Upload
roopeshdesai
View
219
Download
0
Embed Size (px)
Citation preview
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
1/48
Editorial Manager(tm) for Magazine of Concrete Research
Manuscript Draft
Manuscript Number:
Title: Economic Design Of Beams With FRP Rebar Or Prestress
Article Type: Paper
Corresponding Author: Ioannis Balafas, Ph.D.
Corresponding Author's Institution: University of Cyprus
First Author: Ioannis Balafas, Ph.D.
Order of Authors: Ioannis Balafas, Ph.D.;Chris J Burgoyne, MA, MSc, Ph.D.
Abstract: Fibre-Reinforced Polymers have not captured the market for reinforcing bars, or found useseven as external prestressing tendons, despite their obvious advantages in terms of lack of corrosion.
This paper considers the constraints that underlie the optimal design for beams reinforced or
prestressed with FRP, and compares the result with designs for beams with steel reinforcement. This
is done by means of a diagram that compares the depth of the beam with the area of reinforcement, on
which most of the governing constraints can be plotted. The method allows generic conclusions to be
drawn about the governing principles.
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
2/48
Ioannis BalafasDepartment of Civil andEnvironmental Engineering
University of Cypruse-mail: [email protected]: 00357 9671654915 June, 2011
ICE PublishingMagazine of Concrete Research40 Marsh Wall
London
E14 9TP
UK
Dear Sir/Madam,
This letter follows the submission of the article: Optimal design of beams with FRP bars orprestress to Magazine of Concrete Research. The paper studies the initial cost of beamsreinforced and prestressed with Fiber-Reinforced Polymers.
Fiber-Reinforced Polymers have not captured the market for reinforcing bars, or even
prestressing tendons, despite their obvious advantages in terms of lack of corrosion. Theconstraints that underlie the optimal design for beams reinforced or prestressed with FRPare determined, and the results are examined with designs for beams with steel. This isdone by means of a diagram that compares the depth of the beam with the area ofreinforcement, on which most of the governing constraints can be plotted. The methodallows generic conclusions to be drawn about the governing principles.
The paper is about 5000 words and includes five tables and fourteen figures.
Looking forward to hearing from you, concerning the reviews.
Ioannis Balafas Chris Burgoyne
ver Letter
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
3/48
Economic Design Of Beams With FRP Rebar Or Prestress12
I. Balafas*
and C. J. Burgoyne3
4
5
Fibre-Reinforced Polymers have not captured the market for reinforcing bars, or6
found uses even as external prestressing tendons, despite their obvious advantages in7
terms of lack of corrosion. This paper considers the constraints that underlie the8
optimal design for beams reinforced or prestressed with FRP, and compares the result9
with designs for beams with steel reinforcement. This is done by means of a diagram10
that compares the depth of the beam with the area of reinforcement, on which most of11
the governing constraints can be plotted. The method allows generic conclusions to12
be drawn about the governing principles.13
14
INTRODUCTION15
When high-strength fibres (glass, aramid and carbon) were first introduced to the16
structural engineering world it was assumed that they would be most suitable for use17
as prestressing tendons, which are the most highly-stressed of all structural elements:18
this would make maximum use of the high strength that was being provided so19
expensivly. It was considered that use as reinforcement would seldom be justified,20
since deflections would govern and the fibres would rarely be stressed to a high load.21
Experience has shown, however, that there have been few applications of prestressing,22
despite the added benefit that FRP tendons can be placed externally, and the limited23
use that has been made as reinforcement has been using GFRP, which has the lowest24
stiffness and is thus more lightly stressed than AFRP or CFRP bars would be. This25
*Department of Civil & Environmental Engineering, University of Cyprus, PO Box 20537,
1678 Nicosia, Cyprus, [email protected] of Engineering, University of Cambridge, Trumpington Street, Cambridge, CB2
1PZ, UK, [email protected].
n Text
k here to download Main Text: IC_14_6_2011.doc
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
http://www.editorialmanager.com/macr/download.aspx?id=36111&guid=41489e65-c9ec-442c-ad66-1090b3782733&scheme=1http://www.editorialmanager.com/macr/download.aspx?id=36111&guid=41489e65-c9ec-442c-ad66-1090b3782733&scheme=18/2/2019 p75preEconomic Beam Designs Using FRP Rebar
4/48
paper seeks to produce some general principles that explain why this might be the1
case.2
The close fit between the strain capacities of steel and concrete means that they are3
ideally suited to produce a composite reinforced concrete where both materials are4
used economically, while strains and curvatures remain small. Steel is relatively cheap5
(aramid and carbon cost 5-6 times as much per unit of force), but corrosion of steel in6
concrete bridges is a major problem;1,2
by now FRPs should be in widespread use.37
However, applications of FRP are still scarce; the market is hesitant to adopt them,8
partly because of their high cost and partly because the design principles differ from9
those using conventional rebar.10
This paper addresses the question of how, or even whether, their properties should be11
included in the design process. The target is to identify ways in which FRP can be12
used at minimum cost, and to consider why FRPs are rarely used in practice. These13
questions will be addressed through the use of a diagram of depth (d) versus flexural14
bar area (Ap).15
The first sections below will show how the basic diagrams are built up. Additional16
constraints will then be added before specific questions are answered about the17
different options available to designers. Three case studies will be considered before18
final conclusions are drawn.19
20
DEPTH VERSUS AREA DIAGRAM21
The parameters affecting a design can be shown on a diagram of flexural bar area (Ap)22
versus section depth (d), which was first introduced by Al-Salloum et al4 for the23
optimal design of rectangular reinforced concrete beams. They considered only a24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
5/48
limited number of design constraints. The method will be expanded here to include1
many different factors.2
Detailed equations are not given below since the analysis follows standard procedures3
for the design of a cross-section on the assumption that plane-sections remain plane.4
Safety factors are included for ultimate loads, but not for serviceability conditions.5
Sections are designed from first principles, rather than a particular code of practice,6
although some use is made of factors that are also used in codes. Very similar plots7
would result from using codes, and a detailed discussion of the differences would8
distract from the main argument. For the same reason, although all the plots below9
relate to the same loading case, and have been produced as a result of specific10
calculations, details are not given here.11
The following sections show how the basic diagram is built up.12
Ultimate strength condition13
The ultimate strength condition divides the d-Ap diagram into two zones (Fig. 1(a));14
sections that are sufficiently strong lie above a boundary line while those that are15
inadequate lie below it. The boundary line is curved and there is a distinct change in16
slope when the limiting condition changes from failure by tendon snapping to failure17
by concrete crushing.18
Bar snapping condition19
Unlike sections reinforced with steel that can yield, sections with FRP should not be20
allowed to fail by reinforcement snapping, so a minimum limit on the amount of21
reinforcement has to be applied that reduces the size of the feasible zone (Fig. 1(b)).22
This will be referred to hereafter as the balanced-section constraint.23
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
6/48
Deflection condition1
Sections with FRP reinforcement are more flexible than sections with steel rebar, so2
the deflection condition needs to be checked, which leads to a further reduction in the3
size of the feasible zone, as shown in Fig. 1(c).4
Cost Function5
The cost of the beam can be directly related to the cross-sectional area and the amount6
of reinforcement. Allowances can be made for the cost of the formwork, as well as7
for the shear reinforcement. Beams of equal cost are defined by a line of fixed slope8
on the d-Ap diagram; as the cost of the beam increases, this line moves to the right but9
the slope remains the same (Fig. 1(d)). The optimum solution is therefore where the10
cost function line first meets the feasible region; in the example shown in Fig 1 that11
occurs when the bar snapping and deflection constraints meet.12
Other constraints13
Other constraints can also be shown. Crack-width limits can be introduced; for FRP-14
reinforced structures these are associated with appearance, while for steel-reinforced15
structures they are associated with durability. A limit on concrete stress, to eliminate16
the possibility of excessive creep, can also be considered. The latter rarely governs,17
but Fig. 1(e) shows that the crack width is close to the limit for the cheapest design in18
this particular case and it can limit in other cases.19
The method does not specifically include the breadth of the beam. Only rarely is the20
breadth a governing condition; it will be assumed here that it is as small as21
practicable. The breadth does affect the amount of shear steel needed, and is reflected22
in the costs functions, but the details do not affect the main argument and it will not23
be discussed further.24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
7/48
1
APPLICATIONS OF THEd-Ap DIAGRAM2
Having derived the basic diagram, it is now possible to compare different alternative3
designs. This will be done by showing pairs of diagrams for different cases, which4
allows the distinctions to be highlighted.5
Steel versus FRP6
The most fundamental consideration when designing a reinforced concrete section is7
the choice of reinforcement material. To date, steel has been the obvious solution,8
and the d-Ap diagrams show why.9
Fig. 2 compares two designs (for the same loading case), one with steel rebar and one10
with FRP. A number of differences can be noted in the way the constraints apply.11
1. In the steel diagram, the strength condition is more critical than the deflection12condition, except for very shallow beams.13
2. There are minimum areas of steel reinforcement (to avoid snapping or14excessive cracking) as well as maximum areas (to avoid over-reinforcement).15
3. Steel is much cheaper (per unit volume) than FRP, so the constant cost line is16much flatter.17
4. The optimal design for use with FRP is almost always at the vertex where the18deflection (or occasionally the crack width) conditions meet the deflection19
limit.20
5. The flexural optimum solution for steel reinforcement is not at a vertex but lies21on a curved edge. There will thus be a range of designs that have very similar22
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
8/48
costs and the optimum design will be very sensitive to variations in the relative1
costs of labour and materials. This can mean that the optimum solution can2
alter depending on where the structure is built.3
Furthermore, some conclusions can be drawn about the differences between the4
design with FRP and that with steel. Beams with FRP need to be significantly deeper5
than beams with steel in order to comply with the deflection conditions, but the6
amount of reinforcement required is less because the rebar is stronger.7
Prestressed Concrete Beams8
Prestressed concrete beams can be studied using the same basic principles, but9
different parameters become important.10
The most significant difference is the addition of upper and lower bounds on the11
section stresses at the working load. This has the corollary that there is an upper limit12
on the amount of prestress that can be applied, so the feasible zone is now bounded.13
Stress rupture of FRP tendons also has to be considered, but it appears in these14
diagrams indirectly since it affects the allowable stress in the tendon, which in turn15
affects the Ap required.16
Typical results are shown in Fig. 3, (a) (for steel) and (b) for FRP. As with the plots17
for reinforced concrete design, the optimal design for sections with FRP is usually at a18
vertex of the feasible zone (which makes it relatively insensitive to changes in relative19
costs of reinforcement and concrete), while the optimum for a section prestressed with20
steel lies along a curved edge of the zone.21
Deflection criteria are unlikely to govern for either material; the prestress prevents the22
section cracking at the working load so the beam stiffness is now almost entirely a23
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
9/48
property of the concrete and not of the reinforcement, even for sections prestressed1
with FRP.2
Comparison of the two plots for prestressed beams yields conclusions that differ from3
those for reinforced sections. The optimum depth for both sections is similar, which4
is logical given that prestressing allows the concrete itself to carry the load. Less5
material is needed for prestressing with FRP because those tendons are generally6
stronger, while the stress-rupture and ultimate strength of the tendon enter via the7
balanced section criterion.8
9
CASE STUDIES10
The plots developed above are interesting in their own right, but having set out the11
basic principles of how the diagrams are produced, their use can be extended into a12
study which can map the design space and identify factors that most affect the13
economics of design. Three cases will be considered in more detail; a reinforced14
concrete beam and two prestressed concrete beams, one prestressed, the other15
statically indeterminate.16
Materials considered in the case studies17
Various FRP materials have been considered in this study. Glass fibre and carbon18
fibre reinforced polymers (GFRP) and (CFRP) are typically used in the form of19
pultruded rods of various diameters.20
Aramid fibres are available in various forms. Kevlar and Twaron fibres have very21
similar properties and are available in standard modulus and high modulus versions.22
The high modulus version is used in Fibra, which is a braided product embedded in23
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
10/48
resin. The standard modulus version of a similar, but not identical product, Technora,1
is used to make a pultruded rod that is stronger but less stiff than the Fibra. Because2
aramid fibres are tougher than carbon, they can also be used without resin and are3
used as parallel-lay ropes under the name Parafil. These ropes can be fitted with4
termination systems that make them suitable for use as prestressing tendons, but they5
are not suitable for use as rebar.6
Typical values for the strength and stiffness are given in Table 1. Under long term7
loading strength may reduce due to creep rupture; the strength at time t,ft, is given by:8
101 logt i f f B t (1)9
fi is the short term strength, B is a constant determined by fitting to existing creep10
rupture data5,6 and its values are shown in Table 1. Note that t is in hours. Creep11
rupture for steel is insignificant and thus ignored.12
The unit costs for the different materials, obtained from recent quotations from13
manufacturers, are shown in Table 2.14
With these general values as a basis, it is now possible to investigate the effect of15
changing various aspects of the design.16
Case study 1. Reinforced Concrete17
The example chosen is a simply-supported T-beam as shown in Fig. 4. Several input18
parameters assumed in the analysis are shown on Table 3. The width of the web is19
assumed to be 250 mm. The beam and the slab are rigidly connected so the beams20
cross section is T-shaped. The effective width of the slab that acts compositely with21
the beam is difficult to determine. The normal calculations used in codes of practice22
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
11/48
(which typically give effective breadths = span/5) are designed to give a beam with an1
effective value of second moment of area, but the present analysis is more concerned2
with the peak compressive stress, which gives lower effective breadth.13
The value3
used in this analysis is span/7 = 0.85 m.4
Proposals have been made to improve the ductility of beams with FRP by adding5
confinement reinforcement to the compression zone,14,15
but there is little benefit for6
reinforced concrete since the curvatures at failure are very large and the compression7
area after cracking is small, so the addition of confinement reinforcement would have8
little effect.9
Deflections are assumed to be limited to span/250 unless otherwise stated while the10
concrete cover is taken as 50 mm for FRPs and 75 mm for steel due to durability11
concerns. The crack widths are limited to 0.3 mm for all FRPs and 0.2 mm for steel.12
The concrete compression stress under service loads is limited to 40% of the13
characteristic concrete strength.14
Stirrups are provided according to standard practice and are assumed to be made from15
the same material as the flexural reinforcement.16
Deflection limit and material17
Varying the deflection limit from the normal span/250 shows how the governing18
criteria can change. Deflection limits were imposed from span/180 to span/480. All19
the other values were taken as in Table 1. Fig. 5 shows the effect on a beam reinforced20
with relatively flexible Technora bars. In Fig 5(a) the most relaxed deflection limit is21
imposed. The optimal solution is found to lie at the intersection of the crack width and22
deflection constraints. On the other hand, in Fig 5(b) the most strict deflection limit23
(span/480) is applied; the deflection constraint moves higher, reducing beam24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
12/48
deflections, while all the other constraints remain unchanged. Consequently, the1
optimal solution is also at the intersection between crack width and deflections2
constraints.3
For steel-reinforced beams the ultimate load condition governs except when the most4
severe deflection constraint is applied. However, for the FRP-reinforced beams, due5
to their much lower modulus, the optimum solution is almost invariably governed by6
deflection constraints at the working load. For relaxed deflection limits and low7
elastic modulus FRPs like glass and Technora, the optimal solution lies at the8
intersection with the crack width limit, whereas for stiffer FRPs or tighter deflection9
conditions, the optimum condition is usually at the intersection of the deflection and10
balanced-section constraints.11
In Fig. 6, cost ratios relative to the cost of the steel-reinforced beams are shown. Total12
costs include formwork, concrete, flexural and shear reinforcements costs. The figure13
shows that GFRP reinforcement is only marginally more expensive than steel-14
reinforced sections, although of course with much deeper sections. The use of the15
higher-modulus FRPs seems never to be cost-effective for reinforced sections, and in16
general applying more severe deflection constraints increases the cost ratio.17
Sensitivity to price reduction18
It has long been assumed that FRP reinforcement would become more economic as19
the price dropped by the economy of scale theory.16
Fig. 7 shows that, because the20
optimal design lies at a sharp vertex of the feasible zone, reducing the cost of the21
reinforcement (in this case, Fibra) by 40% does not alter the optimal design, although22
it does, of course, reduce the cost of the section. The cost of the reinforcement needs23
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
13/48
to be almost as cheap as steel, on the basis of a cost per unit volume, before there is1
any effect on the position of optimal solution in the diagram.2
Bond3
Bond strength does not normally affect the design of a cross section, provided there is4
enough to grip the rebar at its ends. However, it can have an effect on the crack5
widths, and because FRPs have low stiffness, those limits can be critical.6
Manufacturers have produced bars with various surface characteristics with a view to7
improving bond; representative bond-slip curves, which are assumed to apply to all8
types of FRP, are shown in Fig. 8.17
9
Two d-Ap diagrams are shown. Fig. 9(a) shows the effect of having a smooth surface,10
where the crack-width criterion dominates, while Fig. 9(b) shows the result for a beam11
with braided and indented GFRP bars, which bond very strongly to concrete. As a12
result, the crack width constraint moves to the left on the figure and no longer13
governs. The best solution is then found at the deflection and balanced-section 14
crossing point.15
The effect of the different types of bond are shown in Fig. 10. For beams with low16
elastic modulus fibres, like glass and aramid-Technora, low bond strength is a17
disadvantage, whereas stiffer bars, such as Fibra and CFRP are not affected by the18
crack width criterion.19
Partial material safety factor20
Lack of knowledge on durability and long term behaviour of FRPs has led to various21
proposals on partial material safety factors.18 To assess the effect of the partial22
material safety factor on the optimal solution, values between 1.3 and 2.0 were23
studied and the optimal solutions were found in each case. The partial material safety24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
14/48
factors will only affect the ultimate condition constraints; unity safety factors are1
assumed in service conditions. However, since the balanced section condition is there2
to prevent the bar snapping, increasing the safety factor makes this constraint much3
more severe.4
The balanced-section constraint rotates clockwise on the depth versus bar area5
diagram and the ultimate constraint moves upwards as shown in Fig. 11, which results6
in a substantial increase in total cost when the safety factor is raised. Since the7
deflection constraint still governs, more bars have to be provided to prevent bar8
snapping at failure but the beam can be less deep.9
Concrete strength10
Due to FRP's low elastic modulus, the results up to this point have shown that the11
deflection constraint usually governs the optimal design solution. The use of stiffer12
high strength concrete has been proposed to limit the problem of excessive deflections13
of structures reinforced with these materials.19
14
Analyses have been performed to assess the effect of high strength concrete in FRP-15
RC design. The deflection constraint does indeed yield lower depths but this is16
balanced by a clockwise rotation of the balanced-section line on the d-Ap figure; more17
bars are needed to crush the stronger concrete. In general, the latter affect more than18
offsets the cost reduction due to the stiffer concrete and there is no benefit in using19
stronger concrete.20
Flexural versus total costs21
The d-Ap diagrams give flexural optimum results by inspection, which for expensive22
FRPs are normally located at the intersection between the deflection and the23
balanced-section conditions. The total cost has to include the shear reinforcement as24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
15/48
well and these can be determined from the flexural design. The optimal beam depths1
for low elastic modulus FRPs are normally high to avoid extensive deflections;2
increasing beam depth enhances shear capacity and fewer stirrups may be needed. In3
certain circumstances, however, a default minimum amount of shear links have to be4
provided; consequently stirrup volume may increase.5
When stirrup volume increases with beam depth, the total cost function reduces its6
slope on the d-Ap diagram and the optimal solution may lie at a reduced beam depth.7
The shear demand for the beams reinforced with FRP stirrups was determined by8
assuming truss analogy and stress limitation on the FRP stirrups.20,21 Studies were9
carried out to see the effect of this on the amount of FRP provided, but it was found10
that this would only be significant if FRP stirrups were many times more expensive11
than the cost of straight FRP rebar. This was beyond the scope of this study since it12
depends on the methods adopted to produce the links.13
14
Case Study 2. Determinate Prestressed Concrete15
Prestressed concrete ought to be a logical application for FRPs, and the initial cost16
model can also be used identify ways in which they can best be used in determinate17
prestressed concrete design.18
It has also been identified that elements prestressed with FRP can benefit from19
confinement22
because this provides ductility to the compression zone of the concrete.20
It is assumed here that the FRP prestressed beam's top flange can be confined with21
100 mm diameter AFRP spirals.23
Spirals are placed in regions where the moment is22
highest, and thus where crushing is more likely. The spiral pitch is assumed to be 10023
mm, which is the minimum needed to ensure that the spirals are effective in ensuring24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
16/48
that concrete does not fail explosively. As moment redistribution is not required,1
denser confinement spacing is not needed.2
The structure used for comparison purposes is similar to the reinforced structure3
considered above, but with a 12 m span and an I-shaped cross section. Various inputs4
assumed in the analysis, such as structural and material properties, as well as design5
limits, are summarized in Table 4.6
The prestressing tendons are initially prestressed to 60% for steel and CFRP tendon.7
A lower value is chosen for FiBRA, Technora and Parafil tendons due to creep-8
rupture concerns (Table 4). Concrete creep, shrinkage and tendon relaxation are also9
taken into account.10
Bonding to concrete11
It is not clear whether FRP tendons should be connected to concrete .24,25
Bond12
conditions have a considerable influence on tendon effectiveness and cost.26
The13
method described there was applied to determine the effect of bond on the final14
optimal solution. Three different scenarios were assumed: one with the tendons fully15
bonded to the concrete, one fully unbonded, while the last has partially bonded16
tendons, similar to those tested by Lees and Burgoyne.26 This reduces the local strain17
that is attracted to the tendon, thus reducing the likelihood that the tendon will snap.18
When fully-bonded conditions do not exist, the plane-section-remains-plane19
assumption no longer applies since the tendon sees lower strains than the concrete.20
The effect depends on tendon prestress, tendon profile, tendon length/static depth and21
loading type.27
To allow for these effects, strain reduction factors () are introduced22
for the uncracked as well as for the cracked plane section analysis.3,28
23
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
17/48
Strain reduction factors do not exist in the literature for partially bonded prestressed1
beams. A value for halfway between unity and that determined for the unbonded2
tendons is assumed to represent strain sensitivity in partially bonded beam.3
The plot shown in Fig. 3(a) shows that for steel tendons the optimal solution was4
governed by concrete tensile stresses at the working load and, due to the cost function5
slope, it was located towards higher tendon areas and away from the balanced-6
sections constraint.7
In contrast, FRP tendons are expensive; thus the cost function is steeper on the d-Ap8
diagram. Flexural optimal solutions for the FRP prestressed beams are almost always9
located where the concrete tensile stress limits and the balanced-section conditions10
intersect. Typical plots for a beam with bonded, unbonded, and partially bonded Fibra11
tendons are shown in Fig. 12.12
The differences between the three cases are clear. Most of the constraints do not13
change significantly, but the balanced section constraint is much more severe for the14
fully-bonded case. The ultimate strength, creep rupture and deflection constraints15
shift to lower section depths and tendon areas as the bond increases, while the transfer16
constraints are the same for both cases. For all of these reasons, bonded beams are17
always more expensive than those with unbonded tendons.18
Because the cheapest way to use FRPs in prestressed concrete design is by debonding19
them from concrete, all the FRP beams examined in the sections below are unbonded,20
unless stated otherwise.21
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
18/48
Price reduction1
When designing beams with FRP rebar, the optimal design was at a sharp vertex of2
the feasible region, and very large changes in cost of FRP would be needed to alter the3
optimum. However, the shape of the feasible zone with prestressed beams means4
that, although it still occurs at a vertex, one of the lines at that intersection has a5
similar slope to the cost function. Thus, changes to the cost of the material would6
lead to alterations in the optimal design. These effects are illustrated in Fig. 13(a) and7
Fig. 13(b). A 40% reduction in the cost of CFRP would not alter the optimal solution8
it would remain at the vertex, but for Technora a 40% reduction in price would9
move the optimal solution to a lower depth and higher tendon area (and thus higher10
prestress). It is still clear, however, that significant price changes would be needed11
before the cost of structures with FRP matched the initial cost of structures with steel12
tendons.13
Concrete strength - deflection limits14
Concrete strength has little effect on the total cost. When the concrete strength15
increases, more tendon area is needed to eliminate the tendon snapping mode of16
failure. Thus the balanced-section constraint rotates to require a greater tendon area,17
but this is partly offset by the better shear capacity of stronger concrete.18
The imposition of stricter deflection limits does not affect the optimal solution for19
prestressed concrete since pre-straining flexible FRPs eliminates the problem of20
deflections, as the concrete remains un-cracked. Even for extreme deflection limits as21
much as span/480, the same criteria (tensile stresses in the concrete at the working22
load and balanced-section constraints) still governs.23
24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
19/48
Case Study 3. Indeterminate Prestressed Concrete1
The design of indeterminate prestressed concrete structures differs from the2
determinate beams as more constraints come into play. The amount of prestress has3
to be satisfactory for both hogging and sagging bending, which have an effect both on4
the amount of prestress and the depth of the beam. It is also necessary to ensure that5
the beam has adequate ductility to take benefit from the moment redistribution that6
can occur in such beams. It is often assumed that beams with FRP are less ductile7
than beams with steel, but the large strain capacity of FRPs usually means that large8
rotations can occur as the concrete cracks.9
A continuous 3-span prestressed concrete beam supporting a loaded slab is assumed.10
Structural and section dimensions, as well as loading properties, are shown in Table 5.11
Secondary moments are not considered since these would require the complete cable12
profile to be calculated which is beyond the scope of this paper.13
New design constraints were introduced using the theory of Low29: P2 limits which14
ensure that tendon eccentricities fit in the beam depth and P 3 limits which ensure that15
concordant tendon profiles exist in allowable tendon eccentricities; detailed16
description of those limits can be found elsewhere.317
In particular, for continuous beams prestressed with brittle FRPs, provision of18
confinement in the compression zone is essential.30
The structure has to behave non-19
linearly so that moment redistribution takes place and failure does not occur when one20
of the points on the structure fails. Non-linear behaviour is also needed for design21
calculations to be justified through the safe theorem of plasticity.22
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
20/48
The regions where concrete is likely to fail in compression are confined with AFRP1
spirals similar to those tested by Burgoyne and Leung.30The spirals are placed in the2
top flange in sagging moment regions (mid-span) and in the bottom flange in hogging3
moment regions. The length over which they are extended is chosen to be 70% of the4
span's length for top flange confinement and 30% of the span for the bottom flange.5
Four different spiral pitches were chosen: 100, 50, 35 and 10 mm. A moment6
redistribution of 0, 5, 10 and 20% is also assumed to occur respectively before the7
structure fails. There is not yet sufficient data to allow an accurate calculation of the8
moment redistribution extent according to the spiral spacing provision. No9
redistribution is assumed to take place at 100mm spiral spacing, as this is equal to10
spiral diameter and tests have shown that at this spacing concrete fails in a gentler11
manner, but with limited non-linearity.12
When the confinement is increased, the concrete failure strain increases and thus the13
danger of snapping of tendon before concrete crushing is increased. This can be seen14
on Fig. 14 where two d-Ap diagrams are plotted for beams prestressed with unbonded15
Parafil tendons with 100mm and 10mm spiral pitch. The balanced-section constraint16
moves to greater tendon areas when confinement is increased. In addition, the17
ultimate constraint relaxes to lower section depths and tendon areas as redistribution18
takes place. The net effect is that the tendon area has to increase with a corresponding19
increase in the optimal cost.20
It can be concluded that there is not much benefit in providing confinement in the21
compression zone since it can change the mode of failure, threaten the safety of the22
structure and increase cost considerably. Further research is needed in this area.23
24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
21/48
CONCLUSIONS1
A method has been introduced that allows the optimal design, and hence initial costs,2
of concrete beams reinforced and prestressed with FRPs to be studied. Beams with3
steel and FRP reinforcements were optimised and comparisons were made. Some4
conclusions can be drawn from the analytical results.5
For the reinforced structures:6
the serviceability limit state condition, in most cases deflections or crack7widths, govern the design due to the relatively high flexibility of FRP as8
concrete section reinforcement.9
because the FRP-snapping mode of failure has to be eliminated, the optimum10solution will normally be at the intersection of the above condition and the11
balanced-section constraints12
for the most flexible FRP bars (GFRP and Technora-AFRP) with poor bond13properties (sandblasted or twisted), crack widths may be wide under service14
loads and these will be the govern for aesthetic reasons. For stiffer bars15
(AFRP-Fibra and CFRP) the deflection criterion governs.16
using stronger concrete to diminish the excessive deflections does not seem to17solve the problem. The stronger concrete is harder to crush and more bars are18
needed for that purpose. As stronger concrete is more expensive a cost19
increase was observed.20
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
22/48
beams with FRP reinforcement are usually deeper than their steel-reinforced1counterparts and have higher deflections, but use less reinforcing material2
because it is stronger.3
because the optimal design is at a vertex of the feasible region, small changes4to the design do not reduce the cost. Beams with GFRP are slightly more5
expensive than beams with steel but it is unlikely that beams designed with6
stiffer FRP reinforcement will ever be economic on a first-cost basis.7
For simply supported and continuous prestressed concrete structures:8
optimum solutions are governed by the tension stress limits in the cross9section under working load conditions and the balanced-section constraint, as10
the tendon snapping mode has to be prevented11
in contrast to steel-prestressed bridges, less tendon bond to concrete produces12cheaper solutions. When tendons are bonded they snap more easily and the13
designer will be forced to use more FRPs,14
spiral provision to enhance concrete plasticity triggers tendon snapping and15more flexural bars are needed to avoid this mode of failure,16
for bonded FRP tendons higher prestressing does not necessarily mean17reduced cost,18
designs with unbonded prestressing tendons are the most likely application for19economic use of FRP in concrete20
For reinforced and prestressed structures, the ultimate moment capacity constraint is21
very unlikely to be one of the governing constraints and the optimum structural22
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
23/48
dimensions remain the same even for the most optimistic future FRP price1
predictions. On the other hand, the balanced-section constraint proves to be a2
governing constraint in most cases. Hence, the design calculations should:3
1. determine the material volumes for a balanced section and42. check:5
deflections for reinforced concrete designs or,6
stress limits in prestressed concrete design.7
The outcome will then be very close to optimal in terms of cost.8
From the above study it is clear that the initial cost of FRP reinforced or prestressed9
concrete structures is higher than those with steel, which explains why there are so10
few commercial applications of FRP to date. These additional costs cannot be justified11
unless life cycle costs are taken into account. Whole-life costs for bridges can be12
enormous, especially if the users costs are taken into account as well as the owners,13
and if they are ignored, the cost estimation is far from being complete.3,3114
AKNOWLEDGEMENTS15
The EU-TMR network ConFibrecrete is gratefully acknowledged for funding this16
work.17
18
REFERENCES19
1. FICKELHORN, M. Editorial.Materials and Structures, RILEM, 1990, 23, No. 137,20317.21
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
24/48
2. MAKHTOUF, H. M., ABMADI, B. H. and AL-JABAL, J. Preventing Reinforced1Concrete Deterioration in the Arabian Gulf. Concrete International, 1991, 13,2
No. 5, 65-67.3
3. BALAFAS, I. Fibre-reinforced-polymers vs Steel in Concrete Bridges: Structural4Design and Economic Viability. PhD thesis, University of Cambridge, 2003.5
4. AL-SALLOUM, Y. A. and SIDDIQI G. H. Cost Optimum Design of Reinforced6Concrete Beams. ACI: Structural Journal, 1994, 91, No. 6, 647-655.7
5. GIANNOPOULOS, I. P. and BURGOYNE, C. J. Stress limits for aramid fibres.8Proceedings of the Institution of Civil Engineers: Structures and Buildings ,9
2009, SB4, 221-232.10
6. YAMAGUCHI,T.,NISHIMURA,T.andUOMOTO,T. Creep model of FRP rods based11on fiber damaging rate. Proceedings of the First International Conference12
(CDCC 1998), Durability of Fibre Reinfoced Polymer Composites for13
Construction, eds. B. Benmokrane, H. Rahman, 1998, 427-438.14
7. KALLITSIS CONSULTANTS. GlyfadaAthens. Personal Communication, 2008.158. Republic of Cyprus - Ministry of Public Works. Personal Communication, 2008.169. HUGHES BROTHERS,INC. Personal Communication, 2008.17
10.
PULTRALL,INC. Personal Communication, 2008.18
11. SIREG S.p.A. Personal Communication, 2008.1912. LINEAR COMPOSITES LIMITED. Personal Communication, 2008.2013. HAMBLY,E.C. Bridge Deck Behaviour. 2nd edition, Taylor & Francis, 1991.2114. LEUNG H. Y. Study of Concrete Confined with Aramid Spirals. PhD Thesis,22
University of Cambridge, 2000.23
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
25/48
15. MIKAMI H., KATOH M., TAKEUCHI H. and TAMURA T. Flexural and Shear1Behaviour of RC Beams Reinforced with Braided FRP rods in Spiral Shape.2
Transaction of the Japan Concrete Institute, 1991, 11, No. 1, 119-206.3
16. VARIAN, H. R. Intermediate Microeconomics. 5th edition, W. W. Norton &4Company Ltd, 1999.5
17. COSENZA E.,MANFREDI G.andREALFONZO R. Behavior and modelling of bond6of FRP rebars to concrete.ASCE Journal of Composites for Construction, 1997,7
1, No. 2, 40-51.8
18. Fib-TG-9.3. Chapter 3: Reinforced Concrete. fib Model-Code on Fibre9Reinforced Polymer Reinforcement (Riga Draft). CEB Committee TG9.3, 2003.10
19. NANNI, A. Flexural Behaviour and Design of RC Memebers Using FRP11Reinforcement. ASCE, Journal of Structural Engineering, 1993, 119, No. 11,12
3344-3359.13
20. GUADAGNINI M., PILAKOUTAS K. and WALDRON P. Shear Performance of FRP14Reinforced Concrete Beams. Journal of Reinforced Plastics and Composites,15
2003, 22, No. 15, 1389-1407.16
21. GUADAGNINI M.,PILAKOUTAS K.andWALDRON P. Shear Resistance of FRP RC17Beams: Experimental Study. Journal of Composites for Construction, ASCE,18
2006, 10, No 6, 464-473.19
22. BURGOYNE C. J. Rational Use of Advanced Composites in Concrete.20Proceedings of the Institution of Civil Engineers: Structures and Buildings, 2001,21
146, 253-262.22
23. LEUNG H. Y. and BURGOYNE C. J. The uniaxial stress-strain relationship of23spirally-confined concrete.ACI Materials Journal, 2005, 102, No. 6, 445-453.24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
http://online.sagepub.comhttp/jrp.sagepub.com/content/22/15/1389.abstracthttp://online.sagepub.comhttp/jrp.sagepub.com/content/22/15/1389.abstracthttp://link.aip.org/link/?QCC/10/464/1http://link.aip.org/link/?QCC/10/464/1http://link.aip.org/link/?QCC/10/464/1http://link.aip.org/link/?QCC/10/464/1http://link.aip.org/link/?QCC/10/464/1http://online.sagepub.comhttp/jrp.sagepub.com/content/22/15/1389.abstracthttp://online.sagepub.comhttp/jrp.sagepub.com/content/22/15/1389.abstract8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
26/48
24. BURGOYNE,C.J. Should FRP be Bonded to Concrete?. In A. Nanni and C. W.1Dolan, editors, Non-Metallic (FRP) Reinforcement for Concrete Structures,2
FRPRCS-1, 1993, SP-138, 367-380.3
25. ABDELREHMAN, A. A. and S. H. RIZKALLA. Serviceability of Concrete Beams4Prestressed by Carbon-Fibre-Reinforced-Plastics Bars. ACI Structural Journal,5
1997, 94, No. 4, 447-457.6
26. LEES J.M.andBURGOYNE C.J. Experimental study of the Influence of Bond on7the Flexural Behaviour of Concrete Beams Pre-tensioned with Aramid Fibre8
Reinforced Plastics.ACI Structural Journal, 1999, 96, No. 3, 377-385.9
27. JANNEY,J.R.,HOGNESTAD,E.andMCHENRY,D. Ultimate Flexural Strength of10Prestressed and Conventionally Reinforced Concrete Beams. American Concrete11
Institute, 1956, 27, No. 6, 601-620.12
28. NAAMAN, A. E. and ALKHAIRI, F. M. Stress at Ultimate Post-Temsioned13Tendons: Part 2 Proposed Methodology. American Concrete Institute:14
Structural Journal, 1992, 88, No. 6, 683-692.15
29. LOW, A. M. The preliminary design of prestressed concrete viaducts.16Proceedings of the Institution of Civil Engineers, 1982, 73, No. 2, 351-364.17
30. BURGOYNE C.J.andLEUNG H.Y.Test on Prestressed Concrete Beam with AFRP18Spiral Confinement and External Aramid Tendons, Antoine E. Naaman19
Symposium - Four Decades of Progress in Prestressed Concrete, Fiber20
Reinforced Concrete, and Thin Laminate Composites, SP272-04, American21
Concrete Institute, 2010, 272, 69-86.22
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
http://www-civ.eng.cam.ac.uk/cjb/papers/p36.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/p36.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/p36.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/p36.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/p36.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/p36.pdf8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
27/48
31. BALAFAS I. and BURGOYNE C.J. Economic viability of structures with FRP1reinforcement or prestress,4
thConference on Advanced Composite Materials in2
Bridges and Structures, Calgary, Canada, 2004, Paper 186.3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
http://www-civ.eng.cam.ac.uk/cjb/papers/cp67.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/cp67.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/cp67.pdfhttp://www-civ.eng.cam.ac.uk/cjb/papers/cp67.pdf8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
28/48
LIST OF TABLES1
Table 1. Material Properties .....................................................................................272
Table 2. Material costs.............................................................................................283
Table 3. Parameters for reinforced beam comparisons .............................................294
Table 4. Parameters for prestressed beam comparisons ............................................305
Table 5. Parameters for continuous prestressed beam comparisons ..........................316
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
29/48
Material FormStrength(N/mm2)
ElasticModulus
[kN/mm2]
SnappingStrain
PartialSafetyFactor
B1
GFRPPultruded
rod1200 40 0.030 1.3 0.101
AFRP-
Fibra
Braided
rod 1480 68 0.022 1.3 0.069AFRP-
TechnoraPultruded
rod1900 56 0.034 1.3 0.053
CFRPPultruded
rod2200 140 0.012 1.3 0.016
ParafilParallel-lay
rope1900 120 0.016 1.3 0.069
Steel
rebarRod 460 210 0.1 1.15 -
Steeltendon
7-wirestrand
1700 210 0.1 1.15 -
1determined by fitting on data shown in Giannopoulos and Burgoyne 20095and Yamaguchi1
et al6.
2
Table 1. Material Properties3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
30/48
Material [/m 3 ] [/kN/m]
Concrete1,2 (cf in N/mm
2 ) 48.145e0.0178f
c
steel reinforcement [1],[2] 6700 0.015
steel tendon [1],[2] 20000 0.012GFRP [3],[4],[5] 11500 0.010
FiBRA [6] 82000 0.055TechnoraParafil [6] 82000 0.043
CFRP bars [3],[4],[5] 90000 0.041Formwork [1],[2] [/m
2] 11
[1]7, [2]
8, [3]
9, [4]
10, [5]
11, [6]
12.
Table 2. Material costs12
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
31/48
Structure propertiesdesired lifetime [y] 120
span [m] 6beam web width [m] 0.25
slab depth [m] 0.15
slab width [m] 5
beam top flange effective width [m] 0.85cover [mm] 50 (FRP) 75 (steel)
Loading propertiesslab's short-term live load [kN/m2] 1.0
slab's long-term live load [kN/m2] 1.0
partial dead load safety factor (ultimate) 1.4partial live load safety factor (ultimate) 1.6
partial dead, live load safety factor (working) 1.0
Concrete propertiesstrength [N/mm
2] 30
ultimate strain 0.0035
partial safety factor 1.5
Table 3. Parameters for reinforced beam comparisons12
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
32/48
Structure propertiesdesired lifetime [y] 120
span [m] 15beam web width [m] 0.30
beam top flange width [m] 1.5beam bottom flange width [m] 0.55
beam bottom flange depth [m] 0.2slab depth [m] 0.15slab width [m] 5
cover [mm] 100
Loading propertiesslab uniform live load [kN/m
2] 2.0
slab uniform superimposed dead load [kN/m2] 1.0
partial dead load safety factor (ultimate) 1.4partial superimposed dead load safety factor (ultimate) 1.2
partial live load safety factor (ultimate) 1.6partial dead, live load safety factor (working) 1.0
Concrete properties
strength [N/mm2] 60confinement spacing [mm] 100
partial safety factor 1.5
Initial prestressing (t=0)Steel [%] 60
CFRP [%] 60AFRP (Fibra and Technora), Parafil rope [%] 55
Design limitsdeflections [m] span/250
transfer stress limits [N/mm2] -1 (tension), 18 (compression)
working stress limits [N/mm2] 0 , 25 (compression)
Table 4. Parameters for prestressed beam comparisons12
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
33/48
Structure properties
desired lifetime [y] 120
spans [m] 15 - 20 - 15
beam web width [m] 0.3
beam top flange width [m] 1.5
beam bottom flange width [m] 0.9beam bottom flange depth [m] 0.2
slab depth [m] 0.2slab width [m] 5
cover [mm] 100
Loading properties
slab uniform live load [kN/m 2 ] 2.0
slab uniform superimposed dead load [kN/m 2 ] 1.0
Table 5. Parameters for continuous prestressed beam comparisons1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
34/48
LIST OF FIGURES1
Fig. 1. Depth versus Ap diagram for reinforced concrete sections2
Fig. 2. Beam with: (a) steel rebar; (b) FRP rebar3
Fig. 3. Beam with: (a) steel tendon; (b) FRP tendon4
Fig. 4. Structural configuration5
Fig. 5. Technora reinforced beam with deflection limits: (a) span/180; (b) span/4806
Fig. 6. Cost of RC beams with different materials and different deflection limits7
Fig. 7. Effect of reducing cost of FRP (Fibra) by 40%8
Fig. 8. Bond-slip curves (Cosenza et al 1997)9
Fig. 9. Design curves for GFRP bars: (a) smooth; (b) braided and indented10
Fig. 10. Effect of bond on amount of FRP needed11
Fig. 11. Effect of varying Partial Material Safety Factor for CFRP-reinforced beams12
(a) = 1.3; (b) = 2.013
Fig. 12. Prestressed Beams with Fibra tendons: (a) fully bonded; (b) partially bonded;14
(c) unbonded15
Fig. 13. Prestressed Beam. Effect of price on: (a) CFRP; (b) Technora16
Fig. 14. Indeterminate Prestressed Beam. Technora spirals: (a) 100 mm pitch; (b) 1017
mm pitch18
19
20
21
22
23
24
1
2
3
4
5
67
8
9
10
11
12
13
14
15
16
17
18
1920
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
35/48
0 1000 2000 3000 4000 5000 6000 70000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ap
mm[2
]
p
[
]
Feasible zoneBeams stronger to
design specifications
constraintStrength
Concretecrushing
Barsn
app
ing
Balanced section
Non Feasible zoneBeams weaker to
design specifications
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
mm[2
]
depth
m[
] Feasible zone
sconstraint
trength
Balanced section
minimum reinforcementratio constraint
concrete crushingzone
bar snappingzone
bala
nced
secti
ons
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
mm[2
]
p
[
]
Feasible zone
sconstraint
trength
minimum reinforcementratio constraint
deflectionsconstraint
bala
nced
secti
ons
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
mm[2
]
depth
m[
]
Feasible zone
sconstraint
trength
deflectionsconstraint
Flexural Optimolution
alS
ba
lanced
sections
Costfu
nctio
n
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
mm[2
]
depth
m[
]
Feasible zone
strength
deflections
concrete stresses
crack width
Flexural Optimolution alS
bala
nced
secti
ons
Costfu
nctio
n
Fig. 1. Assembly of the section depth-Ap diagram for FRP reinforced sections
a. Ultimate strength condition b. Bar snapping condition
c. Deflection constraint d. Cost function
e. Complete diagram for FRP-reinforced section
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
36/48
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
(mm2
)
depth(m)
Flexural Optimumsolution
defle
ction
s
Feasible zone
max
strength
Costfunction
b
ala
nced
secti
ons
0 500 1000 1500 2000 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
(mm2
)
depth(m)
Flexural Optimumsolution
Feasible zone
deflections
strength
bala
nced
sect
ions
Costfu
nctio
n
Fig. 2. Beam with: (a) steel rebar; (b) FRP rebar
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
37/48
0 1000 2000 3000 4000 5000 6000 70000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ap
(mm2
)
p
(
)
Feasiblezone
strength
Flexural optimumsolution
transfertension
workingtension
ransfercompression
t
Costfunction
orkingcompression
w
bala
nced
sections
! max
0 1000 2000 3000 4000 5000 6000 0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ap
(mm2
)
depth(m)
Feasiblezone
strength
transfercompression
workingtension
transfertension
Costfu
nctio
norking
compressionw
Flexuraloptimsolution
al
bala
nced
sections
Fig. 3. Beam with: (a) steel tendon; (b) FRP tendon
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
38/48
5.00
15
Ap = ?
d= ?
0.2
6.00
0.85
Fig. 4. Structural configuration
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
39/48
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
A
p
(mm2
)
Feasible zone
Flexural Optimumsolution deflections
crackwidth
creep-rupture
concretestresses
b
ala
nced
secti
ons
costfunctio
n
0 500 1000 1500 2000 20
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
A
p
(mm2
)
depth(m)
Feasible zone
Flexural Optimumsolution deflectionscrack
width
creep-rupture
concretestresses
bal
anced
secti
ons
costf
unctio
n
Fig. 5. Technora reinforced beam with deflection limits: (a) span/180; (b) span
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
40/48
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
l/180
l/250
l/360
l/480
l/180
l/250
l/360
l/480
l/180
l/250
l/360
l/480
l/180
l/250
l/360
l/480
glass FiBRA Technora Carbon
cos
tr
atios
FRP
/stee
l
. 6. Cost of RC beams reinforced with different materials and different deflection
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
41/48
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
(mm2
)
depth(m)
Optimal solution
deflections
crack width
Feasible zone
strength
bala
nced
sectio
ns
curre
ntp
rice
40%barpri
cereduction
ig. 7. Effect of reducing cost of FRP (FiBRA) by 4
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
42/48
ribbed
twisted
braided and sanded
sand blasted
smooth
Fig. 8. Bond-slip curves 17
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
43/48
0 500 1000 1500 20000
0.2
0.4
0.6
0.8
1
1.2
Ap
(mm2
)
p
(
)
Feasible zone
Optimal solution
crackwidth
deflections
strength
bala
nced
sectio
ns
costfunction
creep rupture
concrete
stresses
0 500 1000 1500 0
0.2
0.4
0.6
0.8
1
1.2
Ap
(mm2
)
depth(m
)
Optimal solution
crackwidthdeflections
strength
bala
nced
sectio
ns
Feasible zone
costfunction
creeprupture
concrete stresses
Fig. 9. Design curves for GFRP bars: (a) smooth; (b) braided and indented
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
44/48
0
200
400
600
800
1000
1200
1400
smooth
sandblasted
twisted
ind-braided
ribbed
graincoated
braidedsanded
smooth
sandblasted
twisted
ind-braided
ribbed
graincoated
braidedsanded
smooth
sandblasted
twisted
ind-braided
ribbed
graincoated
braidedsanded
smooth
sandblasted
twisted
ind-braided
ribbed
graincoated
glass FiBRA Technora Carbon
flexuralbararea[m
m2]
Fig. 10. Effect of bond on amount of FRP needed
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
45/48
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
(mm2
)
p
(
)
crackwidth
deflections
strength
Optimal solution
Feasible zonebalan
ced
costfunctio
n
sections
cree
p
rup
ture
concrete stresses
0 500 1000 1500 2000 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Ap
(mm2
)
depth(m)
crackwidth
deflections
strength
Optimal solution
bala
nced
Feasible
costfunctio
n
sections
zone
creep rupture
concrete stresses
g. 11. Effect for varying Partial Material Safety Factor for CFRP-reinforcedams: (a) =1.3; (b) =2.0g g
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
46/48g. 12. Prestressed beams with FiBRA tendons: (a) fully bonded;unbonded
(b) partially bond
0 1000 2000 3000 4000 5000 6000 0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
A (mm2
)
depth(m)
Feasiblezone
streng
th
transfercompression
workingtension
orkingcompression
w
transfertension
balan
cedse
ction
s
Optimal
solution
Costfu
nctio
n
d
eflections
creeprupture
0 1000 2000 3000 4000 5000 6000 70000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
A (mm2
)
dep
th(m) Feasible
zone
Optimalsolution
transfertension
ultimate
transfercompression
workingtensiondeflections
creeprupture
orkingcompression
w
balan
ceds
ectio
ns
Cos
tfunctio
n
(a) (b)
(c)
0 1000 2000 3000 4000 5000 6000 70000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ap
(mm2
)
depth(m)
Feasiblezone
Optimal solution
transfer tension
ultimate
transfercompression
workingtension
deflectio
ns
creeprupture
orkingcompression
wbal
anced
section
s
Costfu
nctio
n
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
47/48
0 1000 2000 3000 4000 5000 60000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ap
(mm2
)
p
()
Optimalsolution
transfer tension
strength
transfercompression
workingtension
deflections
Feasiblezone
orkingcompression
w
current
price
40%barpricereduction
creeprupture
balan
ceds
ectio
ns
0 1000 2000 3000 4000 5000 0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ap
(mm2
)
depth(m)
Feasible
zone
strength
transfertension
transfercompression
workingtension
creeprupture
orkingcompression
w
deflectio
ns
bala
nced
sectio
ns
currentp
rice
40%bar
pricereduction
Optimalsolution
Fig. 13. Prestressed beam - Effect of price on: (a) CFRP; (b) Technora
8/2/2019 p75preEconomic Beam Designs Using FRP Rebar
48/48
0 5
1
1.5
2
2.5
3
depth
(m)
Feasiblezone
Optimalsolution
P3min
Z limits
P1min
P2strength
balan
ced s
ectio
ns
Costfu
nctio
n
5
1
5
2
5
3
Feasiblezone
strength
P3min
Z limits
P1min
P2
creep
balan
ced
sectio
ns
Co
stfunctio
n
Optimalsolution