Professional Documents
Culture Documents
σ1 σ2 ψ 1
beff = ρ ⋅ b
be1 = 0.5 ⋅ beff
be1 be2
be 2 = 0.5 ⋅ beff
b
1> ψ ≥ 0
σ1 σ2 beff = ρ ⋅ b
2
be1 = beff
be1 be2 5 −ψ
be 2 = beff − be1
b
ψ<0
σ1
ρ ⋅b
σ2 beff = ρ ⋅ be =
1− ψ
be1 be2 be1 = 0.4 ⋅ beff
b be 2 = 0.6 ⋅ be1
ψ = σ 2 / σ1 1 1 ψ 0 0 0 ψ 1 -1 1 ψ 3
Figure 2: effective width for internal compression elements (from table 4.1 of [8])
σ1 1> ψ ≥ 0
σ2 beff = ρ ⋅ c
c
bt bc
ψ<0
σ1 ρ⋅c
beff = ρ ⋅ bc =
σ2 1− ψ
beff
ψ = σ 2 / σ1 1 0 -1 1 ψ 3
Buckling factor kσ 0.43 0.57 0.85 0.57 − 0.21ψ + 0.07 ψ 2
beff
1> ψ ≥ 0
σ1 beff = ρ ⋅ c
σ2
c
beff
ψ<0
σ1 ρ⋅c
σ2 beff = ρ ⋅ bc =
1− ψ
bc bt
ψ = σ 2 / σ1 1 1 ψ 0 0 0 ψ 1 -1
0.578
Buckling factor kσ 0.43 1.7 1.7 − 523.8
ψ +17.1ψ 2 1.7 − 5 ψ +17.1ψ 2
23.8
ψ + 0.34
Figure 3: effective width for outstand compression elements (from table 4.2 of [8])
fy
.
σ cr
3 2 α* i − αi
( ) ( )
a α i + b αi + c α i + d = 0 ( ) 14) Erri = 18)
αi
where constant terms a, b, c and d are expressed as: The value of tolerance “tol” must be defined for an acceptable er-
3
a = N Mu ,e ,i ,lim Mv ,e ,i ,lim 15a) ror limit, i.e. if Erri < tol, then αR=αi. Otherwise, a new estimate for
α and a new iteration is needed. It has to be noted that assuming
2
b = − N Mu ,e ,i ,lim Mv ,e ,i ,lim ( Ne ,i ,lim + Ncr ,u ,e ,i + Ncr ,v ,e ,i ) tol=0.001 it is possible to obtain results characterized by a more
- N Ne ,i ,lim ( Mu ,e ,i Ncr ,u ,e ,i Mv ,e ,i ,lim + Mv ,e ,i Ncr ,v ,e ,i Mu ,e ,i ,lim ) 15b) than satisfactory degree of accuracy with a quite limited number
of iterations.
c = N Mu ,e ,i ,lim Mv ,e ,i ,lim ( Ncr ,u ,e ,i Ncr ,v ,e ,i + Ne ,i ,limNcr ,u ,e ,i + Ne ,i ,limNcr ,v ,e ,i ) +
+ Ne ,i ,limNcr ,u ,e ,i Ncr ,v ,e ,i ( Mu ,e ,i Mv ,e ,i lim + Mv ,e ,i Mu ,e ,i lim ) 15c) 2.7 CORRECTION AND NEW ITERATION
If Erri > tol, a new iteration is needed. New value of the load mul-
d =− Ne ,i ,lim Mu ,e ,i ,lim Mv ,e ,i ,limNcr ,u ,e ,i Ncr ,v ,e ,i 15d) tiplier has to be defined in the correction phase. In the present
αi + αi *
αi +1 = 19)
2
Once the new value of the load multiplier is defined, the proce-
dure requires to jump to step 2 assigning i =i+1.
Ae,i
Qi = 0.674 0.807 0.861 0.674 0.926 0.928 0.765 1.000 1.000
Ag
Ie,u,i [mm4] 29095 49433 59949 29095 74244 74584 42318 93022 93022
Ie,v,i [mm4] 6753 11619 14147 6753 17592 17675 9912 22131 22131
3
We,u,i [mm ] 1198.4 1715.9 1954.9 1198.4 2258.7 2265.7 1545.3 2631.1 2631.1
We,v,i [mm3] 549.9 798.6 914.0 549.9 1061.0 1064.3 716.8 1242.1 1242.1
αi* 20877 37061 46604 11483 37221 37485 8205 18977 18977
Nexp
1.159 0.987 1.302
αR ⋅ N
Table 1: key data of the unified approach applied to equal angle specimens
for intermediate as well as last iteration for the shorter and longer
specimens, respectively. Initial Effective Effective Effective
cross-section cross-section cross-section cross-section
5 COMPARISONS WITH EXISTING STANDARDS at 1st iteration at 4th iteration at the final
(8th) iteration
The method herein proposed is different from Eurocode 3 part 1-3 Figure 9: evolution of the effective cross-section for L36F0280 specimen
for these main reasons:
• resistance and stability checks are done at the same time, i.e.
there is no need to perform a resistance check independent of
overall stability check;
• axial force and bending moments define a unique effective cross
section, there are not three different cross sections that are inde-
pendent from each other;
• overall stability is here performed using effective cross section
properties associated with the externally applied loads, while Eu-
rocode 3 part 1-3 uses an hydrid formula which uses gross-section
second area moment and effective section area to compute radius
of inertia to be used in slenderness definition formula;
• in the proposed method, changes in load levels do affect effec- Initial Effective Effective Effective
tive section, while the effective section area and bending moduli cross-section cross-section cross-section cross-section
at 1st iteration at 4th iteration at the final
used by Eurocode 3 part 1-3 are computed irrespective to the ap- (10th) iteration
plied load level. Figure 10: evolution of the effective cross-section for L36F3000 specimen
Ae,i
Qi = 0.774 0.823 0.826 0.772 0.854 0.862 0.773 0.951 0.968
Ag
Ie,u,i [mm4] 402881 407831 408082 402496 409776 410160 401959 411666 411690
Ie,v,i [mm4] 43858 46434 46596 43665 47873 48223 43800 52067 52233
3
We,u,i [mm ] 8281.2 8383.0 8388.1 8256.3 8405.7 8413.5 8253.8 8453.1 8453.6
We,v,i [mm3] 1978.8 2019.5 2021.9 1975.1 2040.2 2045.3 1983.9 2105.1 2107.3
αi* 83489 94459 95233 56700 69484 70754 25336 35746 36067
Nexp
1.052 1.165 1.090
αR ⋅ N
Table 2: key data of the unified approach applied to lipped channel specimens
REFERENCES
[1] Rodhes J., Design of cold formed steel mem- [6] Von Karman T., Sechler E.E., Donnel L.H., The Italian) Costruzioni Metalliche 4/1998
bers, Elsevier Applied Science, 1991. strength of thin plates in compression, Transac-
tions A.S.M.E., APM-54-5, , Vol. 54, 1932, pp. 53-57. [11] Castalia srl: “Sargon©” user’s manual.
[2] Ghersi A., Landolfo R., Mazzolani F.M., Design of
metallic Cold-formed thin-walled members, Spon [7] Winter G., Strength of thin steel compression [12] Popovic, G.J. Hancock and K.J. Rasmussen,
Press, 2002. flanges, Transactions A.S.C.E., paper n. 2305, Fe- Axial compression tests of cold formed angles,
bruary 1946, pp. 527-576. ASCE Journal of Structural Engineering, May 1999,
[3] CEN, ENV 1993-1-3:1996: Eurocode 3: Design of pp. 515-523.
steel structures, Part 1.3: General rules-Supplemen- [8] CEN, pr EN 1993-1-5:2006: Eurocode 3: Design
tary rules for cold-formed thin gauge members of steel structures, Part 1-5: Plated structural ele- [13] Young B. and Rasmussen J.R., Tests of fixed–
and sheeting, European Committee for Standardi- ments, (Draft February 2006) European Committee ended plain channel columns, Journal of Structu-
sation. for Standardisation. ral Engineering, vol. 124, nr. 2, May 1998, pp. 131
- 139.
[4] CEN, pr EN 1993-1-3:2005: Eurocode 3: Design [9] Bernuzzi C., Rugarli P. An approach for the pre-
of steel structures, Part 1-3: General rules-Supple- diction of the load carrying capacity of slender [14] Young B. and Rasmussen J.R., Behaviour of
mentary rules for cold-formed members and shee- cross-section members under Eccentric Axial Load. cold formed singly symmetric columns, Thin-wal-
ting, (Draft February 2006) European Committee Part 2:Validation, 2008, submitted to Costruzioni led Structures, vol. 33,1999, pp. 83 – 102.
for Standardisation. Metalliche for publication.
[15] EN 1993-1-1 Eurocode 3 part 1-1 “Design of
[5] AISI, Cold formed steel design manual, Ameri- [10] Rugarli P., Elastic and plastic flexural properties: steel structures – General rules and rules for bui-
can Iron and Steel Institute, 2004 edition. computerised calculation of generic sections, (in ldings”.
Appendix A: Notation
The meaning of each symbol used in the Greek letters: e2 effective at extreme 2
paper is generally explained when it ap- α load multiplier f flexural
pears first in the text. However, for conven- β effective length factor ft flexural-torsional
ience, symbols are listed below. ρ radius of inertia g gross
β slenderness factor i iteration index
Symbols: ΔM additional moment lim limit
A area φ angle between the principal axes p relative
b width of the wall of gross-section and the ones of pl plastic
f tension, stress effective section red reduced
E Young modulus λ slenderness R load carrying capacity
Err error ψ stress ratio T tension
G cross-section centroid ν Poisson ratio T total
I second moment of area ρ reduction factor u principal axis for effective section
N axial load σ tension v principal axis for effective section
M bending moment Ω section domain y yield, principal axis for
P generic point of the cross-section Index: gross-section
Q form factor c compression x principal axis for gross-section
S vectorial notation for applied com maximum value 1 extreme of the plate element
load cr critical 2 extreme of the plate element
t thickness of the wall d design Special symbols:
Tol tolerance e effective, elastic αi* member load multiplier in
W section modulus eff effective current iteration
X principal axis for gross-section exp experimental k_σ local buckling coefficient
Y principal axis for gross-section e1 effective at extreme 1 λ non dimensional slenderness
• Step 2: Stress computation and assessment of the effective width Reduction factor χ is given by:
(§ 2.2)
1 1
Stresses are constant across the section. In case of first iteration χ1 = = = 0.8219
φ1 + φ12 − λ1
2
0.7713 + 0.7713 2 − 0.6297 2
they correspond to the yielding tension fy, i.e.:
eq. 7)
_
Terms λ 1 and φ1 assume the values:
α1N 111.328 ⋅10000
σ1 = = = 355 N/mm2
Ag 3136 λ1 l0 ρe ,1 4.000 83.135 48.114
λ1 = = = = = 0.6297
λp E 210.000 76.409
Assessment of effective width π π
fy 355
eq. 5)
2
φ1 = 0.5 ⋅ [1 + α( λ1 − 0.2) + λ1 ] = 0.5 ⋅ [1 + 0.34 ⋅(0.6297 − 0.2) + 0.6297 2 ] = 0.7713
σ1 355
λ p,1 = = = 1.0596
σ cr 316.202
The value of the limit axial load on the effective section at the first
eq. 3a) iteration is:
Nc,1,lim = 0.8219 ⋅ 2345.1⋅ 355 = 684240 N
1 ⎡ 1 ⎤
be,1 = ⋅ ⎢1 − 0.055 ⋅( 3 + ψ)⋅ ⎥ ⋅ b =
λ p ,1 ⎣ λ p ,1 ⎦
Term α1* is obtained as:
eq. 13)
1 ⎡ 1 ⎤
= ⋅ ⎢1 − 0.055 ⋅( 3 +1) ⋅ ⋅196 = 0.7478 ⋅196 = 146.572 mm
1.0596 ⎣ 1.0596 ⎥⎦ α1∗ =
Nc,1,lim 684240
= = 68.424
N 10000
• Step 3: Computation of effective section properties (§ 2.3) • Step 6: Check for tolerance (§ 2.6)
Ae ,1 = 4 ⋅ be ,1 ⋅ t = 4 ⋅(146.572 ⋅ 4 ) = 2345.2 mm 2
eq.18)
bp ,1 = b − be ,1 = 196 −146.572 = 49.428 mm
α1∗ − α1 68.424 −111.328
Err1 = = = 0.385
⎡ bp ,1 ⋅ t
3
⎛ b⎞
2
⎤ bp ,1 ⋅ t
3
α1 111.328
Ie ,1 = Ig − 2 ⋅ ⎢ + (bp ,1 ⋅ t ) ⋅ ⎜ ⎟ ⎥ − 2⋅ =
⎢⎣ 12 ⎝ 2⎠ ⎥⎦ 12
being this value greater than limit of tolerance (toll=0.01) a new
⎡ 49.428 ⋅ 4 3 ⎛ 196 ⎞ ⎤
2
49.428 3 ⋅ 4 iteration is necessary.
= 20087125 − 2 ⋅ ⎢ + ( 49.428 ⋅ 4) ⋅ ⎜ ⎟ ⎥ − 2 ⋅ = 16208465mm4
⎣⎢ 12 ⎝ 2 ⎠ ⎦⎥ 12 • Step 7: Correction and new iteration (§ 2.7)
Eq. 19)
Effective ratius of gyration ρe,1
α1 + α1∗ 111.328 + 68.424
α2 = = = 89.876
Ie,1 16208465 2 2
ρe,1 = = = 83.135 mm
Ae,1 2345.1
Iteration 2
• Step 4: Computation of the loads on effective section (§ 2.4) • Step 2: Stress computation and assessment of the effective width (§ 2.2)
α3 N
• Step 3: Computation of effective section properties (§ 2.3) σ3 = = 260.641 N/mm2 < f y
Ag
Ae ,2 = 4 ⋅ be ,2 ⋅ t = 4 ⋅158.301⋅ 4 = 2532.8 mm2
bp ,2 = b − be ,2 = 196 −158.301 = 37.699 mm eq. 5)
⎡ bp ,2 ⋅ t 3 ⎛ b⎞ ⎤
2
bp ,2 3 ⋅ t σ3
Ie ,2 = Ig − 2 ⋅ ⎢ + bp ,2 ⋅ t ⋅ ⎜ ⎟ ⎥ − 2 ⋅ = λ p,3 = = 0.9079
⎢⎣ 12 ⎝ 2 ⎠ ⎥⎦ 12 σ cr
⎡ 37.699 ⋅ 4 3 ⎛ 196 ⎞ ⎤
2
37.699 3 ⋅ 4 eq. 3a)
= 20087125 − 2 ⋅ ⎢ + 37.699 ⋅ 4 ⋅ ⎜ ⎟ ⎥ − 2 ⋅ = 17154494 mm4
⎢⎣ 12 ⎝ 2 ⎠ ⎥⎦ 12
1 ⎡ 1 ⎤
be,3 = ⋅ ⎢1 − 0.055 ⋅( 3 + ψ)⋅ ⎥ ⋅ b = 163.570 mm
λ p ,3 ⎣ λ p ,3 ⎦
Effective ratius of gyration ρe,2
• Step 3: Computation of effective section properties (§ 2.3)
Ie,2 17154494
ρe,2 = = = 82.298 mm Ae ,3 = 4 ⋅ be ,3 ⋅ t = 2617.1 mm2
Ae,2 2532.8
bp ,3 = b − be ,3 = 32.430 mm
• Step 5: Evaluation of the load multiplier α 2* (§ 2.4)
⎡ b ⋅t 3 ⎛ b⎞ ⎤
2
b 3 ⋅t
Limit axial strength is given by: Ie ,3 = Ig − 2 ⋅ ⎢ p ,3 + bp ,3 ⋅ t ⋅ ⎜ ⎟ ⎥ − 2 ⋅ p ,3 = 17572407 mm4
⎣⎢ 12 ⎝ 2 ⎠ ⎦⎥ 12
Nc ,2 ,lim = χ2 ⋅ Ae ,2 ⋅ fy
1 1 Ie,3
χ2 = = = 0.8185 ρe,3 = = 81.941 mm
φ2 + φ2 − λ2 2
2
0.7765 + 0.7765 − 0.63612 2 Ae,3
α3∗ − α3 α4 + α4∗
Err3 = = 0.071 > 0.01( = tol ) α5 = = 77.812
α3 2
Ie,4
ρe,4 = = 81.808 mm Nc ,5 ,lim = χ5 ⋅A e ,5 ⋅ fy = 771141 N
Ae,4
1 1
• Step 5: Evaluation of the load multiplier α 4* (§ 2.4) χ5 = = = 0.8163
Nc ,4 ,lim = χ4 ⋅ Ae ,4 ⋅ fy = 767984 N φ5 + φ5 2 − λ5
2
0.7798 + 0.7798 2 − 0.6403 2
1 1
χ4 = = = 0.8165 eq. 13)
φ4 + φ4 2 − λ4
2
0.7795 + 0.7795 2 − 0.6399 2
Nc,5,lim 771141
α5∗ = = = 77.114
N 10000
eq. 13)
• Step 6: Check for tolerance (§ 2.6)
N 767984
α = c,4,lim =
∗
4
= 76.798 eq.18)
N 10000