In-Situ Load Test of High Capacity Micropiles_Paper

9
IWM 5 th 2002 In Venice Session V In-situ Load Test of High Capacity Micropiles foundation on mountain ground In-situ Load Test of High Capacity Micropiles foundation on mountain ground Masao Sagara * Kouichiro Shito** Yoshinori Igase** Gaku Ohashi** Etsuro Saito* Terukatsu Sasaya* Toshihisa Hatano* Yoshito Maeda*** * FUJITA Corporation, Technology Development Division ** Japan Highway Public Corporation ***Professor of Kyushu Kyoritsu University, Development of Civil Engineering Abstract In general, the caisson type pile method is used on the mountain ground. However, there are some problems in the caisson type pile method. The High Capacity Micropiles method is expected as a method of taking the place of the caisson type pile method. This Paper describes “In-situ Load Test of High Capacity Micropiles foundation on mountain ground”. The following findings were obtained by the experiment. 1) It is possible to construct HMP in the bedrock ground. It has enough bearing capacity. 2) The Bearing capacity is transmitted from the upper part to the lower side. End bearing capacity of HMP was confirmed. 3) The Bearing capacity is transmitted from the upper part to the lower side. 4) More batter piles than the straight piles are stronger. 5) Constructed HMP toe was confirmed by watching. 9/1 Fujita Corporation

description

In general, the caisson type pile method is used on the mountain ground. However, there are some problems in the caisson type pile method. The High Capacity Micropiles method is expected as a method of taking the place of the caisson type pile method.

Transcript of In-Situ Load Test of High Capacity Micropiles_Paper

Page 1: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

Masao Sagara *

Kouichiro Shito** Yoshinori Igase** Gaku Ohashi**

Etsuro Saito* Terukatsu Sasaya* Toshihisa Hatano*

Yoshito Maeda***

* FUJITA Corporation, Technology Development Division

** Japan Highway Public Corporation

***Professor of Kyushu Kyoritsu University, Development of Civil Engineering

Abstract

In general, the caisson type pile method is used on the mountain ground. However, there are

some problems in the caisson type pile method. The High Capacity Micropiles method is

expected as a method of taking the place of the caisson type pile method. This Paper describes

“In-situ Load Test of High Capacity Micropiles foundation on mountain ground”. The following

findings were obtained by the experiment.

1) It is possible to construct HMP in the bedrock ground. It has enough bearing capacity.

2) The Bearing capacity is transmitted from the upper part to the lower side. End bearing

capacity of HMP was confirmed.

3) The Bearing capacity is transmitted from the upper part to the lower side.

4) More batter piles than the straight piles are stronger.

5) Constructed HMP toe was confirmed by watching.

§1. Introduction

In general, the caisson type pile method is used on the mountain ground. The caisson type pile has

the problem in construction such as needing the installation of the digging machine on a narrow site.

Moreover, there is a problem in the environmental, which should be dispose of the dug gravel etc.,

too.

On the other hand, the High Capacity Micropiles method is known generically of the

cast-in-place pile of 300mm or less. The construction machine of the High

Capacity Micropiles method is small. It is suitable for construction on a narrow site.

Because the diameter of High Capacity Micropile is small, the digging soil volume

8/1Fujita Corporation

Page 2: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

is also comparatively few. It is an expected method as a foundation pile in the

mountain ground. We are researching for High Capacity Micropiles method

(hereafter, HMP) which uses the high strength steel pipe, the deformed bar, and

the grout material. We have aimed to examine the applicability of HMP on the

mountain ground. I reported on “Lateral Loading Test for a Model of Micropiles

Foundation on slope ground" in IWM 2001. This Paper describes “In-situ Load Test of

High Capacity Micropiles foundation on mountain ground” and “ Confirmation of constructed

HMP toe ”.

§2. Test pile of HMP and Ground Condition

8/2Fujita Corporation

Table 1  The material of HMPmaterial specification

Steel pipe

The high strength steel pipe (API-N80) 、 Outside diameter =φ177.8mm,Thickness =12.7mm, Standard length =1,500mm, Yield Strength =550N/mm2, Elasticity coefficient =2.0×105N/mm2,

Deformed bar

Yield Strength =490N/mm2(SD490) Calling diameter=51mm(D51), Elasticity coefficient =2.0×105N/mm2,

GroutingCement ( cement water ratio W/C=50% ) , High-early-strength portland cement, Design Strength=30N/mm2, Elasticity coefficient =1.36×104/mm2(result of test)

Bond of HMP

Drilling diameter 205mm, Bond length=4000㎜(vertical pile),4061㎜(batter pile),

D強風化泥岩

8000

4000

4000

bearingstratum

900(5D)

7500

300

9500

1700

2362

4062

1077

2326

1674

8000

4000

4000

2500

1500

9000

6000

500

500500250

1250

a Test pile()(batter pile 10° )

b Test pile()(vertical pile)

311295

1356

1000

weatheringmudstone

crushedmudstone

mudstone

mudstone

crushed mudstone

mudstone

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

0 10 20 30 40 50

50/ 17

50/ 16

50/ 8

50/ 8

50/ 10

50/ 4

50/ 4

50/ 4

50/ 0

50/ 2

50/ 0

50/ 0

very weatheringmudstone

N Value

Figure1   The outline of geological column and the test pile of HMP

Page 3: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

The material of HMP is shown in Table1. The load test of HMP was done in the bedrock ground on

the Nakaissiki bridge construction site in the second Toumei expressway of Japan Highway Public

Corporation. After the load test of HMP, caisson type pile of 13.5m in diameter and 16.5m in depth

will be constructed in the place. Therefore, there is no problem even if HMP is dug up. The outline

of geological column and the test pile of HMP are shown in Figure1. The stratum up to -1.17m in the

underground is weathering mudstone like the sand of the clay mixing. As for the stratum from -

1.17m in the underground to -2.8m, weathering becomes weak more than stratums located up.

However, it is weak mudstone, which has a lot of crack, and becomes crushed gravel easily. The

rock from -2.8m in the underground to -5.6m is fresh. However, there are some cracks. The stratum

at deeper position than -5.6m in the underground is fresh mudstone with little weathering.

8/3Fujita Corporation

500

500

500

500

500

500

500

1000

1500

1000

1300

200

8500

700

700

700

400

550

750

2000

2000

500

200

8500

10゚

508

508

508

508

508

508

1015

1523

1015

1320

203

Sec t ion ①

Sec t ion ②

Sec t ion ③

Sec t ion ④

Sec t ion ⑤

Sec t ion ⑥

Sec t ion ⑦

Sec t ion ⑧

Sec t ion ⑨ Sec t ion ⑪

Sec t ion ⑩

Sec t ion ⑨

Sec t ion ⑧

Sec t ion ⑦

Sec t ion ⑥

Sec t ion ⑤

Sec t ion ④

Sec t ion ③

Sec t ion ②

Sec t ion ① Sec t ion ①

Sec t ion ②

Sec t ion ③

Sec t ion ④

Sec t ion ⑤

Sec t ion ⑥

Sec t ion ⑦

Sec t ion ⑧

Sec t ion ⑨

Sec t ion ⑩

Sec t ion ⑪

Bond

leng

th4

.061

m

Bond

leng

th4

.0m

Bond

leng

th4

.0m

G rou t ing

De fo rmed ba r ,

H igh s t reng th s tee l p ipe

A l te rna t ing Ve r t ica l Load ing Tes t P i leVe r t ica l Load ing Tes t P i le

A l te rna t ing La te ra l Load ing Tes t P i le

A l te rna t ing La te ra l Load ing Tes t fo r a coup le o f ba t te r p i les

Figure2 Test Piles

Page 4: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

8/4Fujita Corporation

0

20

40

60

80

100

0 1000 2000 3000 4000 5000Load (kN)P

Dis

pla

cem

ent

(m

m)

Yield bearing capacity2100kN

Ultimate bearing capacity4100kN

Fig.3 The relation of load-vertical displacement

0

1

2

3

4

5

6

7

8

9

0 1000 2000 3000 4000 5000Axis force (kN)

Depth

(

)m

Fig.4 The axis force

0

1000

2000

3000

4000

5000

6000

0 10 20 30 40 50Relative displacement δ (mm)

Ski

n f

rict

ion s

tress

τ(k

N/m2 )

① ②Section~② ③Section~③ ④Section~④ ⑤Section~⑤ ⑥Section~⑥ ⑦Section~⑦ ⑧Section~⑧ ⑨Section~

ultimatebearingcapacity

yieldbearingcapacity

Fig.5 The relation between skin friction stress and displacement

Page 5: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

§3. Outline of the Loading Test for the HMP

The test piles of the full-scale HMP were constructed, and vertical loading test, alternating vertical

loading test, alternating lateral loading test, alternating lateral loading test for a couple of batter piles

were done. Details of the test piles of HMP are shown in Figure2. The confirmation of the

constructed paling targeted healthy anchor piles. Healthy anchor piles were targeted for the

confirmation of the constructed HMP. HMP were dug up by the casing drilling of 600 ㎜ diameter.

We confirmed HMP toe by watching.

§4. Result of The Loading Test

4.1 Vertical Loading Test

Figure 3 shows the relation of load-vertical displacement. The vertical yield bearing capacity is

2100kN when the pile top displacement is 12.4㎜. The vertical ultimate bearing capacity is 4100kN

when the pile top displacement is 60.0㎜. 60 ㎜ is the amounts of the displacement of about 30% of

the diameter of HMP. The surroundings area of bond length, counted divides ultimate bearing

capacity, and the obtained value is skin friction stress. Skin friction stress of the average is

1524kN/m2. This value corresponds to skin friction stress of hard rock in the ground anchor standard

(The Japanese Geotechnical Society). The axis force chart is shown in Figure4. The value with a

very big warp was indicated in the boundary part of the steel pipe and the deformed bar (⑥

from section ⑤) while becoming near yield bearing capacity. Because it seemed an abnormal value,

the revised value is indicated in figure. It is understood that bearing capacity is supported with bond

part of 5-8m in depth. And, we understood that bearing capacity is transmitted from the upper part to

pile toe in bond part of HMP. The axis force of about 800kN was confirmed in the pile toe, and end-

bearing capacity of HMP was confirmed in the bedrock ground when it is the vertical ultimate

bearing capacity. The relation between skin friction stress and displacement is shown in Figure5.

Skin friction stress indicates a big value in ⑦-⑧ section and ⑧-⑨section at the vertical ultimate

bearing capacity, and it is understood that the HMP is supported in ⑦-⑧ section.

8/5Fujita Corporation

Page 6: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

4.2 Alternating Vertical Loading Test

Result of alternating vertical loading test is shown in Figure6. Both the pulling yield bearing

capacity and the pushing yield bearing capacity is 1200kN. Pulling resistance of HMP is 1800kN

when the pile top displacement is 20㎜. It is the amounts of the displacement of about 10% of the

diameter of HMP. When pulled when the paling is pushed, the inclination of stress-displacement is

equal. The relation between skin friction stress and the amount of relative displacement at the time of

8/6Fujita Corporation

-40

-20

0

20

40

60

80

100

-2000 -1000 0 1000 2000Load (kN)P

Dis

pla

cem

ent

(m

m)

SPullPush

up

down

Fig.6 Result of alternating vertical loading test

0

500

1000

1500

2000

2500

0 1 2 3 4

Relative displacement δ (mm)

Ski

n fric

tion

str

ess

τ

(k

N/m

2 )

① ②Section~

② ③Section~

③ ④Section~

④ ⑤Section~

⑤ ⑥Section~

⑥ ⑦Section~

⑦ ⑧Section~

⑧ ⑨Section~

Pull

yieldbearingcapacity

Fig.7 skin friction stress and the amount of relative displacement

Page 7: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

pulling out is shown in Figure7. Skin friction stress increases in ⑦-⑧ section and ⑧-⑨ section

while the amount of relative displacement is large, and skin friction stress has decreased in ⑤-⑥

section and ⑥-⑦ section when it is pulling yield bearing capacity in Figure7. It is understood that

the bearing capacity has moved from the upper part of bond of HMP to pile toe when the paling is

pulled out.

4.3 Alternating Lateral Loading Test & Alternating Lateral Loading Test for a couple of batter

piles

8/7Fujita Corporation

Table2  The diameter of the HMP toeMeasurement position

(Distance from HMP toe)cm Surroundings

(cm)

Diameter (cm)

(Grout & Deformed bar)(24cm)

----- 24

(Pipe&Grout & Deformed bar) (3.5m)

93 29.6

(Pipe&Grout & Deformed bar) (4m)

94 29.9

Photograph1 Constructed HMP toe

Page 8: In-Situ Load Test of High Capacity Micropiles_Paper

IWM 5th 2002 In VeniceSession V

In-situ Load Test of High Capacity Micropiles foundation on mountain ground

Result of alternating lateral loading test and

alternating lateral loading test for a couple of

batter piles is shown in Figure8. In the model

test (batter pile angle was 15°) of last year, if

the number of the pile is the same, it is

confirmed that the batter piles have about three

times the horizontal bearing capacity than the

vertical piles. It is confirmed that the batter

piles have about eight times the horizontal

bearing capacity than the vertical piles though

the batter pile angle is 10° in this test.

4.4 Confirmation by watching constructed HMP

The diameter of the HMP of 24 ㎝ is confirmed in the pile toe. On the other hand, the diameter of

the HMP of about 30cm was confirmed in the bond part, which included the steel pipe (Table2 and

photograph1 reference). These values are larger than dug diameter (205 ㎜ ). It is thought that the

extension of the dug hole was secured by the pressed grouting.

§5. Conclusions

The following have been understood from this in-site load test.

1) It was confirmed to be able to construct HMP, and to have enough bearing capacity in the

bedrock.

2) It is understood that the bearing capacity has moved from the upper part of bond of HMP to pile

toe.

3) End bearing capacity of HMP was confirmed in the bedrock ground.

4) The horizontal bearing capacity of batter pile is larger than the vertical pile.

5) The bond of HMP was surely constructed by watching in the bedrock ground.

8/8Fujita Corporation

0

200

400

600

800

1000

1200

1400

1600

1800

0 50 100 150 200 250 300Horizontal displacement(㎜)

Late

ral l

oad(k

N)

Vertical pile

Batter piles 10°

Fig.8 Lateral load-Horizontal displacement