an approach to reducing a 3 m3/sec leakage in a more … · grouting in an atypical application of...

CANADIAN DAM ASSOCIATION ASSOCIATION CANADIENNE DES BARRAGES

CDA 2013 Annual Conference Congrès annuel 2013 de l’ACB

Montréal, Québec October 5-10, 2013

Du 5 au 10 octobre 2013

_____________________________________________________________ CDA 2013 Annual Conference, Montréal, Qc, Canada - October 2013 1

AN APPROACH TO REDUCING A 3 m3/sec LEAKAGE IN A MORE THAN 100 METER HIGH NATURAL DAM USING JET-GROUTING TECHNOLOGY

Paolo Gazzarrini P.Eng.- Sea To Sky Geotech Inc.- Specialty Foundation Consultant- West Vancouver, BC, CANADA - [email protected] Stephen Jungaro - Principal - Matcon Excavation and Shoring Ltd - Coquitlam, BC, CANADA- [email protected] Dan Hunt- Project Manager - Matcon Excavation and Shoring Ltd - Coquitlam, BC, CANADA- [email protected] ASBTRACT: The Zeballos Lake Hydroelectric project is an Independent Power Producer (IPP) "lake-tap" project on the West Coast of Vancouver Island. Since the power plant opened, the production of energy has been affected by important leaks in the natural dam that formed the lake more than 300 years ago. The water leaks prevented the utilization of the turbines at their maximum capacity resulting in a significant loss of revenue for the Owner. In May 2011, the Owner decided to investigate the leaks with a geophysical survey method. A potential preferential path of the underground flow of water was mapped and a cut-off wall, approximately 100 meters long, was defined along a small part of the top of the natural dam. In September 2011 the Owner decided to carry out a grouting program to try to reduce the underground flow of water that was evaluated in the order of 2 to 3 m3/ second. A grouting design-build program was proposed and partially accepted by the Owner. The grouting program involved drilling grout holes in a very heterogeneous landslide to a depth of 80 to 100 meters and later grouting them using Jet Grouting technology. Jet Grouting was used not with the intent to create "conventional columns" but with the idea of aggressively attacking the underground flow of water using a high flow and high pressure grout mix. This paper will describe the problem, the possible solutions available, the solution adopted (atypical Jet Grouting), the logistical difficulties encountered, the instrumentation installed, the tests done on site, the results and the lessons learned in a very challenging dam grouting project. RÉSUMÉ: Le projet hydroélectrique du lac Zeballos situé sur la côte ouest de l’île de Vancouver en Colombie Britannique au Canada, a été développé par afin de profiter de la présence d’un lac perché en hauteur qui a été créé suite à un glissement de terrain qui a eu lieu il y a plus de 300 ans. Le lac qui fait office de réservoir est endigué par les débris associés au glissement de terrain, lesquels sont constitués de matériaux hétérogènes incluant des proportions significatives de blocs et cailloux, faisant en sorte que la digue naturelle renferme des zones de perméabilité très élevée au travers lesquelles des débits de fuite de l’ordre de 3 m3/s pouvaient percoler. Ces fuites se traduisant en pertes de revenus pour le propriétaire, qui décide en septembre 2011 de procéder à la réalisation d’un programme de traitement de fondation dans les zones de perméabilité trop élevée, lesquelles pouvaient atteindre des profondeurs de l’ordre de 100 m. Cet article décrit la problématique du projet et présente le programme de traitement de fondation qui a été retenu. Les résultats et les leçons apprises sont également discutés.






1 INTRODUCTION In the fall of 2002, the British Columbia (BC) government released the BC Energy Plan. As part of this plan, BC Hydro, a BC publicly owned electric utility, is restricted from developing new generation facilities other than those in which it has already made significant investments. Furthermore, the BC Energy Plan instructs BC Hydro to acquire new supplies from the private sector. According to the plan, 50% of new electricity needs are to come from conservation; however, much of the focus to date has been on developing new privately owned generating facilities, as in the case of the Zeballos Lake hydroelectric plant. The Zeballos Lake Hydro Ltd. plant is an Independent Power Producer (IPP) "lake-tap" project, located on the West Coast of Vancouver Island, near the village of Zeballos, approx 200 km west of the city of Campbell River. See Figure 1. The Village Voice (a local internet publication) stated proudly: "According to Stats Canada, the population of Zeballos is now 125 people. We are now officially the smallest Municipality in British Columbia!"

Figure 1. Site Location

The hydro plant facility, "lake-tap" type, is composed of an intake structure, a 3 m diameter outlet tunnel, an 1800 mm diameter penstock and a main powerhouse (2 Francis turbine 10 MW), plus a secondary "Fish" powerhouse (1 Francis turbine 3.4 MW) as shown in Figure 2.






Figure 2- The Hydro Plant

2 THE PROBLEM The watershed is unique in that Zeballos Lake is formed by a natural "rubble" dam that was known to be "leaky" and forming, downstream, Maraude Creek. The lake system can experience very large natural water level fluctuations. The flow from the lake was generally stable but occasionally, surface flow associated with major storms occurred in the upper half of the discharge stream. Maraude Creek is mainly created by groundwater seeping through the natural dam from several surges (fissures) of water approx 1,300 meters from the natural dam and 200 meters from the fish powerhouse. The leakage through the dam was probably underestimated and, since the commencement of energy production, the potential capacity of the plant has not been maximized and the missing revenue became a priority for the Owner. A minimum operating lake level, defined by the Ministry of Environmental to preserve the aquatic life in the lake, was also defined as a condition for the operability of the plant. In April of 2011, in order to try to detect the main paths of the underground flow of water in the natural dam, a geophysical investigation (Willowstick® technology) was contracted by the Owner. A "possible" 100 meter long, 60-70 meter deep grout curtain was defined on the crest of the natural dam. The total length of the crest of the dam is approx 400 meters long. See Figure 3.






Figure 3- Water flow in Maraude Creek and grout line location.

In September 2011 the Owner decided to carry out a grouting program and a design-build contract was awarded in March 2012 to Matcon Excavation and Shoring Ltd. Sea To Sky Geotech was hired for the design and QA/QC of the grouting works. 3 POSSIBLE GROUTING ALTERNATIVES AND SELECTED METHOD Despite the geophysical investigation carried out, very little information was available about the characterization of the formation in which the grout line should have been constructed. No cored boreholes or insitu testing were performed. It was clear, from the beginning, that the area to be grouted, being in a landslide, was expected to be very heterogeneous with the presence of cobbles and boulders of large dimensions. No information was available about the characterization of the landslide, and how the flow occurs. The flow at Maraude Creek in proximity of the main powerhouse was estimated at 2.5 to 3 m3/sec, according to tests carried out by the Owner. It was unknown how this flow originated through the natural dam. Billions of small fissures? Millions of medium fissures? Or a few big preferential channels as karstic phenomena? Not knowing these important pieces of information, in the design stage several grouting techniques and grout materials were analyzed, such as conventional grouting with packers (immediately discarded due to the heterogeneity of the formation and the necessity of using a down-stage method), grouting through the cap, Multiple Port Sleeved Pipes (MPSP) and Jet Grouting which is usually used as a soil improvement technique.






What material was to be grouted? Cementitious, High Mobility or Low Mobility grout? Use of accelerant in cementitious grout? Chemical products as polyurethanes? Hot bitumen? The final decision was to start using Jet Grouting in an atypical application of this soil improvement technology. The flexibility of Jet Grouting had the following advantages:

Using a high flow of grout mix (approx 250 L/min) with high pressure (350 to 400 bars), it can be possible to aggressively attack the underground water flow.

Using a double Jet Grouting system, it is possible to add accelerant directly at the grouting point (monitor/nozzles) and be more effective in increasing the grout mix viscosity and setting time. This can reduce the risk of massive escape of grout mix in large fissures that are potentially present. In addition, adding the accelerant at the grouting point, at the monitor, reduces the risk of clogging the grout plant and grout lines.

The formation (landslide) was thought to be comprised of cobbles, pebbles and boulders. Using a Measure While Drilling (MWD) recording system during drilling, it is possible to detect the position of the boulders and avoid them during the Jet Grouting phase. This eliminates the risk of blocking the nozzles and wasting grout mix. In pebbles and cobbles the disaggregation and potential permeation of Jet Grouting can be very effective.

The possibility of changing the parameters of the Jet Grouting if conditions are different than originally expected.

Complete use of the reflow in case of a very open formation. In this case the reflow, mandatory in soil treatment, can be used, and not wasted, for the permeation of the area.

Jet Grouting was used not with the intent to create conventional columns, as in soil, but to permeate the formation using constant volume with the thickest possible grout, that is still pumpable. 4 INSTRUMENTATION Before the beginning of the grouting program, four (4) piezometers (slotted stand-pipes) and two (2) inclinometers, as shown in Figure 4 were installed. The target of the piezometers was to collect data regarding the water table fluctuations with respect to the lake, before and after the treatment. The piezometers were also used to measure, with a special impeller, the vertical flow of the water inside the slotted pipe, to try to localize the most permeable zones. The inclinometers were installed to verify whether movements were detectable in case of a change in the condition of the water table in the natural dam. Downstream, at the location were the fissures were detected, dye, turbidity and pH measurement sensors were installed to verify the path of the underground water and possible communication of the grout mix with the fissures and consequently the creek. As a contingency plan a CO2 plant was installed at the fissures, to treat potential contamination of Maraude Creek, due to communication of cement with the fissures.






5 JET GROUTING DESIGN AND PROGRAM For the 100 meter length of the defined grout line, a classic Space Split Method (SSM) was designed. As it is well known, the SSM decreases the spacing between the holes as the phases progress. The first sequence was to drill and grout 4 preliminary Exploratory holes, equally spaced at 20 meters, 100 meters deep (to try to reach the bedrock at the bottom). See Figure 4. The Exploratory holes were drilled also using Measure While Drilling (MWD) parameters, evaluating the drilling energy, and attempting to characterize the formation, mainly for the presence of boulders and potential voids. In the exploratory holes water tests and dye tests were regularly performed to evaluate the most permeable zone and to see if connection with the fissures was detectable. The water test was done evaluating the quantity of water collected at the surface with constant flow of air.

Figure 4 - Location of the Instrumentation and Exploratory Holes.

Based on the results of the water tests of the exploratory holes the depth of the grouting line was designed to reach depths varying between 70 and 80 meters. Bedrock was not encountered in any of the exploratory holes with the exception of Exp 5 where the possibility of bedrock was encountered at 80 meters. Having considered the exploratory holes as a preliminary sequence with a spacing of 20 meters, the following sequences of Primary and Secondary holes reduced the spacing of the holes to 10 and 5 meters, respectively. Further sequences of Tertiary and Quaternary holes reduced the spacing to 2.5 and 1.25 meters.






6 DRILLING & GROUTING SEQUENCES Considering the unusual depth (80 meters on average for the production holes and 100 meters for the Exploratory holes) to be reached by the Jet Grouting, the drilling/grouting sequence decided for the work was:

8" casing drilling up to the required depth. 100 meters in the exploratory holes and 70 to 80 meters in the production holes.

Installation inside the casing of a 6" PVC pipe, in order to insert the jetting rods, at such a depth.

Extraction of the casing, keeping the PVC pipe insitu.

Introduction of the jetting rods, at the required depth, inside the PVC in place.

Execution of the Jet Grouting, increasing the withdraw speed in case of detected boulders. Knowing the difficult formation to be drilled and Jet Grouted, the contractor used a "special" drill rig which allows for changing the rotary head and working in two different modes: drilling mode (short mast and powerful rotary head) and jetting mode (longer mast- 35 meters- and hollow rotary head). The hollow rotary head allows for carrying out Jet Grouting without the necessity of continuously adding rods. The target was to drill 5 holes in drilling mode, swap the rotary head, elongate the drill mast, and Jet Grout the previously drilled 5 holes. Figure 5 shows the 2 configuration modes of the drill rig.

Figure 5. Drill rig in drilling and jetting mode.

Figure 6 shows the final spacing and depth of the grouting holes at the end of the contract. We can observe that:

The area exceeded the 100 meters considered at the beginning of the work, reaching a total depth of 137 meters, with the addition of a 6th Exploratory hole.

Only 45 holes were completed within the total length of 137 meters.






Not all of the Tertiary holes (final spacing 2.5 meters) were completed.

Only 2 Quaternary holes (final spacing 1.25 meters) were completed.

This final situation differs substantially from what the Contractor/Consultant proposed at the beginning of the work. The initial design criterion was to treat the approx 100 meters with a final spacing of 1.25 m. This criterion was followed until the end of the secondary holes. After that, the criterion chosen was established mainly by the Owner, and not necessarily agreed by the authors of this paper.

Figure 6. Final spacing and depth of the grouting holes.

Some quantities:

45 holes drilled and Jet Grouted

2,800 meters drilled

2,400 meters Jet Grouted

170 meters of boulder drilled (approx 6%)

2,100 m3 of grout mix Jet Grouted

2,270 tons of cement used The grout mix used was variable, between 40 to 60 m3/hole. Using jet grouting technology the volume of grout mix is constant per meter. Where boulders were found the jet grouting was suspended. Consequently we cannot talk about differential grout takes per different depths, as in conventional grouting works.






7 PLUGGING THE SEEPAGE? At the end of the contract, it was difficult to judge whether the "reduced" grouting program was successful in terms of reduction of flow in Maraude Creek. Historical data about its flow and the water table variations in relation to the rain and lake level are missing. Additionally, due to logistic problems, the owner couldn't install a weir or other proper instrumentation to correctly measure the flow in Maraude Creek. Without this very important information, no apparent differences were noticed in the data collected, during and immediately after the completion of the work, in the level of the groundwater measured in the piezometers vs. lake level (see Figure 7) and impeller tests (see Figure 8). All the data should be analyzed over a long term period of a minimum of 2 or 3 years.

Figure 7: Piezometer readings vs. lake level. Comparison of the piezometer readings at the same lake elevation.

The horizontal black line shows the lake level at the end of the grouting treatment. The 2 vertical black lines give the reference to evaluate the level of the piezometers with the same lake elevation.

Figure 7 shows the graph of the piezometer reading vs. the lake level and figure 8 shows the readings of the impeller measurement in one of the piezometers (#4). To be noticed is the consistency of the readings in the impeller tests and their compatibility with the lake level. The higher the lake level, the higher the vertical speed.






Figure 8: Example of impeller measurement data.

A few other facts can be summarized to show the differences before and after the Jet Grouting treatment:

The dye tests done before and after the grouting program gave different results. In the interpretation of the results measured by the sensor at the "fissures" clear hits of dye were detected in the Exploratory holes, and no hits were visible during the tests done in the last holes completed.

The grout holes done as the first and second sequence (Exploratory holes and Primary holes) either did not show any cement returning to surface or the cement returning to the surface was very spotty. There was a totally different situation in the Tertiary and Quaternary holes where cement traces were visible in the cutting, during drilling, and very frequently return of cement reflow was visible at surface. These were clear signs that the grout mix "stayed" in the formation and was not washed away by the underground water flow.

No increase of pH at the fissure was observed during grouting. Despite the fact that it was not possible to measure the differences in flow at Maraude Creek, two pictures (see Figure 9) taken approx the same day, one year apart, show a clear reduction of flow. The two pictures were taken at the same location, from the bridge at the access to power plant. Considering that the location where the pictures were taken is affected also by the rainfall collected by the creek between the fissures and the bridge, an analysis of the quantity of rain in the two years was conducted. The result shows that in 2012 (after the grouting treatment) the quantity of rain, in comparison with 2011, was similar the previous week and higher the previous two weeks.






8 CONCLUSIONS From the grouting point of view, the result was very positive. The initial great concern about possible escape of grout mixes, and ensuing inability to seal the leakage paths, was eliminated with the use of the Jet Grouting technology with the possibility of adding accelerants. We can affirm that in the area treated no large voids were present and the grout mix and grout technology used were effective. From the Owner's point of view the measure of success shall be evaluated through statistical analysis of the future production of the Plant, but for sure the treatment was not completed due to the following reasons:

The final spacing in the 137 meters treated was not adequate; the spacing of the holes is still 2.5 meters or in some cases 5 meters.

The location chosen for the grout line was not necessarily the most effective and the "partial" treatment may not have been long enough. Tests with the impeller show high, or higher, vertical flow speed in 2 other piezometers, quite far away from the area chosen for the grout curtain.

Figure 9. On the left: picture taken in 2012, after the grouting treatment. On the right: picture taken in 2011 before

the grouting treatment. The project was a good start in gathering information about the leaks and finding the right grouting methodology to be used. Its completion will be subject to economical consideration.






To the author's knowledge no Jet Grouting program reaching a depth of 100 meters is documented, and this "unique" project shows, once more, the flexibility of this soil improvement and grouting technique. 9 ACKNOWLEDGEMENT The authors would like to thank Zeballos Lake Hydro Ltd. for their permission to publish this paper. REFERENCES (1) - www.ippwatch.info.com

an approach to reducing a 3 m3/sec leakage in a more … · grouting in an atypical application of...

Documents