Evaluation of mechanical properties of sand …geo.iis.u-tokyo.ac.jp/research/documents/IIS Open...

1
Evaluation of mechanical properties of sand loosened due to piping ソイルパイプを有する土供試体の力学特性 Piping, known as a complex phenomenon of internal erosion, presents a great risk for hydraulic structures. In order to investigate the mechanical properties of sand with internal pipes, artificial piping was created by dissolving glucose column in a hollow- cylindrical specimen of Toyoura sand. In this research, triaxial compression(TC), triaxial extension(TE) and torsional shear(TS) tests were conducted on specimens with vertical pipes before shearing. Piping propagation during water infiltration was observed with Glip Gauges, and gap sensors were used to obtain shear modulus before and after piping effect by conducting small torsional cyclic loadings. Finally, monotonic torsional shear were applied to all the specimens until failure. 本研究に関する担当研究室は桑野研究室です. 部屋は東京大学生産技術研究所B棟3階のBw-304 電話: 03-5452-6843, FAX03-5452-6844 E-mail: [email protected] For further information, contact below. Prof. Reiko Kuwano, #Bw-304, Institute of Industrial Science TEL+81-3-5452-6843FAX+81-3-5452-6844 E-mail: [email protected] 6. Summary 1. Soil pipe Piping, which often occurs in hydraulic works, involving the removal of subsurface soils in a continuous tunnel to a free or escape exit. Landslide caused by piping has drawn much attention recently. Arrangement of Clip Gauges ( Radial strain measurement) Hollow cylindrical torsional shear apparatus 2. Testing apparatus and stress path 3. Testing procedure Specimen with relative densities around 75% were tested, and the number of internal pipe (Dia.=4.5mm) was 0 and 2. Specimens were erected at isotropic stress of 30kPa and then the confining pressure was increased to 60kPa. After isotropic consolidation around 12 hours, 1300ml water was infiltrated into specimen, with the aim of completely dissolving the glucose pipe. Triaxial compression and triaxial extension was conducted while kept p’ constant; for simple shear test, τ was applied at . Small torsional cyclic loading were conducted at the initial dry state, after water infiltrition and during TC (triaxial compression), TE (triaxial extension) and TS(torsional shear) under certain stress state. Monotonic torsional shear test were applied for all the specimens at last. 4. Results Specimen with glucose pipe 盛土や斜面に繰返し雨水が浸透することにより土粒子が流出しパイプ状の水みちができる場合があり、ソイルパイプと呼ばれる。浸透した雨水を速や かに排水する効果が期待されるものの、何らかの理由で水みちが閉塞された場合は一転して危険であるうえ、土構造内に空隙を有することになり構 造上の弱点となりうる。本研究では、中空ねじり試験供試体内にグルコースを溶解させることにより人工的なパイプ生成を試み、供試体の変形強度特 性に対するパイプの数や拘束圧、土の密度の影響を調べた。 Shear modulus variation during triaxial extension Shear modulus vs. Shear Strength 60 61 62 63 64 18 20 22 24 26 G z /f(e)~( ' z ' ) 0.5n NP, n=0.505 2P, n=0.495 TC-NP TC-2P TC-NP-Dry TC-2P-Dry G z /f(e) (MPa) ( ' z ' ) 0.5 (kPa) 52 54 56 58 60 18 20 22 24 26 G z /f(e)~( ' z ' ) 0.5n NP, n=0.355 2P, n=0.171 TE-NP TE-2P TE-NP-Dry TE-2P-Dry ( ' z ' ) 0.5 (kPa) G z /f(e) (MPa) 0 5 10 15 20 25 30 18 20 22 24 26 TS-NP TS-2P TS-NP-Dry TS -2P-Dry G z /f(e) (MPa) (kPa) Shear modulus variation during triaxial compression 0 10 20 30 40 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Increament of vol due to glucose dissolving was around 0.328% in average for specimens with pipe (Initial glucose volume: 0.62% of specimen) Volumetric Strain vol (%) Time(h) TC-NP TC-2P TE-NP TE-2P TS-NP TS-2P Increament of vol due to confining pressure increase was around 0.440% Volumetric strain variation before shearing 0 5 10 15 20 25 30 -10 0 10 20 30 40 50 60 70 TE-2P TE-NP SS-2P SS-NP TC-2P TC-NP (kPa) TC-NP sheared at (100,40,40) peak = 62.4kPa TC-2P sheared at (100,40,40) peak = 59.2kPa TE-NP sheared at (40,70,40) peak = 41.8kPa TE-2P sheared at (40,70,70) peak = 40.2kPa Simple shear-NP sheared at (60,60,60) peak = 56.8kPa Simple shear-2P sheared at (60,60,60) peak = 50.6kPa Shear stress vs. shear strain 0 5 10 15 -8 -6 -4 -2 0 Volumetric strain vol (%) Shear strain (%) Shear at (60,60,60)-NP Shear at (60,60,60)-2P Shear at (100,40,40)-NP Shear at (100,40,40)-NP Shear at (30,70,70)-NP Shear at (30,70,70)-NP Volumetric strain during shearing σ p q Deviator Stress (σz- σr) τ TC TE TS Stress path before shearing Isotropic consolidation at (60,60,60) Gap Sensor ( Shear modulus measurement during small torsional cyclic loadings)

Transcript of Evaluation of mechanical properties of sand …geo.iis.u-tokyo.ac.jp/research/documents/IIS Open...

Page 1: Evaluation of mechanical properties of sand …geo.iis.u-tokyo.ac.jp/research/documents/IIS Open 2015...Evaluation of mechanical properties of sand loosened due to piping ソイルパイプを有する土供試体の力学特性

Evaluation of mechanical propertiesof sand loosened due to piping

ソイルパイプを有する土供試体の力学特性

Piping, known as a complex phenomenon of internal erosion, presents a great risk for hydraulic structures. In order to investigatethe mechanical properties of sand with internal pipes, artificial piping was created by dissolving glucose column in a hollow-cylindrical specimen of Toyoura sand. In this research, triaxial compression(TC), triaxial extension(TE) and torsional shear(TS)tests were conducted on specimens with vertical pipes before shearing. Piping propagation during water infiltration was observedwith Glip Gauges, and gap sensors were used to obtain shear modulus before and after piping effect by conducting small torsionalcyclic loadings. Finally, monotonic torsional shear were applied to all the specimens until failure.

本研究に関する担当研究室は桑野研究室です.部屋は東京大学生産技術研究所B棟3階のBw-304

電話: 03-5452-6843, FAX: 03-5452-6844

E-mail: [email protected]

For further information, contact below. Prof. Reiko Kuwano, #Bw-304, Institute of Industrial Science

TEL: +81-3-5452-6843, FAX: +81-3-5452-6844E-mail: [email protected]

6. Summary

1. Soil pipe

Piping, which often occurs in hydraulic works,involving the removal of subsurface soils in acontinuous tunnel to a free or escape exit.

Landslide caused by piping has drawn muchattention recently.

Arrangement of Clip Gauges( Radial strain measurement)

Hollow cylindrical torsional shear

apparatus

2. Testing apparatus and stress path

3. Testing procedure

Specimen with relative densities around 75% weretested, and the number of internal pipe(Dia.=4.5mm) was 0 and 2.

Specimens were erected at isotropic stress of 30kPaand then the confining pressure was increased to60kPa.

After isotropic consolidation around 12 hours,1300ml water was infiltrated into specimen, withthe aim of completely dissolving the glucose pipe.

Triaxial compression and triaxial extension wasconducted while kept p’ constant; for simple sheartest, τ was applied at .

Small torsional cyclic loading were conducted atthe initial dry state, after water infiltrition andduring TC (triaxial compression), TE (triaxialextension) and TS(torsional shear) under certainstress state.

Monotonic torsional shear test were applied for allthe specimens at last.

4. Results

Specimen with glucose pipe

盛土や斜面に繰返し雨水が浸透することにより土粒子が流出しパイプ状の水みちができる場合があり、ソイルパイプと呼ばれる。浸透した雨水を速やかに排水する効果が期待されるものの、何らかの理由で水みちが閉塞された場合は一転して危険であるうえ、土構造内に空隙を有することになり構造上の弱点となりうる。本研究では、中空ねじり試験供試体内にグルコースを溶解させることにより人工的なパイプ生成を試み、供試体の変形強度特性に対するパイプの数や拘束圧、土の密度の影響を調べた。

Shear modulus variation during triaxial extension Shear modulus vs. Shear Strength

60 61 62 63 6418

20

22

24

26

Gz

/f(e)~('

z'

)0.5n

NP, n=0.5052P, n=0.495

TC-NP TC-2P TC-NP-Dry TC-2P-Dry

Gz

/f(e

) (M

Pa)

('

z'

)0.5 (kPa)

52 54 56 58 6018

20

22

24

26

Gz

/f(e)~('

z'

)0.5n

NP, n=0.3552P, n=0.171

TE-NP TE-2P TE-NP-Dry TE-2P-Dry

('

z'

)0.5 (kPa)

Gz

/f(e

) (M

Pa)

0 5 10 15 20 25 3018

20

22

24

26

TS-NP TS-2P TS-NP-Dry TS -2P-Dry

Gz

/f(e

) (M

Pa)

(kPa)

Shear modulus variation during triaxial compression

0 10 20 30 40-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Increament of vol

due to glucose dissolving

was around 0.328% in average for specimens with pipe (Initial glucose volume: 0.62% of specimen)

Vol

umet

ric

Str

ain vo

l (%

)

Time(h)

TC-NP TC-2P TE-NP TE-2P TS-NP TS-2P

Increament of vol

due to confining pressure

increase was around 0.440%

Volumetric strain variation before shearing

0 5 10 15 20 25 30-10

0

10

20

30

40

50

60

70

TE-2P

TE-NP

SS-2P

SS-NP

TC-2P

TC-NP

(k

Pa)

TC-NP sheared at (100,40,40) peak

= 62.4kPa

TC-2P sheared at (100,40,40) peak

= 59.2kPa

TE-NP sheared at (40,70,40) peak

= 41.8kPa

TE-2P sheared at (40,70,70) peak

= 40.2kPa

Simple shear-NP sheared at (60,60,60) peak

= 56.8kPa

Simple shear-2P sheared at (60,60,60) peak

= 50.6kPa

Shear stress vs. shear strain

0 5 10 15-8

-6

-4

-2

0

Vol

umet

ric

stra

in

vol (

%)

Shear strain (%)

Shear at (60,60,60)-NP Shear at (60,60,60)-2P Shear at (100,40,40)-NP Shear at (100,40,40)-NP Shear at (30,70,70)-NP Shear at (30,70,70)-NP

Volumetric strain during shearing

σ

p

q

Deviator Stress(σz- σr)

τ

TCTE

TS

Stress path before shearing

Isotropic consolidation at

(60,60,60)

Gap Sensor( Shear modulus

measurement during small torsional cyclic

loadings)