Measurement 49 (2014) 42–51

Contents lists available at ScienceDirect

Measurement

journal homepage: www.elsevier .com/ locate /measurement

Performance of single vertical helical anchor embedded in drysand

0263-2241/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.measurement.2013.11.031

⇑ Corresponding author. Post-Doc.E-mail address: ham.niroumand@yahoo.com (H. Niroumand).

Ramli Nazir, How Seng Chuan, Hamed Niroumand ⇑, Khairul Anuar KassimDepartment of Geotechnical Engineering, Faculty of Civil Engineering, Universiti Teknologi Malaysia, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history:Received 21 May 2013Received in revised form 14 November 2013Accepted 18 November 2013Available online 1 December 2013

Keywords:HelicalSingle helixSoil anchorSandUplift response

An investigation into the uplift behavior of single vertical helical anchor embedded in drysand presented. A series of laboratory tests have been conducted to determine the effect ofthe embedment ratio, shaft diameter ratio and sand density against the uplift capacity ofhelical anchor. The laboratory tests were conducted in a small scale model in loose anddense sand. A sand placement technique was utilized over the testing program to achievethe predetermined depth. In this testing program, the uplift load and uplift displacementhave measured. The observation of failure mechanism and the measurement of pulloutload and the vertical displacement analyzed and discussed. A number of graphs will be plotbetween the uplift capacity and the factors to obtain their relationships. From the analysis,the uplift capacity is increase with the increase of embedment ratio and sand density how-ever the shaft diameter ratio is not significantly influent to the uplift capacity. In the obser-vation of failure mechanism, the failure surface proposed for the helical anchor embeddedin loose and dense sand. For loose sand package, local failure surface observed however fordense sand package, a truncated cone shape failure surface was observed. From the anal-ysis, an empirical model was proposed. The proposed empirical model compared withthe experiment result and existing theories to verify the validity of the empirical model.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Slim high-rise structures such as the electric transmis-sion tower, telecommunication tower, street lamp have al-ways being subjected by a high lateral forces especiallyfrom wind load. This lateral load will generate an overturn-ing force to rotate the foundation and affect the stability ofstructures. During slim high-rise structure failure, theoverturning force always higher than the weight of thestructure. Therefore, the lateral loading is the major factorthat must be considered during design. In the design forthe slim high-rise structures, the traditional method is byusing the massive foundation to resist the overturningforces due to wind load. However, it is very heavy, largeand expensive to construct, which not effective in terms

of cost and time. To overcome the problem, Engineers haverecommended a new construction method by usingground anchor systems to enhance the structure stability.Over the years, many types of anchor have been developedand used in the construction of slim high-rise structure.Helical anchor is one of the anchor that is always used inthis type of construction. Helical anchor consist of somesteel shafts with a series of helical steel plates welded ona pitch. During installation, helical anchor was screwedinto the ground by using a standard truck or trailermounted augering equipment. The equipment will applya rotating moment to the steel shafts to screw the anchorsinto ground. The torque resistance of the anchor will bemonitoring along the installation. When the torque resis-tance achieved its designed values, it verified the capacityof anchor achieved. It was thought that the working helicalanchor can provide compression and uplift resistance offoundation between 100 kN and 400 kN depending onsize of helixes, number of helix and the type of soils

R. Nazir et al. / Measurement 49 (2014) 42–51 43

encountered. Therefore, when the friction angle of soil, sizeand number of helix increases, the capacity of uplift andcompression will also increase. Helical anchor system is acost, time effective system for the foundation of slimhigh-rise structure because this system can be installedrapidly, with immediate anchor capacity verification andeasily installed. Currently the helical anchor system isnot widely used as foundation in construction industryand just limited for a transmission tower and pipelines.This is due to the lacking of a rational and reliableanalytical technique to carry out its uplift behavior andfailure mechanism. Many theories have been developedto describe the uplift behavior and failure mode of helicalanchor. Inevitably, there are many variations among thetheories due to different assumption adopted. To producean effective design which with respect to cost and time,the verification of these design methods and a furtherunderstanding of the uplift behavior of failure mechanismof helical anchor in sand are needed.

2. Early background history of helical anchor

The earliest usage history of helical anchor was appliedby a blind English brick maker names Alexander Mitchell.In 1833, he use the helical anchor to design a foundationsupport for a lighthouse. This concept of ‘‘screw pile’’ wasvery successful in the design application but the develop-ment of helical plate foundation was not in further pro-gress [1]. Until 1950s, a power-installed screw anchor forresisting tension load was used in United States and thistype of anchor begins its popularity and widely used inconstruction site. The helical anchor was formed by a steelshaft where one or more helical plates welded to the shaftto create a ‘‘screw anchor’’. Helical anchors are primarilydesigned and constructed to provide the uplift resistanceto the foundation of a structure. However, helical anchorsystem can also provide the compression support to thestructure. Generally, helical anchor can be dividing totwo types namely single helix anchor and multi helix an-chor. Mitsch and Clemence [6] proposed a semi empiricalsolution to predict the ultimate uplift capacity of multihelical anchor in sand. They introduced values for coeffi-cient of lateral earth pressure as a function of embeddedratio, H/D and relative density. Their values were 30–40%less in comparison with those proposed by [5]. They indi-cated that the reduction is due to shearing disturbance ofthe soil during anchor installation. Clemence and Pepe[2] studied the effect of installation and pullout of multihelix anchors to lateral stress in the sand layer. The valuesof lateral earth pressure were measured before and afterthe installation of anchor continuously during the applica-tion of the uplift loads until it failed. From the test, it wasindicated that the installation of helical anchors in drysand causes an increase in lateral earth pressure aroundthe anchor and the pressure was significantly increase indense sand. They concluded that the increase of lateralearth pressure was due to the relative density of sandand the embedment ratio (H/Dh). Ghaly and Hanna [4]finding shows that there are three components mainlycontribute to the uplift capacity of shallow anchor, which

are the selfweight of anchor, weight of sand within the fail-ure surface and the friction along the failure surface. Fromthe experiment result, a theoretical model was developedby using the limit equilibrium technique and Kotter’s dif-ferential equation. In this model, they assume the failuresurface to be a log-spiral shape. In their model, the com-plexity of model has been reduced by developing theweight and shear factors for shallow and deep anchors.The uplift capacity of helical anchors in sand have beenstudied by numerous researchers such as Mitsch andClemence [6] and Ghaly et al. [3] although some research-ers such as [7,8] evaluated the uplift capacity of shallowhorizontal strip anchor in cohesionless soils. Mitsch andClemence [6] have proposed a semi empirical solution topredict the ultimate uplift capacity for multi helical anchorin sand. Based on Laboratory and Field Tests, they have rec-ommended the bearing resistance of top helix, frictionalcylindrical resistance and friction on the anchor’s shaftfor a multi helical anchor. Based on laboratory tests, Ghalyet al. [3] suggested a similar solution with the Mitsch andClemence [6] for the pullout resistance of single helicalanchors in sand. Nevertheless, in their solution, the effectof friction of anchor’s shaft in the uplift resistance has beenignored.

3. Test program

Verification of the previous theories with a series ofmodelling in laboratory will be highlighted in this paper.Through the laboratory test, the failure mechanism anduplift behavior of helical anchor in sand can be carry outto produce a better reference in the designing the helicalanchor system in sand. Results analyzed will be focus onthe physical modelling in the laboratory. Four major vari-ables have been focused for a small model of helical anchortested in laboratory. The sand used in the laboratory test isin dry condition; single helical anchor will be used to stud-ied and tested in the work; limited to vertical pulloutcapacity of single helical anchor only; the study is limitedto the shallow embedment ratio between 1 and 5. In thislaboratory test, the main parameters such as embedmentratio and relative density will tested to obtain the effectsand relationships to the uplift capacity of helical anchorin sand.

4. Physical modelling uplift test set-up

In this experimental work, both full and a half-cutmodel of single pitch helical anchor will be prepared. Thedimensions of the anchor’s shaft diameter ranged between300 mm and 500 mm with helical diameter of 100 mm asshown in Figs. 1 and 2.

Pullout test will be conducted in a bigger glass con-tainer with dimension length 650 mm, 650 mm widthand 1000 mm depth. The glass frame was build using steelangle 50 mm � 50 mm � 6 mm and placed using C channelat a height of 250 mm above floor level to avoid anyhumidity transfer from the base. An electrical motor gear-ing is placed on the top of the frame. The pulling arm, theload cell and Linear Variable Displacement Transducer

Fig. 1. Configuration of single pitch helical anchor model.

LVDT

Load cell

Pulling frame / arm

Model helical anchor

Steel angle

Motor winch

650mm

1000mm

800mm

Screw

Dry sand

Retort Leg

Fig. 3. Arrangement for uplift capacity test.

44 R. Nazir et al. / Measurement 49 (2014) 42–51

(LVDT) were incorporated with electrical motor gearing.The uplift load subject to the models was generated bythe motorized winch pull at a constant rate 0.6 mm/min.All arrangement of uplift test as shown in Fig. 3.

About 300 mm thick of sand layer will be initially fill inthe box with respected packing namely loose and dense. Toobtain a loose sand condition, the sand raining method willbe applied. The sand will rain from about 330 mm heightfrom the top of box. However to obtain the dense sand con-dition, sand have to be compacted for 2 min by using handheld vibrator for each 75 mm thickness of layer. For theloose sand case, the helical anchor model is placed at thecenter of glass box and sand filled into the glass box byusing sand raining method until the predetermine levelof model achieve. The model will incorporated with loadcell and LVDT and place into position. The pulling arm willthen connected to the motor winch. The model anchor at-tach with a pulling arm will be place into the center ofglass box vertically. The model was attached with a steelring of 15 mm � 2 mm to avoid any vibration of sand thatwill affect the verticality of model. After completing thelatter, the sand will be place layer by layer by using a handheld vibrator for each layer 75 mm thick. This step will berepeated until it reaches the predetermine level. A ninechannels data logger will be connecting to the load cell,LVDT where the acquisition of data commenced. The mod-el will imposed by vertical pullout load when the gearingmechanism was engaged using the auto control speed

Fig. 2. Single pitch helical anchor

through potentiometer. Loading will be imposes at aconstant rate 0.6 mm/min and the interval readings takenevery 10 s. The uplift test estimated will be completeapproximately 40 min.

5. Failure mechanism test arrangement

The purpose of a half cut model tested in glass box is togive a visual insight of the failure mechanism occurs dur-ing the progressive uplift movement of the pile. A custom-ised glass box is used for this purpose. The size of a glassbox use for the test is 600 mm length, 260 mm widthand 400 mm depth as shown in Fig. 4. Steel angle of size15 mm � 15 mm � 2 mm is use as its frame. Glass platewith thickness of 2 mm will be placed on both side of thebox. Steel plate of 2 mm thick will be placed at the othertwo side of the box. A geared motor with a speed of0.6 mm/min will be used to pull the half cut model whichis purposely cut vertically into half using cable connection.The arrangement of the system is as shown in Fig. 5.

model and half-cut model.

Fig. 4. The glass box used in the failure mechanism test.

R. Nazir et al. / Measurement 49 (2014) 42–51 45

In this arrangement, the failure mechanisms behaviorwere observed in both loose and dense sand packing. A lay-ers of colored sand with 3 mm thick were placed in thesand medium at every 20 mm intervals. The purpose ofthe colored sand is to enhance the visualization of sandmovement surrounding the half-cut helical model anchor.The position of the half-cut model is mark with a blacktape outside the surface of glass panel. The half-cut singlehelical anchor model placed at the position near the glasspanel with embedment ratio (H/Dh) = 2. The motor winchthen provide a constant rate of uplift load to the half-cutmodel. The half-cut model is then being pulled until it fail.Each prominent movement of the soil will be photo-graphed and geometrically analyzed.

6. Results and discussions for uplift response

In the pullout test series, thirty numbers of tests havebeen conducted to find out the effects of the variation ofembedment ratio, shaft diameter and sand density with re-spect to the uplift capacity of helical anchor. In the pullouttest series, the embedment ratio, (H/Dh) between 1 and 5,shaft’s diameter, Ds ranging between 30 mm and 50 mmwith loose and dense sand packaging were tested. Fig. 6shows the breakout factor, Nq for helical anchor model.The value is increasing for every increment of embedmentratio (H/Dh) in both loose and dense sand. It was as ex-pected that the breakout factor for dense sand is higher

Fig. 5. Arrangement for fai

in comparison with the loose sand packing. The breakoutfactor for model in dense sand increase rapidly than theloose sand. According to results on the loose sand package,the breakout factor is increase approximate 120% for a sin-gle increment in embedment ratio. While in the dense sandpackage, breakout factor increase approximately 130% for asingle increment in embedment ratio. This shows that theembedment ratio (H/Dh) have a significant effect towardsbreakout factor for model helical anchor where breakoutfactor increase with an increment in embedment ratio.

Fig. 7 illustrates that breakout factor, Nq for helical an-chor model embedded in loose sand is slightly decreasewhen the shaft diameter ratio, Ds/Dh increase. A similartrend also occurs for the model that embedded in densesand as shown in Fig. 8. Therefore, it can be conclude thatthe shaft diameter ratio has less significant influence to thebreakout factor. For model embedded in loose sand,breakout factor decreases at approximate 3–6% with anincrement of shaft diameter ratio. However, for modelembedded in dense sand, breakout factor decreasesapproximately 2–3% with an increase in shaft diameterratio. Thus for loose condition, the shaft diameter ratiohas more effect on the breakout factor in comparison withthe dense packing condition.

The results tested for effect of sand density have beensummarized and presented graphically in Fig. 9. The graphcompare the trend of breakout factor for model with vari-ations embedment ratio under loose and dense sand condi-tions. It is observed that the breakout factor of model in thedense sand package is higher than the model embedded inloose sand package. The breakout factor of helical anchormodel in dense sand are between 70% and 90% higher thanthe model embedded in loose sand. Therefore, it can beconcluded that the sand density also provide a significantinfluence to the helical anchor breakout factor. The rateof breakout factor will increase significantly when the sanddensity increase.

The breakout factor show an increment with embed-ment ratio. This condition is due to the overburden stressincrement above the helix. The deeper the embedment ofthe helix the higher will be the overburden stress thus itrequires a higher uplift load to pullout the anchor. The

lure mechanism test.

Breakout Factor of Model in Different Embedment Ratio

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6

Embedment Ratio, H/Dh

Brea

kout

Fac

tor,

Nq

Loose SandDense Sand

Fig. 6. Variations of breakout factor for helical anchor model in varies embedment ratio, (H/Dh) with shaft diameter ratio, Ds/Dh = 0.3.

Breakout Factor of Model in Different Shaft Dimater Ratio

0

10

20

30

40

50

60

0 0.1 0.2 0.3 0.4 0.5 0.6

Shaft Diameter Ratio, Ds/Dh

Brea

kout

Fac

tor,

Nq

H/Dh=1H/Dh=2H/Dh=3H/Dh=4H/Dh=5

Fig. 7. Variations of breakout factor, Nq with different shaft diameter ratio, Ds/Dh in loose sand.

Breakout Factor of Model in Different Shaft Diameter Ratio

0

20

40

60

80

100

120

0 0.1 0.2 0.3 0.4 0.5 0.6

Shaft Diameter Ratio, Ds/Dh

Bre

akou

t Fac

tor,

Nq

H/Dh=1H/Dh=2H/Dh=3H/Dh=4H/Dh=5

Fig. 8. Variations of breakout factor, Nq with different shaft diameter ratio, Ds/Dh in dense sand.

46 R. Nazir et al. / Measurement 49 (2014) 42–51

higher embedment ratio will produce a higher skin frictionbetween the sand and anchor that will increase the upliftcapacity to the anchor. As mentioned earlier, the shaftdiameter ratio effect shows that the shaft diameter has lessinfluence to the breakout factor in agreement with Mitschand Clemence [6] findings. Thus, it is confirmed that the

shaft friction do not provide any significant resistance forthe uplift capacity. While it can be concluded that the den-sity has a significant effect towards the breakout factor.This condition can be explained as when the sand densityis high, the volume of the sand increase, thus increasingits overburden and provide better resistance to the uplift

Breakout Factor of Model with ariation of Embedment Ratio in Loose and Dense Sand

0

20

40

60

80

100

120

0 1 2 3 4 5 6

Embedment Ratio, H/Dh

Brea

kout

Fac

tor,

Nq

Ds=30mm, Loose SandDs=40mm, Loose SandDs=50mm, Loose SandDs=30mm, Dense SandDs=40mm, Dense SandDs=50mm, Dense Sand

Fig. 9. Variations of breakout factor for model embedded in loose and dense sand with different embedment ratio.

R. Nazir et al. / Measurement 49 (2014) 42–51 47

capacity. The dense sand will provide a higher overburdenload to the helix in comparison with the loose sand.Besides, the friction angle of sand will increase with anincrement in sand density. A higher friction angle willproduce a higher frictional resistance which will thenincreasing its uplift capacity.

Fig. 11. Helix continually displaced.

7. Failure mechanism evaluations

Two pullout failure mechanism has been observed toevaluate the uplift failure mechanism of the helical anchormodel. A half-cut model have been used in this test and theuplift movement of the model was captured by the camera.The pullout failure mechanism test have been done in bothloose and dense sand package.

8. Failure mechanism of helical anchor model in loosesand

Figs. 10–12 show the results of uplift failure mechanismfor helical anchor model with embedment ratio (H/Dh) of 2in loose sand. Initially, the helical anchor model embeddedin the right position in the loose sand as shown in Fig. 10.The model is than subjected to uplift load and moveupward shown by the dyed line which start to distorted

Fig. 10. Position of model before subjected uplift load at initial stage.

Fig. 12. Mode of failure for helical anchor model.

upwards. Sand above the helix is than compacted andstarted to move as shown in Fig. 11. Fig. 12 shows the finalmode of failure for the helical anchor model. The mode offailure occurs in a curved shape. The dyed lines surroundthe shaft in Fig. 12 also show that the sand is not givinga high value of skin friction to the model base from the lessdistorted dyed line. The condition explain that the weightof sand and the frictional resistance in the failure zoneprovides a significant uplift capacity to the model thatembedded in loose sand.

Fig. 13. Helical anchor model subjected by uplift load and the modelstarts moving.

Fig. 14. Further displacement of helical anchor model in dense sand.

Fig. 15. Helical anchor model has fail and a failure zone have established.

48 R. Nazir et al. / Measurement 49 (2014) 42–51

9. Failure mechanism of helical anchor model in densesand

Similar to loose sand packing, a failure mechanism ofhelical anchor model which embedded in dense sand withembedment ratio of 2 are as shown in Figs. 13–15. Fig. 13shows the initial position of helical anchor model in thedense sand before subjected to uplift load. As the helicalanchor model initially subjected by uplift load, the sandabove the helix starts to move upwards as shown by themovement of dyed line.

Fig. 14 shows that the model continually pullout up-ward and the sand above helix displaced progressively.The displacement of model can be shown by the move-ment of the dyed line. In Fig. 15, the helical anchor modelcontinue to moves until a failure zone develope. The failurezone has been drawn out in Fig. 15. The failure zone shape

is presented in truncated cone shape in agreement withGhaly et al. [3]. It also shows that the inclination angle ofthe failure surface respects to vertical axis is approxi-mately 30�. The failure angle is approximately about 2/3of the friction angle of dense sand which is in agreementwith Ghaly et al. [3] findings.

10. Results comparisons

An empirical relationship will be carried out from theanalysis of the experimental results. This empirical rela-tionship will consider the effect of the embedment ratio(H/Dh) and the sand unit weight. The effect of shaft diam-eter ratio is not considered in the empirical relationshipbecause the effect of shaft diameter ratio does not giveany significant influence to the uplift capacity of model.The comparison of the empirical relationship with theexperimental result and two other existing theories suchas Mitsch and Clemence [6] and Ghaly et al. [3] will beput forward in this section. Both design method for helicalanchor has been proposed and the uplift capacity of helicalanchor will be computed from these two methods.

11. Empirical relationship for helical anchor

The data obtained from the laboratory tests will beused to produce the empirical relationship. In this work,two empirical relationship will be derived i.e. for looseand dense sand packing. In loose sand, the failure zoneoccurs in curved shape while in dense sand, the failurezone is in truncated cone shape. The failure zone fordense sand packing is always bigger than the failure zonein loose sand. In this work, the empirical relationships arederived from the breakout factor against embedmentratio as shown in Fig. 16. From this graph, the empiricalrelationship for loose sand and dense sand are listedbelow:

Loose sand

Nq ¼ 1:4142HDh

� �2:1378

Dense sand

Nq ¼ 2:1242HDh

� �2:3173

12. Comparisons between empirical relationship andexisting theories

The comparison between the empirical relationship andthe two existing theories are discuss in this section. Twoexisting theories were used in the comparison are fromMitsch and Clemence [6] and Ghaly et al. [3]. Mitsch andClemence [6] defined that the helical anchor with embed-ment ratio (H/Dh) of not more than 5 is consider as shallowanchor. Since the embedment ratio used in this compari-son ranging from 1 to 5, therefore the shallow helicalanchor relationship is used to compute the uplift capacityof helical anchor. The relationship for shallow anchor areas shown in Eqs. (1) and (2).

Breakout Factor in Variation of Embedment Ratio

y = 2.1242x2.3173

y = 1.4142x2.1378

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6

Embedment Ratio, H/Dh

Bre

akou

t Fac

tor,

Nq

loosedensePower (dense)Power (loose)

Fig. 16. Derivation of empirical relationship from graph breakout factor in variation of embedment ratio.

R. Nazir et al. / Measurement 49 (2014) 42–51 49

Q u ¼ pc0Ku tan / cos2 /2

� �D1H2

1

2þ

H31 tanð/2Þ

3

( )þWs ð1Þ

where Qu is the ultimate anchor uplift capacity, c0 is effec-tive unit weight of soil, Ku is lateral earth pressure in upliftfor sands, / is friction angle of soil, H1 is depth to top helix,and D1 is the diameter of top helix.

However Ghaly et al. [3], gives two relationship forshallow and deep anchor. Ghaly et al. [3] defines the shal-low anchor as anchor embedment with embedment rationot more than 11. Therefore, for this section the shallowanchor relationship used is as shown in Eq. (2).

Q u ¼ Pp þW

Pp ¼p2

cH2K 0pBþ H tan h

cos h

� �tan d

W ¼ p3cHðb2 þ r2 þ brÞ ð2Þ

where Qu is the ultimate pullout load, Pp is vertical compo-nent of total passive earth pressure, W is weight of sandwedge within failure surface, c is unit weight of sand, His embedment depth, K 0p is modified coefficient of passive

Comparison between Empirical Formul

0

5

10

15

20

25

30

35

40

45

50

3210

Embedment R

Bre

akou

t Fac

tor,

Qu

Empirical FormulaGhaly et. al. TheoryMitsch and Clemence Theory

Fig. 17. Comparisons between empirical relatio

earth pressure, B is diameter of helix, b is radius of helix,r is radius of influence failure circle on the sand surface,h is surface inclination angle of inverted failure cone re-spected to vertical, d is average mobilized angle of shearingresistance, and / is the friction angle of sand.

Figs. 17 and 18 show the comparison between authorsempirical relationship, Mitsch and Clemence [6] and Ghalyet al. [3] for helical anchor in loose and dense sand packing.These figures show that the authors empirical relationshiphave a good agreement with the Mitsch and Clemence [6]and Ghaly et al. [3]. Fig. 17 shows that the predicted break-out factor for anchor in loose sand has a close agreement toMitsch and Clemence [6] at low embedment ratio. How-ever, as embedment ratio increase the estimated breakoutfactor is closer to Ghaly et al. [3]. Fig. 18 predicted thebreakout factor for dense sand packing is in agreementwith the two existing theories at embedment ratio of lessthan 3. However, the authors prediction become higherthan existing theories when embedment ratio exceed 4.Figs. 17 and 18 show that Mitsch and Clemence [6] predicta lower breakout factor for both loose and dense condi-tions than the authors empirical relationship. While Ghalyet al. [3] gives a good agreement in both loose and densesand packages. It can be concluded that Mitsch and

a and Existing Theory in Loose Sand

654

atio, (H/Dh)

nship and existing theories in loose sand.

Comparison between Empirical Formula and Existing Theory in Dense Sand

0 1 2 3 4 5 60

10

20

30

40

50

60

70

80

90

100

Embedment Ratio, (H/Dh)

Bre

akou

t Fac

tor,

Qu

Empirical FormulaGhaly et. al. TheoryMitsch and Clemence Thaory

Fig. 18. Comparisons between empirical relationship and existing theories in dense sand.

50 R. Nazir et al. / Measurement 49 (2014) 42–51

Clemence [6] findings is more conservative in comparisonwith other theories. The situation is due to Mitsch andClemence derivations from the results of using torqueinstallation method in laboratory test. However, theauthors empirical relationship and Ghaly et al. used sandplacement method in the laboratory test. The differentmethods used in the laboratory tests have provideddifferent values of uplift capacity. The different laboratorytest method cause the breakout factor for Mitsch andClemence to be lower than the empirical relationship andGhaly et al. findings. However, all these design methodshave presented a similar trend for their predicted breakoutfactor in both loose and sand conditions. The breakoutfactor of helical anchor in both conditions increase withan increment in embedment ratio. It also indicate thatthe breakout factor for dense sand is higher than thebreakout factor in loose sand.

13. Conclusions

A research has been conducted to investigate the upliftbehavior of single helical anchor in sand. In this research, aseries of laboratory tests have been carried out to study themain influencing factors that affect the uplift capacity ofhelical anchor embedded in sand. The relationships be-tween the geometrical factors and the uplift behavior ofanchors have been obtained and analyzed in this research.The analysis for the effects of influencing factors has pro-vide a deeper understanding to the behavior of helical an-chor when subjected to pullout load. In this research, upliftcapacity of helical anchor in sand tested with three vari-ables that are embedment ratio (H/Dh), shaft diameter ratio(Ds/Dh) and sand density. Besides, observation has madetowards the mode of failure for single helical anchor thatembedded in loose and dense sand. An empirical relation-ship has been put forward from the analysis by consideringthe effects of embedment ratio and sand density. Findingsalso shows that the embedment ratio providing a signifi-cant influence to the uplift capacity of helical anchor. Theuplift capacity of helical anchor is in linear relationshipswith the embedment ratio. On the other hand, uplift capac-ity increase with an increment of an embedment ratio.Analysis of shaft diameter ratio illustrate that the shaft

diameter ratio has less significant effect to the uplift capac-ity of helical anchor. Only a small differences in the upliftcapacity of helical anchor with an increase in shaft diame-ter ratio. This condition is due to the skin friction sur-rounding the shaft contributes only a minor resistance tothe pullout load. The analysis of sand density shows theuplift capacity of helical anchor embedded in loose sandalways lower than anchor embedded in dense sand. Theanalysis shows that the uplift capacity for anchor in densesand is higher ranging between 70% and 90% than theanchor in loose sand. Therefore, it has been concluded thatthe uplift capacity of helical anchor in sand increase withthe increase of sand density. In the failure mechanism test,the mode of failure for anchor in loose sand and dense sandhave been observed. For the loose sand package, local fail-ure surface has been observed. However, for dense sandpackage, the single helical anchor fails in truncated coneshape. This truncated cone shape failure surface have closeagreement to the failure surface proposed by Mitsch andClemence [6] and Ghaly et al. [3]. In this research, empiri-cal relationship has been proposed based on the resultsobtained from the laboratory tests. In derivation of theempirical relationship, the embedment ratio and sanddensity have been considered. Two sets of empiricalrelationship proposed are for loose sand and dense sand.Comparison between the authors empirical relationshipwith existing theories [3,6], shows that generally it havea good agreement with the Ghaly et al. Theory [3] andMitsch and Clemence Theory [6]. For loose sand condition,the authors empirical relationship predict a similarbreakout factor with existing theories but for the densesand package, the authors empirical relationship is higherthan these existing theories when the embedment ratiomore than 4. Although the work done is not dedicated toany subgrade condition, however it was tought to beinvaluable to look into the insight of the uplift mechanismof single helical anchor embedded in homogeneous sandlayer.

Acknowledgement

The authors would like to thanks to Faculty of CivilEngineering, Universiti Teknologi Malaysia for providing

R. Nazir et al. / Measurement 49 (2014) 42–51 51

laboratory facilities during the course of this researchwork.

References

[1] Chance, Helical Pier Foundation System, 2004. <www.abchance.com>.[2] S.P. Clemence, F.D. Pepe, Measurement of lateral stress around

multihelix anchors in sand, ASTM Geotech. Test. J. 7 (1–4) (1984)145–152.

[3] A. Ghaly, A. Hanna, M. Hanna, Uplift behavior of screw anchors insand, J. Geotech. Eng. Div. ASCE (1991) 773–793.

[4] A. Ghaly, A. Hanna, Ultimate pullout resistance of single verticalanchors, Can. Geotech. J. (1994) 661–671.

[5] G.G. Meyerhof, J.I. Adams, The ultimate uplift capacity of foundations,Can. Geotech. J. 5 (4) (1968) 225–244.

[6] M.P. Mitsch, S.P. Clemence, The uplift capacity of helix anchors insand, Uplift Behav. Anchor Found. Soil ASCE (1985) 26–47.

[7] S.M. Rangari, D. Choudhury, D.M. Dewaikar, Seismic uplift capacity ofshallow horizontal strip anchor under oblique load using pseudo-dynamic approach, Soils Found., Elsevier 53 (5) (2013) 692–707.

[8] S.M. Rangari, D. Choudhury, D.M. Dewaikar, Computations of seismicpassive resistance and uplift capacity of horizontal strip anchors insand, Geotech. Geol. Eng., Int. J., Springer 31 (2) (2013) 569–580.