WIXLIB

Cumulative Rebound30%"Cc:::l20%0.cQ)a: 10%0%mm 0in0502100415062008Average ThicknessFigure 3. Cumulative Rebound with Thickness Placed Particle size distribution depends on the Aggregate blend, Sand, Cement, and Silica fume Velocity of impact depends on the Pressure in the line or air flow, Distance of the nozzle from thereceiving surface, and Orientation of the nozzleA dense, well-placed shotcrete layerwill usually exhibit a good distribution ofaggregate particles including coarse aggregate to a maximum of about 10 mm (3fs in.).The most common aggregate blend seenin international projects is close to theAmerican Concrete Institute's GradationNo. 2 as shown in ACI 506R-90.A well-designed shotcrete mix will alsohelp to reduce rebound. There must be sufficient fine materials (cement and silicafume) to adequately cover the total surfacearea of the sand and aggregate particles.Since there is a high proportion of sandand relatively small aggregate in comparison with conventional cast concrete, the cement contents of shotcretes tend to be high.This is another reason that shotcretestrengths are high. If the cement content istoo low, rebound will again increase sincethere will be too little sticky paste for embedment and dynamic compaction.The relationships between rebound andvelocity of impact are equipment dependent and so specific rebound values are notpresented. In general, however, there willbe an optimum air pressure or volume flowrate, an optimum distance between thenozzle and the receiving surface and theshotcrete stream should be maintained perpendicular to the receiving surface.It is also noted that rebound is a function of the total placed thickness ofshotcrete Rebound is high at the beginning of ashoot, and Rebound becomes lower with increasing thickness.This concept is true for all shotcrete butis far more prevalent when using the drymix process. As the first material strikesthe receiving surface a very high proportion of the coarse aggregate and steel fibers bounces off the surface. Coarser rna-Figure 4. Thickness of ShotcreteMinimum 25 mm (I in.) Average 50 mm (2 in.), 15% Rebound14Irregular rockmass surfaceterial cannot embed itself into the plasticshotcrete until a paste layer has been builtup on the surface. This initial layer is usually rich in cement and is commonly responsible for the good bond developed byshotcrete. Once the coarser material canbecome part of the increasing layer, dynamic compaction starts to take place thatleads to the development of high strengths.As the layer thickness increases further, theinstantaneous rebound drops to almost zeroin the case of wet-mix process shotcrete.The cumulative rebound figures for wetmix shotcrete (the total amount of reboundfor a particular section thickness) areshown in Figure 3.This curve demonstrates the relationship between total anticipated rebound forwell-placed shotcrete and the section thickness called for. Site specific versions ofthischart could be produced by physically measuring the rebound values during pre-construction testing. The curve is based onfield observations and various publishedvalues for rebound.ThicknessThe definition of thickness in mining engineering applications is not at all straightforward . For a smooth walled excavationsuch as a tunnel boring machine (TBM)driven civil engineering tunnel, the concept of a uniform layer of a specified thickness is reasonable. This would be considered to be a "coating" by Windsor andThompson (I 999) (2). The "coat and fill"suggested by Windsor and Thompson ismore likely for mining applications inwhich a minimum coating thickness is required, but in order to successfully achievethis absolute minimum some additionalfilling of hollows in the rock mass isneeded. The "fill " profile, in which all ofthe hollows are filled and the surface between is relatively smooth is unlikely inmining due to high volumes of material required.Thus, in a mining application of shotcrete in a drill andblast excavation, there is nosingle concept of "thickness"that applies. If the design engineer calls for an absoluteminimum thickness of 25 mm(1 in.), then he will likely needan average of 50 mm (2 in.) toprovide the desired cover. Figure 3 shows that this averagelayer thickness would probablyShotcrete Magazine August 1999

be associated with a minimum of 15 % rebound .If the des ign is for a minimum of 50 mm(2 in.), then it is likely that an average of75 mm (3 in .) will be needed and the associated rebound fig ure could dro p to 10%.If the des ign call s fo r a minimum of 75mm (3 in .), an average of 100 mm (4 in.)mi ght appl y and the rebound could dropfurther to 7.5%. The majority of miningapplicati ons of shotcrete call fo r thi cknesses of 50, 75 or 100 mm (2, 3, or 4 in .).It is quite common to see the thickness desc rib ed as a " minimum thi ck ness ,"although estimating volumes based onminimum thickness is not recommended.It is strongly suggested that the notion ofminimum coating plu s some fi lling is themethod used to specify shotcrete in drilland bl ast applications.Figure 4 shows the placement of a minimum coating of 25 mm (1 in.) with localfi lling to 75 mm (3 in.). Thi s illustrates howdiffic ult it would be to assume that a uni fo rm layer thickness could ever be successfull y specified in a mining project.Volume .00Table 2. Vo lume Factorli ke ly th at a rock mass givin g a poorRoughness Factor would onl y require 50or 75 mm (2 or 3 in.) of shotcrete. Similarl y, a high quality rock mass that is wellexcavated would be unlikely to warrant a75 or 100 mm (3 or 4 in.) shotcrete layer.The numbers fo und for des ign purposesfro m this table are similar to those that havebeen developed empi ricall y over the lastfew years by the mi ning industry in Australi a. The process shown in thi s paper allows the engineer to determine where thedifferent components of the overall VolumeFactor come fro m.Planning ProcessSteps in the Planning ProcessIt has been shown that the surface of anexcavation driven by drill and bl ast methods is far from smooth and unifo rm, andthe perimeter of the excavation boundaryis much larger than designed. The amountof overbreak is a function of the inherentcharacter of the ground-rock mass quality-and the actual excavati on processqu a lity co ntro l of the drill a nd bl astmethod. The impact of the pl acement ofshotcrete can also be taken into considerati on at the des ign phase by consideringthe required thickness, as both a minimumand an average. Different levels of reboundcan be attributed to different total secti onthickness, so it is now possible to estimatea Volume Factor for a specific applicationusing site specific values fo r Ground Condition, Drill and Blast Process and Rebound, using Table 2.The Roughness Factors shown in Table2 are taken from the prime diagonal fromTable 1. It is noted that any combinationof Ground Condition and Drill and BlastProcess ca n be used to dete rmin e th eRo ug hn ess Factor, a nd the Re bo undFactor can then be used to generate a specific Volume Factor.The grayed-out zones on Table 2 referto matrix intersections that are not likelyto exist in reality. For example, it is un-The fo llowing outline presents the pl annin g process requi re d fo r es tim atin gshotcrete volumes in a drill and blast mining operation. Select locati on need ing shotcreteShotcrete Magazine August 1999 Estimate volume fac tor and calcul atevolume req uired Instruct contractor accordingly. A typical shotcrete request form should include precise info rmation concerningth e tim e a nd the locati o n of th eshotcreting, and a sketch on a cross section or pl an view of the profile, including overbreak and fall outs . Above all ,such a req uest form must include info rmation on the length, the perimeter, thethickness, the fac tor , and the volumeshotcreted, both as es timated and asshot. It then represents a hardcopy ofthe differe nces, if any, between the es-Continued next page Decide on plain or mesh reinforcedshotcrete or fiber-rei nforced shotcrete(Fibercrete) Determine area to be shot Confirm des ign shape per engineering Determine excavated shape by sitereview Estimate geological facto rs Estimate drill and blast factors Measure effective round/cut length(this may be more than the excavatedcut length depending on loss ofpreviously appli ed shotcrete, forexample) Estimate design area needingshotcrete Estimate ro ughness factor Consider design thickness (minimumand average) Estimate rebound fac tor15

Continued from previous pagetimated and the real quantity applied forevery section of the opening shotcreted.There are a number of other considerations that may come into play when designing a shotcrete program (quality ofscaling walls and back, operator training,fall outs, presence of other support systems,etc.). They may all have an influence onthe amount of shotcrete placed, and mayor may not be under the control of theplacement contractor.ConclusionsOne of the most demanding parts of ashotcrete program in the mining industryis determining what sort of contract to enter into with the placement contractor. Oneprocess commonly encountered in miningis for the design engineer to estimate therequired volume of shotcrete based on thedesign profile of the excavation with asmall allowance for rebound and wastage.Having made this estimate and instructedthe contractor to place a certain volume ofmaterial on the rock mass around the excavation, the engineer is surprised to findthat the volume requested did not cover theground to the desired thickness. Almostinevitably, a conflict starts to develop between the two parties since the engineerand the contractor are looking at the problem from two different viewpoints.It is hoped that by using a process suchas that suggested in this paper, less conflict will arise by making both parties moreaware of the considerations that need to bemade during the design phase of theproject. A clear understanding of the concept of thickness needs to be made beforea decision is taken on how to word the"Payment" clause of the contract. If thecontractor is to assume the majority of therisk, then unit costs will be high . If theowner and the contractor find a way toshare the risk and agree to pay on "cubicmetres through the pot," then both partiesmust use a shotcrete request form, andreach a consensus about the required volume of shotcrete materials to be placed.AcknowledgementsThis paper is based on a presentation madeat the Engineering Foundation's Shotcretefor Underground Support VIII held inCampos do Jordao, Brazil in Ap1il 1999. Iappreciate the opportunity to be offered aforum to explore ideas such as these in anopen framework. I would also like to thankMarnie Pascoe, now with Australian Mining Consultants, for her pioneering workin trying to unravel the difficulties in estimating shotcrete volumes in miningprojects. References1. Throughout this article, the term"shotcrete" is used to include the process of pneumatically placing concreteand the product thus created. It includes"sprayed concrete" and does not inferthe wet-mix or the dry-mix process. Itusually implies a blended aggregatewith up to 7-9 mm C/2- 1/3 in.) sized particles, although it may also be synonymous with the old term "gunite" (a termrestricted to the use of dry-mix processand sand sized aggregate).2. C.R. Windsor and A.G. Thompson, Thedesign of shotcrete linings for excavations created by drill and blast methodsin Rock Support and ReinforcementPractice in Mining, Eds. E. Villaescusa,C.R. Windsor and A.G. Thompson.Balkema, pp 231-242, 1999.David Wood is a registered Professionaland Chartered Engineer practicing in RockEngineering and Engineering Geologywith 25 years experience in consultinggeotechnical engineering programs inNorth and South America, Europe, Australia, Africa and Asia. His experience comprises rock mass evaluation, rock hazardappraisal, rock engineering analysis anddesign, rock slope stability assessment andstabilization design, engineering geological and geotechnical field data-collectionstudies, and report preparation for smalland large-scale civil and mining projects.16Shotcrete Magazine August 1999

ship between total anticipated rebound for well-placed shotcrete and the section thick ness called for. Site specific versions ofthis chart could be produced by physically mea suring the rebound values during pre-con struction testing. The curve is based on field observations and various published values for rebound. Thickness