Use of Geosynthetics in Dams

7/28/2019 Use of Geosynthetics in Dams

1/35

UnitedStatesCommitteeOnLargeDams

UseOfGeosyntheticsInDamsEleventh Annual USCOLD Lecture Series

White Plains, New York April 1991


2/35

BUREAU OF RECLAMATION EXPERIENCES WITH GEOMEMBRANES IN DAM CONSTRUCTION

W . R. orriso on', E.W. Gray Jr.,' 3 . M cyganiewicz, ' and P. G . ~ r e y 'INTRODUCTION

In recent years, there has been a dramatic increase in the use ofgeosynthetic materi a1 s in seepage and pollution cont rol appl i ca t i onsboth in the United Sta tes and worldwide. These ma te ri al s were developedfor use i n appl ic at io ns where conventional construction materi a1s arenot available, or cannot be used because of weather condit ions, timeconstraint s ( for example, minimum downtime to accomplish the work),1imited access etc. Geosynthetic materi a1s include geomembranes(flexible membrane linin gs, pl as ti c linings, et c. ), geotextiles ( f i l t e rfabrics, construction fabrics, etc. ), geogrids, geowebs, synthetic

drainage composites, and er os io n control blankets. T h i s paper describesseveral recent Reclamation geomembrane ins tal l a t i o n s r e 1ated toembankment dams. These i n c l ude:

1. The M t . Elbert Forebay Reservoir, East Slope Power System,Fryingpan-Arkansas Project, Colorado

2. San Justo Reservoir, San Felipe Division, Central ValleyProject, Ca1i forni a

3. Emergency Spi1 way, Cottonwood Dam No. 5, Coll bran Proje ct,Colorado

MOUNT ELBERT FOREBAY RESERVOIR MEMBRANE LININGBackwoundDuring the summer of 1980, Recl amation instal1ed approximately117 hectares of geomembrane in the M t . Elbert Forebay Reservoir nearLeadville, Colorado, figure 1. The purpose of the geomembrane was t oreduce seepage through a previously constructed compacted earth l i n i n g .

The forebay reservoir and adjacent M t . Elbert pumped-storage powerplantare part of the East Slope Power System of the Fryingpan-ArkansasProject. The or ig inal re se rv oi r, bui l t under con tra ct between 1975 and1977, was formed by constructing a small dike i n the open southwestcorner and a 27-m high-zoned earth embankment across the open north sideof a topographic depression. A ridge, composed of glacial depositsoverlying weakly indurated formation materials, forms the south side ofthe reservoir and separates i t from lower Twin Lakes Reservoir. A

'portion of the si de of th e ridge facing lower Twin Lakes had beengeologically mapped as an ancient 1 ndsl ide.

' ~ u r e a uof Recl amat i on, PO Box 25007, Denver, Col orado 80225-0007


3/35

Figure 1. Aerial view taken June 1980, looking south across t he forebayreservoir. The f i r s t po rt io n of placed geomembrane is vis ib le i n the near rigside of the reservoir and the processing plant is located in the center of thereservoir area. A1so, the in1et-outlet structure (upper left-hand c ome r ) ,forebay dam (foreground) , slop e protection material around t h e perimeter of threservoir, and subgrade areas being prepared by the contractor are visible.


4/35

Considerable concern had been expressed that seepage from the reservoirmight reactivate the slide, and a 1.5-m thick compacted earth lining wasplaced under th e ent ir e re se rv oi r. Water was introduced i nt o theForebay to a depth of 7.5 m during the period of November 1977 throughMarch 1978. Uater levels in several of the piezometers and observationwells located in the side of the ridge between the forebay reservoir andthe powerplant began t o rise short ly after completion of this f irs tintroduction of water into the forebay. By the summer of 1979, thewater level had risen nearly 2.4 m in one we11 and up to 2.0 m inseveral others. Since other wells either had not responded or hadexperienced water 1eve1 decreases, the continuous ri s e experienced i nsome wells was considered a t t r ibutab le t o water i n the Forebay, ra therthan cyclical changes in the groundwater level.

Because the original 1.5-m thick compacted lining failed to provideadequate seepage control, a decis ion was made t o dewater the r ese rvo ir

and install a flexible membrane lining over the entire Forebay bottomand side slopes.

A t that time, the installation at M t . Elbert constituted the world'slargest single-cell flexible membrane lining application and representeda milestone in the use of geosynthe tic ma ter ial s i n the United States,i f n o t i n the world. Also, t o meet Reclamation's time schedule forpower on-1ine, the i nst al la ti on had t o be accomplished in oneconstruction season to allow suff ic ie nt time t o f i l l the rese rvoi r andconduct acceptance t e s t s on t he pump-generating units and otheraccessory equipment in the powerpl ant .

Construction

Specifications for the membrane lining were issued in January, 1980(Reclamation 1980 a); the contract for installation was awardedApril 16, 1980, and install ation was completed September 20, 1980. Theprincipal fea tu re s covered by the spec if icat ions fo r the lin ing workincluded: removing a l l ex is ti ng re se rvoi r slope protection; excavatingand processing the top 0.6 m of the compacted earth lining; placing a150 mm processed ear th subgrade; manufacturing, fabr ica ting, t es ti ng,and ins ta l1 ing the geomembrane; plac ing a 0.45 m earth cover over thegeomembrane; and replacing the slope protection materials.

The specifications provided a1 ter na te bidding schedules for instal1ationof one of the following three lining materials: 1.14-mm reinforcedchlorinated polyethylene (CPER), 1.14-mm rein forced chlorosul fonatedpolyethylene (RCSPE), and 2.0-mm high density polyethylene (HDPE). Thecontractor selected CPER because of i t s ava ila bil ity t o meet theconstruction schedule.

This geomembrane was of three-1ayer construction consisting of two equalthicknesses of chlorinated polyethylene (CPE) 1aminated t o a middlelayer of 10 by 10 1,000-denier polyester scrim. The specified physical

properties requirements fo r t h i s geomembrane a re given i n table 1.


5/35

Table 1

Test methods and Physical Properties Requirements for CPERLining

Property Test Method HlnimmRequirenrent

ThicknessASTM: 0

751-79 1.04lar

Breaking Strengtheach direction

Tear strengtheach direction

ASTM: 0 751-79Grab MethodA

ASTM: D 751-79Tongue TearHethodB

Bonded seam ASTM: D 751 -59strength in shear Grab MethodA

strength.

Bonded seam ASTM: D 1876-78strength inpeel

Dimensional stability ASTM: D 1204-78(percent change, 1 hour a t 100 OCmaximum)

Low temperaturebend

Hydrostaticresistance

ASTM: 0 2136-783 nmandrel ;4 hours at -4 O C

ASTM: 0751-79MethodA

Equal s Parentbreaking

No spec requirements

2percent

Pass

2.07 HPa

Ply adhesion ASTM: 0 413-76 1400 N/m


6/35

The membrane 1inin g was factory fabricated in to 'blankets,' each 1300 m2in size and weighing approximately 2268 kg. Two shapes of blanket s werefurnished: 61 by 21 m containing 14 factory seams made with a l e i s t e r tlllhot-air gun, and 30 by 43 m containing 28 factory seams maded i e l e c t r i c a l l y . The latter blankets were installed on the side slopesi n order to avoid making f i e ld seams a t or near the toe of the s lope.

To install the membrane 1ining, la bo r crews unfolded and posit ioned theblankets as shown on figure 2. Adjacent blankets were overlapped aminimum of 150 mm. A three-man crew thoroughly cleaned the contactsurfaces w i t h cleaning solvent and then appl ied the manufacturer'sbodied solvent CPE adhesive t o a minimum width of 100 am. After thefield seams were tested and approved, a cap strip was applied over thefield seam. A 0.45 m protective earth cover material was then placedover the geornembrane, f igure 3.

Details o f the construction work and the quality assurance programconducted dur ing the instal 1a t i on of the flexible membrane 1ining aresumarized i n t he l i t e ra ture l i s t ed i n the references a t the end of th ispaper (Morrison, et. a1

.

, 1981, Reclamation 1980b, Reclamation 1981,Frobel and Gray 1984).Performance

After completion of th e membrane 1 ining i nsta ll a ti on in 1980, thereservoir was r e f i l l ed beginning in January 1981. By June 1981, thereservoir had been fi l l e d t o elevation 2940 m. Since that time, thereservoir has remained above elevation 2935 m except for short periodso f pool drawdown.

Piezometers and observation wells installed in the hil lside south of thereservoir continue t o be monitored. Several of the observation wellswhich began t o r i s e during i n i t i a l f i l l i n g began a gradual decl ine assoon as the reservoir was drained i n 1979. Others continued t o r i s eprimarily due t o time lag and did not show signs of lev el in g of f and-declining u n t i l well a ft er the l inSng was installed and the reservoirrefi l led. A t the time of t h i s writing, the water lev el s i n th eobservation wells in the hillside south of the reservoir continue todecline. The foundation beneath the M t . E l bert Forebay Dam on the northside of the reserv oir i s s t i l l not saturated. Incl i nm e t e r s ins ta l ledalong the south side have not indicated any movement of the oldl ands l ide mass. Also, the r i p r a p on the side slopes has remained stablewith no evidence of sl'ippage.

Included i n the specifications for the work was a 5-year maintenancewarranty period on the membrane l ining. To monitor the performance ofthe 1ining during the warranty period and for l ong- t e rn researchpurposes, a special test section was instal led in the forebay reservoir.The 6- by 30-m test section was installed at a location within thereservoir that would allow periodic access for retrieval of the membrane1ining test coupons.


7/35

Figure 2 . Installation of geomembrane on reservoir side slopes.

Figure 3. PIacement of protective soil material on geomembrane.


8/35

Eleven test panels (or coupons) comprised the total test section. Eachtes t panel was made up of a l l three types of seams used i n the projectwhich included hot air, dielectric, and bodied solvent adhesive fieldseams. The t e s t panels were placed on a 50-mm layer of fine sanddirectly above the M t . Elbert Forebay membrane lining and then coveredwith the same 0.45 m of earth cover. Thus, the panels can be extractedand tested without disturbing the original CPER membrane lining. Testpanels were retrieved on a yearly basis for the first 5 years , i n 1987after 7 years of reservoir exposure, and in 1990 after 10 years ofreservoir exposure.

In addition t o the t es ts l i st ed i n table 1, large-scale hydrostaticpressure r esi st an ce t e s t s were conducted on th e coupon samples. Thistest was developed by Reclamation (Hickey 1969, Frobel , 1981). Theprocedure i s now being adopted by ASTM Committee 0-35, on Geotextiles,Geomembranes and Re1ated Products. For the M t . E l bert evaluation, thecoupon samples were tested over a 10-to 20-mm aggregate subgrade a t ahydrostatic head of 43 m which was the same pressure that was used onsamples o f the unaged membrane 1i ning materi a1 during acceptance testingin 1980.

Original t e s t specimens were taken from the same blanket samples a sthose used t o fabr icat e the t e s t sect ion. Thus, r e sul t s from extractedcoupons can be compared di re ct ly with th e t e s t re sul t s obtained f o r theoriginal bl anket materia l.

Test results are summarized i n tables 2 to 5. Generally, anysignificant changes in the CPER lining and seams occurred within the

f i r s t 3 yea rs of se rvi ce. There has been very l i t t l e change through th eadditional 7 years of reservoir exposure. Specif ic de ta i l s of th e tes tsresults are summarized i n a paper presented ea rl ie r t h i s year a tGeosynthetics '91 (Morrison and Gray 1991).

To obtain additiona l information on the aging ch ar ac te ri st ic s of the H t .E l bert membrane 1 ining, laboratory water immersion t e s t s were conductedon random samples of the 1.14-mm CPER 1 in ing .and the 0.5-mm CPE sheetmaterial used in the manufacture of the membrane li ni ng . The reinforcedmaterial specimens with both sealed and unsealed edges were i m r s e d t odetermine i f th er e were any major di fferences due t o po ss ib le water

wicking through the exposed scrim. The samples were immersed i n Denverlaboratory tapwater f o r 5 years. The temperature of t he runningtapwater varied between 10 and 15 OC during this immersion period.Samples were removed year ly for te s t ing. The same tests l isted aboveexcept breaking strength and large-scale hydrostatic testing wereconducted on the CPER samples. For the CPE sheet material, thefollowing t e s t s were conduced:

1. Weight changes2. Breaking strength (ASTM 0 882)3. E l ongation a t break (ASTM D 882)4. Tear resistance (ASTM

D1004, Die C)


9/35

Test results conducted on random samples of the 1.14-mm CPER during the5-year laboratory water immersion period are summarized in table 6. Thetesting of sealed versus unsealed edges showed no major differences inmechanical p rope rt ie s due t o pos si ble water wicking through t he exposedscrim. Consequently, only th e t e s t r esu l t s for the specimen w i t hunsealed edges are presented i n t a b l e 6. The results generallyparall eled those observed fo r the f i e l d samples. The moisture

absorption for the 1aboratory immersion samples leveled off around21 percent as shown in figure 4.

Test results for 0.5-mm CPE materials are summarized i n t able 7 . Themoisture absorption was si mi la r t o tha t noted for the rein forcedmater ial. I t appeared t ha t the moisture absorption caused somesoftening of the material as reflected i n a reduction in i t s t en si le andtear strength properties.

Summary

Results of studies conducted on the geomembrane installed i n 1980 i n theM t . Elbert Forebay Reservoir indicate that the material is performingsatisfactorily after 10 years of service. The studies involvedcontinuous monitoring of the instrumentation on the ridge between theforebay re se rvoi r and powerpl an t, and periodic retr ieva l of couponsamples from the field test section for laboratory testing andeval uat ion .

Continuous monitoring of the instrumentation on the hillside hasindicated no movement of an old landsl ide mass. Water levels i nobservation we1 1s and piezometers i n the h i l l side continue t o decl inefrom levels reached during or shortly after first fi l l ing of thereservoir following install a t i o n of the geomembrane l ini ng . Results oflaboratory tests conducted on the coupon samples indicate that thelining has experienced some water absorption resulting i n a decrease ini t s strength properties. The water absorption has caused some weakeningof the polyester reinforcing scrim, the bond between the CPE and scrim,and the CPE t o CPE bond. Most of the changes i n the strength propertiesoccurred within the f i r s t 3 years o f service and are not considereddetrimental t o the overall i nt eg ri ty of th e lining. In fa ct , theretained strengths of all the geomembrane's mechanical properties areabove the minimum specification requirements for the original unagedmaterial except for the seam shear st reng th . The lower shear streng thof the seams, however should not affect the integrity of the lining w i t hregard to seepage control.

As part of t he cons truct ion work i n 1980, an extensive qua1 i t y assurance(QA) program was developed and conducted in an attempt t o obta in a topquality installation. Prior to t h i s time, QA programs for flexiblemembrane lining work were generally very minimal, and i n some casesnonexistent. The ins tal lat io n a t M t . Elbert was a major milestone w i t hrespect to advancing the state-of- t he-art on the use of geomembranes forseepage control, both in the United States and worldwide.


10/35

I --.,Table 2. M t Plh er t Teat Sect ion Re ru l t r After Om Veer of Exposure

Property

....-*--...--Hot A i r Sew pbWl - - - - - - - - - - - - - .....--.*--.-Dielectric ream pard - - - - - - -S p t c f f l c a t l o n O r l ~ l n a ldata One year & ta Percent Or lg lna l data One year & tr peRqul rcmcnt (rage) change ( r a g e ) c

burst

PI)

Tear strength

(kN)Pl y crdherion

(kW/m)Breaking s t r m g t h

(kN)

h er tv e f t r l d re wn r h e r r(kN)

i v e f i e l d reem peel(kN/m)

- ---Field ream ulth cap r t r t pf le ld ream without cap r t r l p* - Not rqquired


11/35

Tabte 3. Ht EIbert lest Section Results After Three Years of Exposure

Property

.*......-...- Hot A l r #e m panel - - - - - - - - - - - - - ...----..--..l e l u t r l c #em panel - -- -- - -------Specification Original data Three year data Pe rcmt O r i ~ i n a l ata Three year data PercentR e q u l rmwnt (range) change (range) c h a w

Weight gain (percent) 15.60 15.60

nut t e n burst(kPa)

l ee r strength(kN)

P l y &her ion(kN/m)

Breaklng strength 0.89 1.26 0.W -21.7 1.26 0.66 -47.5(kM) 0.94- 1 .04 0.58-0.75

Bondad seam shear 0.89(kN)

Adhesfve f i e l d seam pae l Nit***(kN/m)

* f i e l d seam 4 t h crp s t r f p** f l e l d stam wlthout cap r t r lp*** - Not rcqufred


12/35


13/35

Table 5. H t E l b e r t Test S e c t lo n R e r u l t r Af te r Ten Years of Expocure

--*.*.-*-..-- Hot Air scan pawl - - - - - - - - - - - - - ..----....--.l e l e c t r l c ream pawl-------------tS p m c f f f c a t f o n O r f 0 i n a l data Tm year data P e r c m t O r i g i n a l data T e n year datm PercentRequirement (range) change ( raga) change

w e i g h t g a i n (p e rce n t )nul lm burst

(kP8Tear strength(kN)

Breaking strength

(kN)Bordad seam shear


14/35

Table 6. Uater lrarrtion Test Results - 1.14 r PER

Property Original On year Percent Three year Percent f ive year Percentdata data chanOc da ta che n~e data change(range) (rawe) (range) (ranse)

Ueight gain(percent

Hullen burst(kPa)

Tear strength

(kN)Ply adhesion(kN/m)

Hot A i r seam shear(W

Hot Air seam peel

(kX/a)Dielectr ic seam shear 1.30(W (1.00-1-41)

Dielectric seam peel 6.74(M/m) (6.01-7.49)

Adhesive seam pet1 5 -38(kNla) (4.01-7.41)


15/35

Table 7. Uater l m r s i o n Test Results - 0.K) a PE1 0 n g i t t J d i ~ ltest d i rect ion

On year Three year Five year

property Original Test Percent T e s t Percent lest percentdata value change value change value change

Ueight gain 16.9 20.4 20-9(percent)

Tensile strength 6.39 5.76 -9.9 5.59 -12.6 5.39 -15.6

(kN/m)Elongation 190.0 692.0 0.4 198.0 1.6 493.0 0.6

(percent)Wodulu~ 1.91 0.93 -51.4 1.21 -34.9 1.63 -11.7

(kNIn)Tear strength 18.68 12.01 -35.7 12.01 -35.7 15.57 -16.7

( W )

Transverse test direction

~ n cyear Three year Five year

Property Or igin al T. st Percent T a t Percent T e s t Perca tdata value change value c h m value JunOe

Ueight gain 16.9 - 20.4 20.9(percart )Tensile strength 5 -53 4.78 -13.6 4-59 -17.1 4.47 -19.3

(kW/o)Elongat im 587.0 587.0 0.0 581.0 -1.0 568-0 -3.2(percent)

I b c b l ~ r 1.69 0.93 -37.6 0.91 -38.8 1.17 -21- 2(kN/m)

Tear strength 18.2C 13.79 -24.4 14.23 -22.0 17.35 -4.9( W )


16/35

15

PERCENT

WEIGHTGAIN10

1 2 3 4

YEARSOF IMMERSION

-9-1.14mmCPER (LAB) -0 - 0.50mrnCPE (IAB) .*TESTSECTION IFIGURE 4. MOISTURE ABSORBTION- LABORATORYWATER IMMERSION

TESTINGANDMT.ELBERT TEST SECTION RESULTS


17/35

SAN JUST0 RESERVOIR MEMBRANE LININGBacksroundSan Ju st o Reservoir near Holl i s t e r , Ca1i fo rn i a, was constructed byReclamation as an off stream stora ge f a c i l i t y to' provide water fo rir ri ga ti on and municipal purposes. Because t he fa ci li ty i s locate d near

both the San Andreas and Calaveras faults, special earthquake designco ns ide ra ti on s were used (Cygani ewicz 1986)

.

Construction of thereservoir was completed in 1985.

The reservoir, shown i n f igure 5, i s formed by two ea rt h f i l l s t ructures:a dam t o the west and a dike t o the north . I t i s fill ed -and re leasesare made by an i n le t -outlet works located in a tunnel through the easts ide of the reservoir . An emergency spillway is located near thisstructure and is provided str ictly as a guard against overfill ing of thereservoir .

Several large beds of clean sand are located within the reservoir site.In addition to loss of water, the increased seepage through the sandbeds could increase the potential for landslides on the downstreamporti ons of natural rid ges which enclose t h e res erv oir . Consequently,th e dec is io n was made t o i ns t a l l a geomembrane over sloping por tions ofthe reservoir containing the impervious sand beds. In f la t te r a reaswhere natural impervious soil covers the sand beds, a supplemental2-meter-thick earthfill blanket of clay was placed in lieu of themembrane lining.

Construction

The following geomembranes were included as options in thespecifications (Reclamation 1984): 1.O-mm high density polye th lene-alloy (HDPE-A), 0.91-mm CPER, 0.91-mm RCSPE, and 1.14-mm polyvinylChloride (PVC). The l in ings were re qu ir ed t o meet the materialproperties as 1i sted in National Sanitation Foundation (NSF) StandardNo. 54, 'Flexible Membrane Liners" (NSF 1985). The contractor selectedthe HDPE-A geomembrane.

Approximately 190,000 m2 of geomembrane were installed for seepagecontrol a t 6 locat ions within the reservoir . An aerial view of severalof t hese s i t e s i s shown i n f igure 6.

The HDPE-A liner was produced in rolls about 6-m wide by 200-m inlength. th e r o l l goods were shipped t o t h e j o b s i t e where they wereunrolled on the prepared subgrade as shown i n f igure 7, and then joinedusing extrusion f i l l e t welding as shown i n f igure 8. The geomembranewas secured at the toe and top of the slope in a v-shaped anchor trenchas shown in figure 9.

During installation some problems were encountered with excessivethermal expansion of t h e geomembrane re su l t i ng i n large waves andwrinkles i n the liner as shown i n f igure 10. This wrinkling and


18/35


19/35

?

figure 8 Field reaming geomembrane using a hot-extruded f i l l e t weld-Worker a t rig ht i s cleaning and dire brushing overlapped area...t o be seamed.


20/35

waviness led to some permanent folds in the liner after the protectivesoil cover was placed. This condition is shown in figure 11. Foldedsamples were included in the coupon monitoring section to determine the

long-term effect of the creases on the performance of the

1ining.

To protect the geomembrane from the elements, mechanical drainage, andvandalism, an earth fill cover consisting of a 0.5-m layer of semi-pervious material, 0.15-m layer of bedding material, and a 0.3-m layerof cobbl es was pl aced over the 1iner.In 1990, an additional 6100 m2 of geomembrane were installed overanother sand lens in the bottom of the reservoir (Reclamation 1990a).For this work, a 1.5-mm HDPE geomembrane was used since HDPE-A is no1onger being manufactured.Performance

In February 1986, unusually heavy rainfall occurred in the reservoir.During and immediately following the rainfall event, sl ippage ofportions of the earthfill cover occurred. Atotal of eight separateareas experienced sl ippage affecting 4 hectares of cover materi a1 andexposing 1.2 hectares of geomembrane. Several areas experiencingslippage are shown in figures 12 and 13.

Astudy was performed to determine the as-constructed slopes on which

the liner failed. In general, failures occurred on slopes steeper than4: 1 (horizontal :vertical )

.

Interestingly, some areas with slopes assteep as 2.5:l did not exhibit stability problems.To support analytical studies and to aidin designing of an acceptableremedi a1modification, a 1aboratory test program was undertaken todetermine the frictional resistance of the soil on the geomembrane.Samples of various types of soils and geomembranes were tested in astandard direct shear apparatus. -Geomembrane materia1s tested included:

1. Original material (smooth).2. Original material scored with a wire brush3. Original material sandblasted4. Texturized material5. Original material with an attached geogrid.

Soil materials used in the test included the original material coveringthe membrane (soil A), material representing the sand underlying themembrane (soil C), and materi a1 s representing proposed cover materia1s(soils B,, andB2) .


21/35

Figure 11. Waviness shown in figure 10 led to some permanent foldsin geomembrane after protective soil cover uas placed.

Figure 12 View showing sl protective soil cover


22/35

The physical properties of the soils used in the test program are shownon figure 14. The results of the testing are shown on table 8. Also,included in the table are some of the details of the test apparatus andtest procedures.

Util izing these t e s t re su lt s, s t a b i l i t y analyses were completed andindic ated t ha t any of th e modi fi ed geomembranes described above would best ab le on the slopes under th e loa din gs imposed. The ana lyse s als oindicated that a thicker cover material generally lowered the safetyfactor .

During the summer of 1987, a remedial construction program was initiatedto repair the exposed areas of geomembrane (Reclamation -1987). Seven ofthe eight areas were repaired by sandblasting the top surface of theexposed 1iner t o enhance the f r i c t i on a l resis tance followed by coveringi t with a 0.15-m layer of pervious sand and gravel bedding and a0.3-m

1ayer of cobbles. Careful qua1 i t y control ensured th a t t her e.

.would be no damage t o the geomembrane dur ing sandblast ing op era ti on s.For the remaining slide area, the existing geornembrane was removed andreplaced with a texturized HDPE geomembrane and covered with bedding andcobbles.

The f i l l ing of San Justo Reservoir was ini t iated in the lat ter part of 1987 and was completed i n March of 1988. A close inspection of thegeornernbrane area af t e r in i t i a l f i l l ing ind ica ted no s igns o f i n s t ab i l i t y .

To monitor the performance of the geomembrane, a coupon monitoringsection was installed i n the rese rvo ir. The monitoring se ct ion co nsi stsof 10 coupon samples (1.5- by 1.5-m ) of the 1.0 mm HDPE-A l iner . Eacht e s t coupon conta ins a f i e l d seam, and as previously mentioned th esamples were placed i n a fol de d condi ti on . One coupon sample wasremoved after 2,3,4 and 5 years of burial, and the following tests wereconducted on both folded and unfolded portions of the sample:-

Breaking strength (ASTM 0- 638)Tear strength (ASTM 0-1004, Die C)

In addition, shear and peel tests were conducted on the seam samples.

Test results summarized in table 9, indicated that there was very l i t t l echange in the tensi le and tear propert ies of the geomembrane after5 years of buri a1. A1so, i t appeared that there was no adverse effectfrom the lining being folded. The folded tensile and tear specimenswere examined under a 5x magnification before being tested and there wasno evidence of cracking. In addition, the field seam exhibited goodretenti on of shear and peel st re ngt h proper ties a f t e r 5 years of burial..

Toobtain additional information on the aging characteristics of the SanJu s t o geomembrane, l abo ra to ry ag ing t e s t were conducted on random


23/35

igure 13. View showing exposed geomembrane after s1ippage ofprotective soil cover.

Figure 14 Gradation of protective soil cover.


24/35

Table 8. Resul ts o f I n t e r f a c e F r i c t i o n a l Te st s

Soi l *

A (wet)

'3,'32

C (wet)

(moist)

( d r y

Sandblasted

F r i c t

Wire brush

ma1 r e s iGeogrid

LanceT e x t u r ized Embossed S o i l4

33(3938(38

(1) Values are g iv en f o r lar ge s t r a i n and peak (parenthesis) re su l t s. Values a t largs t r a i n were used f o r analyses except where una va i l ab le (N/A).(2 ) Test resu l t s us i ng d i re c t shear apparatus on s o i l only.

Test in g de ta i l s: Shear box siz e - 4 i n c h x 4 i n chTime o f sa tura t ion - 1 t o 3 daysNormal applied pressure - 2, 5, 10 ps iPlacement densities - 40 X t o 60 % r e l a t i v e d e n s i tyS t r a i n r a t e - 0.005 i n ches /m inu te

CLSP-SMsP


25/35

Table 9 . Test results for San Justo Geomembrane Coupon Samples

data data data

res i stance, I bfunfolded LT

eaking strength 1bf / infolded)

imate elongation, %folded)

eaking strength, lbf / inunfolded) L

imate elongation Xunfolded) L

T

am shear strength

am peel strengthb/in


26/35

TestDurat ton

Or ig ina l

52 weeks

104 weeks

156 weeks

208 weeks

260 weeks

I

Table 10. Results of Laboratory Aging Studies Conducted on San Ju st o Geomembrane

le s t Cond i tion: 73 O F , 50 X RHTear Strength, Breaking Ult im ate

l b f / ~t,r-rtg,;h, 1 Elongation, % Test Condi tion : Water Imnersion Ult imatel b f St rength, Test Condition: Heat aging a t 100 O FTear Breaking Ulti mateStrength, 1 b f 1 Stength,b f / i n 1 Elongat ion,

F denotes - f o ldedU denotes - unfo lded


27/35

s of the HDPE-A 1 ining. Both folded and unfolded tensile and tearwere subjected to the following aging conditions.

1. Room temperature (23 OC, 50 percent relative humidity)2. 37 OC oven aging3. Water immersion in Denver laboratory tapwater. The temperature

running tapwater varied between 10 and 15 OC during the immersion

e laboratory aging study was conducted for 5 years, with the tensiletear specimens being removed and tested on a yearly basis.

results are summarized in table 10. These results indicated thatwith the field samples there was very little change in tensile and

ear strength properties. Also, the samples exhibited no adverse effectrom being folded.

EMERGENCY SPILLWAY, COTTONWOOD DAM No. 5

ackqround

n response to a Corps of Engineers' survey of non-Federal damsompleted in 1981 which indicated some of the structures had inadequatepillway capacity, Reclamation initiated a study to evaluate theossible use of a geomembrane as one method of increasing the spillwayapacity. Of the more than 63,000 dams inventoried in the Corps'urvey, over 8,800 were examined. More than 2,900 (33 percent) of thexamined dams were evaluated as unsafe. Of these, 81percent wereeficient because their spillways were too small to pass the estimated

aximum floods. This reflects the difference between present-day designlood criteria contained in the "Recommended Guide1 ine for the Safety

nspection of Dams" and the criteria in vogue at the time the dams wereonstructed (Corps of Engineers 1975, USCOLD 1982).mbankment dams are particul arly sensitive to failure caused byvertopping, both during construction and while in service. However,nadequate spillway capacity is not the only cause of overtoppingailure. There have also been many cases where dams were overtoppedecause of gate failure (Londe 1983).

hese potential hazards can be avoided by adding an emergency spillwayith the required discharge capacity. However, in many situations, theost for a conventional concrete-lined spillway or even a rock-1 inedompacted-earth spillway would be prohibitive.

he investigation on using geomembranes started with an evaluation ofhe feasibility of various applications for low-head structures.


28/35


29/35

A synopsis of the spillway design considerations is 1i sted below.1. The spillway was a1 igned to pass through the more plastic soil

materials on the right abutment to provide additional erosion protectionif needed.

2. Two grade s i l l s were provided: One at the upstream end of themembrane l iner t o provide flow contro l and prevent pip ing , th e other a tthe downstream end to prevent head-cutting back into the spillway.

A t the grade s i l l s , th e membrane was at tached t o the concrete w i t hredwood furring strips and nails to distribute the load evenly acrossthe sheets and to prevent separation from the grade sills.

3. The edges of the 1iner along the sides and the upstream edgeof the transverse joints were placed in trenches, and backfilled with

compacted soi 1 .4. Transverse joints between adjacent sheets were not bonded.

This prevented stress buildup in one sheet from being transferred toanother.

5. A protective cover of 150 mm of noncohesive material was placedover the geomembrane t o p ro tect i t from foot, animal, and vehiclet ra f f ic .

6. The alignment was chosen so that there are no discharges alongthe toe of the dam.

7. Inflow design f l o d is the 100-year flooddesign discharge of 1.13 4 s.

8. Flow passes through th e c r i t i ca l depth a tsill; therefore, the flow is super critical over athe flexible membrane 1i ner.

.This results i n a

the upstream -grade17 areas protected by9. The channel bottom width i s 3.66 m w i t h 2:l side slopes and a

depth of 0.91 m (to provide freeboard).

10. The assumed Manning's numbers are 'nu = 0.025 for theprotective cover in place, and " n u = 0.015 fo r the geomembrane.

11. Energy dissipation is t o be provided by a natural hydraulicjump, which should form over the downsteam r i p r ap protection.

12. Riprap i s sized t o re s i s t movement caused by vel oci t ie sassociated wi t h the design discha fge .Some surplus 0.9-mm-Hypalon 1ining material from another nearby job wasavailable for installation in the spillway. A small quantity of

material was purchased t o complete the installation. The physicalproperties o f t h i s geomembrane material are given on t a b 1e 11.


30/35

Table 11

Test Method and Physical P ro pe rt ies Requirements f o r Hypalon L i n i ng

Property Test Method MinimumRequirement

Thickness

Breaking strengtheach direction

Tear strengtheach direct ion

Bonded seam

strength i n shear

Bonded seamst rength i n pee l

Dimensional stabilit y(percent change,maxi mum)

Low temperature

bend

Mul len burst

Ply adhesion

ASTM: D751-79ASTM: D 751-79

Grab Method

ASTM: D 751-79Tongue TearMethod B

ASTM: D 751-59

Grab Method A

ASTM: D 1876-78

ASTM: D 1204-781 hour a t 100 OC

ASTM: 0 2136-783-mm

mandrel;4 hours a t -40 O C

ASTM: D 751-79Method A

ASTM: D 413-76Machine Method

Ply Separationi n p lane o f scr imo r 2.6 kN/m

2 percent

Pass -

1.71 MPa


31/35

Before construction, 2-year water immersion and outdoor exposure testswere conducted on samples of the Hypalon to be used in the field study.A1 though some properties changed, these changes were consideredinsignificant.

An operation test was conducted in July 1986. To provide the necessaryreservoir level t o conduct the te s t , the gate to the primary outletstructure was closed and flashboards were installed i n a weir in thegate chamber t ha t i s used as a ser vic e spillway. Sandbags were placedin the emergency spillway to increase the effective height of thereservoir above the emergency spi 11way crest by approximately 0.5meters.

The operational t e s t was conducted fo r approximately 3 1/2 hours. A tthe beginning of the test, the reservoir level behind the sandbags inthe emergency spillway was about 0.3 meter$. During the test, thedischarge was est imated t o be 0.6 t o 0.7 m/s, and the maximum velocityestimated t o be 6 t o 8 m/s. These velocities are higher than antic -ipated and may be a t t r ibuted t o a Manning's number tha t was lower thanthe assumed value of 0.015. Consequently, some additional studiesshould be conducted to obtain design data for establishing a Manning'snumber for spillways with flexible membrane 1inings.The spillway operated essentially as expected, i . e . , the soil cover waswashed away until the membrane on the bottom was exposed. From then on ,

gradual erosion of the cover on the sides of the spillway continued fora few centimeters up the sides. Even though the flow carried muchabrasive materi a1, stones, and a few cobbles approximately 100 mil1 -meters in diameter, l i t t l e or no erosion of the membrane was observed.Only one small tear, approximately 75 mill imeters long, was found; wesuspected that this occurred during construction. A f is t - sized stonefound under the membrane (original foundation material ) a t th i s locat ionwas probably responsible fo r the te ar. This t ear was v is ible during theoperation of the spi llway but did not appear t o increase i n size.

The overlapped field joints of the membrane functioned well.

Immediately after the test, the overlapped joints were inspected. Theexposed portion of the geomembrane was wet from the flows; however, theportion under the over lap was completely dry. There was no evidence ofaccumulated tensile strain from one sheet to another. As expected, themembrane was instal led with some wrinkles t o help i t conform to thesubgrade. These wrinkles , did not cause any problems during theoperation of the spillway.

Specific observations and results of the field test, in terms of thestudy objectives, are summarized:

1. The flow placed no noticeable serious strain on thegeomembrane, and the over1apped joints helped avoid accumulation ofte ns il e load along th e spillway. Any up1 i f t pressures were accommodatedby the overlapped j oin t s The amount of uplift was


32/35

2. The geomembrane experienced l i t t l e or no abrasive damage fromthe cover material as i t was washed away.

3 . The simple method of securing the membrane i n 1- by 0.5-m

trenches fi l led w i t h compacted soil was successful.

4. The soil cover proved successful i n preventing mechanicaldamage t o the geomembrane fo r the 10 months of exposure as a buriedmembrane 1in ing. The cover was s table on the 2: 1 side slopes.

5. As a precaution, the downstream hydroelectric faci 1i ties wereprotected from damage by the soil cover material by bypassing the turbiflow. However, this was necessary for only a few minutes as the streamquickly cleared up.

6. The velocity exceeded the expected 4 t o 5 m/s and reachedperhaps 6 t o 8 m/s. Even at these higher velocities and with thewrinkled liner damage, distress, or cavitation was not observed.

7. Reasonable care must be taken to prepare the subgrade freeof rocks and stones. If suitab le material i s not available forconstruction of the subgrade, a layer of fine -grained material will beneeded under the geomembrane.

8. Aging and durability were not problems i n the ear ly f ie ld tes t

and none are expected because the normal early aging observed in the2-year materials tests shows adequate retention of materials properties

Future designs should be improved by curved bottom cross sections rathethan the usual f l a t bottom of a trapezoidal sect ion . This wouldminimize the amount of cover washed away a t low flows. This conceptcould be expanded by provi-ding vegetated earth cover that can handleflood flows with minimal erosion and not require recovering the spillwa- af te r each operation. Studies have re ce nt ly been completed in Englandon the reinforcement of steep grassed waterways [Hewlett et.., a1.19851.This may have application i n Reclamation work, but would depend upon

local s o i l and cl ima ti c condit ions. To prevent the membrane from beingtorn by lo gs , t r ees , or branches, ins ta l la t ion of a log boom upstream othe s p i l l way should be considered.

SUMMARY

Recl amat ion has successfully used geomembranes i n several seepagecontrol applications i n embankment dam construction. Besides theinstallations described in this paper, the re have been several othergeomembrane appl ication s. These include:

1 . Installation of a geomembrane as an impervious element i n theraised embankment a t Pactola Dam which i s located i n the Black Hillsof South Dakota. I t should be noted that the use of a geomembrane


33/35


34/35

REFERENCES

Bureau of Reclamation 1980a. Specifications No. DC-7418, Mt. ElbertForebay Reservoir Membrane Lining, Fryingpan-Arkansas Project, ColoradoBureau of Recl amation l98Ob. Construction Considerations,Mt. El bertForebay Reservoir Membrane Lining, Fryingpan-Arkansas Project, ColoradoBureau of Reclamation 1981. Design Summary for Mt. Elbert ForebayReservoir Membrane Lining, Fryi ngpan-Arkansas Project, Colorado

Bureau of Reclamation 1984. Specifications No. DC-7617, San Justo Dam,Central Valley Project, Cal ifornia

Bureau of Reclamation 1987. Specifications No. 20-C0300, Membrane LinerModification - San Justo Reservoir, San Fel ipe Division, Central Val 1 eyProject, Ca1 forni aBureau of Reclamation 1989. Specifications No. DC-7786, Tucson

Aqueduct, Black Mountain Operating Reservoir, Tucson Division, CentralArizona Project, Arizona

Bureau of Reclamation 1990a. Specifications No. 20-C-357, Geomembraneand Drainage, San Justo Reservoir, San Fel ipe Division, Central ValleyProject, Cal i forni a

Bureau of Reclamation 1990b. Specifications No. 1-CC-10-05220, OchocoDam Modi ficat ion, Crooked River Project, Oregon

Corps of Engineers 1975. Recommended Guidelines for Safety Inspectionof Dams, Office of the Chief of Engineers, Department of the Army,Washington D.C.

Cyganiewicz, J. M., 1986. San Justo Reservoir, USCOLDNewsletter March1986 -Engemoen W., and P. Hensley 1989. Reinforced1:l Slopes at Davis CreekDam, USCOLD News1 etter, November 1989Frobel R. K. 1980. Design and Development of an Automated HydrostaticFl exi bl e Membrane Test Facil ty, Bureau of Reclamation Report No.REC-ERC-80-9

Frobel R. K. and E. W . Gray Jr., 1984. Performance of the Fabric-Reinforced Geomembrane at Mt. El bert Forebay Reservoir, Proceedings ofInternational Conference on Geomembranes, Denver, Colorado, vol. 2,

pp 421-426

Hewlett, H.W.M., L. A. Buarman, M.E. Bramley, and E. Whitehead 1985,Reinforcement of Steep Grassed Waterways, Construction Industry Researchand Information Association, Technical Note 120, London, England


35/35

,y M.E. 1969. Investi gation s of Pl a st ic Films fo r Canal Linings,a~ of Reclamation Research Report No. 19r t 1. L. , and 6. G. Hammer 1989. The Use of a Geomembrane i ntening Pa c t o l a Dam, Water Power and Dam Construction, February 1989

de M.P., 1983. Lessons from Earth Dam Failures, News and Views,tish National Conunittee on Large Dams, August 1983

rrison,W . R . , E. W .

GrayJr., D. B.

Paul, andR. K. Frobel 1981.s t a l l a t i o n of Flexible Membrane Lining in M t . E l ber t Forebay

eservoir, Bureau of Reclamation Report No. REC-ERC-82-2

orrison W. R. , and E. W. Gray Jr. 1991. Performance of M t . ElbertReservoir Flexible Membrane Lining after 10 years of Service,

cs '91. At1anta, Georgia, February 1991Sanitation Foundation 1985. Standard No. 54, Flexible Membrane

ners, Ann Arbor, Michigan

mblin, L . 0. Jr., P. G . Grey, B. C. Muller, and W. R. Morrison 1988.mergency Spi 11ways Using Geomembranes, Bureau of Recl amation Report. REC-ERC-88-1

1982. Dam Safety P ra ct ic es and Concerns i n th e United St at es,oston MA, April 1982

Use of Geosynthetics in Dams

Documents

Transcript of Use of Geosynthetics in Dams