Sub-VIs

The sub-VIs to LabVIEW is alike the sub-programs to FORTRAN, which are written for special purposes. The sub-VIs are used to simplify any complicated and huge project, such as industrial automation or testing programs. The testing program could think as the combination of several different kinds of sub-VIs, which have the particular functions. For example, the sub-VIs used in the shearing program are (1) motor movement sub-VI, (2) data acquisition sub-VI, and (3) motor pulse compensation sub-VI. The above three major block sub-VIs wired to each other and formed the completed shearing program. The main sub-VIs used in this testing program will introduce in detail in following paragraphs.

Motor pulse compensation sub-VI

The absolute position accuracy of the D.D. motor is 150 sec, and the pitch of screw is 5 mm. The absolutely position accuracy means the errors after the motor rotated 360°. In order to compensate for the errors, the program should adjust the output pulse number when the motor was in motion. The pedestal will raise 0.1 μm while the motor move 16 pulses, and it needs 51200 times to complete a whole circle.

The error of 150 sec will cause the error of 0.6 μm when the pedestal raising 5 mm.

In other words, the 0.6 mμ will be distributed into 51200 parts equally, and each piece of error is1.171875×10⁻⁵μm. Through a simple calculation, the compensation of the motor pulse is to send 17 pulses for the fifth time and 16 pulses for the other times.

Specimen setup and homing sub-VI

After preparing the specimen, the next step is to set it up on the pedestal and let the top of specimen touch the cap. In traditional triaxial testing apparatus, this step was always controlled by manual. To set up specimen by this way might cause a serious problem, which the pedestal will go too far and squeeze the sample, and any tiny squeezing of the specimen will influence the soil properties at small stain. For that reason, the specimen setup sub-VI was written. The specimen setup sub-VI made the motor move slightly but fast, and after each minute movement the program will check the reading of load cell. Once the increment of the reading of load cell equals to 0.1 N, the motor will stop.

The homing sub-VI is to make the motor go back home position by the maximum speed. This sub-VI can prevent the screw from bumping into the motor and causing the damage. Besides this sub-VI, there are 2 pairs of limit switch to keep the screw from harm. One of the 2 pair limit switch control the motion controller card, and another is to cut off the power of motor directly.

Parameters calculation sub-VI

The acquired data contain the measurement of all sensors, which represent the load, the pore pressure, the displacement, and temperature. The soil parameters, such as strength, strain, and so on depend on these acquired data. But how to transform these data into the parameters we concerned is the reason of this sub-VI being written.

The parameter calculation sub-VI needs to make use of some simple mechanical concepts and theories of soil mechanics. For example, the transform of the load cell reading to the approximate axial stress; the displacement translate to strains; the stiffness of the tested soil specimen; the stress state and stress path, etc. The outputs of this sub-VI are used to plot all kinds of charts and record in files. Another function of this sub-VI is to read, record, calculate, and use the input parameters of soil specimen, which are the size of specimen, all moving rate, and limits.

Digital pressure controller sub-VI

The digital pressure controller is used to provide cell pressure and back pressure.

The user controlled the pressures manually in saturation process. But in most testing procedures, these pressures should be controlled automatically by testing program.

Each digital pressure controller has a control chip in the case and a RS-232 connector, which takes part in sending and receiving data with testing computer. Although the control chip is not the product of NI, but LabVIEW still has the ability of connecting

to the chip. This sub-VI provides a control panel in the testing program and can control two digital pressure controllers directly. The functions of “digital pressure controller sub-VI” contain:

1. Send the required pressure and volume to digital pressure controller.

2. Make the digital pressure controller to fill or empty water.

3. Read the present volume of the digital pressure controller.

4. Stop all motions of the digital pressure controller.

Therefore, the testing program, especially in k₀ consolidation stage, can adjust the pressures freely through this sub-VI.

Calibration sub-VI

This sub-VI provides a semi-automation process to calibrate all sensors. When calibration sub-VI is running, the readings of chosen channel will be shown on the monitor, and press “save” button to save the true value and measured voltage. It’s more convenient than before. Besides the convenience, it also more practical than that recorded manually. Because the motion of record is simultaneously to the change of physical phenomenon, there will be almost no errors. Therefore, the users could calibrate transducers quickly and finely through this sub-VI.

Fig. 4.1 Information from signals

Fig. 4.2 (a) Amplifier case

Fig. 4.2 (b) Amplifier components

Fig. 4.3 Comparison of the attenuation of transfer functions of a real filter and an ideal filter

Fig. 4.4 Response of a step signal under different situations

Fig. 4.5 Sine wave signal sampled at the indicated points

Fig. 4.6 Typical DAQ system

Fig. 4.7 The layout of the screw terminals

Fig 4.8 (a) flowchart of k₀ consolidation before modifying

Read in-situ σ_v^'

Motor move 81pulse

dl＜0.0005 yes

Controller increase cell pressure

0.0001

Read in-situ σ_v^'

Motor move

kPa

<1

Δσ_c ^yes

Controller increase cell pressure

0.0001

-ε₁ >

ε_v

no

yes no

yes

Record no

yes no

Adjust cell pressure

0 - ^'

'

1 σ_v >

σ _STOP

Motor forward slightly

Motor reverse slightly

0.0001

1-ε_v >

ε

Chapter 5

The results of development

The development of the new triaxial testing system was complete, i.e. each component introduced in chapter 2 to chapter 4 was combined into a completed testing system. The photographs of the loading system, the digital pressure controller, all measurement sensors, and the testing programs are going to show in following.

According to the specifics of every single component, the performances of this testing system were determined. And this small strain triaxial testing system is going to compare with other testing systems in different countries.

5.1 Development Results of Hardware

The hardware included the loading system, digital pressure controller, and transducers, which are mentioned in chapter 2 and 3. Since there are only sketches of devices in these chapters, so we will illustrate all devices with their photographs latter.

The first part of testing system is the loading system used to apply the axial load to soil specimen, and is consisted of the D.D. motor, driver unit, high-precision ball screw, and loading frame. The photographs of loading system are shown in Fig. 5.1 to Fig. 5.5. The explanation of each figure is as follow:

z Fig. 5.1—shows the case of loading system. In order to simplify all the tubes of pressure sources surrounding with the base of the triaxial cell, we designed 4 valves on the case and put all the pressure tubes into the case.

Therefore, the user just switches the valve when needed.

z Fig. 5.2—shows the input connectors of the pressure tubes. The de-air water was pressed into the triaxial cell through this connectors.

z Fig. 5.3—shows the switch of power supply and the emergency-stop button.

z Fig. 5.4—shows the side view of the loading system. There are loading frame, D.D. motor, driver unit, ball screw, and power supply.

z Fig. 5.5—shows the bird-view of the loading frame. There are two block in this figure, and they are linear device used to make the shaft of loading system move linearly.

Secondly, the photo of digital pressure controller was shown in Fig. 5.6. All the

components of the digital pressure controller introduced in chapter 2, such as the stepping motor, gearbox, screw, cylinder, control panel, and LCD, were illustrated in this figure. The photos of transducers used in this testing system are shown in Fig. 5.7 to Fig. 5.9. Fig. 5.7 and 5.8 show the radial and axial Hall Effect semiconductor displacement transducer, respectively. And Fig. 5.9 shows the miniature pore pressure transducer.

5.2 Development Results of Software

The software means the testing program here. The functions of testing program were introduced in detail in chapter 4, and we are going to show some pictures of testing program, including the main menu and sub-VIs. Each figure will show and explain in following:

z Fig. 5.10—shows the test sheet in main menu. In this page, user input the specimen size and performs the test he wanted.

z Fig. 5.11—shows the DAQ sheet in main menu. All transducers reading are monitored in this page, and user could set zero for each sensor by press the

“set zero” button.

z Fig. 5.12—shows the function of “set up specimen” and “homing”. Besides those, there are two lights to warn that the screw arrive the limit position.

z Fig. 5.13—the plots were drawn in this page during test. The most common plots are “stress-strain graph” and “pore pressure graph”, which were shown in this figure.

z Fig. 5.14—shows the menu of k0-consolidation.

z Fig. 5.15—shows the sub-VI used to communicate with the digital pressure controller.

z Fig. 5.16&5.17—show the picture of undrained shearing. User input the strain rate and limitary conditions in the input-page, and the program will calculate the rotational speed of the motor. And the current position of motor was shown on the “pulse of motor “page, in Fig. 5.17.

5.3 Comparison between Different Testing Systems

The testing system, which could perform small-strain triaxial tests, was brought out by different research group in different country. But not all the testing systems are

the same in detail, and the difference between them will show up in this section. We can discuss the variety in several aspects, like the loading device, the pressure sources, the measurements for small displacements, and the attainable minimum strain. The testing system developed in this study is going to compare with the system in University of Tokyo and in Italy (modified by Diego C. F. Lo Presti). The comparison is listed in Tab. 5.1.

Tab. 5.1 Comparison of different testing system

Different testing system

Specifics NTUST Univ. of Tokyo Italy

loading device

gearbox pressure chamber

pressure source

Fig. 5.1 The case of loading system Cell

pressure

Back pressure

Upper drainage valve

Filling valve

Fig. 5.2 The case of loading system Connectors

Fig. 5.3 The case of loading system

Fig. 5.4 The inner of loading system (side-view)

Fig. 5.5 The inner of loading system (bird-view)

Fig. 5.6 Photo of digital pressure controller Cylinder Screw

Stepping motor & gearbox Linear device

LCD

Control panel

Fig. 5.7 Radial Hall Effect sensor

Fig. 5.8 Axial Hall Effect sensor

Fig. 5.9 Miniature pore pressure transducer

Fig. 5.10 Main menu 1 of testing system Miniature pore

pressure transducer

Fig. 5.11 Main menu 2 of testing system

Fig. 5.12 Main menu 3 of testing system

Fig. 5.13 Main menu 4 of testing system

Fig. 5.14 Main menu of k₀-consolidation

Fig. 5.15 Sub-VI of digital pressure controller

Fig. 5.16 Undrained shear menu 1

Fig. 5.17 Undrained shear menu 2

Chapter 6

The results of tests at small strain

A series of the triaxial tests using the triaxial testing system developed in this study were carried out on the reconstituted Taipei silty clay, which was sampled from Hsinyi district, to investigate the stress-strain characteristics at small strain. All the sampling sires are located in the Keelung river basin. The detailed test procedures and test results can be referred to the following sections.

6.1 Testing program

After the advanced triaxial testing system equipped with the Hall effect semiconductor local strain transducers, and the miniature pore water pressure transducers was developed, this study drew up a testing program to investigate the small strain behavior of Taipei silty clay. This study carried out a series of small strain triaxial compression tests on the reconstituted clay to investigate the stress-strain characteristics at small strain and the influence degree affected by the initial effective stress and shearing rate. The reason why this study adopted the reconstituted clay to conduct the small strain triaxial tests was that reconstituted clay was a more homogenous material than in-situ clay. Also the stress history of reconstituted clay can be carefully controlled.

6.2 Small strain triaxial tests

(1) Description of the soil sample

The clay with low plasticity sampled from the Keelung River Basin was used. We first crumbled the original dried clay; then took the clay powder that had passed through a 425-micron sieve to churn it well with water to reach the water content of twice of the liquid limit. The resultant slurry was then subjected to pre-consolidation in the odeometer under the final loading pressure of 150 kPa.

The basic properties of the reconstituted clay are shown in the following:

3

(2) Procedures of triaxial tests

A number of reconstituted samples were tested in this study. All the tests were performed to investigate the small-strain stress-strain characteristics of reconstituted clay under the influences of initial axial effective stress and the shear rate. The samples of all tests were saturated and K -consolidated with ₀ different initial effective stress, including 210, 310 and 430 kPa. Then the undrained shearing tests were carried out with a constant shearing rate of 0.2%, 1% and 4%, respectively.

(3) Test results and discussion

The effective stress paths of undrained shearing tests are presented in Fig.

6.1, along with the deduced contours of the developed axial strain. In this figure,

“AES” denotes Axial Effective Stress (unit: kPa) and “SR” denotes Shearing Rate (unit: %/hr). From the results in Fig. 6.1m three important observations can be made.

First, the effective stress paths at the initial stage were nearly straight.

Second, in each test there was a turning point where the paths sharply deviated and then traveled to failure. All the tests except AES2-SR0.2 reached peak deviator stress within the range of axial strain around 0.001 to 0.004. The normalized undrained shear strength S_u σ_v′ is around 0.3.

Third, the maximum effective stress ratio occurred in the position very close to the intersection of the effective failure envelope and the stress path. The maximum stress ratios

(

^σ₁^{′ σ}₃^′

)

_max are around 2.7-3.0, and the corresponding strains are about 0.06.

Fig. 6.2 shows the complete stress-strain relation of undrained shearing tests on logarithmic scale. All tests exhibit the slight strain-softening behavior between the peak deviator stress and failure. The deviator stress fast degraded beyond the failure stress.

As shown in Fig. 6.3, the normalized stiffness of reconstituted Taipei silty clay is compared with North Sea clay and London clay presented in Jardin, et al.

(1984). Also the test result of the undisturbed Taipei silty clay presented in Chin and Liu (1997) is compared in this Figure.

As shown in Fig. 6.4, this study compared the degradation behavior of Taipei silty clay measured in this study and obtained in Chin and Liu (1979) with that of London clay presented by Jardin et al. (1984). The results of small strain triaxial test on the reconstituted clay, undisturbed clay I and undisturbed clay II showed that the trend of the degradation behavior is rather consistent with that presented by Chin and Liu (1997). The initial normalized stiffness of the undisturbed clays varied with the range of 1900-2200.

The comparison of the undrained secant stiffness, E_u S_u , between UC

CK₀ tests and CK₀UE tests is shown in Fig. 6.5. The initial normalized secant stiffness measured by CK₀UC tests is significantly higher than that measured by CK₀UE tests. The ratio of E_u S_u measured by CK₀UC tests to that measured by CK₀UE tests is around 1.5. Although the higher initial stiffness can be obtain by CK₀UC tests, the degradation behavior of secant stiffness of CK₀UC is more significant than that of CK₀UC tests

Fig. 6.1 Stress paths of Taipei silty clay in undrained shear tests

(Axial strains indicated in %)

K⁰consolidation (K⁰ = 0.5)

(Axial strains indicated in %)

K⁰consolidation (K⁰ = 0.5)

(Axial strains indicated in %)

K⁰consolidation (K⁰ = 0.5)

Fig. 6.2 Stress-strain characteristics of the Taipei silty clay

)kPa(2/)(31σσ′−′

1E-005 0.0001 0.001 0.01 0.1

20 40 60 80 100 120 140 160

AES430-SR1 AES430-SR4

AES430-SR0.2 AES310-SR4 AES310-SR1

AES310-SR0.2 AES210-SR1

AES210-SR0.2

AES210-SR4

Axial strain ε_a

Intact and reconstituted North Sea clay 8 ≧ OCR ≧ 1.0

London clay tests

Remoulded and intact OCR＞50

North Sea clay

Axial strain (%)

Upper bound

Lower bound

uusE/

CKoUC test on undisturbed Taipei clay at OCR=1 (Chin and Liu, 1997)

Reconstituted Taipei silty clay

0.001 0.01 0.1

0 1000 2000 3000 4000 5000

0.5

Fig. 6.3 Stress-strain characteristics of the Taipei silty clay

Fig. 6.4 Comparison of normalized stiffness of CK₀UC tests

Chapter 7

ANALYSIS OF SURFACE SETTLEMENT INDUCED BY DEEP EXCAVATION

Damages of buildings adjacent to an excavation site are often caused by the excessive differential settlement induced by excavation. To eliminate or reduce the possibility of such building damages, an effective method is needed to predict accurately the excavation-induced settlement of buildings. Finite element method (FEM) is most often used for predicting the wall deflection and the ground settlement induced by an excavation. It is generally recognized that when a soil model which is not capable of accurately simulating the small-strain behavior of soil is employed in the FEM excavation analysis, the prediction of the ground settlement is generally less accurate, even though the wall deflection can be properly estimated (Jardine 1986;

Burland 1989; Hashash 1992; Simpson 1993; Whittle and Hashash 1994).

Figure 1 shows the typical wall deflection and ground surface settlement computed by FEM and those observed in the field. It is noted that the computed wall deflection and surface settlement are estimated by soil models which can not properly consider the small-strain behavior of soil. The surface settlement near the wall computed by FEM is significantly smaller than that observed in the field. On the contrary, the computed surface settlement far away from the wall is larger than that observed. When a building is assumed to have no rigidity, FEM often underestimates the angular distortion,β, of the building (see Figure 1):

ab L

prediction δ

β = ﹤δ_a′b′ L=β_observation (1)

where δ represents differential settlement between adjacent footings; L represents distance between adjacent footings. Inaccurate estimate of the angular distortion resulted from the inaccurate prediction of ground settlement makes it difficult to accomplish the goals of protecting adjacent buildings and/or assessing the serviceability of buildings.

The difficulty of predicting ground surface settlement in a braced excavation had been well recognized, and over the past two decades, a number of studies were conducted to improve the accuracy of FEM predictions (Jardine et al.

1986; Burland 1989; Whittle et al. 1993; Simpson 1993; Hight and Higgin 1995;

Stallebrass and Taylor 1997). Based on their studies, the predominant factor for less

satisfactory performance in predicting ground settlement, near or far away from the wall, is improper modeling of small-strain behavior of soils.

In this paper, a well-documented excavation case history where the stratigraphy mostly consists of clay was re-analyzed by using the Modified Cam-clay model (MCC) developed by Schofield and Wroth (1968) and the three-Surface Kinematic

In this paper, a well-documented excavation case history where the stratigraphy mostly consists of clay was re-analyzed by using the Modified Cam-clay model (MCC) developed by Schofield and Wroth (1968) and the three-Surface Kinematic

在文檔中 軟弱黏土深開挖及鄰房反應之研究---子計畫二：深開挖粘土橫向異向性通化模式之建置與應用(III) (頁 66-0)