Design of reinforced

concrete Iinings of
pressure tunnels
and shafts
rrof. Dr. A.J. Schleiss
Laboratory of Hydraulic Constructions
Civil Engineering Department
Swiss Federal Institute of Thchnology
Lausanne, Switzerland
Reprinted from
THE INTERNATIONAL JOURNAL ON
HYDKOPOWEK
&DAMS
lssue Three, Volume Four, 1997
Fir;. J. Cradt:d,
rnforccd concrete
!ininr;.
Design of reinforced concrete
Iinings of pressure tunnels
and shafts
A.J. Schleiss, Laboratory of Hydraulic Constructions, EPFL Switzerland
An interactive computational method for reinforced concrete linings of pressure tunnels and
shafts is presented, which considers the effect of the seepage forces in the lining and rock as
well as the deformation-dependent permeability of the lining. The history of the development of
cracks and its influence on the stress distribution in the hoop reinforcement is also taken into
account. Design criteria for reinforced, concrete lined pressure tunnels and shafts are discussed.
T
he development of cracks in the concrete lining
of a pressure tunnel cannot be prevented by
rcinforcing the concrete. The main purpose of
the reinforcement is to increase the number of cracks
and to limit their width. If, compared with the rock,
the permeability of the concrete lining is reduced in
rhis way. water losses from the tunnel will also be
diminishcd. In any case, reinforcement prevcnts
unconrrolled cracking of the lining and local high
water losses, which can cause the washing out of joint
fil!ngs and instability of slopes abo ve thc tunncl.
In practicc. reinforced concrete linings are often
designed oo thc assumption that water pressure acts
only on the inncr surface of the concrete linlng
[Kastner, 1962
1
]. This corresponds to the assumption
that, from the static point of view, thc lining is tight. In
fact this is not the case, because, under interna! pres-
sure, seepage flow into thc rock will develop through
the cracks in the concrete lining. Depending on the
head loss through the cracks, a certain portion of the
interna! water pressure is al so present on the outsde of
the concrete lining. The seepage flow Joads both the
linng and rock by scepage tOrces which must not be
ncglected when calculating deformations and stresses
in the reinforcement and the rock.
Furthermore, mechanical-hy(iraulic interaction will
occur because the width of the cracks in the lining
and. therefore, its permeability are changed by the
deformation of the rock [Schleiss. 19R8
2
J. Moreovcr,
the crack pattem in the lining and, correspondingly,
the distribution ofrhe stresses in the reinforcement are
modified when the internal pressure is increased
[Schleiss. 1990
3
].
Pressure el seep;ge \r-w
Rnck E,,' ,,k
1
Ralirel
lcnmtud;nal
l
l

In this paper, a computation method for reinforced,
concrete-lined pressure tunnels is presented which
takes into account the permeability of the lining and
rock, and the effects of mechanical-hydraulic interac-
tion, as well as the history of cracking of the lining
during increasing of thc internal pressure.
1. Computation model and
assumptions
In view of the static and hydraulic behaviour, thc fol-
Jowng three zones have to be considered separately
(Fig. l ):
crackcd concrete lining;
the rock mass affected by seepage tlow; and,
the rock mass not affected by scepagc t1ow.
Both the defonnation and the permeability of the
cracked concrete lning are intlucnced by the rein-
forcement.
Besidcs radial symmetrical behaviour of the above
zones, the computation method is based on the fol-
lowing assumptions:
Deformability and permeability of the rock mass are
homogeneous and isotropic. Only elastic behaviour of
tlle rock is considered under interna! water pressure.
Seepage takes place only through the cracks; the
permeability of the concrete between the cracks is
neglected.
Regarding the load-sharing between the reinforced
concrete lining and the roe k mass, the reinforcement is
smeared to an cquivalent rhickncss of a steel lincr.
However, when thc distribution of cracks in the Jining
s determined. the srress pauem in thc reintOrcement
between two cracks is assumed to be parabolic
[ Brkenmaier, j. The stress level in the steeJ bars
between two cracks, that is, in the uncracked section,
is int1ucnced by the history of cracking (Fig. 2).
FuJI loading efect (11 = 100 per cent) of the secpage
pressure in the cracked concrete lining and in the rock
mass is assumed [Schleiss, 1986
5

6
].
To determine the load takcn by each of the three
zones, the mcchanical stresses and the water prcssures
at thcir boundarics havc to be known. The unknown
water pressures are derived from continuity con-
ditions. The boundary stresses can be determined by
applying compatibility condtions.
2. Initial cracking of tbe Iining
The lining wll crack as sotm as the tangential stresses
in thc concrete excced ts tenslc strength. Considcring
the scepage forces, the maximum tangential stresses in
Hydropower & Dams lssue Three. 1997
Spacing nf cracks d
Number uf cracks n
cy
Spacing of cracks 1 12 d
Number of cracks 2n
\ 1

QJJ
Spacing uf cracks 1 14 d
Number of cracks 4n
'
\,\ \

1
0
e
_......-Stresses in steai
/a,
Reinfurcement bar
e
Fig. 2. Development ofcracks wul disrribution ofstresses in
the steel bars (slurwn schematicalf.v).
the uncracked concrete lining due to interna! water
pressure are given by Schleis;-ll986
5
j:
(P,, -Pi) ( 2 -V, )
3(1-v,.) .


+ li'r,(l+v,,)!(r;,(2-v.))]+
Jr;At-1 1-(r,!r;.)
+ 2Pr{r:.)
1-(/rj
... (1)
lf the tunncl or the shaft is situated within the
groundwatcr tablc then, as a reasonable approxima-
tion. the acting water prcssure p" on the outside of the
uncrackcd !ning is equal to groundwatcr pressurc
p,. g. h.
If the tunnel or the shaft is abo ve the the groundwa-
ter table, the acting water pressure on the outside of
the lining as a result of the seepagc can be derived
from:
P,_
l+(k,. ln{;,!r,))!(k, ln(R/1;,)) .. 12)
Since the inHuence of p,, on the stresses in the con-
Hydropower & Dams lssue Three, 1997
crete is small, the assumptions R = 2 r(i in the case of
rather pcrvious rock {k,- ;-::: l 00 kJ and R = 1 O ra in the
case of tight rock (k,- k-) give sufficiently correct
results.
The boundary pressure between concrete and rock is
given by Schleiss [1986
5
]:
r _ .-2(2. -v)!((r./r,)' -ll+T'.
(P. P.,)
] L+(l-2v)!(1-/r,,) JI
. [-3(E,(l+v,)/E,(1+v,))p, J
. ' 1

.
l + E,.(l +v,)!E,.(l +v,)+ l- 2 v, J
... (3)
The condition for the formation of the initial cracks
is:
... (4)
Jnserting p,, = p" g b ( groundwater present) or p,
according to Eq.(2) (no groundwater tablc present)
and pr{r .. ) according to Eq.(3) in Eq.(1), the critica!
interna] pressure p;,r at which initial cracks occur in
the lining can be calculated using Eq.(4). In the case
of a tunnel within the groundwater, Eq.(4) gives the
effectivc interna! pressurc exceeding the external
groundwatcr pressure. Thus, the cracking pressure is
PrcT + Pw gb.
3. Head loss of seepage flow
across the cracked lining and
seepage losses
First, the water pressure acting on thc outer side of the
concrete lining, that is, thc head loss of the secpage
flow through the cracks, has to be determined. For rea-
sons of continuity, the losses through the cracked con-
crete lining and into the rock mass must be the same.
3.1 Seepage losses through cracked
concrete lining
Assuming laminar, parallcl flow in the cracks and
knowing the width of the cracks, the water Josses
through the cracked concrete lining can be calculated
using the following equation:
... (5)
3.2 Seepage losses through the rock mass
The water losses through the rock mass for the various
cases considercd (Fig. 3) are given by the following
equations:
For a tunnel within the groundwater rabie [Rat.
1973': Schleiss. 1985
8
]:
(p.Jp, g- h) 21t k,
qoo r
In Lb/r; (1+yl ;
2
/b')]
.. (6)
For a tunnel abovc groundwater leve! [Bouvard, 1975''j:
////
1
/
\ \ . / 1
'-., \ l. ./
"'- ' : /

/ "
)' .. \
1 \
Fig. 3. Fiow parte m
ofsecpage out of/lm-
1/el or siwfl into rock:
ieft. tunnd within
gnmndwater tab/e:
and rigl!i, tmme!
above gmundwutr!r
rabie.
3 \ q
'- r 1 In
p,, g
4 ') 2rr k, 1t k, ;, ''' (7)
For a vertical shaft within the groundwater table
[Schleiss, !985
8
}:
' '(8)
lf no groundwater table is prcsent around thc shaft,
then b:::: O has to be used in Eq.(8). For a cracked con-
crete lining, the reach of the radial-symctrical seepage
tlow can be assumed as follows: R = lO r, in the case
of rathcr pervious rock (k, ;::::IOO k) and R = lOO ro in
the case of tight rock (kr::;; kc).
3.3 Acting water pressure at the outer side
of the concrete lining
The water pressure on the outside of the concrete lin-
ing can be derived frorn the continuity condition, that
is, Eq.(5) equal to Eq.(6), (7) or (8).
3.4 Water losses
Knowing the water pressure on the outside of the lin-
ng, the losses per unit !ength of the tunncl or shaft can
be determincd from Eqs.(6) or 0) or (8), depending on
thc case considered.
4. Load carried by the
reinforcement
Tite loading on thc reinforcement can be obtained from
a compatibility condition. To detemne the load raken
by the reinforcement, it is regarded statically as a steel
lincr with equivalent thickness. This corresponds to the
assumption that. like a steel liner, the reinforcement
exerts a uniform pressure on the concrete rschleiss,
19861 This unif(mn pressure, p,-{r), can be derived
from the following compalibility condition:
u,(r,) u (r,)+u,(1;,)
'" (9)
The sum of the radial deformation of the cracked
concrete lining and of the rock mass has to be identi-
cal to the deformaton of the reinforcement.
These radial defonnations are derived below. For the
case of no surroundng groundwatcr, the dcpth of the
groundwatcr table h is assumed to be zero.
4.1 Deformation of the reinforcement
The radial deformation of the reinforcement can be
calculated from its strain as follows:
u,,(rJ:;:;:E
1
r, ::::m,
2
-r, :;::;:;f110',
2
r
1
1E,
... ( 1 O)
where the tensile force in the cracked section is
Z = cr_,
2
A_,. The associated steel stress is:
'" ( ll)
With a rcdw..:tion factor m, it is considered that the
strain E, and the steel stress as in the reinforcement are
not constant, but have a parabolc distribution and are
dependent on the history of cracking (Fig. 2). The fac-
tor m should be selected according to the sequencc of
formation of cracks:
lst series of cracks: m = l/3 (average stecl stress
cr, cr,, + l/3(cr,,- cr"))
2nd series of cracks: m = 2/3
3rd series of cracks: m = 5/6
nth series of cracks: m = l
Considering the water pressure in the cracks, thc
radial stress in the cracked, pervious concrete lining at
the position of the reinforcemcnt is [Schlciss, I986
5
J:
(J ( r,(p,-pJ(l-(/.l')
r r, 2(.-r - ) l,,
a r,
'" (!2)
4.2 Deformation of the cracked lining
The total compression of the cracked concrete lining
betwccn the inner smtace and the reinforccment is
given by thc sum of the following two values
[Schleiss, 1986
5
]:
wherc:
[r.:- r,
1
-2r,
2
In (1;,/r,)j
'" (14)
Assuming linear distribution of the water pressure in
the cracks (laminar ilow), the water pressure at the
location of the renforccment is cqual to:
P.
'" (15)
4.3 Deformation of tbe rock
The radial deformation of the roe k zone influenced by
scepage is given by the theory of pervious, thick-
walled cylindcrs 1986'-,(l
p, R)C,-p,(R)
e, -(p,(R) a,(;,)) e,
'" (16)
Hydropower & Dams lssue Three, 1997
where:
ri 2v,+(RirJ_+ l
,;,(I+"J_ (RII;J-I 1
2E,(I-v,) !'. . . !.' I-v \j'
+( I- 2v ) ! l+ ' !
, ' \ In (RII;,))
" . ( I7)
e _,;,(I+v,.)(I-2v,.)
,- E
"e
. (!8)
,,'(I +V,) ( I- 2v, + ( Rll;, )')
e
3
E,(R'-r;)
. ( I9)
The external radius of the rock zone affectcd by
seepage is assumed to be the shortcst, vertical reach of
the seepagc tlow above the tunnei[Schieiss, I986
5

6
]:
Tunnel within goundwater table: R = b
Tunncl above groundwater level: R :::::: a
8
(In 2)/rr
where OB :::::: qlk-.
In the case of a vertical shaft. the reach of the seep-
age flow can be assumed to be as given in scction 32.
Besides the water pressure (pa) outsde the lining, the
mechanical boundary pressures at the inner and outer
surface of the rock zonc influenccd by the seepage,
a,(r.,) andp,(R), have also to be considered in Eq.(l6).
The following radial stress is transmitted by the
cracked concrete lining to the rock [Schleiss. !986
5
]:
cr,(;,) p,(J =
l/2 (p,- p,,) (I + ljl,,) +
+p,(r,) rA,
. (20)
The boundary pressure pr(R) between the rock zone
which is influenced by secpage and that which is not,
is obtained from another compatibility condition;
where:
C. . I ' (r 1 R)' + ( R' - :) (1- v,) l
' 2(I-v,)L" 2R
2
In (RII;,) j
. (22)
.(23)
Takng into account Eqs.(20) and (21), the radial
deformation of the rock on the outside of the lining_
according Eq.( 16) is:
u,(,;,)=(p,-h p, g)[c, -e,(e, +e}]+
+ l/2 (p, - P,) ( I + 1 ,) [e, -e, (e, +e)]+
1- p (r,) (rJ,) [e,- e, (C.+ c,)j
... (24)
Hydropower & Dams lssue Three, 1997
4.4 Pressure between reinforcement and
concrete
lnserting Eqs.(!O). (!3) and (24) into the compatibiii-
ty condition as given by Eq.(9), the pressurc transmit-
ted by the reinforcemcnt lO the concrete can be
obtained:
p, (r,) D
1
/D
2
... (25)
where:
D,

A,)-u""
-(p, -b p,. g)[e, -e,(e, +C,)j
-I12(p,- p,)(I + r,ll;,) [e,- e, (e,+ c,)j
D
2
r,
2
!(E, AJ + [(!- v;) IE,jr,
In rJr, + (r,l,) [e,- e, (e,+ e
1
)j
5. Width of cracks in the
concrete lining
5.1 General
Without knowing the width of the cracks in the con-
crete lining, the head loss of the secpage flow through
the lining (that is, p,,) cannot be calculatcd with the
formulae given in section 3. The question is how the
width and the spacng of cracks are influcnced by the
reinforcement. Severa! attcmpts to salve this very
compiex problem have been based on experiments
with reinforced concrete beams and the empiricallaw
of bonding between concrete and sted bars.
Esscntially, the average spacing of the cracks is a
function of stresses in the reinforcement in cracked
conditions, the concrete strength, zone of int1uence of
the reinforcement the thickness and spacing of steel
bars, the concrete cover and thc bond between the
concrete and reinforcement bars .
5.2 Determination of width and spacing
of the cracks
Fig. 4 shows a reinforced concrete Jining which ts
crackcd. According to the calculation model of
Birkenrnaier [1983
4
), the width and spacing of the
cracks are given as a function of the tensile stresscs in
the reinforccment and concrete and of the concrete-
reinforcement bond stress.
With increasing distance from thc crack. the stresscs
in the reinforcement are decreased by rhe bond stress
betwcen the reinforccment and the concrete (see Fig.
4). The reduction of the steel stresses is given by the
following equilibrium condtion:
cr,:::::: a,::::::: as1 + t (dls) .. (26)
The maximum sted stress

between any two

cracks of the first series is withn the rangc:
O< E, lE. .. (27)
Assurning a linear (triangular) Jistrbution of the
steel-concrete bond, the distribution of rhe stresses in
the sted bar between two cracks will be parabolic
fBirkenrnaier, l98J
1
j. Thus, the width of thc cmck is:
.. (28)
Bwd stresses steel-cancrete
Fig. 4. concrete stresses in rein-
{orccmcnt 111 crmaete, hond stress.
It has bcen sllown by experimcnts that the
crete bond stress increases llnearly with the compres-
si ve strength of the concrete [Martn and Noakowski,
1981
11
]. The empirical relatonship for steel bars with
a normal surfucc profile is:
.. (29)
when (2a) s exprcssed in millimetres.
Beginnng with the critica! interna! pressure (see sec-
tion 2) and using Eqs.(26) to (29), the spacing and
width of the f:irst of cracks can be computed by
tria! and error. Thcn the interna! water prcssure has to
be increased un ti! the second series of cracks is form-
ing. This ls the case as soon as the strcsses in the con-
I.Tetc between two cracks excccd the tensile strength
of thc conLTete

:::::: 1 to 2 N/mm
2
):
<
... (30)
After the t!rst series of cracks has formed. the
ing of the is known (see Fig. 2) and the width
of the cracks can be determined directly from Eq.(2B).
However. it has ro be noted that, for every ncw series
of cracks, the steel stress Jistrbution as well as the
spacing of cra<:ks has to be modified (by changing fac-
tor m and ha! vi ng the spacng), whereas the relaton-
ship betwcen cr,. can be assumed as follows:
2nd series of cracks: cr,: = 112
3nd series )f cracks: G-.
1
:::::: 314 cr,,
nth <,crics of cracb;: <J, CL:.
In Eq.(30J, thc area of concrete is not cqual to the
whole section but equal to the zone (that is. the con-
tributing: th;cl\ntss of the lining) in.f1uenced by the
hoop reinforccn1ent, attans in maxirnum 7.5
t mes thc diameter of thc c;tecl bar outsidc of the rein-
fon:ement 1 Sch lciss, J 990-' .
6. Calculation procedure
With the theoretical bases given in sections 3, 4 and 5,
water lo:;ses, stresses in the reinforcement and the
width of crack.s can be computed for a given per-
centage of reinforcement, but not directly because of
the mechancal-hvdraulk interaction. Furthermore,
the cakulation model has to be modified (by changing
factor m in Eqs.( lO) and (2.5)). when a further series of
cracks (2nd crack; 3rd crack: and so on) s fonned.
Therefore, after determining the critica} internal pres-
sure with Eqs. ( 1) to (4). the following step-by-stcp
calculation procedure is proposed:
(A) Assume water pressure on rhe outside of thc con-
CJ'ete linng to be somewhat below internal pres-
sure (p,, < p,).
(B) Calculate the pressure transmitted from the ren-
forcement to the concrete lining by using Eq.(25)
and taking into account Eqs.( 12), ( 14), ( 17), l 8),
(19). (22) and (23).
(C) Determine the stresses in the reinforcement from
Eq.(ll) and considering Eqs.( 12) and (25 ).
(D) Compute width and spacing ofcracks in the lining
for the first series of cracks from Eqs.(26) to (:29),
by tria! and error - for the second series of cracks
and the following series directly from Eq. (28).
(E) CalcuJate the water pressure (pa) on the outside of
the lining using Eqs.(5) and (6) or(7) or (8).
(F) With the above water pressure Pa repeat the calcu-
laon steps B to E until the water pressure fh on
the outside of the lining remains constant.
(G) Increase internal water pressure and control the
stresses in the concrete lining in the section
between two cracks with Eq.(30). As soon as the
tensile strength of the concrete is exceeded, the
next series of cracks will be formed. Repeat the
calculation steps (A) to (F) for the second series
of cracks (that is, the following series), whereas
the spacing of thc crw:.:ks is half of thc spacing of
the first series of cracks (that is, the preceding
series) and the factor m has to be increased
according to 4. l.
(H) Repeat the calculaton procedure until the actual
intemal water pressure is reachcd.
7. Effect of the reinforcement on
the distribution of the cracks
In the foiJowing, the ini1uence of the reinforcement on
the history of development of cracks, their width and
the steel stresscs shall be discussed in an example. The
pressurc tunnel considered is si!uatcd above thc
groundwater table and characterized by the followng
parameters: r, = 1.8 m; r, :::::: l. 9 m; r" = 2.1 m:
E, = 20 GPa,: E, = 4 GPa; E. = 200 GPa, v, = v. = 0.2:
k.= Ht' mis: k. = l(p mis (uncracked): = 1 N/mm':
= 30 N/mm'.
In Fig. 5. the effcct of the spacing of stcel bars for a
certain percentage of reinf(xccment on the width of
thc cracks and rhcr development with increasng
interna! water pressurc is shown. Since the number of
cracks is increased by reducing thc spacing of the steel
bars, the width of thc cracks is reduccd accordingly.
Thus, the spacing of the bars should be made as small
as is practically possible. On thc contrary. the devel-
opment of cracks is intluenced only toan insignificant
degrce by the spacing of the bars. The second series of
cracks with thc internal pressure ( 1 1 bars)
for all bar spacings ccmsidcred. The 3rd series of
Hydropower & Dams lssue Three, 1997
0.35

--o-s=26 mm, 34 cm (n :.241
--o- mm, 20 cm {n "'29)
---o- mm,
'
0.30
1

'
0.25
1
0.20
1
e


'

OJ5
1

w
1
0.10
1
0.05

o
o 10 30 40
50 1
Interna! water pressure (bar)
_j
Fig. 5. Widrh of cracks as ajimaion of interna! pressure for
d!fferenr diameters and spacings ofsteel hars at the same per-
centage ojreinforcement.
cracks develops in the case of smaller spacing of the
bars, somewhat earlicr (42 bars instead of 44 bars).
The crack history is mainly influenced by the
amount of reinforcement as can be seen from Fig. 6. If
the percentage of reinforcement is increased by
ing the bar spacing unchanged, the 2nd and 3rd series
of cracks form at lower interna! pressure. Then, as a
resulr of their larger number, the widths of the cracks
are smaller in the case of a higher amount of rein-
forcement (see Fig. 6 at interna! pressure 30 bar).
Howcver. for the sarne series of cracks. the cracks are
even wider for a higher percentage of reinforcement
(see Fig. 6 at interna! pressure 20 bar). The reason is
that a stronger reinforced concrete lining is stiffer <md,
therefore. takes more load. This on the other hand
rcsults in a lowcr number of cracks, that is, smaller
spacing of the cracks.
For a certain amount of reinforcement, reducing the
bar spacing has no etfect on the stresses in the steel
bars (see Fig. 7). E ven if the amount of reinforcemcnt
is increased, the steel stresses are reduced onlv a little,
as long as the same series of cracks ha ve for a
certain intemal pressure (see Fig. 7 at intcrnal pressure
20 bar. 2nd series of cracks for both perccntages of
rcinforcement).
8. Design criteria for the
reinforcement
The design of the reinforcement, that is, the spacing
and the diameter of the steeJ bars. is governed by the
following criteria:
(1) Limit stresses in the reinforcement; and.
(JI) Lirnit width of cracks in thc lning; and/or,
(III) Limit water Josses from the pressure tunncl.
In rnost cases. the stresses in the steel hner have to
be lower than the permissiblc values (for example 240
N/mm: for Stcd 56) hecause critcria II or III is gov-
erninE.
Jf stability of the tunnel or shaft is endangered by
the washing out of joint fillings as a result of hgh
centrated water Josses, the width of the cracks has ro
be limited fSchleiss, 19B7
10
]. Such erosion should not
Hydropower & Dams lssue Three, 1997
1
0.301
---e- s 20 mm, 20 cm {n 29)
1
--- mm, ds cm
0-25
'
1
0.20 "
1
I
1
0.15

1
1
w
0.10 ---
'
1

0.05

1
1
1
o
o
lnternal water pressure {bar)
Fig. . Width qf'cracks as afunction ofinterntlf pressurefor
different percentagcs of" reinforcement at the same .ljHJcinx of
steel bars.
400
--- mm, dt cm
--o- mm. d, -34 cm
350
-:>-- s-20 mm, ds cm
---&- 18 mm, 16 cm
300
)_. --i = 052 er e en!)
250
200
1
j

150
"
1
1
1
i
1
1
1
1
1
J
100
1
50
1
'
1
o

"
Ni
i
1
T
1
50 1
o 10
'L_
20
Interna! water ressure lbar)
Fig 7. Stresses in the steel bars as afunction of"interna! pres-
l"ltn.' for differenr diametcrs ami spacings of m:cl bars.
occur provided the width of the crack>; is less than
0.3 mm.
1
In spite of this crack width limitation of 0.3 mm, if
thc tunncl is above thc groundwater tabie and n the
case of high roe k mass penneability (kr > w-n m/s)
water Josses rnay be too high. Besides thc purcly
cconomc aspect. the Jeakage out of the tunncl or
shaft, that is. thc saturated rock zone. should not
extend to thc natural ground surface. Seepage Hows
reaching poorly drainng surface deposits may induce
landslides [Schleiss, 1987"'].
The extension of the saturatcd rock zone a-> a rcsult
of seepage is a function of thc permeability of the
cracked concrete lining and can therefore be in1u-
.4. J S'ch!eis5
cnccd by the desgn of the reinforcement.
Nevcrtheless. increasing overburden is normally more
successfuL
The three criteria mcmioned above refer to thc
mechanical and hydraulic bchaviour of the hning. In
addition, the overall stability ofthe tunneL namely. the
bearing capacity of the roe k mass. also has to be con-
trolled. especially in the case of low rock cover
[Schleiss, l987
10
j.
9. Conclusions
Pressure nmncls \vith reinforced concrete linings are
pervious under rhe effect of interna! pressure. Thus,
seepage tlow into the rock mass occurs through the
cracks in the concrete lining. The lining and rock mass
are loaded by these seepage forces. which are a func-
tion of the interna! pressure and the permeabilty of
the lning and the roe k mass. Because the width of the
cracks and therefore the permeability of the lining are
influenced by the deformaton of the rock mass.
mechanical-hydraulic interactions exist. The effect of
these intcractions and the previously mentioned see-
page forces, as well as the hstory of cracking, ha ve to
be considered in the computations to allow for an
analysis of the behaviour of reinforced concrere lined
pressure tunnels which is as realistic as possible. This
al so guaramees a more economical design of the rein-
forccment.
Thc crack widths in a concrete lining are influenced
much more by a reduction n the diarneter. and above
all, by the spacing of the steel bars, than by increasing
the percentage of reinforcement. Heavy reinforcement
consisting of thick steel bars at wide spacing normal-
ly results in a very unfavourable crack pattem and,
consequently, in a larger width of cracks in the con-
crete lning. Thus. the pcrcentage of renforcemcnt
and the :.pacing of the steel bars should be made as
as possible from thc static and practica} points of
VlCW.
The governing design criteria for reinforced, con-
crete lined pressure tunnels are: lmitng the width of
crach. <;tresses in the reinforcement and water Josses.
Besides thesc criteria, consideration of thc ovcrall sta-
bility of the pressure tunneL that is. the bearing capac-
ity of the rock mass, is also essential. O
References
!. Kastncr, H., "Statik des Tunne!- und Stollenb<ll!Cs",
Sprngcr- Vc:rlag. Ber!in/Gbttingen/Heidelberg.
!962.
2_ St.'hlciss, A . of reinfon:ed concrete-!ined prcssurc
tnnneb'' Pmcecdings. InternMiona! Congrcss nn Tunne!s
and Water, Madrid. Span: 1988.
Schleiss, A.. "Bcitrag wr Bemes_\ung von hcwchrtcn
Dmcbto!lenausk!eidungen"_ Bwngenit'ltt No. 65: 1490.
4-. Hirkcnmaier, M .. bcr .:\"achwehe im Gebr:tuchszust;md".
Schwei:;cr lngenieur und An)tekt, :\"\). 6: l9X3.
:'i. Sdlleiss, A., von Druckstollcn. Teil IJ:
der Sicker\!rmung in und Fels,
mechansch-hydr<mlischc Wcchse!wirkungen, Bemessungs-
kritcrien. Mirrei/ung der Vasuchsansra!r jir H-s.\ethau,
Hydmh\r;ie 1md Glozio!ogie. ETH Zri.::h. Switzerland; No.
86, !9K6.
6. SchJeiss. A .. "Design of prcssure tunn.:b" Hrcr
Po'Ni!r & !Jam Consirunion. M ay 19B6.
Rat, M .. ''Ecoulem<:nt d rpartition intersti-
tic!!es autour de<, wnnds .. Bu!!. Liaison Lahormoire des
Ponts u Chausses, NoVccmber am1 Deccmber 1973.
X Schleiss, A .. von J)ruchtollen. Te !JI:
Fclshydraulik imbcsondcre Sickerqrimwng<On dun:h
!\uskkidung und Fds''. Miucihmx dcr Versuchsanstalr fir
H-s,,erhau. ff_vJro!o:,;ic und Gla:iofngie. :--Jo. 7X, ETH
Zrich. Switzerl:_md: 19R5.
9. Bouvard, M .. 'Le'> fuite<, des en charge en terrain
sec. Rle du rcv:!ement. des injection. du terrain". La
f!ouilfe B!anche. No.4; 1975.
1 O. Schlciss, A . "Dcsign criteria for pcrvious and unlined pres-
sure tunne!s". ProccedinRS. lnternational Conferences on
Hydropower, Oslo. Vol. 2: 1987.
11. Martin, H.; Noakowski, P.. "Verbundverhalten von
BetonsU:ih!en. Untersuchung auf Grundlage von
Ausziehversuchen. Deutscher Aussdmss fr Stahlbeton,
No. 319; 1981.
1 Notation
! ih = chmacteristic value of scepage pattcm IBouvard, 1975''1 1
A, = area of radial reinforcement per unit length of tunne! '
(2a) =average crack width )
b = depth of the tunnd be!ow groundwater leve!
1
d =average spacmg: between cracks
1, d, = spadng between the stee! bars
1
E = modulus of e!astidty
, = subscript for rock
= for concrete
1
1
= subscript for
; = acceleration due to gravity
1
k, = permeability of unnacked concrete
k, = permeability of rock mass
1
,
1
,
1
,''... = munber of cracks (= 2nr,/d)
, ::::interna! water pressure
=water pressure on outer of the concrete !ining
p, = water pressure at the position of the reinforcement
p,(r_) = boundary pressure between stecl and concrete
p/r.J= boundary pressure between concrete and rock
Prlr")= houndary pressure between concrete and rock (n the
p,(R) = boundary pressure between saturated and unsaturated
1
i
case of uncracked !ining) 1
rock zones .
= interna! r;dius of lining
=externa! radus of !ining
= radiu'l of reinforcement
=externa! mdius of the rock zone atfected by the seepage
= diameter of steeJ bar
IR
1
q =water !osses per unit !ength of tunne!
u,(r,) =radial deformaton of reinforcemcnt
u,(r,) =radial dcfonnation of concrete lining
l
. u,(rcJ= radial dcformation of rock
, u"'' =total compression of the crackcd concrete lining
, = tensile force in the reinforcernent
1
, B.. = compressive strength of concrete
P = tensile strength of concrete
1
'.: 11, ""loading cffect of seepage prcssure [assumed as
100 per cent 1
11 = pen:cntage of reinforcement (= - r )Id)
1
,
' v = Poisson ratio (subscripts as for E)
v,_ = kinematic viscsity of water
p" = density of water
1
(J, "" tensile stress in !he concrete betwecn two cracks
a,.(r) =radial stress in the cracked concrete !ining
a. = stress in the uncracked concrete section
1
, a,: e:::: stee! strcs:-. in the cracked concrete section
G,-,- = maximum stresses in the uncruckcd
J concrete linin<'
L hond be'tv,-een steel bar and concrete
Prot'. Anton .J. Schleiss gruaduated in Civil Engincering
frorn the Federal lmtitute of Tcchno!ogy (ETH) in
Zurich. Swit;;-crland. in l97R. After joining the Laboratory of
Hydrology and Glacio!ogy at ETH as a research
and senior he obtained a Do<.:torate of
Technical Sciences on the subjcct of pressure tunnel dcsign
in 1986. After that he worked for 1! vcars for Electrowatt
Engincering Ltd and was invo!veJ in. the design of many
hydropowcr projecrs around the world as an expert on
hydnmlics and underground waterways, Un!il recently he
wa, Head of !he Hydraulc Structurcs Se..::tion in the
Hydropowcr Department ;Jt Electrow<ltt. In January ths year
he heunne Head of the Labora!Ory of Hydrau!ic
in the Civil Engincering Depanmem of thc
Swiss Federal Institutc ofTcchnolngy in Lausanne.
Ecok Po!ytcchnique fdrale de Lausannc. DGC-LCH.
, CH- lO J 5 Switzerland.
L
Hydropower & Dams /ssue Three, 1997