6.41 Grouting design and control using the GIN principle By G. Lombardi* and D. Deere** cement grouting of rock masses is derived and presented. The main features unique to th ‘method are: (1) a single, stable grout mix for the entire grouting process (water:cement ratio by weight of 0.67 to 08:1) with superplasticizer to increase penetrability; 2) a steady low-to-medium rate of grout pumping which, over time, leads to a gradually inereasing pres- sure as the grout penetrates further into the rock frae- tures; (3) the monitoring of pressure, flow rate, volume injected, and penetrability versus time in real-time by PC graphics; and, (4) the termination of grouting when the grouting path on the displayed pressure versus total volume (per metre of grouted interval) diagram inter- seets one of the curves of limiting volume, limiting pres- sure, or limiting grouting intensity, as given by the selected Gin hyperbolic curve (a curve of constant p:V, pressure x volume, a measure of energy expended). Experience in several countries at major hydroelectric projects indicates the method to be technically and eco- ‘nomically effective, TT: Grouting Intensity Number (Gin) method for ‘The grouting of rock masses with coment slurries to improve their mechanical and hydraulic properties is a well established practice in civil engineering. This practice, Jong been dominated by rules-of-humb and tilutional experiences, often leading to dog. ‘matic belies, During this decade, at a number of major hydroeleettie projects under consiruction in different countries, the authors have had the opportunity to work with designers, {geotechnical engineers and geologists, and field control engineers in the development of a better understanding af the grouting process by 3 combination of laboratory, theo- retical, und field investigations. Information gained by laboratory studies of the cohesion (yield strength) and dynamic viscosity of different grout ‘mixes, by theoretical studies of grout flow and penetration, id by field monitoring of grout pressures and absorptions (takes), has led to the concept of Grouting Intensity Num: ber (civ), as will be discussed in this paper. For simplicity, this paper will deal only with cement grouting of rock masses, although some of the consider tions presented may also be applied to granular sols and to ‘grouting with other materials. More empiasis is given « ‘grout curtains for dams than to consolidation grouting o- Underground works, in spite of the fact that the GIN method has also been used a few times for these types of works. is not the purpose of this paper to review the present state of the art in grouting, as a number of recent books hhave done that quite well", Rather, a brief theoretical back- _ground will be given, followed by both theoretical and prac- tical concepis leading to the development of the oiN _grouting procedure. The proven effectiveness of the concept in achieving better grouting results and simpler and more {economical grouting operations has suggested to the authors that the method should be presented toa larger audience. 3 Gls Pn 567, hiding Goan oman, {ie coca Stirond Coins lon US Theoretical concepts of grout flow and penetration Unlike Newtonian fluids, such as water oF oil, where reo: logical behaviour can be characterized solely by the param- eler viscosity, a ‘stable’ grout slurry behaves as a Bingham Muid during flow. possessing beth viscosity and cohesion (yield stengih). While both are flow-resistance parameters, the viscosity governs the rate af flow, while the cohesion governs maximum travel distance (For # given applied grouting pressure and a given aperture of rock fis sure). Equations have heen developed sind presented else- ‘where "for computing the maximal travel distance, the ‘maximum volume of injected grout, and the maximum total hydrojacking force exerted on the grouted are nis sufficient here to note that the maximum travel dis- lance achieved by the grout slurry is directly proportional lo the applied grouting pressure and to the aperture of the Fissures, and is inversely proportional to the cohesion of the grout slurry. Thus, for enhancing grouting penetration into the fine rock fissures, it is necessary to inerease the grou 1g pressure of to reduce the grout cohesion, o both, Perhaps the greatest value of the Bingham flow equi- tions is in providing insight into the grouting process, with respect to the factors that influence tke penetration af grout iio a rock fissure and the extent of the splitting forces ere- ated by the grouting process. While stable grouts (defined as those that exhibit in 2 less than 5 per cont of deeant tion of clear water at the top of a 1000 ml eylinder) may ‘proximate the behaviour of a Bingham fluid, itis not sonable to apply the equations to thin, waKery grows. Such thin mixes or slurries are unstable suspensions of cement Particles in water that, during grout flow through rock fis- sures, may be expected to show erratic sedimentation, ero ‘e-suspension. and re-scdimentatinn behaviour, This jour is impossible t predict and characterize with the Bingham flow equations ar any others, In the following section, additional factors that enter into the selection of a design grout mix are presented. Selecting the grout mix The controversy over thick-versus-thin grout mixes (slur= ries) will certainly continue for several more years. The authors, since 1985, have expressed their preference in the literature for thieker mises "The practice of adding 1 to 2 per cent of bentonite, for stabilizing the mix and for ‘educing sedimentation, s progressively being replaced by the use of mixes of higher cement conlens But with super plasticizer additives. These latter mixes are stable, and iecause they possess Tess cohesion, they are moce penitra- ble they also have greater strength upon setting Traditionally, advocates for thicker mixes have noted the several advantages that thick mixes exhibit both during the routing process and during the service life ofthe grouted Fock mass ater the prou has hardened During grouting, a moderately thick, sable grout has the following advantages compared with thin grout ¢# less sedimentation of cement grains during low-flow conditions:‘less bleed water to accommodate as a result of squeezi ‘out oF filtration at narrow zones in the flow pathways, with less premature blockage: fe greaier slabilily over time and distance as a predictable Aid (Bingham fluid with a given cohesion and dynamic viscosity); and, fe less risk of hydrofracturing (also termed hydrospliting, or hydrojacking) and uplift of geological strata, because of fast pressure drop away from the grout hole as a result of the grout cohesion (and the filling of the fracture with a high quality grout in the ease of such an occurrence. During the service life of a hardened grout in rock fis- the thick grout has the following advantages as com- pated with thin grout f less shrinkage during setting, and thus a greater bond along the rock fissure walls and less risk of re-opening; f greater density and higher mechanical strength because of greater cement content, and thus a greater resistance 10 physical erosion and piping: f less porasity, lower permeability, and greater bond strength, and thus a greater chemical resistance to leaching, ida greater durability of the grout curtain over the Tife= time of the dam, Stable grout, because of its cohesion, requires higher outing pressures to reach the same distance of travel compared With thin grout, However, bythe use ofa small mount of superplasticizer, both its cohesion and viseosity- parameters can be dramatically lowered. Curren practice a number of major projects is 10 se a mix ratio of 0.67:1 to G.8:1 (water: cement by weighl) to ‘obtain the desirable higher density and strength ofthe hard- ‘ened grout, and a superplasticizer 1 reduce the cohesion and viscosity during grout placement. Laboratory tests are used (0 determine the flew, sedimentation, setting and strength properties of different grout mixes for different ‘cements and different superplatiizers. Cohesion and viscosity values may be obtained in the Iuboratory using a rotary viscometer with concentric cylin- ders. However, the simple plate cohesion meter may also beused™, This is pices of roughened ste! plate, 100 mm by 100 mm by about 1.5 mm in thickness, which is ‘weighed before and afer diping it for afew seconds into the grout mix. The difference in weight divided by the arca Of the two sides gives the cohesion parameter in shear Steength unils is convenion! to divide the cohesion C by the unit weight of the slurry 7 giving the relative cohesion r= Ch, whieh is the one normally reported. Ic is com- ‘monly given in mm, and typical values are 0.2 to 0.35 mm for thick slurries without superplasticizers. With additives, the apparent cohesion deops to values of 0.08 to 0.15 ram, the recommended range. I is of interes 19 note tht the C> value isin Tact the thickness ofthe rou sutry clinging to tach side ofthe cohesion plate. The cohesion plate must be sulficiently seouted by the cutting of intersecting thin grooves ofits surface, so thatthe adhesion between the Steel surface and grout is greater than the cohesion between the surface layer ef adhered grout and the remaining grou; otherwise all the grout will slide of Other typical laboratory values are: specific weight of the grout, 1.59 t0 1.67 Um? (99.2 to 104.2 Ibi’); Marsh funnel flow time, 29 10 32 s; and, 28 day compressive strength, 15 10.20 MPa (2250 to 3000 ibn’ ‘Another point to Keep in mind is the potential loss of ‘water in grouting dry rock above the water table, Were grout to be thickened for this reason, its cohesion would imerease (and internal rition would build wp) to the point ‘where no more grout could be injected. A pradent practice is to inject water fora period of time, to obtain partial satu- ration of the rock mass just before grouting. Water- retaining admixtures may also be used in the grout. fissures in rock, it should be remem- ‘bered thar penetration of the grout depends more on the size ff the cement grains and flocks than on the dilution of the mix with excess water. Thus, instead of trying to obtain ighcr penetrability by diluting the mix, a finer cement with superplasticizer, together with higher grouting pressures, should he used. Once an acceptable mix has been determined by labora- tory tests, with fis mechanical properties in the desirable ranges, that mix should be used forall the grouting at a pro- Jeo she use of a single mx relly simples the growing procedure Design of grouting works “The design of a grout curtain includes the selection of the following main features: the grout mix, the grout hole spac- ing and dept, the grouting sequences, the grouting proce- Gute (including volume and pressuee limitation) and the field control. A good knowledge ofthe site geology is required in selecting some of these parameters, in paticu- the physical characteristics ofthe rock mass discontinu- to be grouted (Iypes, frequency, fissure aperture, roughness, alteration or iofilling, and exiension). The ins Si sate of sress and the existing groundwater conditions should also be considered In addition o a knowledge ofthe existing geologieal and geotechnical conditions, the changes induced by the project {nthe sires slate, snd hyeraufie pressures and thei val tion with time (sich as during impounding and drawdown Gf the reservoir) must be taken into account. Finally, the foal to be achieve in terms of consoldation or tightening effects should be defined better than is generally the case. “This paper does notatempt to address all these points in detail, but concentrates on several of the more important Development of the cin method. Grouting of wide, open fissures Both practical observations and theoretical studies indicate ‘ha the wider, open fissures in a rock mass are those most readily grouted, The grout travel can also be considerable {a few fens of metres). These ate both practical and eco- Tomical reasons, therefore, to reduce the grout travel and the grout volume injected. There ate three Ways to accom- plish this reduction: by using a less penctrable grout {thicker with greater cohesion}; by limiting the grout pres- sure; or, by limiting the volume of grout injected. Before selecting the limiting criterion, one should con- sider that finer cracks may also exist in the rock interval being grouted, These are more difficult to grout and proba 4 later stage, when the more open, wider cracks have been filled. Nevertheless, uring the fest stage of grouting, tis desirable 10 achieve some filling of these fine cracks. Thus, the grou! mix should not be thickened, bul should remain a moderately thick stuble grout with the superplasticizer additive. The altemative of limiting the pressure is also not very atene= tive, a this would also discourage the grouting of Fine fs- sures, The remaining alternative of placing a volume Timitation appears bs the best approach. “For csp, STA of gt pe mae gf bree (SISS pee)e Shae ig rd {ts weight sami wer: cee ao [£075 wei 18-30 higcomen pr ect fra (ae pt eae.Grouting of fine fissures Aer the wider fissures have been grouted, or if open, ‘wide-aperture fissures were nol present originally in the Zone being prouted, is he grouting of finer issires that ‘becomes the priority. Such grouting can be enhanced cither by using a thinner mix with a lower cohesion, of by groute ing at higher pressures. tis moze convenient to rise the pressure and 10 mainiain the high-quality moderately thick rout mix with the superplasticizer. Beesuse the finer Sures Will have Tess grout travel, and because the grout Pressure diminishes rapidly asi spreads away from the Borehole, the total uplift foree, even at high grout pressures, will, asa rule, be much lower than the overburden weight, hydrofracturing of rock joints and bedding planes is seldom a problem (except in the upper 3-10 m). Consequently, {quite high grout pressures are acceptable, even up to 30 0 40 bar, that is 3 fo 4 MPa (425 to 870 livin?) computed at the grouting interval, provided thatthe grout take is small. 1m considering these ators, an upper limit of prouting pressure is suggested where the grout fakes are smal, in the Fange of 30 1030 bar (3 to 5 MPa), depending onthe geol- ogy (weathering, stratification, weak zones, in-situ state of stress, and soon}, the future water pressure, and the desired intensity of grouting. “The first Iwo elements of the civ principle are thus beginning lo emerge: a volume-linilaion whore the grout enters casily at low pressures, and a pressure-limitation Where the grout enters only with difficulty. I re Tocus on the intermediate ranges. However, before pursuing this intermediate range, itis benelicial to review other routing considerations. Split spacing of grout holes in series In usual practice, primary holes are grouted fist, spaced arly widely (such as 10-12 rm) 30 thal grouting atthe fit rary hole does not interfere with the next. Often, itis Specified that every third or four primary hoe be eilled And grouted before the other primaries, to serve as “exploratory”” primary holes, These holes will often be cored holes, and will be water-pressure tested (0 total depth of 0.75 XH (where H is the height of the future reser- ‘oir atthe point in question). The remaining primary holes may be adjusted in-depth according to the results of the exploratory primary hoes. ‘The next series of holes, the so-called secondary grout holes, are growled next in splitspaced location beiveen the primaries, Since these holes are only or 6m from the primary holes, dhey wil often encounter hardened grout in some of the wider fissures. In general thee ‘Iakes” will be lower than the primaries. Tertiary holes are often called for, again ata split spacing (2.5 0 3.m), usually with even lower takes; eventually, even quaternary holes may he called for (1.25 to 15 mi from each tertiary hole), resulting normally in acceptably Tow final absorption of grout Since the rock gels lighter with each phase of holes, the ungrouted fissures eneountered in the later holes will mostly be finer fissures; higher pressures would thus be of benefit and would produce a more efficient grouting opera- tion “Thus, the third ingredient of the GIN procedure is idemi- fied: a progressively higher pressure as the rock tightens up, $0 a8 fo grout the progressively finer fissures. ‘Water pressure tests (Lugeon) Water-pressure ests (Lugeon tess) have often been used at each grouting interval to help select the grout mix. How ever, experience and theory hive indicated very poor eore- lation of Lugeon values with grout absorptions. Lugeon or similar tests may still be used, however, in the exploratory primary holes, to obtain a general picture of the permeabil- ity throughout the abutment and main dam foundation area Comparisons may then be made with Lugeon tests con- ducted in check or control holes after the grouting is done to see if a sufficient reduction in permeability has been accomplished. The GIN method Specific energy expended To grout a rock zone more intensely, one has to expend mote energy. In a given grouting interval, the energy ‘expended is upproximaely proportional to the prove of the final grouling pressure p'and the grout volume injected V, giving a product p-V. This p-V number is called the Grouting Intensity Number or oi, The volume may be uni tized fo the length of the grouting interval as lites/m (or interchangeably as weight of cement injected in ky, since for moderately thick mixes the numerical values ae similar within 5 to 10 percent). The pressure has been used ta nally as bar, resulting in p-V or GIN units being resi, although other units obviously may be used by snaking the appropriate conversions Advantages of maintaining a constant ciN “The cin method requires tht once a grouting intensity level fas been chosen (for instance, high intensity ctv of 2000 barl/m) this value should be used both for the easily {routed fissures, with large volume absorptions at low pres- Sur, as well os for finer fissures, sith low fakes but with considerably higher pressures. Ths, « constant valu of the ‘iv is maintained. By Keeping the Giv value constant during the grouting process fra grout interval, one obtains a aerly constant Teach of the grout, and almost automatically limits the vo time in an open, wide fissure, but allowing pressures 10 increase in tones of lighter and less grodtaple fissures. ‘Ais, combinations of high pressure and high volume are sliminated a condition that would lead to huge uplift or hydtospliting forecs. Similarly, combinations af lw pres sire with a fow take are eliminated, that would inade quately grout the Fine fissures ‘A constan! GIs valuc, when plotted on a pressure-versus- volume graph, yields hyperbole curve! the higher the routing tensity ar cat vate, the greater the distance of ine curve frm th origin, The cl curve thus completes the missing ingredient for joining the ather two limits dis Cussed in previous sections: the limiting volume line and the limiting pressure line. The combination of the three fives the composite limiting envelope for grouting. Composite limiting envelopes for different intensities of grouting la Fig 1 the p-V paths of thee types of fissures are shown, ‘Aa open fissire of wide aperture is represented by curve 1, Which shows increasing volume of injected grout at only Slighily increasing pressure. Curve 2 represents an average fisure, where the pressure gradually inreases asthe vOl- ume of grout injected increases; only at point a! does the pressure rise more rapidly, as the resistance to grout pene- tration inereases disproportionately. Curve 3 represents a tight fissure, where the grouting presse vises rapidly with nly small grout takes. Similarly, curve 4 represents un extremely tight fissure, with a very small grout take und high refusal pressure.Fig. 2 illustrates five suggested limiting envelopes for Afferent intensities of grouting. The GI Value, dhe limiting pressure and the limiting volume are in fact three more oF fess independent parameters defining the liming envelope for grouting. In the proposed definition, they are related 10 each other, but they need not be. The uppermost envelope represents avery high intensity of grouting with a ON valve 62500 barn, very high limiting pressure of SO bat, and lope Fepresents a very low intensity, with a GIN value of 500 barlim, a maximum pressure of 1S bar and a limiting vol- ume of 100 I/m. For most conditions, the authors would recommend the moderate intensity envelope with a given value of 1500 bar-/m, a limiting pressure of 30) bar, and a limiting volume of 200 Vr For geologically critical areas (near the surfuce and ot steep slopes) the very low curve could apply. It is easily understood, from the combination of Figs. 1 and 2, that the ‘routing process will stop at different poinis depending on the value selected for the grouting envelope. For curve 1 (Fig. 1) the grouting would be terminated at point a for a low intensity of grouting, because of the 150 lim volume-limitation with a resulting pressure of 3 bar. However, ifthe designer had specified the moderate inten- sity envelope, the grouting would continue to point b, the 200 i/m volume limit, the final pressure again being about 3 bar. If high intensity grouting lad been selected, the groul- ing would continve to point c, the 250 lm volume limit the final pressure being about 6 bar. Finally, i€ the very high intensity had been selected, the grouting would continue to point d. The path would not be terminated by the volume limit, but rather by intersecting the 2500 Gtw curve. AL this pon, the (otal injected volume would be 285 Vm and the final grouting pressure would be 9 bar. Thus, there would bbe a range of volume injected of 150 to 285 Wm and a range of final grouting pressure of 3 to 9 bar depending on the specified GIN, For curve 2 (Fig. 1) the volume injected st point a’ would bbe about 60 lim, and the final grouting pressure would be 13 bar. Had the grouting continued to point «representing very high intensity (p-V=2500), the grout take would have increased only 10 90 Vm, but the pressure would have attained 28 bar, considerably less than the limiting value of 50 bar. Similatly, for curve 3, grouting would be halted at any point between a and a" depending on the previously selected ain criteria and the pressure limitation. The portion -y represents an example of a hydrofraciusing or hydros- pliting event, where a joint or bedding plane is suddenly forced open, with a resulting pressure drop and an increase in the absorption rate. One could continue grouting al low rates 10 try (0 arcive at the selected GN curve, but if the pressure increased, there would probably be another hhydrofracturing event at about the same pressure or a litle higher. It is questionable if grouting should be continued in such circumstances. Many times, however, it has been done without any major problem. In any case, the limiting curve is arrived a, sooner or later, and the grouting is stopped. Selecting the civ value ‘The dam designer and his geotechnical and grouting stafT must select the Gin value for the projected grout curtain. While the authors recommend the moderate GIN value of 1500 barra as a starter, geological conditions, the value of fature water losses, and uplift pressures after impounding, are points 10 be considered. The upper pressure limits and the upper volume limits may also be modified by the designers and field control engineers for special cases. Probably the best approach is to conduct one or two grout avian ota ieee) [anv pty average rose -[PeV Fah te tecve path atl tt yotucing ‘Fea gon aD value ous Fig. 1. Example of grouting path carves. | (omg AS see ee os a *ouine of grt need ¥ three) 0G hams ounce v (a) Fig. 2. Proposed limiting envelopes for grouting. ‘est sections before selecting the limiting grout envelope. “The upper pressure limit may be less in the abutments ‘than that selected for the valley bottom, because ofthe dif- ference in reservoir depths. A worihwhile goal isa limiting pressure of at least wie the reservoir pressure, but it may be difficult to achieve this without inducing undesirable hydrojacking, Computer controlled grouting Since re computers can nowadays be installed and easily ‘operated at the grouting site itself, the possibility now exists of continvous real-time controlling of the grouting process. Only vo values need be continuously read by the System: the delual urouting pressure p and the Mow rte 4, from whieh the cumulative volume V per unit length {injected since the beginning of the grouting ofthe depth- interval in question) can be obtained by integrating. If esited, the volume V of grout injected can be measured directly, and the Mow rate obtained by derivation, Starting from the Wwo measured Values of p and 4, 2 twmber of ine graphs ean be shown on the monitor sereen example, pressure, low rate, cumulative (versus time) Fig. 3 shows such a is, the flow rate divided by pressure) versus time. This curve shows eleaely the progressive filling of voids and fissures ‘and the buildup of grouting resistance, caused primarily by the increasing total cohesive force as the grout flow exiendsstopped at *'zer0"” flow rate, with a bh (pote peene. (ot own) (apse. 2 0 Ont final pressure p,-and total cumula- tive unit volutie of injected grout Vp. Fig. 4(b) is also an important real-time monitoring curve. The penetrability qip is plotted versus cumulative grout volume rather than versus time as in Fig. 3(€), giv ing similar type curves, however. AS the curve develops, one nor rmally sees a decline in the penetra- bility, indicating that the grouting elficiency is decreasing. Thus, al & constant grouting pressure, the flow rate is decreasing or, if a'con- PE eee ee ton none lf Serr on O° aking fae. tings dros sur: (grout is lows () rome | stant flow rate is maintained ings and, P= {almost to the end), the grouting pressure is increasing. Which of further along the rock fissures. A hydroftacturing or hyn jacking event is readily observed from the presence of a sharp peak in the graph, The Gin and penetrability-volume curves as practical grouting controls Fig. 4 represens the main contol plots thatthe computer may display. These curves may be used to control the {grouting process. Fig. 4(a) shows the limiting grouting Envelope that has becn chosen forthe project (or for & ziven area of the project), including the Timiting pressure ‘Peo. the limiting Volume Vu. pe unit length, and the flected civ hyperbolic curve, This envelope may be writ ten into the program and ean be displayed upon call In general, these limiting envelope curves are refetred to sim- ply as the cm curves, "The iregular curve 2 shown on Fig. (a) represenis the ctu grouting path, ploted at small increments of time, of the instantaneous grouting pressure versus the cumulative grout volume per unit lengih, The grouting-path iregulur Curve inerseets the GIN eurve al point F, and grouting ey pmetatny cane 4 Grating procs: of sate borehole sane (pea, fre: 1 lining ea rere t2 st [Pong ret tant Po grt ie tet (ip) versus grou tae. F= nal pont ofthe growths; Py = final _oul pressure; and, Vi= actual grow ake. these combinations applies is dependent both on the type of pump and the details of the grouting operation (piping, valving). ‘The decline of the penetrabilty-volume curve indicates that the grouting process is proceeding normally. The ‘grouting pressure must be monitored and conirolled to stop the process at the grouting limits ofthe Giv eurve, ‘As noted in the previous section, the grouting path ‘meet the GIN curve at different points, a funetion of the aperture ofthe rock fissures (wide fissure necr point B and fine fissures near point A, Fig. da). As the grouting path progresses toward the GIN curve, the pumping rates should bbe as low as practicable while sil achieving grout penetra- tion (for example, 500 Im, that is 17.6 f¢/h oF 2. Experience has shown thal these various plots aid greatly in controlling the grouting process in an effective and continu- Application to grout curtains ‘The GIN method as presenied has dealt primarily with con- siderations for a single grouting interval or stage, The ‘method also applies to all intervals ofthe grout hole, and 0 all the primary and splitspaced holes. The spli-spacing procedure in asingle-line grout curtain isa proven efficient method, with sound theoretical reasoning In certain highly permeable or cavernous rocks (lava Rows, some limestones, fractured sandstones), 3-line grout curtains have been constructed, The downsiream line is usually grouted fist followed by the upstream Tine, and finally by the central line, The wo outer lines are consid- ered (o behave as barrier ines, and often only primaty a Secondary holes are grouted, with the intent of filing the majority ofthe larger fissures or voids. The ental line can then be treated as # normal single-line custsin, with pri- maries through to teriaies, and even with quatemnary or dquinary holes if needed {In the split-spacing method, the primary holes will par- ‘ally oF completely itl and plug ony the widest ofthe rock fissures. The next series of secondary holes again, will plug oniy the widest fissures not yet plugged In the first Series, and so on. In Fig 5, the likely fnalpestions of the primary, secondary, terry, and additional holes (quater- nary of check holes) are plotted on the GIN eure, The aver- age grout volume absorbed will decrease from series to series, while, obviously, the average final grout pressure will increase accordingly from series (0 sees. ‘This hap pens automatically when the cit procedure is followed. With a primary hole spacing of 10 10 12 m, i is likelythat both secondary and tertiary series would be required. ‘The tertiary holes would be at 2.5 to 3 m distance from the nearest adjacent hole; these holes could be shorter, depend ing on the geology and the resulls of the secondacy holes, ‘Quaternary holes might or might not be needed, At least some would be required as check holes for performing Lugeon tests, to see if an acceptably low permeability of the rock mass has been achieved: for a tight curtain, requirements may be as severe as 90 per cent of all tests be at or below one Lugeon (1x 10 em/s), with no values ‘greater than 3 Lugeons. Relationship of hole spacing and cin Itis obvious that a relationship must exist beoween the hole spacing andthe GIN required. For instance, if the selected primary spacing is too wide andthe selected GIv is too low, no signifeant decrease of grout take will result from ie Primary series fo the secondary, or even tothe tertiary. In Such a ease, no guarantee for successul curtain cam be given, even though considerable drilling and grouting costs irae been expended ithe primary spacing is too close or the iN is too high after the frst two sees, dhe grout takes will be very low, ad teriary holes would be waste essentially. The cin fy so related tothe distance that the grovttravols and, there~ fore tothe thickness ofthe prouted rock curain or "wall ‘A working rule-of-thumb is 10 select the Values of Gi and the spacing so that the volume of injected grout per metre of Stage grouted reduces from hole scties to hole Series by around 30 per cent (reaisially inthe range of 2 {0 75 per cent). Such behaviour would provide confidence that progressive closing of the curtain i occurring, One or more est grouting sections can be utilized during the design phase or atthe fist part of the grouting contract 0 define better the optimal primary hole spacing and GIN value Criteria for closure If the prout paths for the holes of the,last series (say, the tertiary series) do not arrive at the upper limiting pressure Jine for the selected cin (and preferably in the leftchand half of that fine), additional holes should be grouted on either side of those holes not meeting the criteria. Thus, all parts of the curtain (although not all primary, secondary, and ter- tiary holes) will have been grouted at the maximum limit- ing grout pressure with reasonably low grout absorptions (less than 25 kg/m or 0.18 sucks/f, for example). If non-optimal selection of the hole spacing has been made, the proposed grouting method is, at least to some exten, a self-regulating procedure. This is as x result of the split-spacing techniques, the Gin curve, and the requirement {or the last series of holes to reach the pressure limit with ‘minimal unit takes In conclusion, itis believed that if one follows the con- ccepis or rules presented, a furly optimal distribution of the total grout volume along the grout curtain can be achioved. ‘The procedure almost automatically takes into account the ‘actual irregularities of the geological conditions in the rock ‘mass. In doing so, the benelit-io-cost ratio ofthe grout cur- twin can be maximized Main points of the GIN method Several concepts and procedures are basic in applying the GIN grouting method. These are summarized below under four headings. Basie concepts Use only stable, moderately thick grout mixes: (a) to Fig 5. Example of grouting reals fora grouting cartin, Final pals ofthe rote pts of lhe Bch renting ager reduce sedimentation and premature blockage; and, (b) to ‘obtain dense, resistant, hardened grout, Use, as far as possible, only a single mix for the entre grouting work: (a) © provide a single Bingham fluid with known properties; and, (b) to simplify the grouting proco- dure, thereby improving efficiency and reducing errors. Use the GIN curve to monitor the grouting pressure: (a) 10 allow high pressure to be applied where needed; and, (b) 10 avoid high pressure where it would be harmful or wasteful ‘@ Contfol the grouting process by field computer: (a) to follow in real-time the pressure and flow rate; (b) 10 plot the p-V grouting path on the selected Gin curve; and, (€) 0 indicate the completion of grouting using bath the p-V ‘grouting path and the penetrability-volume curve, Mix design f¢ Use admixtures to obtain the desired grout characte tics: (a) superplasticizer to reduce the cohesion and viscas- ity of the mix, so as to inerease grout penetrability; and, (b) possibly a water retaining agent to reduce water loss during squeezing, ‘© Conduct a comprehensive series of laboratory tests early ‘on several grout mixes, with a Waterseement ratio (by \weight) ranging from 0.7:1 to T:1. This is to: (a) test differ- cent available cements of varying fineness; (b) test different admixtures at varying percentages; and, (c) obtain tet val- es of unit weight of arout slurry, Marsh Funnel apparent viscosity, 2h sedimentation, cohesion, inital and final set ting times, 7 day and 28 day compressive strengths, and water loss in squeeze tes Grow-hole layout # Adopt the normal split-spacing method of primary through to tertiary or quaternary holes: (a) to provide & minimal uniform coverage throughout: and, (b) allow for Closer spaced holes where the geological conditions and ‘grouting results so indicate. {© Conduct grouting field tests cither during the final design Phase ofthe dam or during the frst part ofthe constuction phase: (a) to test ferent parts of the site having different {eological oF topographical conditions (for instance, valley bottom and each abutment; (b) to selet optimum primary hole spacing, so that lator secondary and tertiary holes exhibit a continuing decrease of 25 to 75 per cen pe series (considers preliminary primary hole spacing of 10-12 m)z and, (c) to allow for different GIN curves 16 be examined (for instance, by plotting the p-V grouting path for each routing stag up to the anticipated grouting intasity or up {othe firs, or even second, hydrofiacturing even). Field contro! «© Detine the controlling elements of the GIN curve from theresults of the fest grouting programme, as well as any spe- cial engineering, rock mechanics, or geological considera- tions: (a) (0 ensure that the volume and pressure limits are reasonable for the existing geological features; and. (b) to ‘assess the need for different GIN values at different si locations. ‘© Grout every fourth primary hole first, as exploratory rout holes, except in areas of previous test grouting: (a) 10 allow for bet n of geological and ground ‘water conditions (by rotary core drilling and Lugeon water~ pressute tests to a depth equal to the future reservoir height above the ground point in question); (b) to permit final selection of the hole depth for the remaining primary holes (probably a depia range of 0.5 10 0.8 of reservoir height); and, {c) fo ensure that the selected cin curve is appropriate ‘ Control the grouting process by field computer using the {in curve and penetrability curve: (a) to allow for real-time ‘monitoring of the grouting path; and, (b) 10 allow for the ‘completion of grouting to be anticipated, from the declining penetrability curve, and from the approach of the p versus V grouting path towards the controlling cin curve (inelud- ing the volume limit and the pressure limit portions of the curve) ‘e Pre-injeet water before the grouting of any stage above the water tablo, to part-saturate the rock, s0 as to reduce the risk of water loss from the grout with premature blockage. f Use Lugeon water-pressure tests only in the exploratory primary holes and in the check grout holes to compare the initial and final permeabilities of the rock mass. ‘¢ Summarize the grouting results by appropriate statistical And graphical methods, to ascertain a progressive closing of the rock fissures with a resulting acceptably low residual permeability Example During the lst decade, this grouting procedure was into- ‘diced siep by step at a number of sites in Argentina, A tei, Ecuador, Mexico, Switzerland and Turkey. In the near future, the method will continue to be used in various countries at planned projects. "The on-going grouting atthe Aguamilpa dam in Mexico y be mentioned as an outstanding example. This 180 m- high dam, owned by the Comision Federal de Electrcidad, willbe the highest concrete-faced rocktill dam in the world Its impounding will start during 1993. "The cin method is at present being used both forthe rock consolidation below the foundation plinth of the concrete face and forthe deep grouting curiain, Afier extensive labo- ratory and field test, a single “normal” mix was selected. I Ina the following characteristies: Cement: fine pozzolanie cement, Blaine value about $100 emg (09:1 (somewhat higher than usual because of the high Blaine value) 1.6 per cent ofthe cement weight ‘Water-cement ratio: ‘Superplasticizer: of Sikament NZ Density of the mis: 1.5 to 1.55 giem! Decantation: 4 per cent in 2h Marsh Funnel flow time: 2810 328, Relative cohesion (Ci): 0,08 to 0.2 mm, increasing 100.2100.3 in 2h 910 10. MPa at 7 days Strengih ofthe hardened mi: and 13 to 17 MPa at 28 days, Ii corresponds 0 a stable but very fluid mix, with high pen- ctration properties during the frst hour, and excellent resis- tance both mechanically and against bleaching out For the grout curtain, the follo ral: 18 are specified as a ‘¢ Procedure: 5 m stages from bottom ups 1g method, starting with primary holes at 24m dis- 4 Additional higher rank holes are drilled so long as the take is greater than 25 I/m ‘@ Saturation of the rock mass (above water table) during T hata pressure of 2 bar immediately before the start of the ‘grouting of each 5m stage: ‘© Grouting intensity: 2500 bar, lm (reduced to 1500 below the plinth); ‘@ Maximum pressure limit: 40 bar (reduced nzar the sur- face to 10 bar andl increasing to 40 bar at 20 rr depth, and also, adequately reduced at geologically delicate spots): €¢ Maximum take limit: 400 lm, in fact 2000 lies per 5 m stage (locally reduced to 300 Vim): and, f Stopping criterion: flow rate less than 3 Vm min for a ‘m stage atthe final grouting pressure. ‘These specifications are easy to handle at the site, espe- cially as only one single mix is used. AS an example, Fig. 6 sliows the grouting path of one stage of the consolidation work below the plinth. Readings ‘were taken intermittently at about 5 min intervals, as PC monitoring was not available. In the upper part, Fig. 6 (a), the paths of the pressure p as well as ofthe flow rate q are plotted versus the grout volume V injected. The pressure path may be compared with the civ limiting boundat ‘The grouting target was an intensity of 1500 bar W/m. However, an intensity of 2510 bar was arrived at, and the limiting volume of 300 lim was slightly exceeded. This over-fun was caused by a delay in transmission from the grouting spot to the punip operator, or by slow reaction of the pump operator himself. (This indicates the desirability of real-time continuous monitoring and display by a Pc, as ‘well as of having automatic cutoff of the pump when the controlling ors curve is reached.) In the lower part, Fig. 6 (b), the penetrabiliy is ploted again versus the volume taken V. At the beginning of the sprouting process, the penetrability increases (from 0.2 to 05 Vmin.mn.bar) because of the progressive opening of the discontinuities of the rock mass. Alter a take of 200 Im, the penetrability decreases quite regularly to zero fora vol ume of 320 Im, Indeed, the grouting process wis stopped Just before reaching this value. Tis felt that some irregularities shown in the penetrabil- ity curve are caused by the rounding of some readings by the operators. Also, itis probable that two or more hydro- jacking events occurred that were not complete picked up by the intermediate monitoring, Comparison with usual grouting ‘A “traditional” grouting process consisis of defining a {routing pressure and the use of different types of grou, for example, grouls with successively lower water-cement ratios (41, 3:1, 2:1, I: and so On). The change of the mix fakes place at given volumes of grout taken by the bore= hale, It is obvious that the cohesion of these mixes increases from one to the next nd that, al some point, the increased colsion resistance to low and penctration il slop the grouting process ‘With the ctw method, only one mix is used: conse- quently, the cohesion is constant. As the grout spreads out along fractures, its contact with the fracture walls inereases ind so does ts total cohesion resistance 10 flow. Conse-quently, greater pumping pressures must be utilized to overcome the flow resistance. This is a normal and expected physical phenomenon. The cin method allows for the grouting process (o continue under gradually rising Pressure, until one of three limiting values is reached: limit ing pressure, limiting volume, or a combination of pressure tnd volume represented by the specified p-V curve. Other differences exist beoveen traditional methods and the civ procedure: the GIN method always uses only the best Possible mix with regard to strength, durability, resistance to leaching out and shrinkage; iv avoids injecting large ‘water volumes in the form of thin mixes; the use ofa stable mix avoids or significantly reduces the risk of damaging ie rock by hydrofracturing; errors in the grouting process are greatly diminished: and, the grouting process is simpler and faster, as no time is lost in changing the mi Summary and conclusions Tue grouting process involves steady pumping of the grout ata low to medium rate, with a slow building up of pres- Sure as the grout penetrates further into the rock mass, Grouting is stopped when the volume injected attains a specified limiting value for a grouting interval, when the ‘grouting pressure arrives atthe previously selected limiting Value, of when & given intensity of grouting has been schioved at an intermediate position, less than the limiting vahies of volume and pressure, 2s judged by the a1 curve previously selected. “The distance of the curve from the origin is a funetion of the energy expended in the grouting. A family of curves may be prepared for various intensities of grouting, ranging from very low to very high. One intensity curve may be selected forthe project, oF two may be used, for instance, a high intensity for the valley grouting, and a Tow inten for the shallow, abutment slopes or in geologically weak ‘The complete limiting envelope thus consists of the lim- iting pressure line (in the range 15 t0 50 tar). the lirvting volume line (in the range 100 to 300 lites per mete of routed interval, and a Selecied iN curve connecting the {wo limit lines (with civ values ranging from very TOW 10 very high, for example, 500 to 2500 bar.lim). ‘The GIN method requires close monitoring by rc graphics of real-time curves of pressure versus time, grout flow rate versus time, and total injected volume versus time, plus the derived curve of penetrability (flow rate divided by pres- sure, qip) versus time. This latier curve is of particular value in signifying the approach of grouting refusal, or at least of diminishing absorptions. ‘The GIN method has been proven to be a very useful tool in designing and controlling grouting works. Ii increasing use should improve the grouting works for hydraulic siruce tures. Since the method is flexible with respect to the lirnit- ing pressure, the limiting volume injected, and the Gin value to be achieved, itis probable tht experience gained in curent and future projects will lead to additional refine- sents a References, T'HoUsay, A.C, “Consruction and design of coment grouting”, John Wiley and Sons ne, New YarRChichester, 1992 2 Weaven, K. “Damm foundation grouting, American Society of| Civil Enginesrng, New York, USA: 199 3. Nownen, E. “Growing theory nd practice” Elsevcr, New York, USA; 198, 4. Ewer Fe Springee- Very, ‘ih emphasis on dam ses pie Mes he 9 a2 see eal ee ee ‘trability (q/p) versus volume. 5, Desi, D, and Lombard, G “Grout shies ~ Tick or thin?” Fess in Dar Grouting, Proceedings of the sssion sponsored by the Geotechnical Enginering Division ofthe Arserican Society a {Cit Engineers in conjunction withthe ASCE, Convention it Denver, Colorado; Apri 1985, 6. Loman, G., “The role of the eobesion i cement grouting of| rock", O58 K3, 15th ICOLD Coageess, Lasaase, Switer 18a, 7. Lownauos.G., “Injection des massifs rocheux, Soréié Suisse de Mecanique des Sols des Roches, Jord pines, No. 11S; May 1587, 8. Lowanoy, G., “La roca y ef maciz rocoso — Leyes const as", Aeemis Nacional de Cienclas Exacas, atess ¥ Nath fales, Buenos Alves, Atpentina, Sesiones Clentiiens, “Ine Francico garcia Olan ade cas en It Ingenieria December 1987, May 1989 elton. Majors © Geetechque, No.Si: Apa 10, Loxsayor, G.,Stotfgeyee fr eldfuntatinen von Beonsper- req" 39 Geeinechank-Kolloquium, Salzburg, Avs, October 1000, Oesteretchische Gesellchat fur GeomechanitFelsbas, Jnhygang 0, No: 1901 11, Lownanor, G., The FES rock mass model — Dam Engineering, Vol. : “The FES rock mass adel — Fatt 2", Dam Engitoerg, VoL I, Ise 3, ‘August 1992. 12 Lowarot, G., “Kolnbrein dam: An unusual solution fr an Uinustal problsm™, Water Power & Dam Coustracion; June 91 13, Ocstereicische Draukrafwerke AG, Klagenfurt Carinthia/Aus tris, "Remedial project for Kainbrein aren dame Design aud (Consett uae 19 REPRINTED FROM INTERNATIONAL WATER POWER & DAM CONSTRUCTION, JUNE 1993