Example #2 - Flexural Strength With Unbonded Tendons: 86 Post-Tensioned Concrete - Principles and Practice
Example #2 - Flexural Strength With Unbonded Tendons: 86 Post-Tensioned Concrete - Principles and Practice
Example #2 - Flexural Strength With Unbonded Tendons: 86 Post-Tensioned Concrete - Principles and Practice
Find:
40 (12 in/ft)
Span to depth ratio = = 16 < 35 (this is typical for beams)
30 in
f
∴ Use f =f + 10,000 +
100ρ
fpy & fse + 60,000
7(0.153 in )
ρ = = 0.000374
110 in(26 in)
(note: b = width of the compression face = 110 in)
4,000 psi
∴f = 174,000 psi + 10,000 psi + = 290,951 psi = 291 ksi
100(0.000374)
K. Dirk Bondy & Bryan Allred 87
fpy = 243ksi
or fse + 60,000 = 174 ksi + 60 ksi = 234 ksi < 291 ksi
Assume the compression steel yields & draw the force diagram.
Now use the Whitney Stress Block to find “a” & “c”.
C 403 kips
a= = = 1.08 in < t slab =6 in (rectangular assumption is correct)
0.85f b 0.85(4ksi)(110 in)
a 1.08 in
& c= = = 1.27 in
β 0.85
This example demonstrates that including compression steel in a tee-beam is of very little value. The
neutral axis will typically be in the flange and therefore the compression steel will virtually never be
yielding in compression. Therefore for tee-beams we will ignore the compression steel.
88 Post-Tensioned Concrete - Principles and Practice
C 440.2k
a= = = 1.18 in
0.85f b 0.85(4ksi)(110 in)
a 1.18 in
&c = = = 1.39 in
β 0.85
Calculate Strength.
∴Mn = [189.6 kips(27.5 in) + 250.6 kips(26 in) – 440.2 kips(0.59 in)]/(12 in/ft)
= 955.8 ft-kips
Verify Section Has Minimum Bonded Reinforcement Required per ACI 318 Section 9.6.2.3.
ACI 318 Section 9.6.2.3 – Minimum bonded reinforcement for beams with unbonded tendons:
0.004Act (9.6.2.3)
Where Act = Area of cross section between the tension face and the c.g.c.
Some people refer to secondary effects as “hyper-static” effects. However, we have never heard any of
the legends who taught us how to design post-tensioned concrete use this term and we do not use it
ourselves. When I think of “hyper” I think of my kids when they were young running around the house,
and “static” indicates no movement. So this term seems contradictory and misleading to me. We will be
consistent in this book with the ACI 318 code and only use the term “secondary” to describe the effects
of post-tensioning on an indeterminate system.
How do we find the secondary effects? Look at the “concrete-only” final moment diagram, the moment
diagram generated from the equivalent loads due to the post-tensioning (never with applied loads). The
equivalent load is wequiv acting upward.
Remember, in a statically determinate system we found that the “concrete-only” moment at any point
was always equal to F (the total tendon force) multiplied by e (the eccentricity between the concrete
c.g.c. and the tendon c.g.s.). Therefore, the “primary concrete-only moment diagram” for this example
would look like this:
92 Post-Tensioned Concrete - Principles and Practice
The difference between the final and primary concrete-only moment diagrams is the secondary moment
diagram (remember, only prestressing forces are considered).
Secondary reactions and shears are found from the free body diagrams. Break the spans apart and
determine the individual reactions (shears) for each span:
The reactions at the supports are the sum of the shears from each span at that support.
Again, the sum of the reactions must be zero because simply prestressing the beam does not add any
weight or load to the system.
Remember, for ultimate moment (strength) design we calculate the moments due to the external loads
acting on the system. The secondary reactions are external to the system. They are generated by support
restraint in an indeterminate system. Simply stated, the member wants to deflect due to the prestressing
and the supports resist that movement. This generates reactions, shears and moments in the complete
system. Therefore, the final demand moment diagram is found from the following loading:
K. Dirk Bondy & Bryan Allred 93
Mdemand = Mu + M2
= 1.2MD + 1.6ML + 1.0M2
Secondary effects will generally lower the factored moments over continuous supports (lower the
demand negative moments) but increase the positive moments.
In the previous example we subtracted the prestress load from the applied load to get the “net” load on
the concrete. However, now we will find the moments due to the equivalent prestressing loads only.
Recall that in a statically determinate beam the moment due to prestressing is always “F” times “e”. But
in an indeterminate beam that will usually not be true.
e = 8 in – 4 in = 4 in
F = 186.2 kips
Mprimary = Fe = 186.2k(4 in)/12 in/ft = 62.1 ft-kips
∴ M2 = Mpre final – Mprimary = 289 ft-kips – 62.1 ft-kips = 227 ft-kips (tension on the bottom)
The reactions that cause this moment diagram must look like this:
ΣFy = 0 ΣM = 0
Determine Shears
96 Post-Tensioned Concrete - Principles and Practice
Mdemand = Mu + M2
Notice that by including the secondary moments the final design moment at the interior support has been
dramatically decreased relative to the factored moment diagram, and the positive moments have been
increased but not nearly by the same proportion. This is beneficial to the design because the negative
moment capacity is less efficient due to the smaller compression flange, and the positive moment
capacity is much more efficient due to the large compression flange. So anything we can do to decrease
negative moments and increase positive moments in tee-beam construction makes the design more
efficient.
In practice, with secondary moments at all columns (exterior and interior columns) the secondary shears
are extremely small and are often ignored. This will be demonstrated in a later example.
Now that we have the demand moment diagram, proceed with the flexural design as we did previously.
Determine the capacity of the beam at the critical sections (maximum positive and negative moments).
Determine fps, etc. at each of these sections and add rebar as necessary for ultimate strength and/or
minimum requirements.
K. Dirk Bondy & Bryan Allred 97
Let’s look at the “prestress only” moment diagram previously found with the maximum span moment
values included.
It is theoretically possible to create the same “prestress only” moment diagram that would have no
secondary effects. How? By dividing the moment at each point by the prestress force and calculating “e”
at those points. In other words, creating the profile in which the primary moment equals the final
moment diagram throughout the member. That profile is a “concordant” profile and represents the “C-
Line.”
Let’s look at the interior support. If we divide 289 ft-kips by F (=186.2k) we get an eccentricity of (289
ft-kips/186.2k)(12 in/ft) = 18.6 in. However, remember what the beam section looks like:
With an eccentricity of 18.6 in. the tendon would be 10 in. above the top of the beam! Therefore, we
need to proportion the profile in the beam at all locations until it fits inside everywhere, but when we do
that we will have much less balance load for the same force, F. Therefore, to achieve the same balance
loads and “prestress only” moment diagram we will need to dramatically increase the prestress force, F.
At the middle support B, to achieve a primary moment of 289 ft-kips with an eccentricity of 4 in (8 in –
4 in, or 0.33 ft) the required prestress force, F is 289 ft-kips/0.33 ft = 875.8 kips (an increase of 370%!).
The full concordant profile and force required to provide the same “prestress only” moment diagram is:
This is a remarkably inefficient design. Why spend 370% more on prestressing steel when we have
already shown that the secondary effects actually help us by decreasing the negative moments where the
compression flange is small and increasing the moment in the span where we have a much more
efficient beam section because of the large compression flange? Concordant profiles are definitely only
an academic exercise.
6
Shear Design
Beam Shear
Shear design of beams is covered in ACI 318 Section 22.5.8. For instructional purposes we will assume
that all concrete in the examples is normal weight and we will ignore the lightweight concrete factor, λ
in the equations.
According to ACI 318 Section 21.2 the strength reduction factor, φ for one-way shear is 0.75 (this is the
same for non-prestressed and prestressed beams). The basic required equations for shear design are:
The approach to beam shear design is similar to that of non-prestressed reinforced beams, except in the
calculation of the concrete contribution, Vc. According to ACI 318 Section 22.5.8.2 the designer is given
two choices for determining Vc. The first is the relatively straightforward equation (22.5.8.2(a)):
Vd
V = 0.6 f + 700 b d (Table 22.5.8.2)
M
V d
where cannot be taken greater than 1.0
M
(In ACI 318 this term is Vc, but we have added the subscript “n”)
The second choice that ACI 318 offers for determining Vc is contained in Section 22.5.8.3. Vc may be
taken as the lesser of Vci or Vcw (the “i” refers to inclined flexural cracking and the “w” refers to web
cracking).
99
100 Post-Tensioned Concrete - Principles and Practice
Vci
VM
V = 0.6 f b d + V + (Eqn. 22.5.8.3.1a)
M
where dp need not be taken less than 0.80h & Vci need not be taken less than 1.7 f b d
Note that the lower bound of 1.7 f b d does not make much sense given that we always have the
option to use Table 22.5.8.2 and its lower bound is 2 f b d. Not to mention that we could also ignore
the prestressing altogether and use 2 f b d as a non-prestressed beam.
M = Iy 6 f + f f (Eqn. 22.5.8.3.1c)
(Note: ACI 318 actually uses the term “yt” instead of “ytension”. Our reasons for using a different
subscript will be explained in the example.)
Vi = Factored shear force at section due to externally applied loads (loads applied after the member is
prestressed) occurring simultaneously with Mmax.
Mmax = Maximum factored moment at section due to loads applied after prestressing.
fpe = Compressive stress in concrete due to effective prestress forces only (after losses) at the extreme
fiber of the section where tensile stress is caused by externally applied loads, psi.
fd = Stress due to unfactored dead load at the extreme fiber of the section where tensile stress is
caused by externally applied loads, psi.
Vcw
V = 3.5 f + 0.3f b d + V (Eqn. 22.5.8.3.2)
fpc = Compressive stress in concrete (after losses) at the centroid of the cross section (pure F/A), psi
ACI 318 Section 22.5.8.2 states that the concrete contribution, Vc shall be less than or equal
to 5 f b d , but need not be taken as less than 2 f b d. However this section only includes Table.
22.5.8.2. Section 22.5.8.3 states that Vc shall be permitted to be computed as the lesser of Vci and Vcw.
Since Section 22.5.8.3 does not specifically state an upper bound for either V ci or Vcw, most engineers
and computer programmers have concluded that the upper bound of 5 f b d does not apply to these
K. Dirk Bondy & Bryan Allred 101
equations, despite the fact that the lesser value will be used as V c. Earlier versions of this book limited
all values of Vc to 5 f b d since the Vci equation will approach infinity at inflection points (when has
any Code ever not placed an upper bound on an equation that can approach infinity?), but our view was
widely criticized for going against the standard of practice, so we now follow the herd.
1. Vu & Mu, based upon the ACI 318 definitions do not include secondary effects.
2. Vp, the vertical component of the prestressing force in the V cw equation is difficult or impossible
to determine at reverse curvature of the tendons and is typically ignored in calculations
(including computer programs). It is conservative to ignore Vp.
3. The terms Vi & Mmax include only the factored loading applied after the member is stressed and
stripped of formwork. Vi & Mmax exclude the unfactored dead load shears and moments, but
include the factored dead load shears and moments that occur after stressing, as well as all of
the factored live load shears and moments. Since this can be somewhat confusing and tedious to
calculate, we will take advantage of the fact that typically the ratio of live load shear and live
load moments will be very similar, if not exactly the same, as Vi & Mmax. This is because all of
the post-stressed loading occurs with the top column present (if one exists), and any ratio of V
& M should be similar. Therefore, for instructional purposes in our examples with columns we
will use the ratio of live load shear to live load moment in place of Vi & Mmax. For simply
supported beam examples we will use any convenient ratio of V & M that we have.
4. The ACI 318 Code requires that at least 40% of the total tensile strength will be provided by the
prestressing reinforcement in order to use Table 22.5.8.2. We will assume in our examples that
this is easily met and will not check it.
5. Shear reinforcing (stirrups) is always oriented parallel to the loading (& flexural reinforcing is
always oriented perpendicular to the load.) We’ve heard some engineers say that stirrup
reinforcing is most effective at a 45° angle, but that just is not the case. Keep the following
diagram in your mind and you will never make that mistake.
102 Post-Tensioned Concrete - Principles and Practice
Service Loading:
wd = 3.50klf
wl = 1.25klf
Remember, Vu & Mu are defined by ACI 318 as 1.2D + 1.6L so even for continuous beams the strength
Table 22.5.8.2 will not include secondary effects. However, for the demand shear the secondary shears
should technically be included. We will demonstrate in the two-span beam design in the next chapter
that these shears are relatively small so ignoring them as most designers do is typically inconsequential.
Find: Stirrups and spacing required at a distance “h/2” from the face of the support using:
a) Vcn (Table 22.5.8.2)
b) Minimum of Vci (Eqn. 22.5.8.3.1a) & Vcw (Eqn. 22.5.8.3.2)
c) Vc as a non-prestressed member
Solution:
V d
V = 0.6 f + 700 b d (22.5.8.2(a))
M
K. Dirk Bondy & Bryan Allred 103
“d” is defined as the distance from the compression face to the centroid of all tensile reinforcement, but
not less than 0.80h. It’s typical to ignore the non-prestressed steel and take “d” to the prestressing
reinforcement only.
Somewhat inexplicably, “dp” does not have the 0.80h stipulation in this equation, though it does in all
other equations, as does “d”.
ACI 318 Section 9.4.3.2 states that prestressed concrete beam sections located less than a distance h/2
from face of support shall be permitted to be designed for Vu computed at h/2. In non-prestressed
concrete this critical shear location is defined as “d”. However, the commentary explains that because
“d” varies in prestressed concrete members, h/2 was arbitrarily chosen as the critical section.
The centroid of the prestressing steel at h/2 from the support is:
1.25 ft
d = 9.32 in + (16.68 in) = 10.4 in
20 ft
Mu @ “h/2” = the shaded area under the shear diagram since the moment at the support is zero.
124 kips + 116.3 kips
= (1.25 ft) = 150.2 ft kips
2
= 1,802 in kips
(116.3 kips)(10.4 in)
V = 0.6 4000 psi + 700 (16 in)(24 in)
1,802 in kips
Vd
Remember that the term must be < 1.0
M
Vci
VM
V = 0.6 f b d + V + (Eqn. 22.5.8.3.1a)
M
(Note: In this equation, d is greater than or equal to 0.80h)
3.50klf
∴V = (116.3 kips) = 65.7 kips
6.2klf
This equation can be a little confusing, particularly if you are following along with the ACI 318 Code.
ACI 318 uses the term “yt” instead of “ytension.” We changed the term in this design book because it is
too easy to confuse “yt” with distance from the centroid to the top of the member, and we didn’t want to
have one term that has two different definitions. If you keep in mind that this equation attempts to
predict the moment at cracking under applied loads and which side of the beam will crack at a particular
section (top or bottom) you will hopefully remain clear on this. Remember, all of the section properties
used in the Mcre equation are to the tension face under applied loads at a particular section.
For this example, ytension = yb (distance from the centroid to the bottom of the beam because the applied
loads cause tension on the bottom)
ytension = yb = 20.68 in
The section modulus used in the terms fpe and fd will also be to the tension face under applied loads.
S = Sb = 3,429in3
fpe = Compressive stress in the concrete due to the prestressing only at the extreme fiber where
external loads cause tension. In the case of a simply supported beam, the moment due to prestressing is
equal to F multiplied by e. This is typically not the case in an indeterminate beam.
P M
f =
+
A S
(note: “plus” because prestressing causes compression where external loads cause tension)
fd = Stress due to unfactored dead load at the section where external loads cause tension
K. Dirk Bondy & Bryan Allred 105
w
M (@ "h/2") = (M @ "h/2")
w
3.5klf
= (150.2 ft kips) = 84.8 ft kips
6.2klf
Finally,
M = 70,938 in 6√4000 + 270 psi 297 psi /1000
20.68 in
= 1,209 in kips
= 100.8 ft kips
In this example, since there are no columns above or below occurring at different times in construction,
the ratio of shear to moment is the same for all loading conditions. We will use the dead load shear and
moment for convenience.
V V 65.7 kips
= same ratio in this case as =
M M 84.8 ft kips
= 158.3 kips
Vcw
3.5√4000
∴V = + 0.3(0.212 ksi) (16 in)(24 in) = 109.4 kips
1000
Summary
Vcn = 121.4 kips
Vci = 158.3 kips
Vcw = 109.4 kips
106 Post-Tensioned Concrete - Principles and Practice
Lower Bound (the concrete contribution need not be less than this):
2 f b d = 48.6 kips
ACI 318 Section 22.5.8.2 states that we may use Vcn, or the smaller of Vci or Vcw. For this example we
would benefit most from the Vcn value of 121.4 kips.
Design Stirrups
Reinforcement Contribution, V s
Check Maximum Spacing for Prestressed Members (ACI 318 Section 9.7.6.2.2):
0.75h = 0.75(30 in) = 22.5 in < 24 in but 9.4 in still governs at this location.
ACI 318 Section 9.6.3.3 states that for prestressed members with an effective prestress force not less
than 40 percent of the tensile strength of the flexural reinforcement, Av(min) shall not be less than the
smaller value from the greater of Table 9.6.3.3(c) and Table 9.6.3.3(d) or Table 9.6.3.3(e). In practice
these equations in the range of typical designs will never control the design of a prestressed concrete
beam, but they will be shown for academic completeness.
b S 50b S
A ( ) = 0.75 f (Table 9.6.3.3(c)&(d))
f f
where "t" stands for transverse
A f S d
A ( ) = (Table 9.6.3.3(e))
80f d b
Again, assume 2 legs of #3 stirrups and solve for the maximum spacing, S.
Rewriting the equations in terms of the spacing means that largest spacing from the greater of Table
9.6.3.3(c) and Table 9.6.3.3(d) or Table 9.6.3.3(e) controls minimum requirements.
K. Dirk Bondy & Bryan Allred 107
The maximum spacing based upon minimum requirements is 71.5 in., but we need a spacing of 9.4 in.
for strength at this location. Table 9.6.3.3(e) will typically generate a spacing value that grossly exceeds
the smaller of 0.75h or 24 in., and it is for this reason that these two equations can typically be ignored
in most designs.
c) As a Non-Prestressed Member
1) V u & Mu
2) Vcn (ACI Eqn. 22.5.8.2(a), which has an upper bound of 5 bwd)
3) Vci (ACI Eqn. 22.5.8.3.1a)
4) Vcw (ACI Eqn. 22.5.8.3.2)
5) The required stirrup spacing per the ACI 318 code using the values found in b) through d).
Use 2 legs of #3 stirrups (Av/leg = 0.11 in2) & a lower bound of 2 bwd for all Vc equations.