Behaviour of grouted

anchors

N. Subramanian and Ronald A. Cook

Grouted anchors are often used to connect steel structures and Post-installed anchors
microwave towers to concrete foundations. However, very little Advances in carbide-tipped bits and rotary hammer drills
research has been done on the behaviour of such anchors and have made many new anchors practical because they can
the codes of practices do not provide methods to check the now be post-installed in hardened concrete. There are two
adequacy of such connections. The paper presents a brief review types of post-installed anchors: mechanical and bonded.
on the topic with a discussion on suggested behaviour models.
Also, the results of 229 tests conducted on both headed and un- Mechanical anchors
headed anchors (installed using six cementitious and three Post-installed mechanical anchors, which function on the
polymer grouts) at the University of Florida are compared principles of keying and friction, include undercut, heavy-
with those of the analytical models suggested in the literature duty sleeve, torque controlled expansion and deformation
and a suitable model is suggested for the use of designers. controlled expansion anchors. These anchors loaded in
More tests are necessary to predict the shear and combined tension, apply reaction forces to the concrete at the expansion
shear and tension behaviour of grouted anchors. mechanism, usually near the end of the embedded part of
the anchor. Failures typically occur when the anchor pulls
out a cone of concrete, or slips out of the hole or simply
Fastening systems in the form of anchor bolts are usually breaks in response to the failure of the steel1.
used for connecting steel microwave towers or industrial
steel structures to the concrete foundation through base
plates, Fig 1. These fastening systems can be classified into
cast-in-place systems and post-installed systems 1. Cast-in-
place systems are placed in the formwork before casting of
the concrete, while post-installed systems are incorporated
in hardened concrete. Advancements in drilling technology
have resulted in reliable post-installed systems. Due to their
ease in installation, post-installed anchors are used increasingly
in new construction, as well as repair and retrofit projects.

The design methods of cast-in-place anchors and the recent

developments in their design methods have been presented
in earlier papers2-3. The methods of installation, behaviour
and design of bonded anchors are also discussed elsewhere4.
In this paper, the behaviour and design of grouted anchors,
which are often encountered in India, are discussed.

April 2004 * The Indian Concrete Journal 1

 Bonded grouted anchors
Grouted anchors, Fig 3, may consist of headed bolts, threaded
rods with nut and washer at the embedded end (block-end
anchor) or deformed reinforcing bars with or without end
anchorage. Commonly-used bolt diameters vary from
12-50 mm, although larger bolts are sometimes used. Grouted
anchors are typically installed with a cementitious or polymer
grout in a pre-drilled or pre-formed hole having a diameter
range of 150-300 percent larger than the diameter of the
fastener. Cementitious grouts are primarily composed of fine
aggregates and portland cement. Polymer grouts are similar
to adhesive anchors but with a fine aggregate component.
Some proprietary grouts cater to controlled expansion and
thus develop compression stress in the grout after setting.
This may serve to increase the contribution of friction to the
tension load-bearing resistance of the anchorage.

Bonded adhesive anchors If an end anchorage is provided, the shank may be either
bonded, Fig 3(a) or bonded, Fig 3(b). De-bonded anchors
Post-installed bonded anchors can be classified as adhesive
prelude tension load transfer near the concrete surface. De-
or grouted depending on the bonding agent, anchor type,
bonding is normally achieved by means of a thin sleeve,
and hole dimensions, Fig 2. These types of anchors can be
which prevents the resin from contacting the rod. Grouted
installed with or without a head at the embedded end.
anchors transfer tension loads to the base material by a
Adhesive anchors are installed using an unheaded threaded
combination of bond, friction and mechanical interlock. Belled
rod or a reinforcing bar inserted in a pre-drilled hole — that
or keyed holes, Fig 3(c) and 3(d) will increase the mechanical
is 10-25 percent larger than the anchor diameter — using a
interlock while resisting tensile loads; however they can be
polymer-based bonding agent including epoxies, polyesters,
produced only by placing block-outs in the concrete form
vinylesters, and hybrid systems. The installation methods,
work and as such cannot be considered as post-installed
behaviour and design of bonded adhesive anchors are
anchors 1.
discussed elsewhere4.
The grouted anchors offer an advantage over cast-in-place
anchors, since planning efforts and costs of special formwork
are reduced.

As discussed above, grouted anchors can be distinguished

from adhesive anchors by a larger hole-to-anchor diameter
ratio that can accommodate a headed anchor, which
ultimately affects the load transfer mechanism. Headed
anchors transfer the load to the grout primarily by bearing
on the anchor head. Unheaded anchors installed with
threaded rod take advantage of mechanical interlock between
the threads and the grout. In both cases, the load is transferred
from the anchor to the grout and the grout then transfers the
load to the concrete resulting in one of three potential failure
modes, Fig 4.

Unheaded grouted anchors typically exhibit a failure mode

similar to adhesive anchors; bond failure at the steel-grout

2 The Indian Concrete Journal * April 2004

interface with a secondary, shallow concrete cone,
Fig 4(a). Headed anchors exhibit either bond at the grout-
concrete interface or concrete breakout cone failure depending
on the concrete strength, embedment depth, and grout-
concrete bond strength, Fig 4(b) and (c).

Thus, for anchors grouted in place, four possible failure

mechanisms can be considered5:

(i) tensile failure of the bolt

(ii) full depth concrete cone breakout

(iii) bond failure at the grout-concrete interface with a

shallow concrete cone failure, and

(iv) bond failure at the steel-grout interface with a shallow

concrete cone failure for unheaded grouted anchors.

It is to be emphasised that failure of the bolt needs to be

the objective of the designer. However, practical limitations
on the embedment depth may sometimes prevent full
development of the bolt steel strength.

For polymer-grouted anchor bolts, the load capacity of interface.

the polymer grout is significantly affected by the hole size.
The displacement of the bolt head within the polymer The effect of edge distance, de , may be incorporated into
becomes more significant as the hole size is increased. Bond the calculation of A1. Thus, the following formulae may be
strengths of polymers used vary in the range 6.2-12.4 MPa5,6,7. used.
Since these strengths are greater than the actual shear strength
of concrete at the interface, interface shear failures are When anchor is near a free edge:
expected to occur primarily in concrete, unless good bond
strength has not been achieved due to improper cleaning of
concrete holes prior to grouting5.
2
[
A 1 = π (rs2 – r ) − φ / 2 rs2 + de (r
s
2
]
− de2 ) ...(2)

when the anchor is not near the free edge:

Behavioural models A 1 = π(rs2 – r2) ...(3)
The behaviour of grouted anchors was expected to be similar
A 2 = 2πr[hef – (rs – r) tan θ] ≥ 0 ...(4)
to either cast-in-place headed anchors or post-installed 2 2
adhesive anchors depending on the anchor configuration A3 = π(r – rh )≥0 ...(5)
(headed or unheaded) and material properties. The following
presents a general discussion of suggested behavioural where,
models. θ = the angle of the spall cone from the concrete
surface (θ may be taken as 45o )
Combined cone-bond stress model r s = the radius of the spall cone at the surface
Cones7 and James et al5 have suggested the following model de = the distance from the anchor centre line to
for calculating the ultimate resistance of grouted anchors the free edge
based on the mechanism shown in Fig 5.
hef = the embedment length
Nu = ft A1 + uA2 + f g A3 . . . (1) φ = ar tan c/( √ (rs2 – de 2) /d e ) = the angle in radians
where, subtended in plan view by the intersection of
the failure cone with the free edge
A 1 = the horizontal projected area of the spall cone
r h = the radius of the anchor head or the anchor
A 2 = the area of the exposed vertical cylindrical
body in the case of headless anchors
surface of the grout
A 3 = the horizontal projected area of the grout r = the radius of the hole for grouted anchors.
annulus around the bolt head respectively The areas A1 and A2 depend on the depth of spall, Z. This
ft = tensile strength of concrete has been derived by James et al5 as
fg = the tensile strength of the grout
Z = 1/2(u/ft – 1/ tanθ) dh tan2θ ...(6)
u = the shear strength of grout-concrete

April 2004 * The Indian Concrete Journal 3

 The above equation shows that, for a fixed spall angle, θ, hef = effective embedment length, mm
the spall depth is dependent only on the bond strength, the k = 15.5, for mean failure load of cast-in-place
tensile strength of concrete and the hole diameter. With headed fasteners
θ = 45o and using typical values of u/ft, from 3-5, the predicted
theoretical concrete spall depth should be between 1-1.5 times Uniform bond stress model
the diameter of drilled hole. The uniform bond stress model was developed to predict the
failure loads of adhesive anchors in uncracked concrete by
It has to be noted that the above equations are based on assuming a uniform bond stress throughout the embedment
the simplifying assumptions that the stresses on conical area length of the anchor system10,11,12. This model assumes that
A 1 are pure tensile stresses at failure. The presence of the failure surface could occur either at the steel-adhesive or
significant shear stresses on this spall cone will increase the adhesive-concrete interface. As the hole-to-anchor diameter
pull out resistance. ratio for adhesive anchors is close to unity, the nominal anchor
10,11,12
diameter can be used. Cook, et al showed that for
If the mechanism is assumed to be a simple one as shown adhesive anchors the uniform bond stress is product
in Fig 5, then the ultimate tensile capacity will be smallest of dependent and its value, τ, must be determined
the following values. experimentally. Grouted anchors can also develop failure
surfaces at the steel-grout or grout-concrete interface but the
Pu1 = ft A1 hole-to-anchor diameter ratio is generally larger than 1.5.
Pu2 = 0.87Fut As Therefore, the bond strength of each product should be
evaluated at both potential failure surfaces. The uniform bond
Pu3 = πDh Le τbu
stress model equation is as follows:
where,
Fut = the ultimate tensile strength of the anchor Nbond = ô ð d hef ...(8)
bolt
A s = the cross-sectional area of anchor
Nbond, do = ô o ð do hef ...(9)
L e = embedment length
Dh = diameter of grout hole and
where,
τbu = ultimate bond stress at the interface between
the grout and the surrounding concrete. Nbond = mean tensile strength for a steel-grout failure,
N
The quantities ft and A1 have already been defined. Tests Nbond,do = mean tensile strength for a grout-concrete
have demonstrated that the bond between grout and concrete failure, N
can be lost due to cracking.
t = uniform bond stress at the steel-grout
interface, MPa
Concrete capacity design method (CCD)
t0 = uniform bond stress at the grout-concrete
The CCD model evolved from a series of concrete cone interface, MPa
models that were developed for fasteners that were observed
to have full concrete cones at failure. These behavioural models d = diameter of the anchor, mm
assumed that the concrete failed in tension and that a full d 0 = diameter of the hole, mm
concrete cone formed from the embedded end of the anchor
hef = effective embedment length, mm
to the top of the concrete. There are several versions of the
concrete cone model, but the CCD method is widely accepted. Review of earlier experimental programme
The CCD method evolved from the Kappa-method and
To determine the strength and behavior of grouted anchors
predicts the ultimate load of an anchor loaded in tension or in
an extensive experimental programme with 229 tests on both
shear8. This method was developed for cast-in-place headed
headed and unheaded anchors installed using six cementitious
anchors and post-installed mechanical anchors installed in
and three polymer grouts was undertaken at the University
uncracked concrete that developed a full concrete cone at
of Florida, USA11,13. The investigation included parameters
failure. The CCD equation, used to predict the tensile capacity
typically encountered during design and installation including
of a single anchor installed in uncracked concrete, is as follows8:
binding agent (cementitious or polymer), anchor
configuration (headed or unheaded), anchor and hole
Ncone = k f ck hef1.5 (7) diameters, embedment depth, and concrete strength.

where, Test results for unheaded anchors

Ncone = mean tensile strength of concrete cone, N As hypothesised, the observed failure mode for unheaded
fck = concrete compressive strength (200 mm side anchors was a bond failure located at the steel-grout interface
cubes), N/mm2 with a secondary shallow cone. From all the test series

4 The Indian Concrete Journal * April 2004

Table 1: Steel-grout bond stress and coefficient of variation for represented by a uniform bond stress model calculated using
unheaded anchors the grout-concrete bond strength of the product, τo, applied
Product CA CB CC CD CE CF PA PB PC to the bonded area at the grout-concrete interface. This is
Average bond 20.5 21.8 7.3 21.1 21.6 15.9 17.8 19.4 17.8 given by:
stress, MPa
Coefficient 0.11 0.18 0.22 0.08 0.09 0.20 0.06 0.09 0.13 Nbond, do = τo π do he ...(10)
of variation
Note: Products starting with the letter ‘C’ are cementitious grouts and with the letter
For higher bond strength grouts, a full concrete breakout
‘P’ are polymer grouts failure occurs and the mean concrete breakout strength
developed by Fuchs et al8 is appropriate. This is given by 9:
evaluated in this test programme, only one test series (four
anchors) exhibited a failure mode at a location other than at Ncone = 15.5 fck hef1.5 ...(11)
the steel-grout-interface. This series was installed with product
CE and produced a failure mode at the grout-concrete
interface. Two other test series were performed using the The predicted mean strength of a headed grouted anchor
same product, but they developed a steel-grout failure mode. is determined by the lower value of equation (10) and
The only difference between these tests series was the equation (11).
dimensions of the anchor system. Therefore, transition from
one failure mode to another can be explained by observing A total of nine different products were tested which
that the anchors in the test series exhibiting bond failure at consisted of six cementitious grouts and three polymer
the grout-concrete interface were installed using large grouts. Test results showed that out of the 108 tests included
diameter anchors. This allowed the anchor system to develop in the headed grouted anchor test programme, 61 (56 percent)
the ultimate bond strength of the grout-concrete interface
before it could develop a steel-grout failure mode.

Product variability and anchor strength

Table 1 provides a summary of the steel-grout bond stress,
τ , and coefficient of variation for the products tested with
un-headed anchors. For the entire unheaded data set, the
mean bond stress was 18.4 MPa with a coefficient of variation
of 0.27. As shown in Table 1, the variation between all products
is greater than that within any individual product. This
indicates that the unheaded bond strength is product
14
dependent. (It is of interest to note that Lang reported an
ultimate bond strength of 8MPa for a concrete with
fck = 20 MPa for capsule type anchors utilising polyester resin
with a quartz sand filler. For high strength concrete the bond
strength was found to be 12 MPa or more).

Behavioural model comparison

The observed failure loads for all data sets as a function of
bonded area are illustrated in Fig 6(a). Failure loads shown in
Fig 6(a) were normalised to the mean steel-concrete bond
stress, τ, of 18.4 MPa by multiplying actual failure loads by
the factor 18.4/τproduct. The solid line shown in this figure
represents the mean value for the uniform bond stress model
based on the bonded area and τ = 18.4 MPa. Fig 6(a) shows a
linear relationship, indicating that the uniform bond stress
model is appropriate for unheaded grouted anchors. Also
shown in Fig 6(a) is a 5 percent fractile boundary based on a
coefficient of variation of 0.20 and a large database. Fig 6(a)
indicates that out of the 121 anchors tested, only 2 anchors
(1.6 percent) fall below this 5 percent fractile boundary line.

Test results for headed grouted anchors

For headed grouted anchors, two failure modes are possible.
For low bond strength grouts, bond failure at the grout-
concrete interface may occur. This failure mode can best be

April 2004 * The Indian Concrete Journal 5

Table 2: Grout-concrete bond stress and coefficient of variation Comparison of behavioural models
for headed anchors
The presence of two failure mechanisms in headed grouted
Product CA CB CC CD CF PA PB PC
anchor systems requires the use of two different behavioural
Average bond 10.2 7.8 4.8 9.1 8.4 7.9 7.7 11.1
stress, MPa models to predict behavior. For grouted anchors that
Coefficient 0.12 0.15 0.12 0.21 0.24 0.10 0.06 0.07 exhibited a concrete cone breakout, the failure loads were
of variation compared to the CCD method, equation (7). Anchors that
exhibited a grout-concrete bond failure were compared to
anchors developed a bond failure at the grout-concrete the uniform bond stress model for failure at the grout-
interface and 47 (44 percent) anchors developed a concrete concrete interface, equation (9). The results of these
tensile failure that resulted in a full concrete cone. This comparisons are presented in Figs 6(b) and 7.
confirms the assumption that headed grouted anchors can
develop either a concrete cone failure mode or a bond failure Fig 6(b) shows a graph of the tensile failure load versus
at the grout-concrete interface depending on the properties bonded area for all headed anchors that developed a bond
of the grout and the dimensions of the anchor system. failure at the grout-concrete interface. In Fig 6(b), actual failure
loads were normalised to τ0 = 8.3 MPa, which is the mean
value for τ0 for all tests that exhibited grout-concrete bond
Product variability and anchor strength failure. Fig 6(b) shows a linear relationship, indicating that
The strength of a headed grouted anchor system is dependent the uniform bond stress model for failure at the grout-
on the grout-concrete bond strength and the concrete cone concrete interface, equation (9) is appropriate for headed
breakout capacity of the concrete. Therefore, headed anchors grouted anchors that exhibit grout-concrete bond failure. Also
that produced a bond failure were analysed separately from shown in Fig 6(b) is a 5 percent fractile boundary based on a
those that developed a concrete cone. Table 2 illustrates the coefficient of variation of 0.20 and a large database. Fig 6(b)
average bond strength and corresponding coefficient of shows that out of the 59 anchors that exhibited the grout-
variation calculated for the different products that exhibited concrete failure mode, only 2 anchors (3.3 percent) fall below
bond failure at the grout-concrete interface. For all eight this 5 percent fractile boundary.
products tested with headed anchors and exhibiting grout-
concrete bond failure, the mean grout-concrete bond stress Fig 7 shows a comparison of the headed grouted anchor
was 8.1 MPa with a coefficient of variation of 0.30. This tests that were observed to develop a full concrete cone
indicates that the variation between all products is greater breakout failure to the CCD method, equation (7), as
than that within any individual product. Therefore, there is represented by the solid line in Fig 7. As shown by Fig 7, the
enough evidence to indicate that the unheaded bond strength test data typically fall above the solid line indicating a
is product dependent. conservative model. It is believed that the conservative results
indicate that the threaded rod used in the majority of the
The other failure mode observed in headed grouted headed anchor tests may have contributed to the increased
anchors was a full concrete cone failure. As indicated by capacity due to a combination of thread-grout interlock and
equation (7), the capacity of this failure mechanism depends bearing at the head of the anchor. A dashed line representing
on the strength of the concrete and embedment length of the the 5 percent fractile associated with the CCD method is also
anchor. The compressive strength of the concrete for the shown in Fig 7.
headed anchor tests ranged from 27.4 MPa to 63.7 MPa and
the embedment length ranged from 102 mm to 178 mm.
Recomended design model
Equations (8) to (11) provide predictions for the mean
strength of grouted fasteners. For design purposes, these
strengths must be reduced. For load and resistance factor
design, the determination of design strength from behavioral
models which represent mean strengths is typically based on
establishing a nominal strength (some lower bound fractile
of the mean strength) and then applying a capacity reduction
factor, φ, to limit the probability of failure.

In current US and European design standards, the nominal

strength is commonly taken as the lower 5 percent fractile of
the test data. The 5 percent fractile represents the value where
it would be expected that 95 percent of the tests performed
would exceed the specified nominal strength. The
determination of the 5 percent fractile depends on the number
of tests available and the scatter of the test results. The scatter
of the test results is typically expressed as the coefficient of
variation (V) which is defined as the standard deviation of

6 The Indian Concrete Journal * April 2004

the test results divided by the mean. This leads to the following
for nominal bond strengths:

τ’ = τ(1 – αV) ...(12)

τ’0 = τ0(1 – αV) ...(13)

The selection of the a factor depends on the number of

tests available.

The selection of an appropriate capacity reduction factor

(φ) for bond can be based on detailed studies of probability of
failure and/or on what φ factors are used for similar failure
modes in existing building codes. Bond failure can be
compared to shear-friction since it involves slip along an
interface. In ACI 318, the φ factor for shear-friction and shear
is 0.85. A capacity reduction factor. φ , for bond of 0.85 is
recommended for designs controlled by bond failure.

Various behavioral models for both edge effects and

group effects for grouted fasteners are being studied in both
the US and Europe.
• mix proportioning (primarily manually-mixed
products)
Factors influencing the strength of
bonded anchors • fire resistance

The evaluation of both the mean bond stress (τ and τo) and • wet (submerged) hole
design bond stress τ’ and τ’0) must be based on product • maximum torque
approval tests that include the effects of installation and in-
service conditions. As noted by Cook et al 14 there are • repeated load
significant differences between adhesive products. Basic tests
• seismic load
for mean bond stress in clean, dry holes at room temperature
indicate that the mean bond stress can range from 2 MPa to • environmental effects (chemicals)
20 MPa for adhesives and 7 MPa to 21 MPa for grouts. The
• cracked concrete (static cracks and moving cracks)
coefficient of variation for these tests can vary between 0.05
and 0.25. In many cases, products that exhibit high bond • other possible factors:
stress in clean, dry holes at room temperature are inadequate
• age of concrete
under typical installation and in-service conditions such as
damp holes and elevated temperatures. It is mandatory that • presence of oil (compressed air cleanout of holes)
designers require product testing for expected in-service and
• hole size
installation conditions prior to the final design.
• hole drilling
The following provides examples of the factors influencing
• UV radiation.
bond strength that need to be considered for product
approval tests of grouted fastener products12,15: As can be observed from the above list, a product
approval standard for bonded fasteners must be quite
• concrete mix (equal concrete strength does not ensure comprehensive to ensure reliable performance of products.
equal results)
• temperature effects Dynamic behaviour
• damp hole Rodriguez at al studied the dynamic behaviour of tensile
grouted anchors in concrete16. The effect of cracking and
• improperly cleaned hole dynamic loading on maximum load and displacements is
shown in Fig 8. They found that during crack opening, the
• curing time
crack propagated along the interface between the grout and
• freeze-thaw effects surrounding concrete in most of the tests. As a result, the
friction between the grout and concrete reduced considerably
• installation direction (vertical down, horizontal,
resulting in the pull out of the grout plug. Due to this effect,
overhead)
the average maximum capacity under static loading in
• creep (normal and elevated temperatures) cracked concrete decreased by 40 percent compared with the

April 2004 * The Indian Concrete Journal 7

test in uncracked concrete. Due to the reduced dynamic friction 7. CONES, M.A. Analysis of test on grouting of anchor bolts into hardened
coefficient between the grout and the concrete, the capacity concrete, Paper presented at the 1982 Annual American Concrete Institute
Convention held at Atlanta. Ga., USA, 1982.
still reduced under dynamic loading. The grout plugs were
8. FUCHS, W., ELIGEHAUSEN, R. and BREEN, J. E. Concrete capacity design (CCD)
observed to be pulled out of the cored holes, with little or
approach for fastening to concrete, ACI Structural Journal, January-February
even no damage to the surrounding concrete. Further study 1995, Vol 92, No 1, pp. 73-94.
is needed to see if behaviour would be improved in rough,
9. ______Building code requirement for structural concrete and commentary, ACI
drilled holes instead of smooth cored holes. 318M/318RM – 2002, Reported by ACI Committee 318, American Concrete
Institute, Farmington Hills, Michigan, USA.
Conclusions 10. COOK, R.A., KUNZ, J., FUCHS, W. and KONZ, R.C. Behavior and design of single
adhesive anchors under tensile load in uncracked concrete, ACI Structural
The behaviour of grouted anchors is dependent on the Journal, Vol 95, No 1, January-February 1998, pp. 9-26.
product and whether or not the anchor is unheaded or headed. 11. KUNZ, J., COOK, R. A., FUCHS, W. and SPIETH, H. Tragverhalten und Bemessung
von chemischen Befestigungen (Load Bearing Behavior and Design of Adhesive
For most engineered grout products, the behaviour of Anchors), Beton- und Stahlbetonbau, January 1998, Vol 93, H.1, S. 15-19,
unheaded grouted anchors can be predicted by the uniform February 1998, H. 2, S. 44-49.
bond stress model recommended for adhesive anchors, 12. COOK, R.A. and KONZ, R.C. Anchoring with bonded fasteners, International
Symposium on Connection between Steel and Concrete, University of Stuttgart,
equation (8). This model is based on a product’s bond strength,
Germany, September 9-12, 2001.
τ, at the steel-concrete interface. For products with a low
13. COOK, R.A., ZAMORA, N.A., and KONZ, R.C. Behaviour of grouted anchors,
grout-concrete bond stress, τ0, bond failure may occur at the
International Sympysium on Connection between Steel and Concrete, University
grout-concrete interface, equation (9). In general, product of Stuttgart, Germany, September 2001.
approval tests need to be developed to establish both the 14. LANG, G. Festigheitseigenschaften von verbundanker – systemen, Bauingenieur,
grout product’s steel-grout bond strength, τ, and grout- 1979, pp. 41-46.
concrete bond strength, τ 0. The controlling embedment 15. COOK, R.A. and KONZ, R. Factors influencing the bond strength of adhesive
strength can then be determined as the smaller of the strength anchors, AC1 Structural Journal, January -February 2001, Vol 98, No 1
controlled by steel-grout bond failure, equation (8) and grout- pp. 76-86.
concrete bond failure, equation (9). 16. RODRIGUEZ, M., LOTZE, D., GROSS, J.H., ZHANG, Y.-G., KLINGNER, R.E., and GRAVES,
H.L. III, Dynamic behavior of tensile anchors to concrete, ACI Structural
For headed grouted anchors, bond failure at the steel- Journal, July-August 2001, Vol 98, No 4, pp. 511-524.

grout interface is precluded by the presence of the anchor 17. ZAMORA, N.A., COOK, R.A., KONZ, R. and CONSOLAZIO, G.R. Behaviour and
design of single, headed and unheaded, grouted anchors, ACI Structural Journal,
head. For headed grouted anchors, embedment failure can
American Concrete Institute, March-April 2003, Vol 100, No 2, pp. 222-230.
occur by bond failure at the grout-concrete interface, equation
(9) or more likely by a full concrete cone breakout failure as
Dr N. Subramanian is the chief executive of
occurs with cast-in-place headed anchors, equation (7). For
Computer Design Consultants, Chennai. The
headed grouted anchors, the controlling embedment strength highlights of his professional career of 21 years include
should be determined as the smaller of that determined by designing several multi-storey concrete buildings,
grout-concrete bond strength, equation (9), or concrete cone steel towers, industrial buildings and space frames.
breakout strength, equation (7). He also worked with the Technical University of Berlin
and the Technical University of Bundeswehr, Munich
Due to dynamic loading, the friction between the grout for 2 years as Alexander von Humboldt Fellow. Dr N. Subramanian
and the concrete is reduced considerably and results in pullout has contributed more than 150 technical papers in seminars and
of the grout plug. Hence if the anchor will experience dynamic journals and published 17 wide selling books. He is a life fellow of
loading, it is preferable to reduce the static pull-out strength several professional bodies, including the American Society of
by 40 percent. Civil Engineers. He is the past vice president of the Indian Concrete
Institute and the Association of Consulting Civil Engineers (India).
He is the recipient of ACCE-Nagadi Award (for his book on Space
References Structures) and Tamilnadu Scientist Award.
1. ______Fastenings to concrete and masonry structures – State of the art report,
Comite Euro – International du Beton, Thomas Telford, U.K., 1994, pp. 249. Prof Ronald A. Cook has been involved with
anchorage to concrete for over 25 years. Dr Cook is a
2. SUBRAMANIAN, N. and VASANTHI, V. Design of anchor bolts in concrete, The
Bridge and Structural Engineer, September 1991, Vol XXI, No 3, pp. 47-73.
professor of civil engineering at the University of
Florida, USA. He has conducted several research
3. SUBRAMANIAN, N. Recent developments in the design of anchor bolts, The Indian projects related to anchorage to concrete, primarily
Concrete Journal, July 2000, Vol 74, No 7, pp. 407-414. in the area of adhesive and grouted anchors. Dr Cook
is chairman of ACI 355 Anchorage to Concrete and a
4. SUBRAMANIAN, N. and COOK R.A. Installation, behaviour and design of bonded
member of ACI 318 Subcommittee B Reinforcement and
anchors, The Indian Concrete Journal, January 2002, Vol 76, No 1, pp. 1-10.
Development and ACI 349 Concrete Nuclear Structures. He is
5. JAMES, R.W., GUARDIA, C.D. and MCCREARY, C.R. Strength of epoxy grouted also a member of ASTM Committee E06.13 Structural
anchor bolts in concrete, Journal of Structural Engineering, ASCE, December Performance of Connections in Building Construction, the
1987, Vol 113, No 12, pp. 2365-2381. International Association of Bridge and Structural Engineers
Association Working Commission on Concrete and the Federation
6. MAILHOT, G., et al In-place concrete strength – New pull-out methods, ACI
Journal, 1979, pp. 1267-1283. International du Beton Task Group on Fastenings to Concrete and
Masonry Structures.
•••

8 The Indian Concrete Journal * April 2004