6.

3RD INTERNATIONAL CONFERENCE ON

GROUTING AND GROUT TREATMENT

GROUTING OF ROCK MASSES

Invited key note lecture by

Giovanni Lombardi Dr Eng. PhD h.c.
(with additional comments)

Our Ref.: 102.1-R-147A

Minusio, October 2002

Complements March 2003

102.1-R-147A

Grouting of Rock Masses

Giovanni Lombardi1

Abstract
The grouting of rock used to be a quite empirical constructional technique aimed to
improve somewhat the physical properties of the masses, which are actually part of
the civil engineering project.
Different "grouting schools" or "ways of thinking" did develop and a considerable
number of recipes were dictated by "great old men" and enforced later on by regulations of any kind.
Only in the last decades the grouting process started to be studied in a more scientific
way and a number of well-founded results are now on hand.
Nevertheless, improvements are still possible and even required so to be put into
practice in order to optimise the grouting process in adapting it better to the actual
rock conditions as well as to the real objectives of the project.
Introduction
The theme I was asked to deal with was quite simply "Rock grouting". It covers a
very wide field of theoretical and practical aspects so that some limitations had to be
considered.
In addition, a certain number of concepts are understood differently by the various
authors who treat this subject. A definition of a few fundamental notions are thus
likely to be necessary in order to avoid at least some of the usual misunderstandings.
These differences are due to the fact that during the last two centuries different
"school of grouting" did develop and drifted apart each from another mainly due to
historical reasons.

1 Giovanni Lombardi, Dr Eng. PhD.h.c., President Lombardi Engineering Ltd., Via

R. Simen 19, 6648 Minusio (Switzerland), tel. +41-91-744 60 30, fax +41-91743 97 37, info@lombardi.ch, www.lombardi.ch
-1-

102.1-R-147A

In spite of the more scientific recent achievements, it appears, that research, studies
and new improvements of theory and practice are still necessary and desirable to better fit the needs of the construction industry in the fields of foundation, rock mass
stabilisation, tunnelling and construction of water retaining structures.
Due to personal experience, emphasis will be given to consolidations of foundations
and to grout curtains for dams. For time and space limitations, the question of the optimal orientation of the boreholes in relation to the geological features, nor the drilling and grouting equipments will be dealt with in the following presentation.
The theme announced should thus be cut down and changed into "Cement grouting
of jointed sound rock masses for dam foundations", which concept will be defined in
more details later on.
The jointed sound rock mass
The first delimitation will be obviously against granular or fine soils, which present a
high percentage of voids and thus a high deformability. They require the use of special techniques different from the ones we will deal with hereafter.
But, also highly weathered rock masses like for example hydro-thermally altered
granite completely or partially turned to sand, or decomposed dykes, or completely
crushed zones in faults as well as highly porous rock lay outside of the field of our
discussion.
Karstic phenomena in sound rock require equally a special treatment as for example
a filling of the cavities with mortar, concrete or other matters before the very grouting of the zone can take place.
On the other hand absolute massive unjointed rock masses cannot, nor need to be
grouted.
In between these limits there are a great number of real cases of rocks, which are
sound in the matrix - that means that they possess an appreciable ultimate compressive strength as well as noticeable deformation moduli - but are subdivided by strata,
joints, cracks, fissures, faults or by any kind of discontinuities.
Indeed, these rock masses are more like a pile of blocks, which do fit perfectly each
to another; their interfaces presenting properties of friction, cohesion and viscosity.
Occasionally, some spotwise continuity of the rock across the joints can be observed.
A special kind of discontinuities needs a particular mention, which may be described
as "potential joints". They are in fact weakness surfaces in layered or schistous rocks,
which may open so easily that one doesn't know whether they were open from the
beginning or whether they just did open during the grouting process. "Hydrofracturing" is then called in, of which we will talk about later on.
In the aforesaid conditions of a jointed, sound rock the net of discontinuity surfaces
plays the most important role in the grouting process, or is even the only aspect to be
seriously taken into account. So one should rather talk of "grouting the discontinui-

-2-

102.1-R-147A

ties " than the rock. This explains also the fact that quite similar techniques can be
used to repair cracks in concrete blocks (Turcotte 1994).
The "opening" of the joints, the nature of their surfaces, their stiffness, and frequency, their continuity as well as the interconnection between them represent a system - that means a complex of elements - which the grouting process must correctly
take into account.
The RQD (Rock Quality Designation) (Deere 1968) was a first quite original and
useful attempt to quantify the properties of the fissured rock masses, which depend
on said discontinuities.
The empirical relationship between deformability and RQD is shown in Figure 1.

Figure 1: Relationship between RQD and a deformability factor defined

as the ratio: modulus of deformability Ed to modulus of elasticity ET 50 (Coon and Merritt 1970).
The FES-Rock mass model
To correctly understand the grouting process, a model for the rock mass is thus necessary (e.g. Hssler 1992).
In recent times a number of such models were set up, but the so-called FES (Fissured, Elastic. Saturated) rock mass model appears to be still very useful (Lombardi,
1989, 1992)2 to solve problems in different fields of rock mechanics (settlement,
grouting, tunnelling, etc.).
For a simple uniaxial compression of a sample of dry rock loaded perpendicularly to
the joints the relationship shown by Figure 2 holds quite convincing.
2 The FES Model was first set up and developed to solve the problems of the settlements occurred in 1978 at the Zeuzier arch dam in Switzerland due to the drainaging effect of an exploratory adit.
-3-

102.1-R-147A

Figure 2: Strain stress relation for a dry fissured rock sample loaded perpendicularly to the joints.
OC : Elastic deformation of the matrix with E* =
OD
OF
OA
AB
0

E (1 )
1 2 2

: No tension (joint opening)

: Initial compression modulus E0 for RQD = 50. (Figure 1)
: Actual progressive closing of the joints ( = closing rate)
: Possible elastic deformation after complete closure of the joints.
: Total closing of the joints in contact (or possible joint aperture
until loosing contact) referred to the height of the sample (h).

If an interstitial water pressure is considered, the FES model can be completed as

shown by Figure 3. A point P on the graph corresponds to a given total stress, to a
certain interstitial water pressure and to a degree of opening of the joints.
This simple representation considers an uniaxial homogeneous loading case, but,
meanwhile the method has been developed to a complete 3-D Finite Elements Model
(including the desired number of joint systems of any orientation and properties) apt
to solve all the problems, which may be of interest in Civil Engineering. However,
there is no need for our considerations to enter in more details.
Grouting matters
A number of matters can be used to consolidate jointed rock masses.
Among them: cements, with or without adjunction of ashes, fillers and silica fumes,
resins, silicates and at least theoretically many other chemicals may be mentioned.
However, in the following only cementitious grouts will be discussed because they
represent the most popular and economical way to improve the properties of the rock
masses.
Different types of cement can be considered, which may differ one from another
mainly in matter of granulometry and chemical resistance. Also the addition of dif-4-

102.1-R-147A

ferent kind of chemical admixtures can be taken into account to improve the properties of the slurry or of the hardened grout.
To mention is finally the possible combination of different products in the same project.

Figure 3: Example of a FES model for a given rock mass.

=strain; t=total stress; p=neutral water pressure; =degree of closure
( =1 fissures completely closed, =0 fissures completely open):
A ( 0, 0)= point of total closure at nil water pressure; e=effective
stress; P( , t, p, e, a) means: general strain, total stress, neutral pressure, effective stress and closing ratio at point P.

Cement grout
Definition of the water-cement ratio
Essentially the cement grout, or slurry, is a mixture of cement and water.
It is a well established rule in the concrete technology to define the ratio W/C that is
"water to cement" by weight (e.g. W/C=0.5: means 0.5 kg water added to 1.0 kg cement), thus considering the cement as the base of the mix. Strange enough, in the
field of grouting the habits are quite unstable. They refer to this ratio, but also to its
inverse, from time to time they use the weight but also the volume of the components.
It is felt that a conformity with the concrete technology should be enforced by any
mean to avoid additional confusions and that only the W/C ratio by weight should be
used, as shown in Figure 4 (Deere 1982, Houlsby 1982).
In the following only said water/cement ratio will apply.
-5-

102.1-R-147A

Figure 4: Definition of the water to cement ratio (W/C).

The role of water
The discussion about the type of slurry to use: "thick or thin", is quite old (Deere,
Lombardi 1985).
Fundamentally the observed discrepancies in opinion are due to the various roles the
water is supposed to play during the grouting process:
The first role of the water is to hydrate the cement, and there is obviously no discussion on the necessity of this first amount of water (let say W/C=0.3 to 0.4);
A second role, achieved by an additional quantity of water, is to produce a fluid
mix with reduced cohesion and viscosity so that the slurry can propagate into the
joints of the rock mass. This function can however be taken over, at least partially,
by some admixtures like plasticizers. An additional of, let say, W/C=0.3 is nevertheless necessary and quite normal (depending on the Blaine Value).
The scope of this additional water amount is, in fact, to avoid the direct contact
between the single cement particles and thus the building up of an internal friction
in the slurry. This friction would cause an exponentially increasing pressure along
the joint so to stop immediately the progression of the slurry (Lombardi 1985).
A third amount of water in the mix is intended to compensate for possible water
losses during the grouting process due, for example, to water absorption by a dry
rock. It is not possible to estimate the quantity of water required for this function.
However, this risk does not exist below the ground water table and can be eliminated above it in saturating adequately the rock mass just before the grouting
starts.
An additional role attributed to water is that of a "joint opener" to facilitate the
penetration of the slurry in thin cracks. But, this role can again be taken over by
the existing ground water or the saturation water, which will automatically be
pressed into the joints in front of the grout itself, without having to be included in
the slurry and thus without affecting its properties.
-6-

102.1-R-147A

When using a normal Portland cement a total water-cement ratio of the order of 0.6
to 0.7 is a practical minimum but also a quite adequate value in a great majority of
cases. See Table 1. For micro-fine cements however a higher W/C ratio is required,
e.g. up to 1.0 or 1.2 (Bremen 1997).
Dam

W/C

Fluidifier

Paute (Ecuador) upper part

Alicur (Argentina)
El Cajn (Honduras)
Clyde dam (New Zealand) 2nd part
El Chocn (Argentina) repairs
Sir (Turkey)

0.6
0.67
0.7
0.6
1.0
0.7
or 1.0
0.59
0.7
0.7
1.0
0.63

Intraplast 1.4%
Intraplast 1.2%
Bentonite 0.2%
Intraplast ~ 1%
Bentonite 0.5%
Puzolanic cement, Mistra 1%
Bentonite 1.2%
Cement + ash. Conplast 1.5%
various
Rheobuild/Viscocrete 0.7-0.8%
Bentonite 2%
Intraplast 1%

Katze (Lesotho)
Pichi Picn Leufu (Argentina)
Potrerillos (Argentina)
At Hamou (Morocco)
Casecnan lower tunnel (Philippines)
Table 1:

Few examples of unique thick mixes used in a number of

important dams and tunnels (sometimes: mainly used mix).

In spite of these considerations, there is still a historical concept often referred to,
which considers the water as being mainly a vehicle for transporting the cement grains
and introducing them into the joints of the rock mass (also for washing them in).
It is also often believed that the penetration of the grout will be the higher, the more
water is added. However, a difference must be made between useless penetration of
water and desired penetration of cement grains into the joints.
Thin to thick mixes
Said historical concept led to develop the recipe of progressive thickening of the
grout mixes. A number of quite personal rules were developed. They are of the kind:
"you inject 200 litres of a stained water of the type W/C=20, then 200 litres of coloured water of the type W/C=10, then again 200 litres of a dirty water with a ratio
W/C=5, and so on, until you possibly reach, by chance, a real grout mix of the order
of more or less W/C=1".
The basic idea is to try to start the grouting with something like water of very low
yield stress and to increase its cohesion step by step using liquids of higher and
higher viscosity until a thick mix is possibly injected in the joints.
In the background of these recipes there is the mentioned idea that thinner mixes will
enter into finer cracks better than thicker can do.
The main difficulty with this, let say "classical", procedure is firstly that one doesn't
know from the beginning when the process will end and thus which mix will be the
last one and consequently which will be the quality of the works carried out.

-7-

102.1-R-147A

In fact any stage will be grouted with a mix of different quality.

Additionally, the mixing of the different slurries in the joints is quite erratic and absolutely unpredictable.
So, the writer could observe, for example, in an adit excavated in an already
"grouted" rock mass, that the lower parts of the vertical joints were filled with a
white flour formed by loose hydrated cement grains, while their upper parts were
simply void. Only "coloured water" had been used to grout the rock.
As proved many times, the final strength and the chemical resistance (durability) of
the mix harden in the joints decreases sharply with its water content, that is with the
excess of water introduced in the slurry (Houslby 1982).
Also it is not true that the new thicker mix will wash away the thinner ones and replace them.
As shown by Figure 5, the velocity of penetration of the grout into a joint is not uniform. It is higher in the centre and reduced, even to nil, along the joint walls. Consequently, at the end of the grouting process the walls of the joints will be covered by
weak, more or less white matter corresponding to the thin mixes, while the core of
the joint will possibly be filled by a stronger grey cement body, resulting from the
hardening of a thicker mix.

Figure 5: Velocity distribution of the grout in a joint.

1) Newtonian fluid (e.g. water);
2) Binghamian body (e.g. cement slurry)
2.e = stiff kernel.
The non-uniform distribution of velocity makes that the former slurry
sticks on the walls and the following one progresses further in the
centre.

This layering of the grout may be often observed in boreholes cores drilled in a
grouted rock mass.
Unfortunately, the mechanical and chemical properties of this "sandwich" are dictated by the outer weaker white layers, corresponding to the thinnest mix used. Obviously, some additional mixing of the different slurries may also take place, so the
situation is not always that clear. (Also the presence of ground water or saturation
water may produce such a layering of the grout set.)

-8-

102.1-R-147A

In matter of penetration of the grout into fine joints, one may observe the Figure 6
and see that there is an illusion to believe that thinner mixes will enter the joints better than the thicker do. In fact, the diameter of the grains and the clusters they may
form in relation to the aperture of the joints are determinant, not the amount of water,
which will have to flow around the grains and to continue its way in entering the
joints. There is thus something like an inverse "filter criterium" (to the one used for
fill dams) to be taken into account.

Borehole under grout pressure P

Actual joint
"Potential" joints
Cement grains filter
Water
"Presso-filtration"
Hydro-jacking
Possible hydro-fracturing due to
water pressure
9 Possible hydro-fracturing due to
tensile stresses
1
2
3
4
5
6
7
8

Figure 6: Penetration of a thin grout under pressure in a joint.

The theory of "presso-filtration" developed decades ago in order to explain this process and to try to find some assumed advantages in it, is apparently outdated.
In fact, to ease the penetration of the grout from the borehole into the joints there are
a number of possibilities available as:
adding a plasticizer to the grout to avoid the forming of clusters of cement grains
due mainly to electrical forces;
using finer cement (if economically justified);
using higher grouting pressures, and
counting, at least to some extent, on the pressurized water in front of a thick mix
to open the joints and thus the way to the cement grains.
In conclusion, there are a number of good reasons, to abandon the so-called classical
cumbersome technique of "thin to thick slurries series" and to go somewhat more
modern grouting ways.

-9-

102.1-R-147A

One of the main drawbacks with thin mixes is that they are not stable. Their progression in the joints is unpredictable (Figure 7). Also they will shrink more at setting
than thicker mixes do.

Figure 7: Experimental grouting tests in a joint of constant thickness (by

Deere).
Stable mixes
The reasons to use thick mixes, called "stable mixes" (with limited bleeding, e.g. less
than 5% in two hours) are numerous and related to the final properties of the grout mix
after setting in the rock cracks, that is the quality of the grouting works carried out.
The main advantages over thin mixes are:
complete filling of voids and joints by cement avoiding bubbles due to an excess
of water;
high mechanical strength (like in the concrete technology);
reduced shrinkage potential so to avoid, or at least to limit, micro-cracks in the
hardened grout;
good or better bound to the rock surfaces (due also partially to the higher grouting
pressures required);
high resistance against chemical leaching out (Houlsby 1982);
predictability of the grouting process.
Due to the good results obtained, at many dam sites, for the grout set in the cracks, it
appears that stable grouts should be preferred. In order to simplify the process only
one type of slurry should be used for any single project: the most adequate one, or
the "best" one, which by definition is unique3.

3 At the Paute Dam (Ecuador) the lower part of the grout curtain was grouted follow-

ing the classical rules with thin to thick mixes. In the upper part only thick mixes
were used. The grout cement taken per metre borehole was about the same; the
amount of water uselessly injected in the lower part of the grout curtain was obviously much higher, leading to a weaker grout when set.
- 10 -

102.1-R-147A

Properties of the slurry

To judge a slurry, two groups of properties must be considered, which are obviously
not independent each from another.
The first group refers to the fresh grout, which is indeed a "suspension" of grains in
water and is supposed to follow the Bingham's body law, while water is a Newtonian
body (Figure 8).

Figure 8: Rheology of a cement slurry vs water.

= shear stress,

dv
= shear velocity
dx

1 Newtonian body (viscosity only): (water)

dv
dx

2 Binghamian body (cohesion and viscosity):

(approximation for grout mix)

=c+

dv
dx

- - - - real stable mix; c = cohesion (yield point)

=dynamic viscosity; B=plastic viscosity; 1=apparent viscosity

The first group includes mainly

density;
bleeding;
viscosity;
cohesion (yield point);
set time.
The second group of properties concerns essentially the hardened grout and refers to
mechanical strength,
resistance to chemical agents, and to
permeability.

- 11 -

102.1-R-147A

The rheological properties of the fresh grout, which are mostly of interest and even
decisive for the actual grouting process, can be influenced, modified or even determined in using a number of admixtures available today on the market.
Just to recall an already quite old example, Figure 9 shows the influence the addition
of water, bentonite or of a chemical admixture may have both on the final mechanical strength and on the apparent viscosity of the fresh slurry, or better said, on the
flow time from a Marsh Funnel.
Starting from a W/C ratio of 0.67 (Point A) the addition of the Intraplast plasticizer
results in a sharp decrease of the flow time but also of the mechanical strength.

Figure 9: 28-day strength vs Marsh Funnel flow time (Am).

When using bentonite (B/W=1%) and starting again from a W/C ratio of 0.67 (Point
B1) the increase of this ratio has a similar effect tending to the viscosity and the
"strength" of pure water, which is nil.
If a 2% B/W ratio is used, the reduction of the strength is quite the same but the flow
time remains higher.
Consequently, the effect of added bentonite is just the contrary of that of lubrification
it is often believed to be (Bremen 1997). Indeed, adding bentonite to a cement slurry
is an indirect way to have the excess of water sought out of the mix4.
For a given required final strength of the hardened grout, a much greater reduction of
viscosity is obtained by the addition of an adequate plasticizer than of bentonite.
In the same way, a higher final strength is obtained for a given required Marsh flowtime.

4 However, it may be the intention of the designer to get a cheap mix just apt to fill

voids in the rock mass, while mechanical and chemical weaknesses of the result
are not main concerns (Bremen 2001).
- 12 -

102.1-R-147A

In conclusion, the required properties of the fresh mix to allow an easy penetration,
which are mainly: cohesion, viscosity and set time can be obtained using adequate
additives without jeopardising the final properties of the slurry and thus the quality of
the grouting works carried out, as it happens when the W/C ratio is increased.
Obviously, as mentioned here-above, a minimum water content of the slurry is always required to avoid the building-up of an internal friction. An optimal balance between the two groups of properties must thus be set up.
The objectives of grouting
Main aspects
The principle of grouting is to fill the open voids existing in a rock mass in introducing, by pressure through boreholes, a certain amount of a "liquid" matter, in fact a
suspension, that will harden later on. The properties of the grouted rock complex
should be modified in the desired way.
The main expected improvements are well known. They are:
reducing the permeability of the rock mass;
reducing its deformability, and
increasing its strength especially against shearing forces.
The relative importance of these three objectives of the grouting works depends obviously on the type of construction dealt with.
However, there are also other aspects to be considered like:
the feasibility;
the expansion of the rock mass produced by the grouting process;
the durability of the expected beneficious effects;
the economics of the treatment including the costs and the time required to carry it
out.
Limitations to grouting
There are a number of conditions and circumstances, which may prevent the grouting
process, except some special measures are taken.
So flowing water may require the adjunction of some anti-wash additives. Also too
low temperatures may be of harm.
A very high sensibility of existing buildings or structures (e.g. underground steel
lined pressure shafts or a system of drains) may make grouting work inadvisable.
Considerations of different kinds related to the environment may not permit the use
of certain types of grout.
The nearby presence of springs used for drinking water may preclude the use of any
kind of grout and might impose a temporary freezing of the groundwater instead of a
grouting of the rock mass for example during the time period, an underground construction is carried out.
- 13 -

102.1-R-147A

Expansion of the rock mass

Another aspect of the grouting is sometimes ignored. This is the expansion inevitably
suffered by any rock mass when grouted, as grouting is done by definition in using
pressures, which will and must open the joints. This expansion may produce a heaving of the ground, which is often limited by some specification, but is in fact a prerequisite for any grouting (Figure 10). The problem is then to limit the heave should it be of harm - but in no way to avoid it completely.

Figure 10: Unavoidable heave of the ground surface above a grouted zone.
There is however a further aspect of this expansion process, which is seldom taken
into consideration. In some cases a successful grouting is not feasible because such
an expansion is not possible. As an extreme example let us consider the following
situation (Figure 11), which may be encountered in some fractured zones in a massive very strong rock.

Figure 11: Due to the restraint by the stiff rock only few of the fine joints
can be grouted.
grout penetration

- 14 -

102.1-R-147A

A heavily jointed rock zone may be strongly fitted in between two massive blocks.
Suppose a frequency of 10 thin or flat joints per meter. In using a certain cement
grain size, the joints should be opened by 0.3 mm each to be grouted and correctly
filled with cement. This would result in an expansion of the rock mass of 3/1000 in
the corresponding direction. If the modulus of deformability of this mass is 20 GPa
the grouting would lead to a final transversal compressive stress in the rock mass of
60 MPa to achieve an uniform opening of all the joints, as well as an even higher
grouting pressure.
As such an extremely high pressure is practically not feasible, only few joints out of
the entire series can be opened and thus grouted. If a grouting pressure as high as
6 MPa is used only 1 joint out of 10 might theoretically be grouted.
The real situation is even worse because the joints are not absolutely identical, so one
of them will open first and wider than necessary, while the others will tend to close
and can no longer be grouted (Lugeon 1933).
Consequently, the objectives of the grout process can not be reached and a factual
limitation to grouting must be accounted for.
To solve the problem finer cement or chemical products might possibly be used instead of normal cement.
Durability of the treatment
It was observed that some grout curtains for dams had practically disappeared after a
number of decades (Houlsby 1982). The too thin grout mixes used were washed out
by leaking water. By the way, this fact could be considered, in some cases, as a proof
that the curtains were not necessary at all!
The attack of the set cement grout requires two conditions:
a weak material, weakly bounded to the joint walls, but also
the possibility for the water to seep along the grouted joints;
this last condition is fulfilled when grouting pressures lower than the final water
pressure were used or when the grout could shrink enough to open a way for the water along the surface of the joints, or if voids left by bled water from the grout did exist.
Obviously, this phenomenon may not always be of concern, for example if only a
temporary tightening of rock masses is required during the construction of permanent
structures. In many cases however, the necessity of repeating the grouting works after a number of years may have heavy consequences on the economy of the project.
These aspects should be adequately considered by the designer of the grouting
works, while it is recognized that the problem is not easy to be quantified and that often only an engineering judgement and experience help to solve it.

- 15 -

102.1-R-147A

Water pressure tests

It is customary, since almost one century to use water pressure tests to evaluate the
permeability of a rock mass before and after its treatment by grouting. Among them
the Lugeon test is well known and widely used (Lugeon 1933). Obviously, it furnishes only a quite rough criteria but it is, at same time, a quite simple and useful
testing procedure.
However, some too simplistic assumptions regarding these tests as well as old grouting habits justify some scepticism. For example, a rule of thumb was in use, which
stated that a grout curtain is necessary where the Lugeon values is higher that a certain limit; this regardless of the kind of dam, its height and the type of foundation
which are concerned nor the depth below ground where the test is carried out.
Another quite misleading interpretation of these tests is the temptation to set-up a statistical relationship between the Lugeon values and the volume of take to be expected
at grouting stage.
Indeed, the flow rate of water is put in relation with the grout volume disregarding
also the fact that water is a Newtonian body, while the cement slurry is a suspension
of grains of a certain size that follows, at least approximately, the Binghamian body
laws.
A high frequency of fine joints may give the same Lugeon value as an unique wide
crack. In this last case the grout take can be very high, while in the first one no cement at all can enter the fine joints.
In such theoretical exercises on statistical correlations, the real conditions of the
grouting process are often ignored.
In fact, the experience shows that water pressure tests may, at the best, give an approximate indication of the reduction of permeability obtained by a grouting work,
but are practically useless to define the grout take to be expected, nor to give a correct indication even on the groutability itself of the rock mass.
Indeed the degree of groutability can only be defined by grouting tests.
It is thus felt that water tests carried out at every grouting stage do simply represent a
systematic waste of money, without any beneficial effect on the grouting process itself. Even more they may have an unfavourable effect because they may cause damages in re-opening already grouted joints.
Hydro-jacking and hydro-fracturing
Two concepts, which are often imprecisely understood, are "hydro-jacking" and "hydro-fracturing". This confusion may lead to wrong decisions on the grouting site.
It is believed that hydro-jacking should refer to the opening by the grout of preexisting joints in the rock mass, while hydro-fracturing describes the forming of new
cracks due to an increased pressure of water or slurry.

- 16 -

102.1-R-147A

In fact, hydro-fracturing is a relatively seldom event during normal civil engineering

grouting works. Not every one of the observed drops in the pressure during grouting
can be interpreted as a hydro-fractioning event. Indeed, pre-existing joints may also
open suddenly. According to Figure 12, a kind of elastic instability takes place.
In the current practice, true hydro-fracturing is mainly related to the already mentioned "potential joints" like weak inter-stratigraphic or beddings planes.
The opening of such planes is due to the tensile stresses induced in the compact rock
by the nearby pressure of grout or of water.

a) sudden extension of the open stretch

of a fissure at grouting
b) spreading out of pressure
To a) and b):
1 before

hydro-jacking
2 after
q flow rate of the grout
p pressure
R Reach from the borehole

Figure 12: Hydro-jacking or claquage as a kind of "elastic instability" or

bifurcation.
This happens more frequently when the borehole is parallel to the planes of weakness, while hydro-jacking do happen more or less in the same way regardless of the
angle between the borehole and the joint (see Figure 6).
Combinations of hydro-jacking and of hydro-fracturing are also possible as shown on
the same figure. However, the risk is clearly higher when water or a thin mix is
pressed into the fine "potential" joints where a thick mix can hardly penetrate.
The main question is however whether a hydro-fracturing is always of harm or not,
while the hydro-jacking is fundamentally the expression of an effective grouting,
which does more than just fill - at no pressure - opens voids in the rock mass.
If a strong slurry with good bounding properties to the rock is used, a hydrofracturing is seldom harmful from the technical point of view; except it produces an
excessive heave at shallow depth. But, the possible useless waste of grout and thus
the related additional costs due to important hydro-fracturing events causes unfavourable economical consequences.
The same can obviously be said of any excessive hydro-jacking of existing joints.
The often used French term of "claquage" may apply to both cases and refers mainly
to the sudden opening of a new way for the grout, which will concentrate along certain surfaces as interpreted by Figure 12.

- 17 -

102.1-R-147A

Hydro-jacking and hydro-fracturing are both related to splitting forces across the
joint considered.
These forces are obviously the integral of the pressures acting on any single elements
of the joint surface. They are thus function of the pressure applied in the gout hole,
but also of the extension of the surface submitted to pressure. It can be assumed, as
an acceptable approximation, that said surface is in some way related to the volume
of grout already pressed in, obviously only as long it has not yet set.
The penetration of grout
The process of the penetration of grout under pressure into the rock joints depends on
a number of factors, the first one being the geometry of the joint walls, that means
their shape and variations in their opening, their extension and the interconnections
between them. Very complicated cases can be and were analysed from a theoretical
point of view. Nevertheless, the most important aspects of the grouting process can
be studied in a simple but quite reliable way on the basis of very simple models, like
open flat joints of constant thickness. Figure 13 (Lombardi 1985) shows the theoretical model, which gives the relationship between grouting pressure, opening of the
joints, cohesion (or yield point) of the slurry and maximal reach of the grout.
Accordingly, the higher the pressure, the wider the joint and the lower the cohesion,
the greater will obviously be the reach. More refined geometrical models may represent the reality somewhat better, but cannot change significantly this relationship.
Additionally, the reality will always be different from any model chosen and is also
different from spot to spot of the same rock mass.

Figure 13: Pressure distribution and forces in a two-joint system, during

grouting.
B pressurised borehole. 1 Joint closes; 2 Joints opens;
p = pressure; c = cohesion; 2t = opening; R = reach;
F1, F2 = splitting forces.

- 18 -

102.1-R-147A

The opening of any single joint - no matter how such opening may be defined - is
different, so that the grout will penetrate more easily in some of them and more
hardly in others, as already said.
From the practical point of view this means, roughly speaking, that any grouting
stage will fill mostly, or at least to a greater distance, only the main, wider, not already grouted joints, while the thinner ones will have to be grouted later on. This
leads to classical grouting procedures in steps with various successive series of splitspacing boreholes and possibly with increased grouting pressures from series to series.
The possible closing of the thinner joints due to the expansion of the main joints may
just be recalled at this place.
Also the relationship between grain size and joint opening has to be taken again into
account. Due to this fact, a differentiation must be done not only between wider and
finer joints, but also between different spots of the same joint, which obviously does
not show a constant opening along its surface. In the thinnest part of a joint only water will enter, while the cement paste will stay in the wider spots, and follow preferential "channels", along the joint (Figure 14). The beneficial effect of hydro-jacking
is that the joint will open and the cement can then enter any spot of the joint, increasing thus the proportion of the surface, which will be really bounded.
The mentioned relationship between, joint opening, cohesion of the slurry, grouting
pressure and reach does apply obviously only to stable mixes, where no excess water
exists, which may separate from cement.

Figure 14: Zones around contact areas ungrouted or poorly grouted because the local aperture is too small in relation to the grain size
of the cement used.
1 contact areas;
2 ungrouted or poorly grouted zones;
3 well grouted zones; - - - - main flow-lines of the grout.

- 19 -

102.1-R-147A

As already shown in Figure 13 the pressure of the grout will decrease from the borehole away, so that the extension of the pressurised surface will be limited and the average pressure acting on it can be estimated to be about only one third of the grouting
pressures applied.
The fear of heaving the rock surface is thus strongly reduced - at least by three times
- respect to the usual formulation, which limits the grouting pressure to the simple
weight of the overburden disregarding thus the mentioned fast decrease of the pressure with the distance from the borehole. This pressure drop is sharper and thus the
reach smaller the higher the cohesion of the slurry is. Consequently, at shallow
depths higher pressures as usually assumed may be applied.
The requirement to limit the heave, that is to reduce the pressure and to increase the
cohesion, is just contradictory with the requirement of a good penetration that means
high grouting pressures as well as low viscosity and cohesion of the slurry. Therefore
these limits should be placed as high as possible.
The grouting intensity
The grouting pressures should be as high as possible to increase the reach of the
grout, but low enough to avoid hydro-fracturing.
A way out of this dilemma appears to be shown by the concept of "grouting intensity". It is known by experience that limiting only the grouting pressure and the grout
take is not sufficient to avoid or even only to reduce the risk of hydro-jacking and
hydro-fracturing and thus a possible waste of grout.
The El Chocn dam in Argentina required important rehabilitation works because of
the internal erosion of the clay core, which had taken place (Aisiks E. 1991 and
Vard O. 1991). It was necessary to consolidate the rock below and laterally of the
core, but also the contact surface between rock and core, as well as the core itself.
The works had to be carried out at almost full reservoir, as a drawdown was not possible. The risks involved in these works were high. A hydro-fracturing of the rock
near the dam foundation or of the core itself had to be absolutely avoided.
Based on the consideration presented here-above the notion of "grouting intensity"
was introduced, about twenty years ago to solve said problem, and developed in the
following years.
This parameter is simply the product of the grouting pressure by the volume taken
any time the grouting process is stopped and the grout flow rate is nil.
As it may be seen on Figure 15, this intensity represents a rough estimate of the Energy pressed into the rock mass. If the processes were absolutely elastic - this means
linear - and when the internal energy losses due to the viscosity of the grout, the slippages of the grout along the walls as well as to the inelastic deformations of the rock
mass are disregarded, the intensity would be the double of the elastic energy introduced by the pumps and theoretically accumulated in the rock mass.
In the real grouting process the conditions are significantly more complex, but nevertheless the intensity, as previously defined, proved to be a quite useful tool to manage the grouting process.

- 20 -

102.1-R-147A

Figure 15: The grouting process

quired or by reaching
situation as "refusal"
GIN, represented by

can be stopped at any final pressure reany required GIN value (there is no such
by the rock). The successive values of
the rectangles O Pi Ai Vi O are an apV

proximation of the energy E = p dV injected.

O

actual grout path;

--------- successive points of equilibrium at nil flow rate q.

The GIN value is evaluated at a nil grout flow-rate; that is at pumps stopped.
During the grouting process however, head losses have to be overcome, so that the
manometric pressure will be higher than the final one required by the GIN rule. This
overpressure may be of the order of 10 to 20% of the final one.
The essential definition of the GIN value is shown in Table 2.

- 21 -

102.1-R-147A

Starting for joints of constant aperture, from

R=

p e
2 c

(reach achieved) and

V = R2 e (volume taken)
and defining
GIN = p V
one obtains for real joints
GIN = p V = 2

n kp kv c R3 (intensity independent of joint opening)

where
GIN =
c
=
R =
p
=
V =
e
=
n
=
kp =
kv

Grouting Intensity Number

cohesion (yield point) of the slurry
average reach of the grout
final grouting pressure
grout take (per m borehole)
opening of the theoretical flat joints
number of main joints (per m borehole)
coefficient to consider pressure losses due to the rugosity of the rock
walls of the joint and the variation of the aperture
= coefficient to consider volume increase due to undulations as well as
to variations of the opening of the joints

so that
R 3 GIN
or
R = Rt 3 GIN / GIN t
Rt and GINt being test values.
Table 2: The GIN concept.
The GIN-principle
The most immediate application of this concept consists of taking into account, during the grouting process, the intensity value reached at any instant. This leads to the
definition of a so-called "Grouting Intensity Number" or GIN-value, which is used to
limit the grouting process accordingly to Figure 16.
The main result is to avoid the simultaneous occurrence at high pressures and high
grout takes, which combinations lead to the dangerous zone for hydro-fracturing or
hydro-jacking indicated in the figure.
Consequently three limits are to be taken into account in designing a grouting work:
the maximum pressure;
the maximum take or grout volume, and
the maximum intensity.
- 22 -

102.1-R-147A

The maximum pressure must be in someway related to the water pressure to be expected at that spot during the future life of the structure. A ratio of 2 to 3 in respect of
this water pressure appears reasonable.
The volume limit should not be seen as an absolute boundary, but more likely as an
indication of the necessity to take a decision, which could be:
continue the grouting;
stop the grouting definitively;
stop the grouting for a time period and restart it later on;
abandon the borehole and drill another nearby;
add, for example, an anti-wash product to the mix, or
take any other adequate measure.

Figure 16: The limitations of the grouting process.

a) traditional method: limit ADG
b) GIN-method: limit ABCG
Vmax=maximum take;
pmax=maximum pressure;
GIN=limiting curve (p V= const.);
In the upper corner danger of hydro-fracturing exists. Grouting paths
like F and H are not allowed. The traditional method is a special case
of the GIN method when GIN pmax Vmax.

The GIN-value is mainly an indicator of the average distance reached by the grout. It
depends essentially on the requirements of the project; the GIN-value itself being approximately proportional to the third power of the reach. These relationships may be
seen also on Figure 17.

- 23 -

102.1-R-147A

Slurry:
c/ = Cr = 1.0 mm

GIN 5

Relative cohesion

= 17 kN/m3

Density

3'000

bar l/joint

Figure 17: Example of grouting of a single joint.

Relationship between: p=pressure; V=volume taken; e=opening of the
joint; R=reach of the grout. In this example it was considered that a
thinner than 0.4 mm joint is not groutable with the cement used.

It is interesting to notice that any reach corresponds to a GIN value independently on

the actual opening of the joints. A thinner joint absorbs less grout volume but requires a higher grouting pressure in the inverse proportion.
A limiting factor of the grouting is clearly the risk of excessive hydro-jackings of the
joints. Figure 18 shows that hydro-jacking phenomena do occur at constant intensity
when they start. The optimal value of the GIN must be defined on the site by grouting tests in order to adapt it to the local conditions of any homogeneous rock domain.

Figure 18: The GIN value is the governing factor of a hydro-jacking event. In
this case an extremely high GIN value of 9000 bar l/m was used.
Figure 19 gives some reference to practical values dictated by the experience gained
at many grouting sites, where sound jointed rock was present. They are called "standard" limiting curves or guidelines.

- 24 -

102.1-R-147A

Figure 19: Set of "standard" GIN-limiting curves, which may be helpful as

starting guidelines.
The fundamental principle of the GIN-Method is to reduce the grouting pressure in
function of the volume of the grout injected. 5
In using successive series of grouting holes one may observe how in following the
GIN principle the take diminish, in average, from one series to the next more or less
in the proportion of 2 to 1, while the final pressures increase in the inverse proportion
(Figure 20); confirming that any grout hole series fills practically only the widest
joints not yet filled by former series6.
If the maximum pressure is reached at very small takes, one may consider the grouting of this stages as tests, which are much more effective than any water test in
showing the low groutability of the rock mass.

5 The principle was already applied in 1985 in the headrace tunnel of the Ruzizi 2

power plant in Zare, in order to limit the losses of grout through wide-open joints.
The reduction of the pressure took place at that time stepwise and not continuously
as the GIN principle achieves it automatically.
6 Deere and Lombardi (1985) proposed to increase the grouting pressure in the bore-

holes of a grouting curtain as follows: primarily holes p, secondarily 1.5 p, tertiary

2.0 p, quaternary 2.5 p. A similar more flexible progression is automatically
achieved by the GIN-Method.
- 25 -

102.1-R-147A

Figure 20: Example of grouting results for a grouting curtain. Final points
of the grouting paths of all the borehole grouting stages (typical).
The grouting of any stage of the boreholes of a grout curtain will, as a rule, result in a
decreasing penetrability with time as shown by Figure 21, where this factor is defined as the ratio of the flow-rate to the grouting pressure.

Figure 21: Grouting process of a single borehole stage (typical) where:

1 = limiting curve, pressure versus grout take;
2 = actual grouting path, pressure versus grout take;
3 = penetrability (q/p) versus grout take.
F = final point of the grouting; pF = final grout pressure; and
VF = actual grout take.
The rectangle OCFDO is defined as the "grouting intensity"

- 26 -

102.1-R-147A

Monitoring and recording grouting works

Given the present high level of the grouting equipments available, no important
grouting activity should any longer be carried out without a corresponding real time
monitoring and a complete recording of the main parameters of each grouting stage.
The most important parameters to be analysed and plotted are:
pressure vs time;
flow rate vs time;
take vs time;
pressure vs take;
penetrability vs take;
the two last being considered the most important ones.
The real time interpretation of the plots of these functions allows to characterise the
grouting process and to detect at early time any incipient hydro-jacking or hydrofracturing and thus to adapt the grouting process to the real conditions of the rock
mass.
There are today a number of software applications available in this field, which
proved to be very effective.
The GIN-Grouting Method
GIN was originally a simple numerical parameter of the grouting procedure.
However, with the years a number of rules to be followed in grouting fissured sound
rock masses were developed and implemented. Finally this complex of rules got the
name of "GIN-Method".
The corresponding principles, which are intended to produce a quality grouting while
simplifying the grouting process and avoiding excessive hydro-fracturing, are summarised in Table 3.
The most important aspects are:
use an unique recipe for the mix, the most adequate one, defined on the base of
extensive laboratory tests, in particular to ensure the durability of the works;
use a GIN limiting function (with the three parameter pmax, vmax, GIN), a borehole
distance, a length of the stages and a cement grain size that best fit the properties
of the rock and the requirements of the project, on the base of field tests and rockmechanics considerations;
follow the grouting process on an electronic display and record the data of the
grouting process;
analyse continuously the recorded data in order to optimise the grouting process;
avoid water pressure tests in the already grouted rock mass.
The main advantages of the method are:
simplifying the procedure in using an unique slurry and eliminating or at least reducing the wastage of unused mixes;

- 27 -

102.1-R-147A

reducing or even avoiding the risk of excessive hydro-jacking and possibly of hydro-fracturing in eliminating the combination of high takes at high pressure;
equalising approximately the reach of the grout at any grouting stage regardless of
the rock quality, thus making the process more predictable;
producing a set of coherent data, which allows to judge the progress of the process
as well as the result achieved and thus to optimise it;
but, first of all the GIN-procedure is a self-adaptive one which compensates for an
appreciable part the scatter of the natural conditions met in the rock mass.
Principles of the GIN-Method

Define exactly the scope of the grouting works.

Do design, not specify, the grouting process.
Find out the "best mix" for the project by laboratory tests both from a technical
and an economical point of view. Only stable mixes, generally with superplasticiser, should be used. (Thick stable mixes have been favoured by European grouting experts since various decades.)
Use only a single mix, "the best possible one", for all grouting stages in order to
ensure the quality of the results and to simplify the procedures. This will also
reduce wastage of grout.
Define the parameters of the GIN limiting curve: pmax, vmax and GIN-value =
p V, taking into consideration all determinant geological and rock mechanics
parameters as well as the scope of the works and the related economy of the
project.
Confirm the studies by field tests and check the works carried out by additional
test groutings.
No water pressure tests to be carried out during the grouting works; they are
useless and dangerous.
The split-spacing method for the boreholes is not a new technique, but it is used
in the GIN-Method as a self-adaptive and self-regulating procedure.
The use of stage lengths, increasing with depth below ground, is progressively
recognised as a way of speeding up the grouting works and of making some, albeit small, savings.
Inject water in dry, water-absorbing rock formations above the ground water table, shortly before grouting, as a mean of avoiding a sudden blockage in the
grouting process due to water losses from the mix.
The necessity of a new borehole and its length (e.g. in a grout curtain) is decided on the base of the volume of grout taken by the nearby ones.
Computer controlled procedures are an obvious pre-requisite for optimal grouting works. Many pieces of information can be obtained from these plots. Very
interesting statistical relationship can be set up. Unfortunately, at several projects the graphs were plotted without trying to draw the due conclusions.
Table 3: The generally accepted principles of the "GIN-Method".
- 28 -

102.1-R-147A

The basic idea of the GIN to reduce the grouting pressure in function of the take is
not that exotic as it may appear at a first glance.
Indeed, if one considers not just the grouting pressure itself, but the so-called "normalised pressure"7, that is the dimensionless ratio between said pressure and the cohesion of the slurry an interesting relationship can be set up as shown by Figure 22.
Common to the traditional method and the GIN-Method is the intention to avoid uselessly high takes of grout. In the first case the limitation is obtained in increasing the
cohesion from the thinner to the thicker mixes (in function of the volume already
taken).
In the case of the GIN-Method said objective is achieved by a constant cohesion in
reducing the target pressure in function of the volume taken.

Figure 22: Comparison, by way of an example, of the traditional and the

GIN grouting methods. The "normalised pressure" pn=p/c is decreasing with increasing take, both for the traditional and the
GIN method.
7 For stable mixes the value of the normalised pressure may usually be comprised

between 104 and 5106. This value tends to infinite for thin mixes or water.
- 29 -

102.1-R-147A

In both cases the result is obtained in reducing stepwise or progressively the normalised pressure; the fundamental difference lies in the quality of the results obtained
and in the predictability of the process.
Additionally, the GIN-Method allows higher maximum pressures and higher maximum takes at no risk.
Finally, the GIN-Method can be seen as being a simple way to achieve an excellent
grouting by avoiding low strength and easy to leach out mixes, while ensuring a
complete filling of the voids and a homogeneous treatment of the rock mass (as far
this is physically possible). Additionally, the risk of hydro-fracturing and of excessive heave of the ground are practically eliminated.
An example of application
The recently very successful rock groutings below the plinth of various Concrete
Faced Rockfill Dams may be mentioned as an example of the implementation of the
GIN-Method.
A sketch of consolidation and tightening of the rock below the plinth is given in Figure 23.

Figure 23: Requirements to and extension of the grouted zone (consolidation and curtain) under the plinth of a CFRFD.
DP = differential water pressure between upstream and downstream
of the grouted zone.
Staggered grouting stages of increasing length with depth are designed.

- 30 -

102.1-R-147A

Indeed, there is no reason to make a distinction between consolidation and tightening; there is simply a zone where the rock properties need be improved for any reason what so ever. The extension of the zone must be defined at design stage on the
base of geotechnical and hydraulic considerations.
Said figure shows also in a schematic way the overall requirements for the project.
Immediately below the plinth the improvement required is the most important because the hydraulic gradient is quite high (e.g. of the order of 15).
At lower elevation the gradient is diminishing. So the requirements do. At the very
tip of the grout curtain the differential pressure from upstream to downstream is nil
and thus also the requirements for the grouted zone.
This diminution of the treatment with depth is achieved by a number of design features as
reduction with depth of the grouting rows from three to two, and finally to one;
each row of grouting is formed by successive series of primary, secondary, tertiary and where necessary of quaternary holes of different depths;
the length of the grouting stages is increasing with depth as schematically shown
in the figure. This time and cost saving feature ensures that with increasing depth
only the wider and wider joints are grouted accordingly to the actual requirements
of the design;
at the contrary the grouting intensity referred to by the GIN-value and measured at
the borehole mouth is kept constant. The actual effective pressure increases obviously with the depth of the stage below ground. This increasing pressure is considered apt to compensate, more or less, the increasing total stresses in the rock
mass due to the overburden as well as the increasing future water pressures with
depth. It has to be considered that the density of the grout is comprised in between
that of the water and that of the rock.
An additional comment to be done refers to the length of every borehole. Traditionally the depth of the holes is rigidly staggered so that secondary holes are shorter
than primary and again tertiary are intended to be shorter than the secondary ones.
In fact, there is no stringent reason to do so. It is felt that the result may be better and
obtained at lower costs if the decision to drill any hole and to define its depth is taken
in function of the absorption of grout of the nearby holes, as shown in Figure 24.
The limit of take or the "critical take" must be carefully defined. One logical way is
to set it equal to the theoretical take at the maximum pressure for a GIN reduced to
half the normal value.
Obviously, a preliminary overall investigation through longer holes is necessary.
Another aspect to be considered from the point of view of the economy of the project
is the possibility to approach the best possible balance between the drilling costs and
the cost of the quantities of slurry injected to reach the same result.
The question may also be raised whether the distance between the boreholes must be
kept absolutely constant or not. It is felt that, for instance, the tertiary holes could be
placed somewhat nearer to the primary than to the secondary ones. However, this is a
marginal aspect of design.

- 31 -

102.1-R-147A

Design of a grout curtain

A

Traditional design

Possible design adapted at each stage

to the measured grout take in the
previously nearby grouted holes

Figure 24: Design of a grout curtain.

In summary, the optimisation of the distance between the boreholes as well as of the
length of the single grouting stages, but also the grouting pressures to be used, may
provide interesting possibilities of saving.
Unfortunately, a number of these possibilities of optimising the design of a grouting
project are often overlooked or not accepted because old-time specifications are
blindly enforced.
Flow rate of grout8
An important question frequently addressed to, is how to manage the flow rate of
grout during the process and even more how to reduce it in nearing the GIN limit so
to finally reach the target intensity.
Depending on the actual conditions, there are different manners to proceed. However, the following one appears to be the most convenient. It is represented on Figure 25.
Being GIN the target intensity, two new limits at, let say, 0.95 GIN and at 1.05 GIN
as shown in the figure should be defined.
The selected basic grout flow rate "q", is kept as constant as possible all along the
grouting process until the 0.95 GIN limit is reached. The flow is then reduced stepwise to 2/3 q0, 1/3 q0 and zero at crossing each of the three limits.
A similar rule can be used in respect to the maximum pressure, while for the maximum volume the process can be stopped at one time.
8

Section added after the presentation at the Conference

- 32 -

102.1-R-147A

Obviously the number of successive limits and thus the number of steps in the pressure reduction, as well as the width of the fringe defined in said way may be adapted
to the local conditions.
The basic flow rate depends of course on the equipment available, but has to be
checked in any case by grout tests in function of the scope of the grouting works. As
a rule however a medium or moderate flow rate is preferred to a very high one.

q = q0 = basic flow rate

q = 2/3 q0
q = 1/3 q0
q = 0
= 0.95 GIN and 0.95 pmax
= GIN and pmax
= 1.05 GIN and 1.05 pmax

Figure 25: Management of the flow rate of grout (typical).

GIN = Target grouting intensity
q0 = basic rate
q = q0
if pV 0.95 GIN
q = 2/3 q0 if 0.95 GIN pV GIN
q = 1/3 q0 if GIN pV 1.05 GIN
q =0
if pV 1.05 GIN
Conclusions
The technique of rock grouting is an already old one. Unfortunately, the results of the
treatment carried out can not be observed directly and thus are difficult to be exactly
assessed.
With the decades, this situation led to a number of historically founded beliefs based
on single experiences but seldom cross-checked. So, while each one of them can be
true under given conditions, more general correlations were missing.
The heterogeneity and anisotropy of the rock mass and consequently the great scatter
of its properties, do not really favour the definition of clear general concepts.
- 33 -

102.1-R-147A

In this paper a few simple notions and ideas were presented, which any grouting project should refer to.
The suggested grouting method takes these concepts into account, it is based on a
critical examination of the current habits and tries to replace subjective beliefs by
physically founded notions. In no way, it does represent the final stage of the technique of rock grouting and future developments are to be expected.
Meanwhile, it is however felt that the GIN-Method represents a clear step in the right
direction of better optimised and even of more economical grouting processes.
The GIN-Method has obviously to be solely intended as a design support, which
does not replace a sound analysis of each specific project. In particular the parameters to be used must result from this analysis as well as from laboratory and field
tests.
It is however well known that the implementation of new ideas and working methods
is always slow because there is worldwide a number of people who prefer to stay
comfortably with the old rules enforced by specifications taken over, without any
change, from former tender documents. They thus decide not to follow new ways of
thinking and to skip the necessary efforts to understand new developments.
Main errors to be avoided9
The main points of the GIN Method are summarised in Tables A to H.
According to Table H (or 3 of the main paper), the GIN Method is based on 12 rules.
There are thus at least 12 manners to make a wrong use of it!
Some of the most usual mistakes are listed hereafter.
1st mistake
Decide on grouting works just to comply with some old tradition, not with the real
necessities of the project.
2nd mistake
Specify from the beginning a GIN Number and forget to confirm it by grouting tests.
The selected value may not fit the actual geo-mechanical conditions and hydrofracturing may happen too frequently, or the reach may result insufficient.
3rd mistake
Specify the distance between the boreholes and forget to confirm it by grouting tests.
The classical split-spacing method is not wrong in itself and helps a lot, but may not
represent always the optimum.
4th mistake
Shift to a second mix with higher cohesion and viscosity just to comply in a formalistic way with some specification on grouting pressures.
9

Comments included in the oral presentation.

- 34 -

102.1-R-147A

The scope of grouting is to inject a certain volume of grout and to reach a certain distance not to arrive at a given pressure at any grouting stage.
5th mistake
Specify rigidly a single, double or triple row curtain all the depth down, while the requirements to the curtain are decreasing at depth; so the number of rows should do.
6th mistake
Specific a fix stage length, while the requirements are not constant all the depth
down.
7th mistake
Believe that a relationship can exist between water pressure tests and groutability and
continue to carry out useless Lugeon tests at every grouting stage.
8th mistake
Continue to pump grout in, after a hydro-fracturing did show up on the display or the
GIN number has been reached. Doing so, you certainly will confirm again that hydro-fracturing is governed by that intensity, but you just waste money.
9th mistake
Specify from the beginning the lengths of the boreholes for a grout curtain and stick
to them without taking into account the actual rock conditions, borehole by borehole
and stage by stage.
In fact it is not a drama if at the end of the day tertiary holes will be deeper than the
secondary and even the primary, provided they are required by the local rock conditions.
10th mistake
Record the data, but analyse them later on in your cosy office when you don't have
nothing else to do.
The recorded data must steer the grouting process and must thus be looked at immediately on the site. Practical simple rules must in any case be set up, so to be easily
followed by the operator. Nevertheless, unexpected situations may occur, which require some thinking and decisions.
11th mistake
Think (and possibly write also in technical papers) that the GIN method implies high
grouting pressures, which may increase the risk of hydro-fracturing, while it is your
responsibility to choose the grouting intensity and the maximum pressure the rock
mass may resist at no or at reduced risk.
12th mistake
Forgot to adapt the grouting intensity (GIN), in increasing or decreasing it, if the
conditions of the rock mass do not correspond to the expected ones, or at the contrary
change continuously the GIN value so that nobody, including yourself, can under- 35 -

102.1-R-147A

stand what kind of job was really carried out, because an interpretation of the data
recorded is no longer possible.
Acknowledgements
Many thanks are due to numerous engineers who helped to analyse in a critical way
the problems related to grouting of jointed rock masses and to implement new procedures. Among them Prof. Don U. Deere deserves a special mention.
Thanks are also due to Dr R. Bremen for the help provided in preparing this paper.
References
Aisiks E.G., Giuliani F.L., Tipping E.D., Vard O.A. (1991), "Remedial works at El
Chocn dam", Transaction 17th ICOLD Congress, Vienna 1991, Q.65, R.67, Vol.II,
pp. 1211-1230
Amberg W. (1996), "Modle pour les massifs rocheux fissurs et saturs", Journe
Fondation de Barrages, 15 October1996 l'Ecole Polytechnique Fdrale de Lausanne
Bremen, R. (1997), "Use of additives in cement grouts", The International Journal on
Hydropower & Dams, Volume Four, Issue One, pp. 71-76
Bremen R. (2001), "Rehabilitation of the Langetental flood diversion tunnel",
AITES-ITA 2001 World Tunnel Congress. Progress in Tunnelling after 2000, June
10-13, 2001, Milan
Cambefort H. (1964), "Injection des sols", Eyrolles, Paris
Coon R.F., Merritt A.H. (1970), Predicting in situ modulus of deformation using
rock quality indexes, "Determination of the in situ modulus of deformation of rock",
ASTM STP 477, pp. 154-173
Deere Don U. (1968), Geological consideration, Chap. 1, "Rock mechanics in engineering practice", Stagg & Zienkiewcz, eds. John Wiley & Sons, N.Y., pp. 1-20
Deere Don U. (1982), "Cement-bentonite grouting", Grouting in Geotechnical Engineering, Proceedings of Conference sponsored by the Geotechnical Engineering Division, ASCE, New Orleans, pp. 279-300
Deere Don U., Lombardi G. (1985), "Grout slurries - Thick or thin ?", Issues in Dam
Grouting, Proceedings of the session sponsored by the Geotechnical Engineering Division of the American Society of Civil Engineers in conjunction with the ASCE,
Convention in Denver, Colorado, April 1985

- 36 -

102.1-R-147A

Hssler L., Hkansson U., Stille H. (1992), "Classification of jointed rock with emphasis on grouting", Grouting Soil Improvement and Geosynthetics, ASCE and
ISSMFE, New Orleans, La, 25-28 February 1992, Vol. 1, pp. 449-460
Hssler L., Hkansson U., Stille H. (1992), "Computer simulated flow of grouts in
jointed rock", Grouting Soil Improvement and Geosynthetics, ASCE and ISSMFE,
New Orleans, La, 25-28 February 1992, Vol. 1, pp. 461-473
Hoek E. (1983), Strength of jointed rock masses", Rankine Lecture, Geotechnique
33, N 3, pp. 187-223
Houlsby A.C. (1982), "Optimum water: cement rations for rock grouting", Grouting
in Geotechnical Engineering, Proceedings of Conference sponsored by the Geotechnical Engineering Division, ASCE, New Orleans, pp. 371-331
Houlsby A.C. (1992), "Construction and design of cement grouting", John Wiley &
Sons Inc., N.Y./Chichester
Lombardi G. (1985), "The role of cohesion in cement grouting of rock", ICOLD,
CIGB, Quinzime congrs des Grands Barrages, Lausanne, Q.58 R.13, pp. 235-261
Lombardi G. (1987), "Injection des massifs rocheux", Publication de la Socit
Suisse de Mcanique des Sols et des Roches (extract), Journe de printemps, 7 mai
1987, Bienne, N. 115, pp. 29-41
Lombardi G. (1989), "The F.E.S.-model and foundations for concrete dams", "Victor
de Mello Volume" Editora Edgar Blcher Ltds, So Paulo, Brasil, pp. 270-283
Lombardi G. (1990), "La permabilit et l'injectabilit des massifs rocheux fissurs", Revue Franaise Gotechnique, N. 51, April 1990, pp. 5-29
Lombardi G. (1991), "Sull'iniezione di massicci rocciosi fessurati. On grouting of
fissured rock masses", Societ Italiana Gallerie, Il consolidamento del suolo e delle
rocce nelle realizzazioni in sotterraneo, Soil and rock improvement in underground
works, Milano, March 18-20, 1991, Estratto Atti, Vol. I, pp. 405-414
Lombardi G. (1992), "The F.E.S. rock mass model - Part 1", Dam Engineering, Vol.
III, Issue 1, February 1992, pp. 49-72, "The F.E.S. rock mass model - Part 2", Dam
Engineering, Vol. III, Issue 3, August 1992, pp. 201-221
Lombardi G. (1996), "Selecting the grouting intensity", The International Journal on
Hydropower & Dams, Volume Three, Issue Four, pp. 62-66
Lombardi G. (1997), "GIN principle revisited", Water Power & Dam Construction,
October 1997, pp. 33-36
Lombardi G. (1999), "Iniezioni di masse rocciose con miscele cementizie. Il concetto
GIN", Rivista italiana di geotecnica, Anno XXXIII, No. 2, pp. 40-52
Lombardi G. (1999), "Grouting of rock with cement mixes", ICOLD Symposium
Antalya (Turkey) 23rd September 1999, Dam Foundations: Problems and Solutions
- 37 -

102.1-R-147A

Lombardi G., Deere Don U. (1993), "Grouting design and control using the GIN
principle", Water Power & Dam Construction, June 1993, pp. 15-22
Lombardi G. Deere Don U. (1993), "Diseo y control del inyectado empleando el
principio GIN", Tomado de: Water Power & Dam Construction, Por Comisin Nacional del Agua, Mxico
Lugeon M. (1933), "Barrages et Gologie", Librairie de l'Universit de Lausanne
Nonveiller E. (1988), "Grouting theory and practice", Elsevier, N.Y. USA
Turcotte L.,Savard B., Lombardi G. , Jobin H. (1994), "The use of stable grout and
G.I.N. technique in grouting for dam rehabilitation", Annual Meeting Canadian Dam
Safety Conference CDSA and CANCOLD, Winnipeg, Manitoba, 2-6 October 1994,
pp. 137-161 (Bibl. VIII 426)
Vard O.A. (1991), "Presa El Chocn - Tareas de reparacin", IX Panamerican
Conference on Soil Mechanics and Foundation Engineering. Sociedad Chilena de
Geotecna, Via del Mar, Chile - Volume IV, pp. 1523-1556
Weawer K. (1991), "Dam foundation grouting", American Society of Civil
Engineering, N.Y. USA
Additional notes
To avoid frequent misunderstandings, one should consider the following in respect to
the paper "Grouting of rock masses". The Tables A to H summarise for clarity the
main points presented in the paper.
1. Actually, one should talk of "grouting the discontinuities" rather than "grouting
the rock".
2. The notion of "best mix" used, e.g. on points 3 and 4 of the following Table H
(or Table 3 of the paper), refers obviously to the "best mix found and defined in
the frame of the project dealt with under given circumstances", and has obviously not to be understood in a general absolute sense.
3. In matter of "potential joints" it must also be noticed that the risk of hydrofracturing depends, at least to some extend, also from their direction to the borehole (Refers also to Figure 6).
4. To Table G (or Table 2 of the main paper), notice should be taken that the GIN is
not a simple arbitrary number but a function of various factors. In particular, the
reach is related to 3 GIN .
Lower GIN value means thus reduced distance between the boreholes, that is
more boreholes and additional costs.
5. To Figure 16 one may additionally remark that the GIN Curve has firstly to be
understood as a "warning line" against excessive hydro-fracturing and hydro-

- 38 -

102.1-R-147A

6.
7.

8.

9.

jacking. The target of a grouting process is to introduce in the rock mass a certain amount of grout not to reach a given pressure.
Figure 17 refers to the contact surface "core to rock" at the El Cochn dam (Argentina).
To Figure 18. Indeed the hydro-fracturing is governed by the GIN value not that
much by the grouting pressure (which of course is necessary to produce the
grouting intensity). It can be assumed that a high enough grouting intensity always exists to cause a hydro-fracturing. This value may however lay outside of
any practical or reasonable range as this was the case for Figure 18.
To Figure 19. Notice must be taken that the "standard GIN Curves" do in no way
mention any maximum nor minimum limit for the intensity or the pressure to be
used at a given project. It is the responsibility of the engineer to define the values
to be used in any single case.
To Table H (or Table 3 of the main paper). The previous saturation of the
discontinuities to be grouted is ever since usual in the grouting of contraction
joints of concrete dams.

TABLE A
The role of water
1.
2.
3.
4.

Hydrate the cement

Produce a fluid mix (avoid internal friction)
Compensate for water losses from the mix
Open the joints in front of the grout mix

TABLE B (Table 1 of the main paper)

Dam (Examples of unique mix)

W/C

Paute (Ecuador) upper part

Alicur (Argentina)
El Cajn (Honduras)
Clyde dam (New Zealand) 2nd part
El Chocn (Argentina) repairs

0.6
0.67
0.7
0.6
1.0
0.7
or 1.0
0.59
0.7
0.7
1.0
0.63

Sir (Turkey)
Katze (Lesotho)
Pichi Picn Leufu (Argentina)
Potrerillos (Argentina)
At Hamou (Morocco)
Casecnan lower tunnel (Philippines)

- 39 -

Fluidifier
Intraplast 1.4%
Intraplast 1.2%
Bentonite 0.2%
Intraplast ~ 1%
Bentonite 0.5%
Puzolanic cement, Mistra 1%
Bentonite 1.2%
Cement + ash. Conplast 1.5%
various
Rheobuild/Viscocrete 0.7-0.8%
Bentonite 2%
Intraplast 1%

102.1-R-147A

TABLE C
To ease the penetration of grout
1.
2.
3.
4.

Add plasticizer
Use finer cement
Use higher pressures
Count on water in front of the mix

No reason any longer for "thin to thick" mixes

TABLE D
Advantages of stable mixes
and high pressures

Complete filling of joint; no water bubbles

High mechanical strength
Reduced shrinkage
Better bound to rock
High chemical resistance
Predictability of the grouting process

TABLE E
Properties of the mix
Fresh mix

Density

Bleeding

Decantation

Viscosity

Cohesion (yield point)

Set time

Set mix

Shrinkage

Mechanical strength

Chemical resistance

Permeability

- 40 -

102.1-R-147A

TABLE F
Expected improvements of the
rock mass by grouting

Reducing the permeability

Reducing the deformability
Increasing the strength
Aspects to be considered

Feasibility
Expansion due to grouting
Durability of the beneficious effects
Economics of the treatments

TABLE G (Table 2 of the main paper)

The GIN concept
For joints of constant aperture, from
R=

pe
2c

reach and

and defining GIN = p V

V=

R2 e (volume taken)

one obtains for real joints

GIN = p V = 2

n kp kv c R3

(the grouting intensity is independent of the joint opening)

where

so that

GIN
c
R
p
V
e
n
kp
kv

=
=
=
=
=
=
=
=
=

Grouting Intensity Number

cohesion of the slurry
average reach
final grouting pressure
grout take (per m borehole)
opening of the theoretical flat joints
number of main joints (per m borehole)
coefficient for pressure losses
coefficient for additional volume

and
R 3 GIN
Rt and GINt being test values.

R = Rt

- 41 -

GIN / GIN t

102.1-R-147A

TABLE H (Table 3 of the main paper)

Principle of the GIN Method
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.

Define the scope of grouting

Design the grouting process
Find out the best mix
Use a single mix (the best)
Define the GIN limits
Confirm by tests
No water pressure tests
Split-spacing as self-adaptive process
Variable stage length
Previous saturation of dry rocks
New boreholes steered by grout take
Real-time grouting control

Notice
The paper "Grouting of Rock Masses" was published on page 164 to page 197 of
Volume 1 of the Proceedings of the Third International Conference on
GROUTING AND GROUND TREATMENT
on February 10th to 12th, 2003 New Orleans LA, USA
edited as "Geotechnical Special Publication N 120 by
Lawrence E. Johnsen
Donald A. Bruce
Michael J. Byle
on behalf of
ASCE, DFI (Deep Foundation Institute) and GEO-Institute.

- 42 -