HK Poly '10 NB4, Pre-grouting - Barton

7/25/2019 HK Poly '10 NB4, Pre-grouting - Barton

1/79

PRE-GROUTING AND WATER CONTROL SOME OF

,BEHIND ITS SUCCESS IN TUNNELLING

NB # 4

PRE-GROUTING FOR STABILITY AND OVER-BREAK CONTROL

PRE-GROUTING FOR WATER CONTROL

COMPARING PARTICLE SIZE WITH JOINT APERTURE LUGEON TESTING FOR APERTURE ESTIMATION (e and E)

TYPICAL HIGH-PRESSURE GROUTING QUANTITIES

1


2/79

-water inflow in THE NEXT THREE or FOUR ROUNDS OF

TUNNEL ADVANCE

2


3/79

PRE-INJECTION FANS FOR PROTECTING THE NEXT 4

TO 6 ROUNDS

3


4/79

THIS ELKEM EXAMPLE HAS OUTER BLOCKER-GROUT TO

ALLOW HIGHER PRESSURE WHEN LOW COVER

4


5/79

PRE-INJECTION MAY ALSO PREVENT PROBLEMS LIKE THIS

- CONCRETE VOLUMES LATER WHEN LINING

5


6/79

PRE-GROUTING TOREDUCE FUTURE

....AND CURING

PROBLEMS

6


7/79

LOW COVER HIGH WATER TABLE DEEP WEATHERINGAN APPROACHING RIVER CROSSING.

GOOD PRE-GROUTING IS IMPERATIVE!

7


8/79

SKETCHES OF WHY

PRE-GROUTING IS

IF TUNNELS(AND STATIONS) ARE

THEREFOREIN WEATHERED,PERMEABLE,LESS STABLEROCK MASSES !

8


9/79

OBVIOUSLY LEAKING JOINTS ARE EASY TO

GROUT.. BUT THE TIGHT ONES THAT ALSOLEAK.?

9


10/79

The dilemma is how to get blocks

(i.e. particles) that are too large in

joints that are too tight.

10..smaller particles! .. wider joints!


11/79

Before leaving large blocks and

concentrating on cementparticles.note problems

-

in mine ore passes!

Blockage if D< 4.d95

Boundary layer

Wall roughness

Slow particles

11

problems with large blocksand wall roughness


12/79

IN TUNNEL WALLS

with grout ???

12


13/79

SOME FUNDAMENTAL PHYSICAL ATTRIBUTES OF ROCK JOINTS

13


14/79

HEAD LOSSES IN JOINTS AND BOUNDARY LAYER EFFECTS

14


15/79

EFFECTIVE STRESSES IN WATER AND GROUT INJECTION

The existin a ertures can be chan ed b fluid ressure which is ver

15

advantageous in grouting!


16/79

WATER INJECTION TESTS LUGEON METHOD

16


17/79

Pay special attention to % of zero flow sections

AVERAGESPACING OF WATER CONDUCTING JOINTS

17% of zero flow stages means 1.8 conducting joints per test lengthon average

17


18/79

SOME MORE SIMPLIFICATIONS ARE NEEDED!

BECAUSE WE ARE ENGINEERS WE NEED A SOLUTION!

This is how we modified from Snow 1968

18

visualize the conducting network of joints


19/79

WE WILL NOW WORK WITH TWO DIFFERENT JOINT APERTURES

19


20/79


21/79

THEORETICAL INTERPRETATION (IN 3D) OF LUGEON RESULTS

,

permeability of one smooth parallel plate : k = e2/12

permeability of 1 set of parallel plates : K1 = e2

/12 x e/S

permeability of the conducting rock mass (3 sets) : Kmass 2e3/12 S

1 Lugeon 10-7m/s, 10-7m/s 10-14 m2, laminar flow

3D interpretation of Lugeon tests e (6LS x 10-8)1/3

e an n m me ers, s average ugeon va ueeac app y olocal domain, rocktype, or borehole

21


22/79

The equation e (6LS x 10-8

)1/3

looks like this for typical S-values of 0.5 to 3.0m

Obviousl it is difficult to in ect cement articles into e. . < 0.1 Lu eon rock masses

22

unless E ( the physical joint aperture ) >> e ( the hydraulic aperture )


23/79

SOME OPTIONS FOR INJECTING TIGHT JOINTS

It usually helps to start injection with a high water/cement ratio (i.e. > 1.0)

It may help to use ultrafine or micro cements (with micro silica andplasticisers) if the joints are very tight

It usually helps to use much higher grouting pressure (P ) than water pressure(PW)

=. . G W

In Norway we use final grouting pressures in the range 5 to 10 MPa(50 to 100 bars), when pre-injecting ahead of tunnels, successivelyreducing the w/c ratio. The grouts are stable (little shrinkage or water

separation).

BUT FIRST WE MUST INVESTIGATE IF WE NEED HIGH PRESSUREGROUTING what h sical a ertures do we have??

23


24/79

Here we have Lu eon results from 4 de th zones at a ermeable dam site

(e) and (S) have been interpreted as previously explained

(e) is converted to (E) using JRC (the joint roughness coefficient)

24 Note that the Grout-Take estimate (from this 1978 example) assumes the

same grout pressure as the Lugeon test.. i.e. PW= 1 MPa


25/79

HOW TO ESTIMATE . JRC THE JOINT ROUGHNESS COEFFICIENT

Can be done on core from ahead of the face, or in the tunnel close to the face

25


26/79

26


27/79

JRC value ranges derived from tilt tests and direct shear tests

0

27


28/79

. ,

2850, 100 and 400m limits are simpler to remember


29/79

IF THE AVERAGE APERTURES (E) ARE NOT LARGE ENOUGH ? then

(very exaggerated) the pressure must be increased!

Local aperture increase (0.1 to 0.3mm ?) with higher pressure

29


30/79

GROUTING BETWEENTHE JOINT SETS IS

ACHIEVED WITH HIGHER PRESSURES

30


31/79

Even this is possible (all 3 cements can be injected) due to E>e, and

maybe due to use of high injection pressure causing deformation

If the joints were smooth this would be impossible without higher

31


32/79

32


33/79

, or per aps ..BUT THE ROCK MASS IS STILL WET!

33


34/79

THE MECHANISM OF JOINT OPENING WITH REDUCED EFFECTIVE. . . ..

SEE BB-MODELLING EXAMPLES

a y rau c aper ure c anges

34


35/79

b) Physical aperture changes

35


36/79

EXAMPLE OF INPUT DATA, AND INITIAL CALCULATIONS WITH BB-EXCEL

The initial unstressed physical aperture is estimated from an empirical equation :E0 = JRC0/5(0.2c/JCS0 0.1) mm

36


37/79

BUT. a false idea of permeability will be given with too high water pressure testing

37


38/79

Note that E can be almost doubled from 30 to 60m now OK for ULTRAFINE?

38


39/79

The difference between E and e in a UDEC-BB model (Makurat, 1988: Oslo Tunnel)

39

SOME RECENT ESTIMATES OF (e) from LUGEON TESTS


40/79

SOME RECENT ESTIMATES OF (e) from LUGEON TESTSand (E) from Jr to JRC conversion.

Large (E) due to high Lugeon values...and roughness.

(ORANGE and RED and PINK are easiest to grout)

YELLOW (OK)GREENBLUEneeds higher pressure or finer micro/ultrafine

40


41/79

BUT ROCK MASSES ARE VERY VARIABLE.WHAT ABOUT SHEARING ???

41


42/79

IF THE STRESSES ARE ANISOTROPIC

42


43/79

LOCAL SHEARING CAUSES LOCAL DILATION AND EASIER GROUT

PENETRATION IN JOINTS THAT MAY HAVE BEEN TOO TIGHT

43


44/79

A TYPICAL SITUATION WHERE JOINT SHEAR AND DILATION MIGHTOCCUR DURING HIGH PRESSURE PRE-GROUTING

Future tunnel wall on ri ht rinci al stress vertical.but no h draulic fracturin

44

HIGH PRESSURE PRE-INJECTION COULD CAUSE LOCAL HYDRAULIC


45/79

HIGH PRESSURE PRE INJECTION COULD CAUSE LOCAL HYDRAULICSPLITTING IN ROCKS WITH LOW TENSILE STRENGTH

but the presence of joints in other directions (not parallel to principal stress) willoften prevent this

45

DIFFERENT JOINT DEFORMATION MECHANISMS CAN OBVIOUSLY


46/79

DIFFERENT JOINT DEFORMATION MECHANISMS CAN OBVIOUSLYOPERATE (LOCALLY) DURING HIGH PRESSURE INJECTION..

due to the reduced local effective stresses

46

BUT HIGH PRESSURE GROUTING IS UNLIKELY TO CAUSE UNWANTED


47/79

DEFORMATION i.e. uncontrolled hydraulic opening.ue o ew on an u pressure ecay mec an sms

47

THE LOGARITHMIC PRESSURE DECAY IS A SAFETY MECHANISM for

high pressure grouting.while flow is still occurring.. and is extraeffective with cohesive + frictional fluids like grout


48/79

48

Logarithmic pressure decay with radial, laminar or turbulent flow (e.g. Cruz, 1979)


49/79

DEFORMATION THAT MAY OCCUR IF FLOW HAS CEASEDAND PRESSURE IS MAINTAINED

49


50/79

WHAT HAPPENS WITH TOO LOW PRESSURE, TOO TIGHT JOINTS, TOO

LARGE CEMENT PARTICLES..AND UNSTABLE GROUTSTHAT BLEED WATER?

50


51/79

- -...... ,

BUTTHE NEXT SCREENS SHOW WHAT OCCURS WHEN DOING

SUCCESSFUL INJECTION when the grout penetrates as expected

51


52/79

NOTE DRY ROCK BEHIND JUMBO ALREADY PRE-

GROUTED)SEE SEVERAL EXAMPLES THAT FOLLOW

52


53/79


54/79


55/79


56/79


57/79


58/79

58


59/79

59

STABLE NON-BLEEDING GROUTS ARE ALSO ESSENTIAL FOR


60/79

STABLE NON BLEEDING GROUTS ARE ALSO ESSENTIAL FORPREVENTING WATER-SICK ROCK

60


61/79

WHAT ABOUT PROPERTYIMPROVEMENT?

CAUSED BY SUCCESSFUL GROUTING

61


62/79

mu t -probe-

measure

ment ofgrouting(Quadros

an

Correa,

1995)

62


63/79

reduced in magnitude due to grouting Sin le hole inter retation shows from 1 to 4 orders of ma nitude

improvement e.g. 10-7 to10-8 m/s up to 10-4 to 10-8 m/s

With 3D interpretation across 4 to 8 m of rock mass (this was thehole spacing), the improvement is 1/17 for Kmax, and 1/11 for Kmin

This emphasises the need for closer spacing (e.g. split-spacing) of

The results suggest that individual joint sets are successively

Norwegian experiences suggest pressure plateau . as each joint

set is in ected? Can we take this further and think of rock mass property

improvement ???

63

Suppose the following small improvements occur to individual Q-parameters


64/79

64

These could give the following improvements in rock mass properties ?


65/79

65


66/79

SOME OF THE

EQUATIONS

Q-value androck massproperty estimates

a seen n ear erlectures)

66

.


67/79

.

THE EFFECTIVE Q-VALUE CAN BE IMPROVED

67


68/79

REDUCED TUNNEL DEFORMATION. WOULD ALSO BE SEEN IN MODELS

68

IF Q-VALUES CAN BE IMPROVED SO TOO WILL SEISMIC VELOCITY


69/79

(see steep-diagonal behaviour due to consolidation effect)

69

RELATIVE TIME FOR TUNNEL EXCAVATION AND SUPPORT


70/79

RELATIVE TIMEFOR TUNNEL EXCAVATION AND SUPPORT

potential benefits of pre-grouting, especially if Q 0.1

70


71/79

potential benefits of pre-grouting, especially if Q 0.1

71

CONSEQUENCES OF-


72/79

ON COST.....IF EFFECTIVE

CAN BE INCREASED

TRY TO ELIMINATEMOST OF THE LOWQ-VALUE ROCKi.e. Q< 1

THEN GET LOWERCOST BECAUSE OFLESS PROBLEMS

WITH CONSTRUCTION

72


73/79

Take care if cover (depth to surface ) is limited !

73

kg cement / bored meter?

ra n en = g m we s : oo ow pressure: - ars ns ruc onf E i (1 d l ith t i j ti t t ti )


74/79

from Engineer. (1 year delay with post injection, contractor won compensation) Arlandabanen = 16.9 kg/m (rail) Tsen = 24 kg/m Svartdal = 80 kg/m (very bad rock and low cover, mainly for stability) Lundby = 5.9 kg/m (Swedish: too low pressure extra injection rounds

needed)

Storhau = 8.0 k /m Bragernes = 68 kg/m (deformable volcanic rock) Baneheia = 9.0 kg/m Jong-Asker JA1: = 13 kg/m, JA2: = 19 kg/m (4 to 8 MPa)

ese are mos y orweg an unne s, w ere sys ema c pre- n ec on was usefor many kilometers. Pressure range 2 to 9 MPa, cover 2 to 150 m)---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Line 4, screen 4

11 kg/m (mean 134 kg/hole,12m long holes using 13 holes in arch only)(pressure 10- 60 bars, mean 32 bars

AREA OF TUNNEL per 12 m grout fan 12.D x 12.10 x 180 m2

1,735 kg/180m2 = 9.6 kg/m2 (THIS IS LOW, BUT SEE FIGURE FOR GNEISS

74

(assumed 6 m cylinder of grout gives 1.6 l/m ...typical for tight gneiss, granite)-----------------------------------------------------------------------------------------------------------------


75/79

How many litres of grout per m3 of rock mass?

with 6m cylinder assumption usually 1 to 6 liters/m .

75

Some pre-grouting volumetric results


76/79

Some pre-grouting volumetricresults

Pre- routin data derived from ndal et al. 2001.

Rock type kg/m2 tunnel surface kg/m3 litres/m3

gneiss 11.0 to 16.5 1.8-2,8 1.7-2.6

granite 12.0 to 52 2.0-8.7 1.8-8.3

phyllite 26 4.3 4.1

. - . . - .

syenite (dike) 30 to (186) 5.0-(31) 4.7-(28.7)

fracture zone 19 to 50 3.1-8.3 2.9-7.7

ASSUME 6M THICK CYLINDER (ON AVERAGE) IS GROUTED

76

AN UPDATE FROM 2006 NUMEROUS RECENTNORWEGIAN TUNNELLING PROJECTS


77/79

NORWEGIAN TUNNELLING PROJECTS

77

Estimating Liters-of-grout / m3 of rock mass (as previous)


78/79

78

SUGGESTION FOR ALTERNATING SHORT AND LONG HOLES(to give better over-lap, and act as probe-holes without extra drilling)


79/79

( g p, p g)

79

Documents

HK Poly '10 NB4, Pre-grouting - Barton