6 A Structured Approach

to the Design and Analysis of Beams in Shear

Viktor Sigrist* and Britta Hackbarth**

*Professor, **Research Assistant
Institute of Concrete Structures
Hamburg University of Technology, Germany

Abstract: Even after many years of in-depth research, the determination of shear strength of
structural concrete beams still is in discussion. The reasons for that are, on the one hand, the
complexity of the problem and, on the other hand, further developments of structural systems
and building materials as well as new questions arising from the evaluation of existing struc-
tures. In this contribution a structured procedure including design, detailed analysis and
elaborate assessment of structural concrete members subjected to shear is presented. The fo-
cus is on the Generalised Stress Field Approach which particularly enables a profound analy-
sis and a general evaluation of a structure. As a rule, such calculations have to be performed
iteratively but the equations given here may easily be used in engineering practice, neverthe-
less. Shear strengths calculated on the basis of this method are compared to experimental
findings as well as to results according to fib Draft Model Code 2010 and Eurocode 2. It
shows that the different methods are partially in good agreement but yield deviating strength
values in certain ranges of reinforcement ratio.

6.1 Introduction
Stress field analysis and strut and tie models, respectively, permit visualisation of the force
flow and proportioning of a member [Marti (1999)]. They indicate the necessary amount, the
correct position, and the required detailing of the main reinforcement. Comprehensive contri-
butions and monographs on these methods are available [Marti (1985), Schlaich and Schäfer
(1987), Muttoni et al. (1997)]. The background of these developments is mainly formed by
limit analysis methods as are well summarised in an IABSE state-of-the-art report (1979), in
Thürlimann et al. (1983) and in Nielsen (1984). However, the problem of determining the
shear strength remains open to discussion. In recent years, research has focused on the en-
hancement and refinement of the above mentioned methods and several compression field ap-
proaches which consider equilibrium as well as compatibility conditions and more general
stress-strain relationships have been presented. As a rule, the application of such a refined ap-
proach requires specialised knowledge and computer programs and therefore, is not prevailing
in engineering practice. For the design and analysis of a structural concrete member calcula-
tion procedures have to be easy to understand and to use, and the engineer should be able to
give physical significance to the parameters involved and understand their importance [Bentz
et al. (2006)].

In the present contribution a structured approach to the design, analysis and assessment of
reinforced concrete members subjected to shear is described. First, the well known design
procedure associated with limit analysis is expounded and next, the Generalised Stress Field

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 1/12

Approach (GSFA) which is an extension of the former. Finally, the (modified) Cracked
Membrane Model is briefly outlined and it is shown that it comprises the other two methods.

Fig. 6.1: Geometric definitions: (a) cross-section, control section and strain profile; (b) web element.

Stress fields are models for describing the stress state within a structure at ultimate. Usu-
ally, in the analysis concrete stresses are assigned to nodal zones, fans, compression bands
and chords whereas steel stresses refer to tension bands and chords. Fig. 6.1(a) shows the
support region of a beam for which (as well as possibly for other locations) design for shear
has to be carried out. As defined in several codes [e.g. fib (2010)] a control section at the dis-
tance z from the face of the support may be assumed, z denoting the effective shear depth and
the distance between the top and the bottom chord axes, respectively. Shear forces are resisted
by an internal load bearing system consisting of the steel bars (with cross-sectional areas Asz
and Asx) and the inclined concrete struts, see Fig. 6.1(b). In webs of beams the forces in hori-
zontal or x-direction are normally taken by the chords and hence, the horizontal reinforcement
is of minor importance. For vertical stirrups and a given value of the stress field inclination θ
equilibrium considerations yield the well known design equations

Asz
VR , s = f y z cot θ and VR , max = f cebw z sin θ cos θ (6-1a) (6-1b)
sz

VR,s denotes the shear resistance provided by the stirrup reinforcement, sz the stirrup spac-
ing and fy the yield limit of the reinforcing steel. This resistance is limited to the strength of
the concrete struts which is defined by Eq. (6-1b) or rather the effective concrete compressive
strength fce. In the context of design the angle θ may be chosen whereas in the analysis of an
existing member θ has to be computed. In any case minimum and maximum limits have to be
respected, thus

θmin ≤ θ ≤ θmax (6-2)

For θmin the resistance is defined by Eq. (6-1a), for higher θ values by VR,s = VR,max and for
θmax by Eq. (6-1b). The reduction of the concrete compressive strength is due to strain effects
as well as the brittle failure behaviour of concrete in compression:

f ce = k c f c with k c = ηc k f (6-3a) (6-3b)

Analytical expressions for kc, ηc and kf are given in the following chapters.

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 2/12

 Fig. 6.2: Structured approach to design (limit analysis), detailed analysis (GSFA) and assessment
(Cracked Membrane); results for fc = 40 MPa (or fcm = fc + 8 MPa for assessment), fy = 500 MPa,
εx = 0.5⋅10-3 and steel N (normal ductility); .

The application of the above equations in the context of the models presented in the fol-
lowing chapters form a structured approach to design, analysis and assessment of members
subjected to shear. The resulting systematics is exemplified in Fig. 6.2; the material strengths
are fc = 40 MPa and fy = 500 MPa, the longitudinal strain at mid-depth of the web εx is 0.5⋅10-3
and characteristics according to normal ductility steel (steel N with ft /fy = 1.08, εsu = 50⋅10-3)
are assumed. The diagram shows the shear strength τR = VR /bw z as a function of the mechani-
cal reinforcement ratio ρz fy /fc where ρz denotes Asz /bw sz.

The three curves illustrate the idea of differentiating engineering tasks with respect to ef-
fort and expected or required precision. For the conception and the "design" of a new struc-
ture a procedure according to limit analysis can be used; the results are below those of more
accurate calculations. Generally, in the design of structures not all information about material
properties, loads and service conditions is available and therefore, a conservative strategy is
advisable. However, for a complex design situation or a general evaluation of a structure a de-
tailed "analysis" using the Generalised Stress Field Approach might be necessary; in these
cases, more effort in procuring data and computing is reasonable. Eventually, for the elaborate
"assessment" of a structure calculations on the basis of the Cracked Membrane Model may be
carried out. For the application of this model, full particulars of material properties, structural
geometry and loading states are required, but the highest precision is attained. Note that
changes in concrete strength as well as in the strength and ductility characteristics of the rein-
forcing steel may be taken into account. Hence, as visualised in Fig. 6.2 with the choice of an
increased concrete strength value, analysis and assessment do not depend on the same set of
data.

6.2 Design, Analysis and Assessment

6.2.1 Design of Members in Shear

For the design of beams in shear it is reasonable to use a clear procedure which can be de-
rived directly from limit analysis. The basis for this is the inclined stress field (see Fig. 6.1(b))
and the corresponding equilibrium equations. Even if compatibility of strains is not consid-
ered, the strength of the inclined stress field has to be reduced to the effective concrete com-

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 3/12

pressive strength fce. A possible reduction, used in the Swiss design code SIA 262 (2003), is
found by applying Eq. (6-3) in the following form:
1/ 3
⎛ 30 ⎞
f ce1 = 0.6 k f f c with k f = ⎜⎜ ⎟⎟ ≤ 1 .0 [MPa] (6-4a) (6-4b)
⎝ fc ⎠

To account for strain effects, a constant reduction ηc = 0.6 is assumed. With the factor kf
the concrete compressive strength is diminished for fc > 30 MPa; in this way, the more brittle
behaviour of such concrete types is taken into account [Muttoni (1989)]. By considering the
use of steel with limited ductility (see Chapter 6.2.2) the following limits for θ should be sat-
isfied:

30° ≤ θ ≤ 45° (6-5)

Within these constraints the inclination θ of the concrete struts can freely be chosen. Rear-
ranging Equations (6-1a) and (6-1b) and using the expressions τR = VR /bw z and ρz = Asz /bw sz
yields the following equations:

τ R , s = ρ z f y cot θ min (6-6)

ρz f y
τ R = ρ z f y ( f ce − ρ z f y ) with tan θ = (6-7a) (6-7b)
f ce − ρ z f y

τ R max = f ce sin θ max cos θ max (6-8)

In the context of limit analysis Eq. (6-7a) is termed as web crushing criterion [Braestrup
(1976)]. Results from the above equations are shown in Fig. 6.3. Note that a minimum rein-
forcement ratio ρz,min of 0.12( fc)1/2/fy [MPa] is presupposed in both cases. The three parts de-
fined by Eqs. (6-6) to (6-8) can clearly be identified as a linear function for low, a curved line
for medium and a horizontal branch for high reinforcement ratios. The corresponding stress
field inclinations are depicted in Fig. 6.3(b); again, a slightly curved transition line from the
lower limit θmin to the given maximum value θmax is visible.

Fig. 6.3: Design of members on the basis of limit analysis: (a) shear strength for fc = 40 MPa and
fc = 80 MPa, fy = 500 MPa; (b) corresponding stress field inclinations.

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 4/12

6.2.2 Analysis of Members in Shear

For a general evaluation of an existing structure or for treating a complex design situation
some more effort and precision may be adequate. One possibility to increase the accuracy of
the calculation is the use of the Generalised Stress Field Approach (GSFA). Fig. 6.4(b) shows
a centred fan that is defined by the two inclined discontinuity lines and by the parallel chord
axes. The uniformly distributed load q is equilibrated in vertical direction by the stirrup forces
σsz Asz /sz acting on either side of the fan. The concrete within the fan is uniaxially compressed
along straight trajectories that intersect outside the beam; the stresses σ2 vary hyperbolically
along these lines. The highest value is found for the bottom right corner of the fan where the
smallest angle θ as well as the shortest width occurs. A fan with θ = θr becomes a stress field
with parallel trajectories. In this case, the concrete stresses σ2 are constant; for the assump-
tions σsz = fy and -σ2 = fce the Eqs. (6-6) to (6-8) are found.

Fig. 6.4: Generalised Stress Field Approach: (a) cross-section; (b) free body diagram of centred fan;
(c) beam segment with sectional forces.

The generalization of the stress field analysis is accomplished by introducing special limits
for the angle θ as well as by adopting a more general equation for determining the effective
concrete compressive strength fce. For the latter, Eq. (6-3) can be used, but ηc has to be evalu-
ated on the basis of experimental results. Kaufmann (1998) proposed a relationship for mem-
brane elements that has been modified to include beam elements as well and adjusted to the
notation used here [Sigrist and Hackbarth (2010a,b)]; this yields

f ce 2 = ηc k f f c with (6-9a)

1/ 3
1 ⎛ 30 ⎞
ηc = and k f = ⎜⎜ ⎟⎟ ≤ 1 .0 [MPa] (6-9b) (6-9c)
1.2 + 55ε1 ⎝ fc ⎠

The principal strain ε1 (as defined in Fig. 6.1(b)) is found with help of Mohr's circle of
strains to ε1 = εx + (εx - ε2) cot2θ. The principal strain ε2 may be taken as the peak strain -εc0 at
reaching fce. This value is affected by numerous influences, including the sequence of loading.
Though, comparative calculations prove the relatively small implication of this parameter on
the predicted strength and therefore, in good approximation a constant value εc0 = 0.002 may
be used. Note that longitudinal strains of webs vary linearly over the depth which results in
slightly curved stress trajectories with somewhat flatter inclinations and hence, higher com-
pressive stresses at the compression chord. However, the concrete rather crushes at the ten-
sion chord where the reduction of the concrete compressive strength is more pronounced. In-
cidentally, it can be shown that after the onset of yielding of the stirrups initially curved
cracks become almost straight [Kaufmann (1998)] as assumed in the stress field analysis. For

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 5/12

reasons of applicability, it is recommended to adopt the idea of taking the strain value εx at
mid-depth of the web as being decisive (Fig. 6.1). This is an arbitrary definition, but it helps
harmonizing design rules as well as the analysis of tests.

Fig. 6.5: Analysis of members with help of the Generalised Stress Field Approach (GSFA): (a) shear
strength for fc = 40 MPa and fc = 80 MPa, fy = 500 MPa, εx = 0 and 1.0⋅10-3 and steel N (normal ductil-
ity); (b) corresponding stress field inclinations.

Limits of the stress field inclination according to Eq. (6-2) can be determined on the basis
of the kinematic condition for membrane elements [Baumann (1972), Collins and Mitchell
(1980)]; from Mohr's circle of strains one finds

ε x − ε2
tan 2 θ = (6-10)
ε z − ε2

and by rearranging and inserting -ε2 = εc0 and εz = εsm

1/ 2 1/ 2
⎛ ε + εc0 ⎞ ⎛ ε + εc0 ⎞
θ min = arctan ⎜⎜ x ⎟⎟ and θ max = arctan ⎜ x ⎟ (6-11a) (6-11b)
⎜ε +ε ⎟
⎝ ε smu + ε c 0 ⎠ ⎝ smy c0 ⎠

By respecting θmin the average stirrup strains are limited to values below or at rupture of
the steel, i.e. εsmu = κ1εsu, and by complying with θmax to the onset of yielding, i.e. εsmy = κ2εsy.
The relevant bond coefficients κ = εsm /εsmax can be computed with the Tension Chord Model
[Sigrist (1995), Marti et al. (1998)]; in a first approximation κ1 = 0.25 and κ2 = 0.8 may be
taken, indicating that reduction of the steel strain due to tension stiffening is 75% at rupture
and 20% at the onset of yielding. Equation (6-11) is rather inappropriate as long as bond coef-
ficients and maximum steel strains are not known and therefore, a simplified expression
might be helpful. Provided that the yield limit of the steel fy is between 400 MPa and 550 MPa
and that -0.5·10-3 ≤ εx ≤ 1.5·10-3 the following limits are recommended:

20° + 5'000 ε x ≤ θ ≤ 35° + 5'000 ε x (steel N) (6-12a)

16° + 4'000 ε x ≤ θ ≤ 35° + 5'000 ε x (steel H) (6-12b)

Equations (6-12) correspond to linear approximations of Equation (6-11). In this contribu-

tion for steel N (normal ductility) the values ft /fy = 1.08 and εsu = 50⋅10-3 are assumed and for

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 6/12

steel H (high ductility) ft /fy = 1.2 and εsu = 100⋅10-3. Note that for fy different from the values
given above, steel L (low ductility) and values εx > 1.5·10-3 Equation (6-11) has to be applied.

In Fig. 6.5 results from calculations according to the Generalised Stress Field Approach are
shown, i.e. from applying Eqs. (6-6) to (6-8) and considering Eqs. (6-9) and (6-12); basically,
this is an iterative procedure. The curves visualize the differences in shear strength τR for the
strains εx = 1·10-3 and 0; these longitudinal strains comprise the most relevant range for beams
in shear. A value εx < 0 refers to a compressed or prestressed member whereas εx > 1·10-3 sig-
nifies a member whose tension chord is at the onset of yielding; note that with Eq. (6-11)
cases with higher strain values i.e. plastic hinge regions can be treated as well.

6.2.3 Structural Assessment

The background of the Generalised Stress Field Approach is the Cracked Membrane
Model. This model was presented by Kaufmann (1998) and Kaufmann and Marti (1998) and
modified by Sigrist (2011). It is based on the Tension Chord Model which allows treating
problems of crack formation and deformation capacity, and extends the concept to states of
plain stress. For this purpose, the web of a beam is considered as a composition of membrane
elements which are orthogonally reinforced (Fig. 6.1(b)) and subjected to in-plane stresses. In
the course of the loading process cracks occur which, in a fully developed crack pattern, have
an average spacing of srm. Assuming a stepped, perfectly plastic bond shear stress-slip rela-
tionship [Sigrist (1995)] and applying the general equilibrium equations for the cracked
membrane as well as the kinematic condition (6-10) and adequate material properties srm, θ
and τR can numerically be determined. Corresponding computations yield the shear strength
diagrams plotted in Fig. 6.6.

Fig. 6.6: Elaborate structural assessment of members using the Cracked Membrane Model: (a) shear
strength for fc = 40 MPa and fc = 80 MPa, fy = 500 MPa, εx = 0 and 1.0⋅10-3 and steel N (normal ductil-
ity); (b) corresponding stress field inclinations.

The assumptions described above imply that the directions of the principal stresses and
strains coincide, i.e. crack faces are stress free, able to rotate, and perpendicular to the princi-
pal tensile direction of the average strains. Thus, θ is a variable rather than a fixed angle. In
such a model other influences have to be considered through the definition of the effective
concrete compressive strength; since there is good agreement with experimental evidence, in
Sigrist (2011) it is proposed to use Eq. (6-9).

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 7/12

 The results of Fig. 6.6 apply to elements with fc = 40 MPa and 80 MPa, fy = 500 MPa and
steel N (normal ductility) and again, the two longitudinal strains εx = 1·10-3 and 0. The curves
can be subdivided into three parts. For low reinforcement ratios rupture of the stirrups is deci-
sive. For higher reinforcement ratios crushing of the concrete governs failure, where up to
shear strengths of approximately 12 MPa and 19 MPa the stirrups are yielding. Eventually,
for high mechanical reinforcement ratios the stirrups remain elastic. The two failure modes
can be distinguished also in Fig. 6.6(b) where the corresponding angles θ are shown. It might
be of particular interest that the values of θ quite strongly depend on the longitudinal strain
and the ductility characteristics of the steel.

It has to be emphasised that a change of one or more of the input values has an influence
on the results but does not bias the general findings. For computing the curves of Fig. 6.6 re-
alistic (rather unfavourable) values are chosen so that the calculated strength values are typi-
cal for practical application; consequently, these curves can be used as reference. As the
method is general, each value can be adjusted to the actual situation and effects like reduced
strength or bond properties due to corrosion of the steel or degradation of the concrete may be
taken into account. For the assessment of a structure such refinements are of special interest.

6.3 Comparison of Results

6.3.1 Comparison to Experimental Data

The definitions of Eq. (6-9) rely on experimental data. In the course of a study presented in
Sigrist (2011) 113 experiments, 56 of which are beam and 57 are panel tests, have been ana-
lysed with help of the Generalised Stress Field Approach. Only selected experiments are in-
cluded in the evaluation. Beam tests with flexural or anchorage failures as well as panel tests
with applied axial stresses or anchorage failures are excluded.

Fig. 6.7: Comparison of predicted and observed values: (a) observed shear strengths and approximated
envelope of predictions; ratio of observed to predicted shear strengths as a function of (b) the rein-
forcement ratio and (c) the stress field inclination.

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 8/12

 Geometric data, material properties and failure loads are acquired from the reports and all
the static key figures such as stresses and strains are computed on the basis of the equations
given above. For the beam tests the longitudinal strains are found with the top and bottom
chord forces

M V
F=∓ + cot θ (6-13)
z 2

from which, by considering cross-sectional areas and stiffness of the chords, the strains εx
can be approximated; in doing so, tension stiffening in the tension chord is taken into account.

In Fig. 6.7(a) the observed shear strengths are divided by kf fc and plotted versus the nor-
malised mechanical reinforcement ratio ρz fy /(kf fc); thus, the results are independent of the
concrete compressive strength. In the beam tests the longitudinal strains εx are in the range of
0.1·10-3 to 1.5·10-3 (with one exception) and the calculated θ values are between 14.3° and
39.3°. Therefore, the curved line drawn in the diagram can be regarded as an upper boundary
for all experiments with ρz fy /(kf fc) < 0.24. In the panel tests the strain values are between
0.5·10-3 and 43·10-3 (maximum 5·10-3 for high reinforcement ratios) with θ values in the
range of 20.3° and 45°. For low and medium reinforcement ratios the straight line, corre-
sponding to θ = 45°, can be seen as a lower boundary. For values of approximately ρz fy /(kf fc)
> 0.3 upper and lower boundaries may be defined by εx = 0.5·10-3 and 5·10-3. In all, the calcu-
lated lines of Fig. 6.7(a) represent the approximated envelope of predictions and impressively
visualise the accuracy of the method.

However, it has to be noted that the experimental results are not uniformly distributed
within the ranges of the parameters. In Fig. 6.7(b) the ratios of observed to predicted strength
values are depicted. The reinforcement ratio ρz fy /(kf fc) varies widely but most of the experi-
ments refer to values smaller than 0.2 and in this range the average of the strength-ratios re-
veal a less accurate prediction compared to the overall trend. The scatter of the plotted points
suggests that especially for low reinforcement ratios additional factors contributing to the
strength might be important, such as dowel action of the reinforcing bars or the resistance of
the compression chord. As the calculation of the inclination of the stress field involves most
of the parameters related to the shear problem it is of special interest to plot the shear strength
ratios as a function of θ; the results are more or less uniformly distributed in this diagram
which again supports the general findings. Overall, an average of the ratio τR,exp/τR,pre of 1.03
and a coefficient of variation of 11.5% are found; the corresponding values for the beam tests
are 1.09 and 10.6% and for the panel tests 0.98 and 9.8%.

As can be seen from the comparison to experimental data, the relationships given here
cover a wide range of parameters and include the behaviour of beams as well as that of pan-
els. On the one hand, this unification strongly supports the reliability of the method but on the
other hand, it is debatable since the state of strain in panels in pure shear differs from that in
beams subjected to flexure and shear. Nevertheless, good correlation of experimental data and
the results from Equation (6-9) is obtained in both cases. Within the framework of research
further differentiation might be made and some of the parameters should individually be ac-
counted for; of special interest are concrete strength, member size and contributions of the
chords.

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 9/12

6.3.2 Comparison to Code Equations

In Fig. 6.8 results from the Generalised Stress Field Approach (GSFA) are compared to
those from Level I, II and III calculations according to fib Draft Model Code (2010) and from
Eurocode 2 (2004) (EC2). Level I and III equations are based on the general idea that the
shear resistance of a reinforced concrete member is provided by the shear reinforcement (stir-
rups) VR,s and the contribution of the concrete VR,c due to frictional forces transferred over the
cracks; this results in

VR = VR , s + VR , c (6-14)

For given θ values VR,s can be calculated with help of Eq. (6-1a), and for VR,c the expres-
sion kv (fc)1/2 bw z is used; the factor kv is defined as 0.15 for a Level I and as 0.4/(1+1´500εx)
for a Level III Approximation. In applying these methods θ values of 36° and (29°+7´000εx),
respectively, have to be assumed. In a Level II Approximation kv is set to 0, i.e. the contribu-
tion of the concrete is not considered explicitly. This method refers to limit analysis as de-
scribed in Chapter 6.2.1, however, for the choice of θ the lower limit of (Eq. 6-5) is replaced
by (20°+10´000εx). In any method, the shear resistance is limited to the values of Eq. (6-1b)
where according to fib Draft Model Code kc = 0.50kf (Level I) and 0.55kf (Levels II and III)
have to be inserted.

Eurocode 2 (2004) is based on conventional limit analysis (Chapter 6.2.1) and Eqs. (6-1a)
and (6-1b) apply. The limits for the stress field inclination are given to 21.8° ≤ θ ≤ 45° and
the reduced concrete compressive strength is defined as 0.6(1-fc /250) fc. Note that for com-
parison purposes the results from Eurocode 2 have to be transformed to the control section,
see Fig. 6.1 [Sigrist and Hackbarth (2010b)].

Fig. 6.8: Comparison of Generalised Stress Field Approach (GSFA) to Level I, II and III calculations
according to fib Draft Model Code (2010) as well as to results from Eurocode 2 (2004) for fc = 40 MPa,
fy = 500 MPa and εx = 0.5⋅10-3 and steel N (normal ductility).

In Fig. 6.8 it can be seen that the curve of the Generalised Stress Field Approach is close to
or lies above the other lines in the entire range of reinforcement ratios. This finding is true for
longitudinal strains εx higher than approximately 0.5·10-3; for lower εx values the Level III re-

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 10/12

sults exceed the others. In a similar manner, the Level II results partly overlap those from
Level III. This is due to the angle θmin which is a model parameter in the methods based on
limit analysis. For low reinforcement ratios this does not cause great problems, however, for
medium and high reinforcement ratios as well as for high longitudinal strains Level I and III
results seem to be somewhat conservative. Apart from that, as can be concluded from Fig. 6.8
it is appropriate to increase the upper limit (e.g. to 0.6kf fc) given by the fib Draft Model Code
for high reinforcement ratios. Regarding Eurocode 2, it is evident that for low and medium re-
inforcement ratios results are rather high but basically do not contradict the ones from the
Generalised Stress Field Approach. For high reinforcement ratios they agree well with the re-
sults from Level I analysis.

For members without shear reinforcement or with ratios below ρz,min only Level I and III
calculations yield a shear resistance which for the latter is given by

0 .4 1300
VR 0 = ⋅ f c bw z [MPa] (6-15)
(1 + 1500ε x ) (1000 + 0.7 k dg z )

where kdg accounts for the influence of the aggregate size. This equation (or an equivalent
expression) might be combined with the Generalised Stress Field Approach and with Level II
Approximations, respectively. By doing so, the gap in the diagram of Fig. 6.8 could be closed
and the respective procedures would be amended to cover all possible cases.

6.4 Conclusions
The structured approach presented in this contribution covers design, detailed analysis and
elaborate structural assessment of beams in shear. The models involved are consistent and can
be deduced from one another. Evidently, the effort in calculating the shear strength is low for
designing and high for evaluating a member but accuracy is increased as well and moreover,
insight in structural behaviour is significantly deepened.

The procedures discussed here are based on rational models, i.e. on models that reflect the
physical background of structural behaviour. Such models do not (and never will) cover all
aspects of the actual relations but enable - provided that parameters are chosen carefully - the
safe design of structures. With respect to engineering practice, the strength of a rational model
also lies in the possibility of extending it to new design problems apart from the few that are
discussed in engineering text books.

6.5 References
Baumann, T. (1972). Zur Frage der Netzbewehrung von Flächentragwerken (On the Problem
of Net Reinforcement of Surface Structures). Bauingenieur, Berlin, Vol. 47, No. 10, pp. 367-
377.
Bentz, E.C., Vecchio, F.J., Collins, M.P. (2006). The Simplified MCFT for Calculating the
Shear Strength of Reinforced Concrete Elements. ACI Structural Journal, V. 103, No. 4, pp.
614-624.
Braestrup, M.W. (1976). Plastic Analysis of Shear in Reinforced Concrete. Magazine of Con-
crete Research, Vol. 26, No. 89, Dec. 1974, pp. 221-228.

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 11/12

Collins, M.P., Mitchell, D. (1980). Shear and Torsion Design of Prestressed and Non-
Prestressed Concrete Beams. PCI Journal, V. 25, No. 5, pp. 32-100.
Eurocode 2 (2004). EN 1992-1-1: Design of concrete structures - Part 1, General rules and
rules for buildings. CEN - European Standard, 225 pp.
fib (2010). Model Code for Concrete Structures - first complete draft. fib Bulletin, No. 56, pp.
312.
IABSE (1979). Colloquium on Plasticity in Reinforced Concrete. Copenhagen, International
Association for Bridge and Structural Engineering (IABSE), Introductory Report V. 28, 179
pp. and Final Report V. 29, 360 pp.
Kaufmann, W. (1998). Strength and Deformations of Structural Concrete Subjected to In-
Plane Shear and Normal Forces. Report, Institute of Structural Engineering, ETH Zürich, No.
234, 147 pp.
Kaufmann, W., Marti, P. (1998). Structural Concrete: Cracked Membrane Model. Journal of
Structural Engineering, ASCE, Vol. 124, No. 12, pp. 1467-1475.
Marti, P. (1985). Basic Tools of Reinforced Concrete Beam Design. ACI Structural Journal,
V. 82, No. 1, pp. 46-56.
Marti, P., Alvarez, M., Kaufmann, W., Sigrist, V. (1998). Tension Chord Model for Structural
Concrete. Structural Engineering International, V. 8, No. 4, pp. 287-298.
Marti, P. (1999). How to Treat Shear in Structural Concrete. ACI Structural Journal, V. 96,
No. 3, pp. 408-414.
Muttoni, A. (1989). Die Andwendbarkeit der Plastizitätstheorie in der Bemessung von Stahl-
beton (The Applicability of the Theory of Plasticity in the Design of Reinforced Concrete).
Report, Institute of Structural Engineering, ETH Zürich, No. 176, 159 pp.
Muttoni, A., Schwartz, J., Thürlimann, B. (1997). Design of Concrete Structures with Stress
Fields. Birkhäuser, 141 pp.
Nielsen, M.P. (1984). Limit Analysis and Concrete Plasticity. Prentice Hall, 420 pp.
Schlaich, J., Schäfer, K. (1987). Towards a Consistent Design of Structural Concrete. PCI-
Journal, May-June 1987, pp. 75-150.
SIA 262 (2003). Concrete Structures. Code, Swiss Society of Engineers and Architects,
Zürich, 90 pp.
Sigrist, V. (1995). Zum Verformungsvermögen von Stahlbetonträgern (On the Deformation
Capacity of Structural Concrete Girders). Report, Institute of Structural Engineering, ETH
Zürich, No. 210, 159 pp.
Sigrist, V., Hackbarth, B. (2010a). Stress Field Analysis of Structural Concrete Beams. Pro-
ceedings, 3rd International fib Congress 2010 and Annual pci Convention, Washington D.C.,
USA, May 29 - June 6.
Sigrist, V., Hackbarth, B. (2010b). Querkrafttragfähigkeit von Stahlbetonträgern - Bemes-
sung, Überprüfung, Beurteilung (Shear Capacity of Structural Concrete Beams - Design,
analysis, Assessment). Beton- und Stahlbetonbau, Wiley - Ernst & Sohn, V. 105, No. 11.
Sigrist, V. (2011). Generalized Stress Field Approach for the Analysis of Beams in Shear.
ACI Structural Journal, accepted for publication 2011
Thürlimann, B. et al. (1983). Die Anwendung der Plastizitätstheorie auf Stahlbeton (Applica-
tion of the Theory of Plasticity to Reinforced Concrete). Institute of Structural Engineering,
ETH Zurich, Switzerland, 252 pp.

Sigrist, Hackbarth / fib Bulletin 57 (chapter 6) / October 2010 pg. 12/12