 ∙ (εfct_L12 - εfct_L11 + εfct) + 0.5 ∙ (ffctd / ffctd,L1) ∙ (εfct - εfct_L12)

Sfct = εfct_L112 / 3 + (εfct_L12 - εfct_L11)2 / 2 + [(ffctd,L1 - ffctd) / ffctd,L1] ∙ [(εfct - εfct_L12) / 2] ∙ [(εfct - εfct_L12) / 3 + εfct_L12] + (ffctd / ffctd,L1) ∙ (εfct - εfct_L12) ∙ [(εfct - εfct_L12) / 2 + εfct_L12]

The reduction factors are than calculated as follows

ηft = Afct / εfct

βft = Sfct / (Afct / εfct)

λft = 1.0

Estimation of the reinforcement stress for calculation of the compression zone

Stress in upper and lower reinforcement is estimated for the calculation of first iteration as "starting point". The stresses are generally set to yield strength of reinforcement.

IF mEd = 0

THEN σs+(-) = max(-εc2(3) ∙ Es+(-); -fyd+(-))

ELSE σs+(-) = -fyd+(-)

σs+(-) = -fyd+(-)

IF mEd ≥ 0

THEN σs+ = -fyd+ and σs- = fyd-

ELSE σs+ = fyd+ and σs- = -fyd-

Calculation of the balanced normal force

The balanced normal force nRd,bal is force in the cross-section, when balanced state is adjusted.

nRd,bal = - xbal1 ∙ λ ∙ b ∙ η ∙ fcd + (h - xbal1) ∙ λft ∙b ∙ ηft ∙ fctd,L1 + As+ ∙ σs+ + As- ∙ fyd-

Calculation of the height of compression zone

Three possibilities for calculation of the height of compression zone are available

xc = (-nEd + λft ∙ b ∙ h ∙ ηft ∙ fctd,L1 + As- ∙ σs- + As+ ∙ σs+) / (λ ∙ b ∙ η ∙ fcd + λft ∙ b ∙ ηft ∙ fctd,L1)

βtin = 1 - βft

A = λ ∙ b ∙ η ∙ fcd ∙ β + λft ∙ b ∙ ηft ∙ fctd,L1 ∙ βtin

B = -λ ∙ b ∙ η ∙ fcd ∙ 0,5 ∙ h - λft ∙ b ∙ ηft ∙ fctd,L1 ∙ h ∙ (2 ∙ βtin - 0,5)

C = |mEd| + As- ∙ σs- ∙ zs- + As+ ∙ σs+ ∙ zs+ - λft ∙ b ∙ ηft ∙fctd,L1 ∙ h2 ∙ (0,5 - βtin)

D = B2 - 4 ∙ A ∙ C

if D ≥ 0

THEN xc = (-B - √D) / (2 ∙ A)

ELSE xc = -999 and error message is shown

eccentricity: e = -|mEd / nEd|

βtin = 1 - βft

A = λ ∙ b ∙ η ∙ fcd ∙ β + λft ∙ b ∙ ηft ∙ fctd,L1 ∙ βtin

B = -λ ∙ b ∙ η ∙ fcd ∙ (0,5 ∙ h - e) - λft ∙ b ∙ ηft ∙ fctd,L1 ∙ h ∙ (2 ∙ βtin - 0,5 ∙ h + e)

C = As- ∙ σs- ∙ (zs- - e) + As+ ∙ σs+ ∙ (zs+ - e) - λft ∙ b ∙ ηft ∙fctd,L1 ∙ h2 ∙ (0,5 ∙ h - βtin ∙ h - e)

D = B2 - 4 ∙ A ∙ C

if D ≥ 0

THEN xc = (-B -(+) √D) / (2 ∙ A) (- when e < 0; + when e > 0)

ELSE xc = -999 and error message is shown

The calculation of the resistance of the cross-section

Calculation of the inner lever arm for concrete and reinforcement from centre of gravity

Inner level arm of compressive forces in concrete from centre of gravity

IF mEd ≥ 0

THEN zcc = 0,5 ∙ h - β ∙ xc

ELSE zcc = -0,5 ∙ h + β ∙ xc

Inner level arm of tensile forces in concrete from centre of gravity

IF mEd ≥ 0

THEN zct = - xc - (h - xc) ∙ βft + 0,5 ∙ h

ELSE zct = + xc + (h - xc) ∙ βft - 0,5 ∙ h

IF xc < 0, THAN zct = 0

Calculation of the stress in the reinforcement according to height of the compression zone

Three possible states for recognition of stress in reinforcement, based on comparison of xc with d and as.

IF mEd ≥ 0

THEN σs+ = -fyd+; σs- = fyd-

ELSE σs+ = fyd+; σs- = -fyd-

σs+ = fyd+; σs- = fyd-

IF mEd = 0

THEN σs+ = max(-εc2(3) ∙ Es+, -fyd+) ; σs- = max(-εc2(3) ∙ Es-, -fyd-)

ELSE σs+ = -fyd+; σs- = -fyd-

Calculation of the resistance

The resistance of the cross-section is based on calculation of forces in compressive and tensile concrete and in reinforcement and than these forces are multiplicative by inner level arms to obtain final resistances.

Acc = xc ∙ b

Act = (h - xc) ∙ b

σcc = -fcd

σct = fctd,L1

Fcc = Acc ∙ η ∙ σcc

IF xc > 0

THEN Fct = Act ∙ ηft ∙ σct

ELSE Fct = Act ∙ fctd,L2

Fs-(+) = As-(+) ∙ σs-(+)

nRd = Fcc + Fct + Fs- + Fs+

mRd = - Fcc ∙zcc - Fct ∙ zct - Fs- ∙zs- - Fs+ ∙ zs+

The calculation of the unity check

Final unity check is calculated according selected type of method for calculation

UC = mEd / mRd

UC = nEd / nRd

UC = max(mEd / mRd, nEd / nRd, (nEd2 + mEd2)0,5 / (nRd2 + mRd2)0,5)

Final value of unity check can be drawn in 3D scene in SCIA Engineer.

Check of the compression strut

On addition to capacity check is the check of compression strut. The internal forces which are calculated in the direction of the concrete strut have to be also checked. The internal forces have to be carried by concrete in compression.

The depth of the compression zone is calculated for each surface as average values from user defined reinforcement for both directions

xc,avg± = 0,5 ∙ (xc,1± + xc,2±)

where

xc,1±

Depth of the compression zone calculated from user defined reinforcement in the first direction

xc,2±

Depth of the compression zone calculated from user defined reinforcement in the second direction

Design value of resistance of concrete compressive strut is calculated according formula

nRd,sc± = -Acc± ∙ η ∙ Redfcd ∙ fcd

where

Acc±

The area of the compression concrete at upper (lower) surface

Acc± = xc,avg± ∙ 1 m

η

The coefficient for reduction design compressive strength of the concrete from concrete setup

Redfcd The coefficient for reduction strength of the concrete in compressive concrete strut from concrete setup
fcd The design compressive strength of concrete

Unity check is than calculated as follows

UCsc = max(nEd,sc+ / nRd,sc+, nEd,sc- / nRd,sc-)

The value of Unity check for compression strut is also part of capacity check UC

UC = max(UCcapacity, UCsc)