THE GROUT LINE

Grout Line
Overture

Here we are again for the twelfth appointment with our Grout Line.
For this issue we have 2 articles, and
a short comment related to one of the articles. Both articles are related to previous articles published in past editions of
the Grout Line. I remind you here that
all the past articles of the Grout Line can
b e d ow n lo a d e d a t w w w. g r o u t
line.com/Articles.htm.
The first article is the answer of Dr.
Lombardi (Lombardi SA Engineering

Limited, Minusio, Switzerland,

info@lombardi.ch) to the article published on the Grout Line- Geotechnical
News Volume 25 Number 3- on September 2007 by Dr. Shuttle and others.
This was consequent to a previous
article of Dr. Lombardi published on the
Grout Line- Geotechnical News Volume 25 Number 3- on September 2007.
Dr. Shuttle also sent me a short final
comment that you can find at the end of
Dr. Lombardis article.
The controversy about the GIN

(Grouting Intensity Number) is continuing!

The second piece is Jim Warners
clarification of the article published in
the March 08 issue related to the difference between Low Mobility Grout and
Compaction Grout.
I wait for your grouting papers, articles or comments. Please send them to:
Paolo Gazzarrini, fax 604-913-0106 or
paolo @paologaz.com, paologaz
@shaw.ca or paolo@groutline.com.
Ciao!

Misunderstanding of GIN Confirmed

Giovanni Lombardi
Introduction

In a paper published by Geotechnical

News Grout line [1] titled GIN
Again Misunderstood the writer presented some comments to the paper
Penetrability Control of GIN Mixes
during Fractured Rock Grouting by
Vafa Rombough, Grant Bonin and
Dawn Shuttle [2]. In response to these
comments the paper GIN Distilled by
the same authors was published, also in
Grout Line with date September 2007
[3].
As this last paper contains again
some misunderstandings, the following
considerations will refer to them, but
1

will provide also some considerations

of more general nature.
GIN as Limit

In the intent to show the presumed limits of GIN, figure 2 from the Antamina
Dam, was presented in paper [3]. The
same is reported hereafter with its original legend, as Figure 1.
Unfortunately, its meaning was
again misunderstood and was the occasion to invent the amazing tale of multiple GIN closures. It was actually
assumed that the so- called closure
was reached 4 times at the points A, B,
C and D, where the GIN-line was

touched by the pressure path.

The question is thus: why only 4
times?
In fact the GIN line could have been
easily reached and even crossed not
only 4 but 40 or more times1. Even
more, if said GIN line would have been
followed continuously from a point of
high pressure - low volume to a point
of low pressure - high volume, to
some final value, then the two lines
would have coincided an infinity of
times.
Fortunately, the 4 mentioned points
are not closure points at all, because
the flow rate at these instants was about

In fact the same grout path crosses easily 40 times a somewhat smaller GIN line as the one shown on figure 1.
Geotechnical News,

June 2008

THE GROUT LINE

a) = Pressure vs. Volume curve

b) = Measured flow rate vs. injected volume
Example grout injection data from Antamina Dam showing
multiple GIN closures [234BP 27.6-32.6 m] adapted
from Richie and others, 2003).
Figure 1. As presented in GIN distilled [3] as figure 2; with
its original legend.
20./12./9. and 6. L/min respectively.
As one should know - and this is repeated again - the flow rate at the closure point should and will be nil. Only
then one can talk of closure.
From the graph of Figure 2 one may,
in extrapolating the flow-rate tendency,
easily conclude that the nil flow rate
would have been arrived at for a volume
injected of about 400 l/m. Therefore the
closure would have occurred at the
point E (or nearby), which point was
however not reached with the grouting
works, and which is the unique real
c lo s u r e p o in t in s p ite o f th e
mentioned tale.
In the case of figure 1, the grouting
was stopped too early, that is before
reaching the point of closure E. This
means simply that the concept of the
4

Geotechnical News,

June 2008

Figure 2. The real closure for the example shown on

figure 1 takes place at the point E. In fact, the unique closure
would have taken place at the point E, or near of it, at a flow
rate nil. The grouting was stopped before reaching the real
closure.

GIN-method was not really understood,

already at the time of grouting. Only
300 l/m were injected instead of the
about 400 l/m, which should have been.
The real closure point was missed by
33% of volume.
This point outlines again one of the
significant aspects of the GIN value and
doesnt need many comments, except
that it was overlooked both by the grouting engineer at site and the writers of the
paper from which figure 1 was taken.
By the way, this point E could have
been reached in following the GIN line
(which task requires however some skill
and an adequate equipment, but which
could be considered to be the most
logical one). The point E could also
having been reached from below or

from above the GIN line (provided, in

this last case, that the volume taken
would not overpass the corresponding
estimated value, e.g. 400 l/m) (see Figure 3).
Indeed the GIN line is a target to be
reached not a wall which could not be
crossed. However, frequent cases are
met where certain GIN values represent
a physical limit, which can hardly be
crossed. This happens when an important hydro-fracturing, hydro-splitting
or hydro-jacking takes place.
One knows, from the theory and the
grouting practice, that when the pressure suddenly decreases at an almost
constant flow rate, a phenomenon of
this kind does occur. Said decrease of
pressure, often at a constant GIN value,
may be of a limited extent. This means

THE GROUT LINE

. . . the quite usual path

__________the possible theoretically best path
a possible but risky strategy, and
++++++++a possible path if the final take can be estimated (see fig. 2)
In nearing the point E it is always advantageous, in any
case, to overpass somewhat the GIN line so not to delay uselessly the conclusion of the grouting works.
Figure 3. The ways or strategies to reach the unique, a priori
unknown, closure point E.
that the hydro-fracturing discontinuity
itself is of a limited length because cut
by other more stable fractures or by different rock layers.
During a grouting process a series of
such local limited hydro-fracturings
can therefore be observed on always
higher intensities, that is on higher GIN
values.
In some cases, however, a major hyd r o - f r a c tu r in g alo n g a m a in
discontinuity can take place.
The meaning of the GIN line as a
limit for hydro-fracturing was de-

Figure 5. Indication of hydro-splitting.

Case 1 = closure to be expected (normal grouting)
Case 2 = hydro-splitting quite likely to occur

scribed in paper [7], from where Figure

4 is taken.
The fact that hydro-jacking takes
place by increasing volume and decreasing pressure is easily explained,
because the GIN value - that is the
grouting intensity - represents roughly
the energy pumped into the rock mass,
and because - like many other instabilities - the hydro-jacking event takes
place at a practically constant energy
level (see [7]).
It may also be mentioned that, as
shown in Figure 5, when the pressure

a) The historical grouting limits for pressure and take.

b) The GIN limits for pressure, take and intensity. The main scope of limiting the
intensity is to reduce the risk of hydrofracturing and to avoid useless grout losses.
Figure 4. The limitations of grouting.

path follows the GIN line two cases may

happen:
Case 1: the flow rate tends to diminish with increasing volume injected. In
this case a normal grouting takes place.
Case 2: the flow rate remains more or
less constant, then an important hydro-splitting is likely to occur.
In this last case the corresponding
GIN line represents a barrier, which
cannot, or at least should not, be
over-passed. Even more the continuation of the grouting can practically take
place only along said GIN line. Increasing the power of the pump leads mainly
to an increase of the flow rate, while the
pressure still decreases.
Reducing that power leads to a reduction of the flow rate and soon to a
stoppage of the injection. Consequently, the GIN value has an additional
meaning when hydro-jacking or hydro-splitting takes place.
It should thus be evident that the conclusions drawn in paper [3] do not hold,
and that an additional merit of the GIN
concept has to be recognized, in the
sense of a protection against excessive
hydro-fracturing of the rock mass.

Geotechnical News,

June 2008

THE GROUT LINE

Figure 6. The grout process can be stopped at any final pressure prescribed or by reaching any required GIN value (there is
no such situation as refusal by the rock).
________actual pressure path
- - - - - - theoretical pressure path at flow rate nil (the viscosity related term of the pressure vanishes and only the cohesion related one remains)

GIN Closure

It appears, further that in the background of many statements about the

GIN-method, the opinion prevails that
grouting by that method must take place
in using a constant, possibly very low
flow rate. It is thus stated in [3]: hence
in practice GIN becomes unreliable to
control practical grouting. Fortunately, this is an additional confirmed
misunderstanding.
It was not seen by many that figure 3
of paper [5] titled Actual grouting path
and grouting intensity (GIN) as well as
the concept of over-passing it by, for example 10%, have just the intent to explain the question. (Possibly, the
explanation was not clear enough) (see
Figure 6).
It is said also, by the authors of paper
[3], that in a typical GIN grouting project you pump at a constant rate2.
For sure, there are worldwide a number of designers, who specify a constant
and even a low flow rate, when referring
to the so-called GIN grouting. (Hopefully, they know the reason of!). But,
such a specification and similar ones are
independent of the GIN principle and
2

are by far not always optimal.

The right way to proceed is shown at
the reverse in paper Grouting design
and control using the GIN principle [5]
published already in 1993. It consists to
steer the grouting process in taking into
account the penetrability defined by
q/p (flow rate divided by grouting pressure), which value has obviously to be
shown to the operator at real time on the
display (see Figure 7).
In fact, the penetrability must be nil
at the closure point; thus also the flow
rate. As already shown, this principle
was not understood in the case of
Antamina shown (as well as in a number of other jobs).
The principle to follow consists simply in adapting progressively the pressure and thus the flow-rate, when the
GIN value (at nil flow rate) accordingly
to the flow path, approaches the desired
value.
This way of doing allows a certain
extrapolation in time of the decrease of
the penetrability so there is no need to
select an extremely low flow rate long
before stopping the process (see again
figures 2 and 3).

This was apparently not the case in Antamina according to figure 1!

Geotechnical News,

June 2008

Figure 7. 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 grouting
pF final grouting pressure
VF actual (final) grout take
To avoid new misunderstandings
some additional explanations appear to
be necessary.
An example is shown on Figure 8 for
the case the pressure is maintained constant, which means that it is changed
only stepwise.
No matter how the grouting did proceed until the point A. At that instant,
the pressure p1 and the flow rate q1 were
arrived at and the volume V1 was already injected.
If the pressure were kept constant, its
line would cross the GIN line at the
point B with a value VB for the take.
However, a check shows that the decrease of the flow rate would take place
quite earlier and the extrapolation of it
would lead to a stop at the volume VC,
which is smaller than VB. The two values V B and V C should obviously
coincide.
The pressure must thus be increased
at the point 2 so that the corresponding
volume, VB, will decrease while the expected one, VC, would increase reducing the gap (VB - VC) between them.
If necessary, a further adjustment of
the pressure can be done a while later.

THE GROUT LINE

Figure 8. Example of a way to steer the grouting process.

p = grouting pressure
q = flow rate
VB = volume theoretically taken at a pressure p1 (VB = GIN/p1)
VC = volume take estimated in extrapolating the decrease of the flow rate
(better the decrease of the penetrability = q/p)
If VC<VB then increase the pressure, so VC increases and VB decreases, reducing the gap (VB-VC).
At the reverse, if VC would appear to
be greater than VB, the pressure should
be reduced.
The procedure can be improved in
referring to the penetrability, (q/p), instead of the flow rate, q, itself. In doing
so, at least certain irregularities are
smoothed and the process made more
precise and easy to steer (refer to figure
7).
A way to speed up the procedure, as
already presented in paper [7], consists
to target a somewhat higher (10%) GIN
and to stop when the flow rate falls below a certain small value.
It may be easily understood that the
described procedure can be automatized in using an adequate software.
Similar procedure can also be developed if different grouting methods were
adopted.
By the way, the following consideration is due.
The GIN value will be decided on the
base of some grouting tests in a rock
mass similar to the one, which has to be

actually grouted; but, obviously not for

any single borehole nor for any single
grouting stage. Therefore it is not necessary to try to reach some theoretical
absolute precision and therefore some
tolerance has to be accepted. Other
means, as for example additional boreholes series, are available to compensate for too important scatters in the
rock mass properties.
This way of doing is by far not a
poor guidance and the statement in
practice GIN becomes unreliable to
control practical grouting does not
hold. In fact, the practical experience on
many sites confirms the good, if not excellent, guidance it provides. GIN is actually a target, which can be missed as
many times as one wishes. However,
this is a problem of the shooter not of
the target!
Also there are, for sure, many people
who dont know how to steer correctly
the GIN grouting process.
At the end of the day, the actual
physical penetration distance reached

by the grout depends on the pattern of

the discontinuities, on the grout volume
absorbed, and on the cohesion-related
term of the pressure at the moment the
procedure is stopped. Indeed, when the
pump is turned off the flow rate becomes nil and the viscosity related
terms disappear.
This is exactly the definition of the
GIN value.
In fact, the actual pressure path, the
flow rate and their variations during the
first part of the grouting process are of a
quite secondary importance, except
they may influence the duration of the
procedure.
In fact, in accepting the GIN principle, higher grouting pressures, and thus
higher flow-rates as usual can be used
as long as some distance from the targeted GIN value is kept. This allows
speeding up the grouting process. The
second part of the grouting process was
already discussed.
Definition of the GIN Value

It should be clear that due to the extremely high number of unknowns existing in the actual rock mass, as types
of discontinuities, orientation, opening,
deformability, rugosity, frequency,
interconnections, infill, etc. no possibility does exist to define a priori on a theoretical base the parameters to be selected for the grouting process.
The only realistic way is to carry out
test fields for each zone of the rock
mass, which may reasonably be considered to be homogeneous, as explained
in paper [6].
The split-spacing method should be
used in any case both for grout curtains
and for foundation consolidation. The
criteria to judge the results are the decrease of the take from series to series of
boreholes while keeping a constant GIN
value.
In matter of the grout reach it was not
always clearly distinguished between
the GIN value computed for a single
joint of constant opening and the value
obtained by a grout test in an actual rock
mass.
Therefore the statement GIN overestimates the grout reach cannot have
any meaning when referred to a grout
test, in a real rock mass, nor when reGeotechnical News,

June 2008

THE GROUT LINE

Figure 9. FES model of a fissure along its plane (for hydraulic mathematical analysis).
a) = pattern of channels and islands
b) = cross section of a channel
Q1 = flow through a channel
1. = contact zones
2. = flow channel
3. = semi-rigid core (in the case of a cohesive fluid)
V = velocity in the case of a Binghams grout
the same happens successively with

ferred to a single joint as used for theoretical considerations. In fact, the formulae of paper [3] confirm the reach
given by paper [4] in the theoretical
case.
The figure 4 of paper [3] showing the
grouting of two parallel joints of different openings with different reaches,
which was already included as figure 12
in paper [4] refers simply to the
fundamental rule that the grout reach (at
nil flow rate) depends linearly on the
opening of the discontinuity. Said figure is thus meaningless in the present
discussion and just confirms an additional misunderstanding.
It should be known that in the case of
an actual treatment of a rock mass;
the primary holes will fill mainly the
widest joints to the largest distance;
the intermediate secondary holes
will fill the widest of the remaining
joints to quite a distance because,
due to the minor take, the same GIN
value will lead to a higher grouting
pressure;
3.

the intermediate tertiary, quaternary

and possibly further grouting series;
the success of the grouting is shown
by the decreasing take from boreholes series to borehole series;
if such a decrease, from borehole series to borehole series, doesnt take
place, an escape of grout, e.g. to a
Carstic cavern, or a very important
hydro-jacking (or fracturing), is
quite likely to have occurred.
Radial against Strip Flow
Computations

The channelling of grout or water

along the rock discontinuities is by far
not a new concept. It was, for example,
analysed in the paper The FES Rock
Mass Model 1992 [8]. See Figure 9.
However, this effect cannot be called
in to justify the use of a strip of constant
width to analyse the grouting process,
except the channels were all parallel
and independent each from another
without any bifurcation or crossing.
The experience shows that, as a rule,

the joints and fissures are filled by the

grout in all directions. The model of a
radial flow is thus a bit more realistic
than the one of a strip flow3.
The weak point of the strip model is
that the specific flow rate , that is the
flow by unit width is assumed to be constant all along the discontinuities from
the borehole to the border of the grouted
zone, while it is obvious that this specific flow-rate decreases strongly with
the distance from the borehole. (Approximately with 1/r in function of the
actual reach.) As consequence of the
strip-model the pressure gradient appears to be constant and thus the pressure to decrease linearly from the
borehole, while in fact the pressure
drops very strongly near the borehole
and is very significantly reduced at
distance.
The linear distribution is reached
only when the process is stopped and
the flow rate is nil everywhere.
For this reason, at this moment, the
theoretical final reach and the pressure
distribution are the same, whether computed with the strip or with the 2-D
model.
In the reality, the grouting process is
sometimes more a 3-D, than a 2-D problem; but for sure not a 1-D one.
In any case the statement the GIN
method does overestimate the reach of
the grouting is the consequence of the
wrong assumption that the method implies a low; near nil flow rate, that is a
low grouting pressure and thus a long
grouting time. The contrary is true because in keeping a sufficient distance
from the dangerous-zone, accordingly to figure 4, higher grouting pressures can be used at the beginning of the
process, which will be successively reduced, as already explained.
One of the consequences of the assumption made in using the strip model
is that the splitting force produced by
the grout is wrongly computed, that is
overestimated until the end of the grouting, where the flow rate nil corresponds

To consider the effect of channelling, a factor k1 is introduced in the formula for GIN so to take care of the longer path to reach a given point
than the radial distance from the borehole. To take into account also the difference between the volumetric averaged and the hydraulic averaged
opening of a discontinuity a factor k2 is also considered. Neverthless, as already said, the optimal value of GIN must be defined on the base of
grout test fields.

Geotechnical News,

June 2008

THE GROUT LINE

any proposal in order to improve the

GIN procedure. However a realistic
way to do so has not been suggested yet.
For sure, the improvement can not
consist to come back to historical criteria as, for example, to reach any a priori,
arbitrarily specified constant pressure
in ignoring the risk of hydro-fracturing
the rock mass and thus to waste any
amount of grout if such a phenomenon
does occur.
Also, any arbitrarily given maximum take can not be of any help if not
put in relation with the pressure required to reach it, that is to the grouting
intensity.
Conclusions

Figure 10. Pressure distribution along the joint at different time intervals during
grouting.
to the value obtained by the radial
flow-model.
The strip model is thus misleading in
evaluating the risk of hydro-fracturing
the rock. See [1].
Flow Rate Computation

In the paper by Dawn Shuttle and others

it is claimed that the analytical formula
for a radial flow does not exist. It could
at the best be developed under very restrictive assumptions (e.g. constant
pressure or flow rate). In fact, such a
formula is not necessary at all because
the following steps can be easily carried
out
at a given grouting flow rate the
specific flow slows down from the
borehole with the ratio (1/r), provided the possible opening of the
discontinuity due to the pressure itself is disregarded (no hydro-jacking
considered);
consequently, the gradient of the
pressure drop can be computed at
any point using the properties of the
grout that are cohesion and viscosity
of the Bingham body;
in integrating that gradient from the
momentaneous border reached by

the grout towards the borehole the

corresponding grout pressure can be
obtained;
the flow rate can then be changed as
required, for example in order to respect the characteristics of the grouting pump or any other required
steering procedure, in particular
when the grouting process is
stopped;
as the integration is carried out numerically, any kind of grouting procedure can be considered, in contrast
to any analytical formula, which
could, at the best, take into account
only a predefined one.
It is agreed, that the procedure is less
simple that an analytical formula, but
has a much more general meaning and
is significantly more flexible.
So, the example shown in paper [4],
1985 and represented in Figure 10 was
computed accordingly.
It is thus quite difficult to imagine
how to give Bingham a try as paper
[3] proposes in its conclusion, because
in any case the computation of the pressures and of the flow rate is based on the
properties of the Bingham body.
Of course, there is nothing against

It is well known, that there are worldwide quite a number of people who do
not like and do not use the GIN method
of grouting. They have, of course, the
right to do so, even if this is not in the
interest of the job, nor in that of the
owner. The reasons of this fact are indeed numerous.
A first one is that GIN disturbs certain, sometimes century old, habits and
ways of thinking.
Very interesting historical cases are
also referred to, without taking into account how normal or exceptional they
were.
Procedures and ways of thinking developed for unstable mixes continue to
be applied when using stable ones.
Some operator prefers indeed to follow the principle dont think, just
click or introduces new rules so to get a
simplified procedure, as for example
to define a GIN value on the design
board instead of carrying out the
necessary grouting tests on the site.
A list of frequent mistakes can be
found in paper [7].
It is also quite usual to make any kind
of errors during the grouting operation
and then to criticise the GIN method instead of recognising the mistakes made.
The case of Antamina is clearly one of
them.
Often also, the GIN method is used
to grout carstic or similar formations for
which it is not intended nor applicable.
It should be used only at a second stage,
when the carstic cavities have already
been filled, e.g. with mortar.
Geotechnical News,

June 2008

THE GROUT LINE

On a given dam the lack of communication between the pumping station

and the borehole mouth led to frequent
excessive grouting in over-passing the
GIN line because the pump was turned
off too late. As usual, not the malfunctioning of the communications but the
GIN method itself was considered responsible for the mishaps and this
w r o n g c o n c lu s io n was la rg e ly
published.
Finally, people, even very specialized in some particular field, try to lead
a grouting campaign in spite of a lack of
an overall knowledge of this kind of
work in the theoretical field or in the
practical operation.
Indeed, grouting is a quite special activity, which requires both theory and
practice, as well as some thinking.
The overall conclusion is thus the
counsel to try to apply seriously GIN;
one will finally like it. Indeed, the only
problem with GIN is that it requires to
be understood.
Bibliography

1. G. Lombardi, GIN again misunder-

10

Geotechnical News,

June 2008

stood, Geotechnical News-Grout

Line, Vol 25, Nr. 2, June 2007.
2. V. Rombough, G. Bonin and D. Shuttle, Penetrability Control of GIN
Mixes during Fractured Rock
Grouting, Proceeding of Sea to Sky
Geotechnique, 59 t h Canadian
Geotechnical Society Conference
October 2006-pp 528-535.
3. D. Shuttle, V. Rombough and G.
Bonin GIN distilled, Geotechnical
News-Grout Line, Vol.26 Nr.3,
September 2007-Geotechnical
News.
4. G. Lombardi, The Role of Cohesion
in Cement Grouting of Rock,
ICOLD. 15th Congress on Large
Dams, Lausanne, 1985, Q.58 R.13,
pp. 235-261
5. G. Lombardi and D. Deere, Grouting Design and Control using the
GIN Principle, Water Power &
Dam Construction, June 1993, pp.
15-22
6. G. Lombardi, Selecting the Grouting Intensity, The International
Journal on Hydropower & Dams,

Volume Three, Issue Four, 1996, pp.

62-66
7. G. Lombardi, Grouting of Rock
Masses, 3rd International Conference on Grouting and Ground Treatment, New Orleans, February 9-12,
2003
8. G. Lombardi, The F.E.S. Rock Mass
Model - Part 1, Dam Engineering,
Vol. III, Issue 1, February 1992, pp.
49-76, Part 2, Dam Engineering,
Vol. III, Issue 3, August 1992, pp.
201-221
Dr. Lombardis article references
several articles published on the Grout
Line and from himself. You can download these articles (of course for free!)
from: www.groutline.com/Articles.htm
(article published) - www.lombardi.ch
(publications).
Dr. Eng. Giovanni Lombardi, Lombardi
SA, Via R. Simen 19, 6648 Minusio (TI),
Switzerland, Tel: +41 91 744 60 30,
Fax: + 41 91 743 9737
email: info@lombardi.ch

THE GROUT LINE

An Alternative Viewpoint on GIN

Dawn Shuttle
Vafa Rombough
Grant Bonin
We would like to commend Giovanni
Lombardi [1] on this more complete explanation of the GIN method. A problem with GIN is that it is extremely difficult to separate the writings of
Lombardi (covering a range of grout
topics) with an actual GIN methodology that can be applied by a grouting
engineer in the field. This new presentation of ideas by the author contains the
most complete summary of GIN we
have seen, including the effects of grout
injection rate which was poorly clarified before. We are sure it will be of use
to the grouting community.
We are pleased to find that Lombardi
now agrees with the primary technical
point raised in our original conference
papers [2, 3] and subsequent Grout Line
discussion [4] specifically that grout
injection rate does matter. Contrary to
the misleading tone of [1], there is no
technical point on which Lombardi dis-

agrees with our previous work. That

said, we do continue to disagree with
Giovanni Lombardi on some fundamental issues.
Grout Penetration Geometry

Real fracture networks are neither ideal

radial flow nor perfect linear flow (although both of these simplifications are
equally useful for illustrating the effect
of flow rate on grout penetration).
Discussion on the relative merits of radial flow versus strip flow is irrelevant.
The way forward is to realistically represent the true fracture network. This is
now possible using discrete fracture
network software (e.g. reference [5]).
This approach, applied to a fractured
rock grouting project, is illustrated in
Figure 1.

Figure 1. Cross section of transmissivity distribution through

grouted discrete fracture network (grout curtain shown as dark
grey) (from [5]).

Misleading Guidance on GIN

If applying the GIN approach, the methodology proposed to obtain the nil
flow rate required by GIN in Section 3
of Lombardis article [1] is misleading.
Specifically the suggestion, illustrated
in Figure 6, that by overshooting GIN
by 10% and then reducing the pressure
you will easily intersect the specified
GIN curve at nil flow rate. The effect
of the 10% overshoot was previously
discussed in Shuttle et al. [4] using the
Antamina grout data of Ritchie et al [5],
and the same figure is also included in
Lombardis article as Figure 2 (and also
reproduced here as Figure 2). If the 10%
lower GIN is set as our target GIN it is
clearly seen that this GIN is exceeded at
A, B, C and D, and as confirmed by

Figure 2. Example grout injection data from Antamina Dam

showing multiple GIN closures [234BP 27.6-32.6m]
(adapted from Ritchie et al, 2003).
Geotechnical News,

June 2008

THE GROUT LINE

Figure 3. Example penetrability (q/p) versus grout take record from Antamina.
Lombardi [1], even at D this is not at
nil flow rate and hence not at GIN.
Therefore the practicalsuggestion that
by overshooting GIN by 10% you can
quickly obtain the required nil flow
rate is erroneous.
Similarly, Lombardis Figure 7 of
the grouting process of single borehole
stage (typical) is grossly simplistic.
The figure suggests that it is (a) typical
to intersect the GIN curve at zero flow
rate, and (b) that all penetrability (q/p)
versus grout take curves show a gradual
decay with increasing take. This is untrue. As previously shown in Figure 2,
and stated at length in [1], most first intersections of the GIN curve do not coincide with nil flow. Figure 3 shows a
penetrability curve plotted for an
Antamina grout injection stage. Two
mixes were used. The first mix was a
standard low viscosity grout typically
used in GIN grouting. Negligible, if
any, reduction in penetrability was observed and the grout flowed away.
The second mix was a higher viscosity
grout, and a reduction in penetrability
with continuing injection is seen. Except, unlike the idealized version in Figure 7 of [1], the penetrability suddenly
drops to zero as refusal is met. There is
no gradual reduction, and the recommendation to extrapolate the penetrability curve to zero flow rate is
erroneous.

Geotechnical News,

June 2008

We would also note that GIN recommends a single grout mix [7]. For the
grout stage shown in Figure 3, mix A
(the lower viscosity grout) flowed
away and continued injection of this
too thin grout would have been costly.
We believe a more economic solution
would have been achieved by modifying the mix and volume injected based
on water pressure test results.
Lombardis Figure 8 builds upon
Figures 6 and 7, so although the engineer knows that GIN and a nil flow
rate must be reached coincidentally,
there is no confidence that the methodology proposed for reaching this state is
achievable.

be achieved (without hydrofracture) the

more economic the project hence
there is significant advantage in minimizing the use of low flow rates. There
is really no reason to obtain a condition
of nil flow rate at the end of injection.
The question for the grouting engineer is the connectivity of the fracture
network intersected and the fracture apertures. Because each grout stage intersects a specific part of the fracture
network, with particular rather than average properties, an optimum grouting
approach is to relate the unit take criterion to the measured penetrability and
grout rheology. This is a very different
strategy to the GIN approach to grouting. It is not a question of misunderstanding GIN.
Despite the new clarifications of
what the GIN method comprises, the
basic limitation of GIN remains. Zero
flow rate is confused with the desired
radius of influence, often leading to extended injection periods at very low
flow rates to no effect. In contrast, basing a grouting protocol on the measured
penetrability of each grout stage provides the grouting engineer with a
clearly defined basis to bring a curtain
to closure. Necessarily this requires
some understanding of the fracture network being grouted, but we do not consider this a bad thing. Thus we assert
basing an entire grouting program on
GIN has the potential to mislead the
grouting engineer. It is not a misunderstanding of GIN GIN can indeed mislead.

A Misunderstanding of GIN? Or
Does GIN Mislead?

References

Lombardis reply to our publications

suggests that the lack of success of the
GIN method is solely down to a misunderstanding of GIN, and that any competent engineer should be able to understand and apply GIN successfully. We
do not agree.
The purpose of rock grouting is to fill
apertures within a pre-defined radius of
the injection grout hole, the radius being based on the chosen grout hole spacing. For a given radius and a set of fracture apertures, this equates to a grout
volume to be injected. Flow rate and
pressure are merely the means to get
this volume in place. The faster this can

1. Lombardi, G. (2008), Misunderstanding of GIN confirmed, The

Grout Line, Geotechnical News,
June 2008.
2. Rombough, V., G. Bonin, and D.A.
Shuttle (2006). Penetrability control of GIN mixes during fractured
rock grouting. Proceeding of
Sea-to-Sky Geotechnique: 59th Canadian Geotechnical Society Conference, Vancouver, Canada, 1-4
October 2006, pp. 528-535.
3. Shuttle, D.A., V. Rombough, and G.
Bonin (2007), Impact Of Grout
Rheology on GIN. Proceedings of
the mini-symposium on grouting,

THE GROUT LINE

Geo-Denver 2007: New Peaks in

Geotechnics, 18-21 February, Denver, CO.
4. Shuttle, D.A., V. Rombough and G.
Bonin, (2007) GIN Distilled. The
Grout Line, Geotechnical News,
September 2007.
5. Shuttle, D.A., W. Dershowitz, E.
Glynn, S. Burch, and T. Novak
(2000) Discrete Fracture Network
Analysis of Foundation Grouting.
in Girard, Liebman, Breeds and Doe
(eds) Proceedings of the Fourth
North American Rock Mechanics
Symposium, Pacific Rocks 2000. 31
July-3 August. Balkema, Rotterdam,
pp 1369-1376.
6. Ritchie, D.G., Garcia, J. P., Amaya, F
and Jefferies, M.G. (2003) Curtain
Grouting for the Antamina Dam,
Peru: Part 2Implementation and
Field Modifications. Proceedings
of the third international conference

on Grouting and Ground Treatment Geotechnical Grouting and Deep

Mixing. Edited by L.F. Johnson,
D.A. Bruce and M.J. Byle, pp.
929-940, 10th to 12th February 2003
in New Orleans, Louisiana.
7. Lombardi, G. (2003), Grouting of
Rock Masses, Proceedings of the
third international conference on
Grouting and Ground Treatment Geotechnical Grouting and Deep
Mixing. Edited by L.F. Johnson,
D.A. Bruce and M.J. Byle, pp.
164-197, 10th to 12th February 2003
in New Orleans, Louisiana.
Dawn Shuttle (corresponding author),
Consultant, White Rock, Canada V4B
1R2, dshuttle@civil.ubc.ca
Vafa Rombough, Junior Engineer,
Golder Associates Ltd, 500 4260 Still
Creek Drive, Burnaby B.C, Canada

V5C 6C6, vrombough@golder.com

Grant Bonin, Intermediate Engineer,
Golder Associates Ltd, 500 4260 Still
Creek Drive, Burnaby B.C, Canada
V5C 6C6
I have a final short comment about this
controversy about the GIN method.
The paper and further articles/comments from Dawn Shuttle and others are
directed only to criticize drastically the
GIN method. I have a short question for
them: what is their proposal for procedures, pressures, grout mixes, flow etc
for grouting in rock? Also this can be a
great contribution to the grouting community, and specifically to the grouting
engineer on the field!
It is easy to criticize, but what alternate do they have to offer?

Low Mobility Grout (LMG); Compaction

Grout (CG) Clearing the Confusion
Jim Warner
As so clearly illustrated by Vanderpool,
Norris, and Elfass in the March, 2008
Grout Line (Warning: Low Mobility
Grout (LMG) May Not Equal Compaction Grout (CG), not all low mobility
grouts can be effective for soil
densification by compaction grouting.
The ASCE Geo-Institute Glossary of
Grouting Terminology (2005) defines
low mobility grout (LMG) as Low
slump grout, such as compaction-type
grout, that does not travel freely and
that becomes immobile when injection
pressure ceases. Whereas compaction
grout is a low mobility grout with special properties, and is typically site
batched and mixed, much low mobility
grout is centrally batched and transit
mixed, being delivered to the jobsite in
ready mix trucks. Although this offers
the advantage of easy availability and
reasonable cost, it limits the grout composition to ingredients normally

stocked and used in the production of

concrete, which requires properties
quite different from those typically desired for grout. None the less, it is quite
easy to compound a low mobility grout
using available constituents, including
viscosity modifying and concrete
pumping aid admixtures. Though of
low mobility, most such mixtures will
behave as a fluid if injected in the
ground however, resulting in hydraulic
fracturing of the soil, and precluding
controlled compaction. Thus, such easily obtained and economical low mobility grouts are usually not satisfactory
for compaction grouting.
Compaction grouting involves injection of a special stiff mortar-like grout
into soil. It does not fill voids but rather
expands as a globular mass under pressure so as to uniformly displace and
densify the adjacent soil. Should the
grout behave as a fluid, hydraulic frac-

turing of the soil will likely occur, precluding controlled compaction. In loose
soils, some densification will occur
even with fluid behavior of the grout
and resulting hydraulic fracturing, but
the improvement will be only a small
portion of that obtainable by proper
compaction grouting, as so vividly
demonstrated in the cited work. To
achieve the compactive mechanism
thus requires a much more precisely
compounded low mobility grout.
This does not mean that plain low
mobility grouts should never be injected in soil. As pointed out by Gomez,
Robinson, and Cadden in the June 2006
Grout Line (Use of Limited Mobility
Grout for Shallow Foundations in
Karst) The LMG technique, often erroneously referred to as compaction
grouting, can help meet the needs of the
project and the engineer in a cost-effective manner and improvement can be
Geotechnical News,

June 2008

THE GROUT LINE

achieved through densification or displacement of the very soft soils (emphasis added) above and near the rock
surface (in Karst terrain). LMG supply
and injection is simpler and less costly
than compaction grouting and while it
will not provide near the degree of improvement to the soil, it is a valid application where very loose materials exist
and a high standard of improvement is
not needed.
Production of compaction grout that
will behave as a growing solid in the
ground severely limits the mix constituents, and any that might contribute to
fluid behavior must be avoided. Unfortunately, fluid behavior is the very property that is desired in concrete pumping
and thus provided by most viscosity
modifiers and concrete pumping aid admixtures. The Geo-Institute Glossary of
Grouting Terminology (2005) defines
Compaction Grout as Grout injected
with less than 1 (25 mm) slump. Normally a soil-cement with sufficient silt
sizes to provide mobility together with
sufficient sand and gravel sizes to develop sufficient internal friction to
cause the grout to act as a growing mass
as injection continues under pressure.
The grout generally does not enter soil
p o re s ( ex c e p t , p e r h a p s , wh e re
open-work boulder gravels are present)
but remains a homogeneous mass that
gives controlled displacement to compact loose non-plastic soils, gives controlled displacement for lifting
structures, or both.
Extensive research and experience
has found that merely providing a low
slump will not provide a suitable compaction grout, though many practitioners have failed to recognize this. The
case described by Vanderpool, Norris,
and Elfass is unfortunately a common
example of performance, and had the
researchers not recognized the inappropriate grout material used in the first instance, and insisted in retaining a
competent contractor to re-grout the
pile tip, it would no doubt have been
incorrectly concluded that compaction
grouting was not capable of improving
shaft capacity.

History of Low Mobility Grout

Use

As a young contractor in 1952, I had

constructed a small structure in a flood
control channel. Unfortunately, some
sand filter material eroded from under
the structure in an early storm. The resulting void had to be remediated, and it
seemed reasonable to fill it with a cement-sand mortar. Neither concrete nor
mortar pumps had yet been developed
so a crude hand powered piston pump
was fabricated and the void was filled
using a typical masonry mortar. This
early experience led to development of
a powered piston pump and further
work filling sub-surface voids. As experience was gained, it was found that
structures as well as pavements could be
accurately lifted or jacked to proper
grade by such injection. The stiff mortar
would tend to stay near its point of injection allowing jacking of even differentially settled and heavy structures to a
close tolerance.
An ever expanding range of applications for pressure injected mortar (now
LMG) evolved, all the way from filling
small voids in concrete and masonry, to
filling of large tanks, abandoned pipelines, and annular spaces, Figure 1.
With the expanding range of application, a variety of different mix designs
were made, but always of a low mobility

plastic consistency. Fortunately, I was

unaware of pressure grouting as then
practiced and most importantly the well
established rule that if you cant pour
it, you cant pump it.
In 1957, our now established grouting firm was engaged to raise one end of
a swimming pool in a courtyard completely surrounded by an apartment
building. The site had formerly been
home to a hospital and the pool was located in what was previously a basement boiler room. The plans called for
the bottom slab to be perforated and the
entire basement to be filled with compacted granular soil, however, the slab
had apparently not been perforated and
the fill was neither granular nor compacted, but a low cohesion clayey silt.
The deep end of the pool was situated at
about the old floor elevation, while the
shallow end was founded on several feet
of the fill. The shallow end settled causing cracking of the pool shell, allowing
pool water to saturate the fill which was
contained in the old structure.
Grout holes were placed on a grid of
about four feet throughout the bottom
of the pool, Figure 2. The geotechnical
engineer had requested holes in the
shallow end be extended to the old floor
slab. Recognizing the then prevailing
opinion that fine grained soils couldnt
be grouted, he reasoned a number of

Figure 1. Concrete pumps had yet to be invented in 1960 so four custom pumps
were used to place read.
6

Geotechnical News,

June 2008

THE GROUT LINE

Figure 2. Water squeezed from the soil ran into pool as fast as
grout was injected (1957).

Figure 3. Extraction of injected grout masses.

small holes filled with grout might provide enough support to prevent further
settlement. Water ran from the lower
grout holes into the pool as soon as injection began, and the inflow continued
for several hours. The job required an
amount of grout equal to more than ten
times the calculated volume, before lift
occurred. A horrible mess was made by
the removal of the muddy water with
buckets which had been spilled all the
way from the pool to an outside catch
basin. Everyone involved was obviously upset and not pleased except for
the geotechnical engineer, who exclaimed you have squeezed the water
out of the soil
.this is not supposed to be
possible
.its wonderful. The only explanation was that the stiff grout had
somehow expanded in masses that
forced the soil grains into tighter packing, expelling the water that was
otherwise trapped in the old basement.
To better understand the phenomena
a test program was conducted wherein
trial injections were made and extracted
for evaluation, Figures 3. Several hundred holes were injected using a variety
of different grout mixtures and injection
parameters. The grout mix composition
was found to have a major influence on
the shape and volume of the resulting
injected mass. In general, the best results were achieved with the stiffest,

harshest consistency grout that could be

pumped. Conversely, poor results were
experienced with easy to pump grouts
containing clay, entrained air, and/or

pozzolanic materials such as fly ash,

with frequent fracturing of the soil and
leakage to the surface.
Because the nature, and gradation of

Figure 4. Original Brown and Warner envelope of acceptable gradation for

compaction grout aggregate.
Geotechnical News,

June 2008

THE GROUT LINE

Figure 5. Denver Grouting Services performed many large scale demonstrations in

the 1990s to evaluate grout performance with varying aggregates, additives, and injection methods.
the aggregate was found to be a significant contributor to the shape and volume of the injected mass, trial injections
were made using aggregates and aggregate combinations from many different
sources. A particle size distribution envelope of gradation that provided good
performance was prepared and is provided here as Figure 4. Clay or other
pumping aids which were found to
cause fluid behavior were prohibited.
The results of the test evaluations including the acceptable envelope of aggregate gradation were reported by
Brown and Warner (1973).
Several research efforts of Denver
Grouting Services in the 1990s also included retrieval of grout masses result-

ing from different aggregates, additives,

and injection procedures (Figure 5).
These generally confirmed the data
from the earlier Brown and Warner
work but also determined that injection
control could be improved and the deleterious effect of clay lessened by extending the top aggregate fraction,
in c lu d in g g r ave l s iz e p a r ticles
(Bandimere, 1977). The upgraded envelope is provided here as Figure 6.
Many thousands of projects have been
successfully completed with grout containing aggregate falling within the limits of both gradation envelopes. While
some of the most experienced contractors continue to use the original Brown
and Warner aggregate, the more recent

Figure 6. Preferable gradation envelope for compaction grout aggregate.

Geotechnical News,

June 2008

gradation envelope containing gravel,

has proven to be even more suitable and
is recommended in sensitive situations
where high performance is required.
Examples would be when grouting in or
near unrestrained downslopes, or in water retaining embankments where hydraulic fracturing of the soil cannot be
tolerated. Most compaction grout is a
simple combination of the appropriate
aggregate combined
with 8% to 12% common cement.
No other constituents or admixtures are
needed or should be used. Cement is not
an absolute requirement, but if eliminated, a larger fraction of silt size particles will be required.
In addition to the above formal research, many hundreds of injected
masses have been extracted or exposed
in a variety of different soils and using
grout of varying composition. The results have been amazingly consistent;
grouts containing the appropriate aggregate create massive inclusions in the
ground as illustrated in Figure 7.
Conversely, inclusion of clay or
pumping aids nearly always results in
hydraulic fracturing of the soil with thin
wings of grout extending out from the
point of injection as shown in Figure 8.
It is unfortunate that the reputation
of compaction grouting has been badly
maligned due to poor work from
grouters that refuse to follow the well
established material requirements and
use an appropriate grout mix. As in the
case reported by Vanderpool, Norris
and Elfass, many contractors, especially large geotechnical contractors
claim to be experts in compaction
grouting but once on a project, resort to
ready mixed grout with inappropriate
aggregate, high levels of round grained
fly ash, and a variety of additives such
as bentonite, chemical viscosity modifiers and concrete pumping aids. The result is all too often, insufficient
improvement of the treated soil and a
bad name for compaction grouting.
Though some commonly claime that
conforming aggregate is not available,
responsible compaction grouters have
managed to procure acceptable
materials for projects around the world.

THE GROUT LINE

is nearly always obtainable. In this regard, I have hauled silt considerable distances, but as it comprises only 15-30%
of the total aggregate mix, the cost is
still manageable. Acquiring the appropriate aggregate does require more effort than simply purchasing a ready
mixed compound but it is requisite to
proper performance.
Summary and Conclusions

Figure 7. A variety of massive grout shapes resulting from aggregate conforming

to the gradation envelopes in Figures 4 and 6.

Low mobility grout is any composition

that does not travel freely and that becomes immobile when injection pressure ceases. It is often batched and
supplied from ready mix plants. Compaction grout is a special type of low
mobility grout that is compounded so as
to not behave as a liquid when injected
into soil. There is no question that acquiring proper aggregate for compaction grouting frequently requires more
effort and cost, however failure to do so
will result in a high probability of unsatisfactory performance, and should not
be allowed.
References and Further
Reading

Figure 8. Hydraulic fractures result from grouts with clay or pumping aids.
So How Do We Find Appropriate
Aggregate?

Silty sand which is the basic material required, will virtually never be found at a
ready mix plant because concrete constituents must be clean and free of fines.
Occasionally, producers will have silty
sand overlying cleaner aggregates in a
quarry. Perhaps the best place to locate
appropriate material is top soil suppliers
to the landscaping industry. They not
only often stock satisfactory silty sands
but nearly always have portable screening plants that can be used to remove
oversize particles from natural deposits.
On many projects, I have found the na-

tive site soil appropriate by simply

screening out the oversize fraction.
Whereas locating satisfactory silty
sands free of clay and conforming to the
Brown and Warner envelope is usually
possible, finding such deposits also
containing pea gravel is much more difficult.
Gravel is nearly always stocked by
aggregate producers and ready mix
plants however, so it is readily available,
and can be easily blended with the basic
silty sand. In those rather rare instances
where silty materials free of clay are not
available, it might be necessary to separately mix silt into common sand which

Bandimere, Samson, (1997) Compaction Grouting; State of Practice

Grouting: Compaction, Remediation and Testing, Geotechnical
Special Publication No. 66, ASCE,
New York, pp. 18-31.
Brown, D. and Warner, J., (1973)
Compaction Grouting, Journal of
Soil Mechanics and Foundations Division, ASCE, Vol. 99, No. SM8,
Proc. Paper 9908, pp. 589-601.
Garner, S., Warner, J., Jefferies, M., and
Morrison, N., (2000), A Controlled
Approach to Deep Compaction
Grouting at WAC Bennett Dam,
P ro c e e d i n g s , 5 3 rd Ca n a d ia n
Geotechnical Conference, Montreal,
Quebec, Oct. 15-18, 2000.
Gomez, J., Hobinson, H., and Cadden,
A., (2008) Use of Limited Mobility
Grout for Shallow Foundations in
Karst, Geotechnical News, pp.
51-56.
Grouting Committee of the Geo-Institute (2005) Glossary of Grouting
Ter m in o lo g y , Jo u r n a l o f
G e o te ch n ic a l a n d G e o e nvironmental Engineering, Vol. 131, IsGeotechnical News,

June 2008

THE GROUT LINE

sue 12, pp 1534-1542.

Warner, J., Compaction Grouting
Mechanism What Do We Know?
(1997), Proceedings, Grouting
Compaction, Remediation, Testing,
Geotechnical Special Publication
No. 66, ASCE.
Warner, J., Schmidt, N., Reed, J.,

10

Geotechnical News,

June 2008

Shepardson, D., Lamb, R., and

Wong, S., (1992), Recent Advances
in Compaction Grouting Technology, Proceedings, Grouting, Soil
Improvement and Geosynthetics,
ASCE, New Orleans, Geotechnical
Special Publication No. 30, pp.
252-264.

Warner, J., (1992), Compaction Grout;

Rheology vs. Effectiveness, Proceedings, Grouting, Soil Improvement and Geosynthetics, ASCE,
New Orleans, Geotechnical Special
Publication No. 30, pp. 229-239.