Geothermics (i97o) - SVECIALXSSU~a

U. N. Symposiumon the Developmentand Utilization of GeothermalResources,Pisa x97o. Vol. 2, Part 2

Factors Controlling Borehole Performance

R. ]AMES *

ABSTRACT sine gases in the steam have an influence on the de-

Geothermal boreholes have been studied to estimate signed condenser pressure and therefore on the power
maximum discharges when drawing on (a) a source of pres- output, while knowledge of dissolved chemicals can
surized hot water, (b) a source of dry or superheated steam.
The case is initially considered where infinite permeabi- bc used to estimate the temperature prevailing in the
lity exists at depth (which is common in areas similar to that deep reservoir and also whether deposition of calcite or
at Wairakei, New Zealand) and in which only geometry of silica will take place within the borehole or in the
the hole controls the flow-rate.
Also investigated was the effect on discharge of increased source rocks, either of which could be most important
hole size throughout the depth or over fractions of th~ depth. for the economics of a scheme.
with resultant flow increases quantitively established. The
relationships derived, enable fall in borefield discharge to be
predicted for known decline in temperature and pressure within M e a s u r e m e n t of f l o w a n d e n t h a l p y
the reservoir; conversely the actual decline in field discharges
can be used to confirm such changes. The measurement of the flow and enthalpy of dry
For boreholes such as (a) above, the enthalpy may be steam bores has presented no difficulty, as conventional
accurately determined by a method using the maximum run.
ning pressure, the wellhead pressure at which imminent col- means are available which have long been developed
lapse of the steam-water mixture occurs, and the bottom hole in industry (AsME 1959). The density of the steam can
pressure. A critical discharge pressure tapping located at the be determined from pressure and temperature figures
pipe outlet face may then measure flow-rates direct over a
range of wellhead pressures and hence supply the data required and the flow then metered using a nozzle when dis-
to plot the borehole characteristic curve. charging to the atmosphere, or an o r i f i c e m e t e r if the
The conditions were also considered where rock per-
meability controls the flow into the bottom of boreholes either fluid is being transmitted by pipeline to a power house
(c) from radial fissures or (d) from permeable-porous media. or process. In the case of bores discharging steam-water
Because ~ draw-down, (drop in bottom hole pressure under mixtures, however, no such equally simple devices were
discharge) is proportional to the square of flow for case (c)
and directly as the flow for case (d), it appears that nearly all available, and the development of new ones had to be
boreholes operate with discharge supplied from fissures even undertaken specifically for two-phase flow.
if these are often very narrow.
In the meantime, cumbersome conventional prac-
tice, involving calorimeters or phase separators, were the
General
only alternatives which could be used. Ii was found,
Geothermal boreholes can generally be divided however, that the discharge from powerful bores could
into two types; those producing a single-phase dis- not be measured using calorimeters in which the whole
charge such as the dry or superheated steam found in mixture was condensed to form one phase because the
the fields of Larderello and California, and those typi- size of the equipment required together with the amounts
fied by wells in N e w Zealand which emit steam-water of cooling water, was prohibitively large. Instead, the
mixtures, usually, but not always, dcrivcd from a deep mixture was passed into a separator and the discrete
single-phase all-water source. phases measured by orifice meters. Although there was
Test information required from a borehole includes no difficulty in measuring the steam phase, the water
flow-rate, enthalpy, draw-down (downhole pressure fall phase presented some problems as it left the separator
under discharge), pressure recovery, downholc tempera- at boiling point under the appropriate pressure and it
ture variations, quantity and quality of chemicals dis- was believed that the intrusion of an orifice plate into
solved in the water phase and of non-condensible gases the flow would lead to local steam flashing at the plate,
in the steam phase. These data are necessary not only to hence falsifying the meter readings. To suppress this
evaluate the performance for power potential but are local boiling, cooling of the hot water pipeline had to
useful for comparison purposes when measurements are be introduced intermediate between the separator and
taken at timed intervals; this may lead to predictions the metering orifice which would reduce the water tem-
of the field life under exploitation and to estimates of perature by about 20-30°C and hence permit single-
the volumetric capacity of the reservoir. Non-conden- phase flow through the plate with reasonable accuracy
in measurement. Since then, it has been found that the
* D.S.I.R., Taupo, New Zealand. trouble at hot water orifices is due not to local flashing

1502
but to a small percentage of steam being carried over tank introduced into the pipeline after the separator,
from the separator with the water phase. Even very and is due to vortex formation within the exit water
small quantities of the order of ½% would cause rela- pipe; a condition which is rather difficult to eliminate
tively large errors in the estimated flow (at 180 psig with certainty, as this can take place even when sight
this would result in an over-estimation in the water gauges on the vessel indicate a high water level, above
flow by 30%). This occurred even with a level-control the water exit point.

K
0.2 0.3 0./, 0.50.6 0.8 1-0 Z.O 3"0 4"0 6"0 8"0

I
d ,~oi-d kCANO,4E'rER TAPS

p.s.i.a.
1.0
O 9 [[E~
08
dm
"V --IN --
0 " 7 ~ Orifice
" 0"6 (sham-edged)

~0 0'4
"5
~ i | -

of dischcurge
~c = pl:~, incrms.
din= Diameter of orifk:l~ inch
i

d =~ of main i~pe,
~ches
0"3 K = Factor depqndent p~n
orifice m pipe rcmo .
Meter ~ , r n m Hg
2.0 Pc: ~ ~ ot outlet
NOTE: This graph applies _of dischan~ P'I~. p.s.i.a.
when the ratio of discharge P! Pressure Ot p-s.tream.
manometer t ~ x r ~ p s u :
pipe to main pipe ho= $tacjnation enthalpy of
dc • 3 stec~n-water mixture,
d ' ~ which gives Btu / Ib
1.0
09 convenient values of the _~_
0.8 critical pressure Pc. H o w e . , -
07 for a given flow-rote, values_
ot' critical pressure are re4ated
06
to discl~rge pipe diameters
05 by the fol(owing ratio
I~ O" 4 . F- 7 "°"
.~-&
Pc= Ldcld
0'3-Thus the critical pressure when
an "unsuitable"discharge pipe is
,.e

used, may be a m e n d e d to that

0"2 _ for the correct size by
of this ratio.

01
10 20 30 40 50 60 70 80 100 200 300 400 6OO CO0
Pv p.s.i.o.

FIG. 1. - - Graph used to evaluate stagnation enthalpy o] steam water mixtures.

1505
 Besides these inherent difficulties, the disadvan-
tages using heavy equipment are obvious, especially
when it has to be transported to the site of new bores
when these are ready for test, or alternatively, full STEAME ~ U t A U S ~
separation, cooling and metering facilities built at each
bore. Pcpsm
The technique finally adopted and now in use,
avoids the problems mentioned and has the benefit of
being applicable without modification to both dry steam
as well as steam-water bores. Based upon the fact that
when a large flow of a compressible fluid such as steam
(or a steam-water mixture) flows along a pipe towards a I
region of low pressure, in this case, the atmosphere,
sonic velocity is attained at the exit and the pressure 0, \~1J ~--.-.-'L~-- ' V~TF.R
~ ' ' - ~OUTLET
; ~ ' - ' ~FLOW
'"
along the pipe falls to a value at the discharge point
which is above atmospheric and which is directly re-
lated to the flow-rate and stagnation enthalpy of the 0"1 ' ~ W
flowing fluid. Tests conducted by lAMES (1962) on 3",
6" and 8" diameter commercial pipes, have led to the °' \\
following empirical equation:
G ho'''m
pc0.~ - 11,400 (1)

where G --- flow in Ib/ft' see; ho = stagnation enthal-

py BTU/lb, P~ = critical discharge pressure, psia.
This equation applies where the critical discharge
pressure is taken from a 1/4" hole drilled at a centre-
line distance of 1/4" from the discharge outlet face Fl6. 2. - - Enthalpy and flow Jrora critical discharge pressure
and weir readings.
of the pipe.
This formula was found immediately applicable
for measuring the flow of hot water at known enthalpy dry, neither an orifice meter nor a weir box measure-
rejected from the separators of Wairakei production ment would be necessary as the critical discharge pres.
bores, but could not on its own be used to measure the sure alone would suffice so long as a thermometer is
discharge from new bores not having a separation plant. used at the wellhead to ascertain the steam temperature
However, when used in conjunction with an orifice me- (and hence, the enthalpy) in case it is superheated.
ter, it was found ()tAMES 1965) to be capable of solving As the critical discharge pressure is fundamental
for both flow and enthalpy for bores discharging hori- to these devices, a correction has to be introduced when
zontally to the atmosphere. A graphical solution for en- employed on bores having a high gas content (up to
thalpy is shown in Figure l; equation (I) may then be about 0.20 by weight). To correct P: in Figures 1 and
used to calculate the flow. 2, the gauge reading is first amended to absolute pres-
Blowing a bore horizontally to the atmosphere is
sure, psia, and then multiplied by ( 1 - =-~) where y
not altogether suitable, however, because of its adverse
effect on vegetation (chemical coating of leaves leads is the CO~ content in lb gas/lb' steam, "":asdetermined
to loss in transpiration followed by death), hence most by analysis of an extracted sample.
New Zealand bores now have permanent concrete si- Another method of finding the enthalpy of steam-
lencers designed in the form of twin vertical towers as water bores is to use the <<gas method >> as described
shown in Figure 2. These not only silence the noise by M~moN (1966), which relies upon the ratio in gas
and save the vegetation but also separate the mixture concentrations of two steam samples each taken at
at atmospheric pressure and allow the water to be widely different pressures on either side of a throttle on
measured by a conventional weir situated in the silencer the discharge pipe. This method is fairly reliable on
outlet. Coupled with a critical discharge pressure, it bores with high gas content when the tapping points
can be used (lAMES 1966) to determine both the enthal- are selected with care. It can be used in association
py and flow as in the arrangement employing an orifice with the critical discharge pressure Pc and here it may
meter, but is simpler and more convenient than the lat- be preferable to sample at the wellhead at pressure P,,.
ter system. The curve of Figure 2 is used to evaluate and at the location of P: as these give a maximum dif-
the enthalpy which is then inserted in the equation for ference in pressure and hence greater accuracy. If steam-
total flow (also on Figure 2). Of course, if the bore is water samples are taken at these positions and the gas

1504
content analysed, then the following relationships are ~o0
found to hold:
mw_ xc (2)
me Xw "

h,~= (ww + xwLw)=(w~ + xcL~) + ( V ~ 2 - - V w ~) ~ " ~4~ss flow of Siedm-woter

of ~=ng~Oy 473 8 ~ [ I b
(3)
700
where: r I s~,~ ,,,r. I \
m -- gas concentration, lb gas/lb steam
x = steam dryness fraction, lb steam/lb mixture
w = sensible heat of water, BTU/Ib
L = latent heat of steam, BTU/Ib
V = svecific volume of steam-water mixture, homogeneous, 1:
fta/lb, suffices: w and c at wellhead and critical pres-
sure tapping respectively.
By using equations (1), (2) and (3), the stagnation en-
thalpy ho and flow G may be determined. As a first 300

trial, it is necessary to neglect the kinetic energy term

(-~-~-~-)~ (V~2- - V,~2) to gain an approximate figure for 200

the ho.
~0
When taking a steam sample at the wellhead, a
water sample should also be retained for analysis of
...... ~ I. . . . . . . . I I x
the dissolved chemical. It will be found difficult to 1'~ 29O 2t0 ~OO
do this at the critical discharge pressure tapping be- I~'B~ m P,S,I.~

cause the velocity is very high (Math 1) which results

in stripping of the water from the pipe wall leading to FIG. 3. ~ Comparative discharges ]rom specimen 8" boreholes.
steam dilution of the water sample. The gas method
when used correctly should give not only enthalpy and
flow but also chemical and gas analyses of the bore
fluids. As these data are often required at the same time curve of such a hole (together with that of a dry steam
(to avoid transients in enthalpies), this technique has hole). Hence, once the enthalpy is known, discharge
potential advantages which have been marred somewhat at various wellhead pressures may be estimated by not-
by difficulties in obtaining consistent test results under ing the appropriate values of Pc and using equation (1).
apparently constant conditions. In order to find the enthalpy, it is necessary to
close the bore and to measure the stagnation pressure
Po at depth L near the hole bottom, below the water
Enthalpy determined by maximum discharge pres- surface (this is usually accomplished by lowering a
sure
recording instrument on a long wire through a sealing
When the discharge from a borehole is being pro- gland at the wellhead). The bore can now be discharged
gressively reduced by throttling of the wellhead valve, and the MDP noted; at this point the flow is assumed
the wellhead pressure rises until any further closure to be zero (this is discussed later) which eliminates
of the valve results in the collapse of the suspended frictional pressure-drops from the calculations. Fig-
steam-water mixture over the vertical extent of the bore; ure 4 shows the conditions prevailing within the bore-
at this point the wellhead pressure passes through a hole under MDP with only water existing over a dis-
maximum. This is known as the maximum discharge tance L]. At the flash point, the pressure would be the
pressure (MDP) and occurs when the pressure dif- saturated steam pressure Ps which is associated with
ference between the bottom and the top of the borehole the bore stagnation enthalpy ho; for example, at enthal-
just balances the weight of the column of mixture. If py of 467.4 BTU/lb, Ps = 580 psia from the steam ta-
the MDP is fairly sharply defined and is reproducible, bles of KEEMAN, KEYS, HILL, MOORE (1969). It is
with a swift recovery (order of a day), then its value seen that the only data required are Po at depth L and
may be used to estimate the bore enthalpy. For many the wellhead pressure Pw at MDP in order to calculate
bores which tap an all-water source, the enthalpy re- the bore enthalpy, as will be shown.
mains constant throughout the range of discharge as Over the length L, only water exists and under
indicated in Figure 3 which shows the characteristic essentially static conditions the pressure difference be-

1505
tween the ends of the column are due to the weight enthalpy has to be chosen and the calculation repeated.
of the fluid alone, hence This is continued until agreement is reached.
Actual test results indicate that MDP does not
L1 strictly occur at zero flow, flows of 200,000 Ib/h in 8
P o - - Ps -- - - (4)
144 Vw inch pipe have been usually registered, but correcting
where Vw is the specific volume of water at enthalpy the above method by the introduction of frictional pres-
ho. Similarly, over the length L2 only a steam-water sure-drops due to such flows only raises the calculated
mixture exists, and enthalpies by as little as 1 BTU/lb and can therefore be
considered as negligible.
L,
P~Pw -- - - (5)
144 Vsw Downhole temperatures

where Vsw is the unknown specific volume of the mix- When drilling a borehole, water (from the drill-
ture at enthalpy ho and average pressure (over length ing mud used) can spread outwards into fissures or
into porous-permeable media and hence reduce the sur-
P s + Pw rounding rock temperatures below their original values.
L2) of As L, + L~ = L , which is known,
2 In fact, the very act of drilling alters the whole thermal
equation (4) and (5) may be used to derive: pattern in the neighbourhood of the hole.
- - T h e hole may intercept different permeable
L ~ 144 Vw(Po-- P~)
V,w = (6) horizons with different fluid temperatures and when
144 (P~-- Pw)
the hole is closed, the cooler and denser fluid at the
To determine the enthalpy, a trial calculation is em- higher horizon will flow down the hole in the uncased
ployed using equation (6). The procedure is to assume region and outwards at the lower temperature horizon.
an enthalpy ho and from the steam tables take the as- This results in complete distortion of the true thermal
sociated saturated steam pressure, P~ and water spe- situation in the ground which existed prior to drilling
cific volume, Vw. Using these values, V~ can be esti- and which probably still exists at some distance from
mated and if the assumed enthalpy was the correct one, the hole. This flow within a closed-in bore may attain
should agree with the steam-water specific volume cal- quite high values (a few gallons/minute) so that on
P~+Pw being closed for a time of say 6 months, can result in
culated at average pressure - - and at the assumed an enormous quantity of cooler water flooding into
2
enthalpy. (To simplify this latter calculation, Figure 5 warmer regions. Days or even weeks of discharge may
may be used which correlates V,w at various average be required to eject this fluid, and to heat up the sur-
pressures and over a range of enthalpies). If these two roundings to a more truly representative condition.
values of V,w do not agree, then a new figure for One method of dealing with this situation is to
blow the bore for a while, then close it and lower an
electric resistance thermometer (or similar instrument)
• ~ 1' p.
to the bottom. The hole is then filled with drilling mud,
thick enough to inhibit convection between horizons
within the closed-in bore, but not thick enough to be
baked hard at the temperatures estimated to exist in
the ground, otherwise it will be difficult, if not impos-
mixture o ~ r L 2 a t m~'oge prmmum o' sible, to withdraw the instrument.
It is clear that it is not easy without taking con-
siderable care to obtain reliable temperatures particular-
ly in steam-water holes; hence the difficulty in obtaining
an enthalpy which matches that of the discharging bore
even when it is drawing from an apparently all-water
source surrounding the uncased lower portion of the
hole
Vw : sl:~N:ific vobJam Of wQier aver L I
Maximum flow from boreholes
The maximum discharge from a borehole takes
place when the well is flowing vertically, without re-
striction at the wellhead or at the well bottom, in other
words, the flow is dependent only on the bore geom-
- T -~ Po. % etry, such as length and diameter, and not upon limita-
Fro. 4. - - Borehole at b r i n k o] collapse: discharging with
tions imposed by the impermeability of the rock in the
m a x i m u m w e l l h e a d pressure. uncased hole, or to the tightness of fissures supplying

1506
 P p.s.i.a.
10 100 1000
10

10

i 0.1

0"01
10 100 1000
P p.s.i.o.
FIG. 5. - - Speci/ic v o l u m e ol h o m o g e n e o u s steam-water mixture.

the flow. (These effects will be discussed later). If the MAXIMUM FLOW FROM DRY STEAM BORES
pressure and enthalpy are taken below the water sur-
TO estimate the maximum flow possible from a
face, at a measured depth near the bottom of a closed free flowing vertical bore which taps a source of dry
borehole it is possible to estimate the maximum pos- or superheated steam, the curve of Figure 6 gives the
sible flow which the well can discharge. Also, the ef- best solution. This is derived from the theoretical study
fect on maximum discharge due to falling bottom hole of LAPLE (1943) and its applicability for dry steam was
temperatures or pressures can be evaluated and trends experimentally confirmed over a large range of pipe
predicted. diameters and lengths by lAMES (1964). With the pres-

1507
 1000
I I ~
I I i , IlI I t 1 ,1. I I , ,
P

Stagnation Specific Volume ft 3/ Ib

Stcsgr~tion Pressure P.S.I.A.

o/vo Po
G
L

d
•

.
°

,
Critical Flow Ib/ft 2 st,<:.
Pipe Length ' n :
Pipe Diorneter
d
Consistent
Units
~ . ° Fo'nr~ Friction Factor

100
Ilil
"t
II
! I
II
III
I1 "

10
I

III

I
f,

1
0.1 1 %) 100

FtG. 6. -- Single-phase, critical ]low. capacity curve.

sure Po and temperature measured at depth L in a 124 G d

closed bore, the stagnation specific volume Vo is derived R= = (7)
~J-s
from steam tables. The Fanning friction factor [ is taken
initially at 0.003 for a first trial, in order to estimate where d is in inches and Ix~ is the viscosity of dry steam
in centipoise taken from Figure 7b. The friction factor
4 ] which is used in the figure to evaluate G -~o can now be calculated from an equation derived from
the Moody chart as given by PERRY (1963)
and hence G. The flow-rate G is then used to correct
the value of friction factor assumed, by first evaluating 0.0344
the Reynold's number ! = R o.~o5 (8)

1508
 1.0

z
~- 0.I

=L

0.01
lo ~ looo lo.oo0
002

0 01
10 100 1000
P I~.$. i. a.

FIG. 7 (a, b). - - a) Viscosity o[ steam,water mixtures, b) Viscosity o[ superheated steam.

which applies to commercial steel pipe. Only the one mon in fields tapping hot water aquifers, the following
trial is required. The value of G so determined can be method is very useful in that, not only can it be used
compared with that on a vertically discharging dry to estimate the maximum possible flow for known down-
steam bore by attaching a critical discharge tapping at hole conditions, but faU-off in field discharge can be
the pipe outlet face and measuring the actual critical related to changes in the aquifer. Alternatively, changes
discharge pressure Pc which is related to G by equa- in the maximum discharge can be used to estimate pres-
tion (1). If the measured flow is significantly less than sure and temperature changes in the aquifer or to con-
that calculated from Figure 6, then draw-down will be firm instrument measurements of bottom hole condi-
taking place at the bottom of the hole where the pres- tions.
sure will be falling below the reservoir value of Po. Figure 8 is a diagrammatic sketch of a borehole of
This assumes, of course, that bottom pressure recovery length L ft discharging without restriction vertk:ally
occurs on closing the bore and that the fall in pressure from a source at the bottom of the hole of pressure Po
is not due to gross depletion within the reservoir, but and enthalpy ho which are considered as unvarying
is caused only by the restrictive effect of the surrounding stagnation conditions supplied by an infinitely perme-
media. As this can be checked by downhole measure- able reservoir. This is usual at Wairakei and elsewhere
ments of pressure recovery, there is no difficulty in dif- where it is found that continuous measurements of the
ferentiating between these two conditions. bottom hole pressure indicate no significant fall over
the range of flows that it is possible to test with an
MAXIMUM FLOW FROM STEAM-WATER BORES
instrument in position.
Over the lower length L~ the flow is of water and
This is a more difficult problem than for dry steam, it is only when this has risen to a level where the pres-
particularly where two flow regimes operate within a sure has declined to the saturated pressure P~ that
borehole, namely all-water vertical flow over the lower steam can exist. This is known as the flash point and
fraction of the hole followed by steam-water two-phase above this point the flow is of a steam-water mixture
flow over the higher fraction. As this case is quite com- with increasing steam fraction as the fluid rises to higher

1509
levels at lower pressures. The mixture velocity increases
along the pipe until it discharges at the exit with sonic The kinetic energy term ~
(c)2 ~Vwis only of the order
velocity and at a critical discharge pressure Pc just of 1 psi so may be ignored, while the Fanning friction
within the outlet face. factor is found to be close to 0.003 for boreholes at
high flows and may be safely taken at this figure (the
value of L~ calculated in equation (9) is not sensitive
to its value).

S t e a m - w a t e r f l o w over l e n g t h L .~

! As in the above case of single-phase flow, the

sptc~ic votume of stea~-wot~
overall pressure-drop P , - Pc equals the sum of the
VSw :
mFtu r e ~ w r L2 m ~ ~ c hydrostatic, frictional and kinetic energy pressure drops.
I rs+r~ [¢fbd ec~thOIpy ho,
L. 2 .J L2
The hydrostatic pressure-drop is 144 Vs-"-'-~where V , is
the homogeneous steam-water specific volume taken at
-~ flash - po~t
the average pressure of Ps +2 Pc over the flashing length
L~ and at enthalpy ho. The frictional pressure-drop is

Vw : specH-- volume of woter over L 1

( G ) " /''~L'V"
d where /~'wis theFanning
factor for homogeneous steam-water mixtures. The pres-
sure equivalent of the change in kinetic energy between
P., and P~ is "--'(q~) ' (V~--Vw)where Vc is the homo-
geneous specific volume of steam-water mixtures at
FIG. 8. --Borehole with unrestricted discharge to the atmos- pressure P~ and enthalpy ho. L~ may now be given as
phere: infinite permeability at the bottom.
for equation (9):

W a t e r flow of l e n g t h L, (P. -- PD -- (V¢-- Vw)

For vertical flow, the pressure-drop (Po--P~) L ~ = ( G )~ ],wV,w 1
equals the sum of the hydrostatic pressure-drop, the
frictional pressure-drop and the pressure-drop due to d + 144 V~w (10)
the increase in kinetic energy within the pipe. The hydro-
The value of/,w is found to be close to 0.003 as for
L~
static pressure-drop is 144 V---'--~psi where V,,. is the spe- the single-phase portion of the hole and will be initial-
ly assumed at this figure. It may be checked finally
cific volume Of the Water. The frictional pressure-drop
by calculation from equations (7) and (8) where the
can be transposed from the Fanning equation to the
two-phase viscosity l~,w is used in the former equation
/L G, '~
following more convenient form~l~.92)2/Wd Vw and taken from Figure 7a at enthalpy ho and average
where /w is the Fanning friction factor, G the flow,
Ps+Pc
pressure ~ . These plotted values of two-phase
lb/ft z see, and d the pipe diameter, inches. The pressure-
drop due to the kinetic energy increase at the pipe en- viscosity are based on the following equation for homo-
geneous conditions:
trance, is transposed to the form ( ~ _G )2Vw. The
overall pressure-drop equation can now be given as: 1 = ~
x "l-
(1 - - x)
Ix~w I~s l~w (11)

Po -- Ps -- 144 V-----~+ \ 13.92] d + ~ V,~, where t~s and tx~ are the single-phase viscosities of steam
and water taken from ASME (1959), X is the steam dry-
ness fraction (weight basis)at enthalpy ho and pressure
(Po--P~)-- ~ V,,.
Ll = Ps + P_______~.It will be noted that L, + L~ = L (the bore
G)'-' /~.Vw I 2
T + 144 V'----~ (9) depth, which is known); hence from equations (9) and

1510
(10) and ignoring the negligible kinetic energy term in 446 BTU/Ib and a saturated steam pressure of 485
the former: psia, hence the amount of pressurisation, '~P, is 400

~'~--'~/--(~; ( Vc--Vw 1
G f 0.O03Vw 1
d ~- 144 Vw ~) ~ d -~ 144 V~,,,

To solve this equation for maximum flow, we require psi. From the chart, Pc = 75 psia and the flow may be
to know the pipe diameter d, the stagnation pressure determined from equation (1) to be G = 855 lb/ft2sec
Po at known depth, and the stagnation enthalpy ho. The and through a pipe of 7.825 inches diameter, flow--
pressure can be obtained down a closed bore by lower- = 1.02 (10) 8 lb/h. Suppose that after a number of
ing an instrument through a wellhead sealing gland years, this hypothetical bore has been found to decline
with depth L measured from the casing head flange. in bottom hole temperature and pressure to 220°C and
The enthalpy can be determined by one of the methods 536 psia, with associated water enthalpy of 405.4
already described. With this information, steam tables BTU/lb and saturated steam pressure of 336 psia, then
give P~ and V,, leaving unknown G, Pc, and Vc. The AP = 200 psi and from the chart, Pc = 43 psia. From
procedure is to first assume a figure for Pc which per- equation (1), the new value of G = 562 Ib/ft2sec and
mits G to be calculated from equation (1). From Fig- the maximum discharge has declined to 0.675 (10) 6
ure 5, V,w and V~ are identified at enthalpy ho and ib/h. It is clear, therefore, that the chart may be used
to predict the effect on discharge of various combina-
pressures Ps +2 Pc and Pc respectively. With these values tions of bottom hole temperature and pressure; this is
inserted in equation (12), L may be obtained and should particularly important where measurements indicate de-
be equal to the actual depth taken when measuring Po. cline in these factors as it permits future estimates of
If agreement is not met, then a new value of P: must discharge in time to consider remedial action.
be tried and the calculation repeated until equality is Figure 9 is suitable for geothermal fields which
attained. Although equation (12) looks formidable, it are of the pressurised hot water type so long as the
can be solved in about 30 minutes using a slide rule. average pipe size does not vary from the Wairakei aver-
It will be noted that these calculations depend age of 7.825 inches internal diameter by more than
on the assumption of homogeneous steam-water flow about __+ 0.2 inches, in determining the value of Pc. And,
over the flashing length of pipe and this, in fact, gives of course, the measurements of temperature and pres-
good agreement with test results for maximum vertical sure should be taken at the same depth of 2200 ft
flow through boreholes. It has not been found necessary outside the cased portion of the hole (i.e. within the
to take into account the concept of ' s l i p ' (differential region of slotted casing); this is to make sure that the
velocity between the phases)postulated by some work- values truly represent reservoir conditions at this level.
ers operating with two-phase mixtures flowing horizon-
tally in small diameter pipes, hence the method adopted
here is less complicated than it was at one time thought
likely to be. In order to study the effect of bottom hole 16o!
pressures and temperatures on the discharge from bores
tapping a source of pressurised hot water under condi-
tions of maximum flow, the dimensions of a typical Wai-
rakei bore are taken as an example. With an internal
diameter of 7.825 inches and a depth from the casing
head flange of 2200 ft, bottom hole temperatures are
taken as 220, 240 and 260°C with bottom hole pres-
sures either at the saturated point or pressurised to an
amount AP above this, with values of 0, 200, 400 and 40.
600 psi (the value of AP ---- 0 psi coincides with the or-
dinate through L, = 0, where L~ is the single phase 20,

length as given on Figure 8). Using these data, the crit- I I I I I

ical discharge pressure has been calculated for condi- &O0 800 1200 1600 2000
LI Ll
tions of maximum flow and the results plotted on the
I~pth of bocehole = 22001 = (L1, L 2 )
chart of Figure 9. As an example of the use of this chart, Int~'ol diomildet = ?.825"
Sl.~O~l'-pf(lSrjl.lfe : Z~P PSI (arnou~t ofi0clP~ul~ollx~
suppose the temperature and the pressure measured soturoted vopour pressure)
at the depth of 2200 ft is found to be 240°C and 885 FIG. 9. Discharge o] borehole related to reservoir temper-
--
psia, then the steam tables give a water enthalpy of ature and s u p e r - pressure.

1511
 Figure 9 indicates the primary importance of water meable horizon rather that to estimate the length to
temperature, rather than the amount of pressurisation, the flash point L.~ and only increase the diameter over
on the flow of energy from boreholes. This is reflected that portion. However, even the latter is preferable
in the higher values of critical discharge pressure Pc to the case where the smaller diameter extends over
which from equation (1) is seen to be roughly propor- the whole depth (L1 + L._,). Obviously, drilling rather
tional to (Gho) which has units of BTU/ft2sec. The large diameter boreholes would only be undertaken
depth of the flash-point from the wellhead (as shown when there is considerable confidence in the extent and
in Figure 8) may also be obtained from the chart where depth of a permeable reservoir either of the dry steam
L2 = 2 2 0 0 - L~. In the numerical example given, the type or of the pressurised hot water type.
depth of the flash-point has increased from L~ = 1170 It should be mentioned that to be on the side
to L2 = 1670 ft during the period while the tempera- of conservatism, these calculations were performed on
ture and pressure at the hole bottom declined. a bore with a low enthalpy of 418 BTU/Ib and it is
expected that the exponents of equations (13) and (14)
Influence of diminishing well casing diameter with would increase slightly for higher enthalpies with that
depth of equation (13) finally attaining the value of 2.62 when
the bore enthalpy becomes that of dry steam.
Where bores tap a source of dry steam at a ho-
rizon of infinite permeability, the hole diameter and
length alone limit the flow. For a fixed wellhead pres- Boiling point with depth curve (B. P. D.)
sure and fixed source pressure, the flow is proportional
to d262 when the effect on the friction factor of average There is a great deal of circumstantial evidence
steam velocity and pipe diameter is taken into account. that hot water reservoirs are ~<once through )> systems
Obviously, therefore, in these circumstances it is eco- in which a uniform column of high temperature water
nomically preferable to have as large a pipe diameter rises slowly from depth towards the ground surface,
as is possible over the whole length of the hole down passing as it does through regions of decreasing hy-
to the permeable horizon. drostatic pressure. It is only when this ambient pres-
In the case of a bore tapping a source of pressuris- sure exactly matches the vapour pressure associated
ed hot water at a horizon of infinite permeability (as with the temperature of the column that steam can ap-
occurs in the Wairakei field), it is of interest to see if pear in the water. From this point upwards the water
the same relationship is obtained or whether it would is at boiling point with a temperature associated with
be preferable to have a large diameter hole over the the hydrostatic pressure which in turn is controlled by
flashing portion of the hole (L_~of Figure 8) and a rath- the depth of the water from its surface at 100°C, This
er smaller diameter hole over the portion L, where water surface, when extrapolated from the reservoir con-
only single phase all-water flow takes place. It was ditions, is estimated to be quite often very close to the
found possible to solve this problem using equation ground surface as at Wairakei, but it can be much lower,
(12) by increasing the borehole internal diameter over and at Ahuachapan No. 1 borehole in El Salvador it is
the whole length (L~ + L.~) and also by increasing the found to be about 600 ft below the ground surface. An
diameter only over the flashing length L2. integrated B.P.D. curve has been calculated and the
It was found that increasing the diameter from d, following equation derived:
to d2 over the whole pipe length resulted in an increase
in the flow according to the relationship: C = 54.3 H °~s (15)

W~ [ d2 ]TM where C is the temperature in °C when boiling point

= I-W/J (13) exists over a depth H ft from a surface at 100°C. The
equation applies over the depth range H = 100 to
whereas increasing the diameter over the flashing length 10,000 ft.
to d2 while the diameter of the single phase port" ",a L1 It also appears likely that when the deep column
remains at dl results in the following relationsh,l~ be- of hot water rises through a broken rock zone, steam
tween W,.., (the flow through the two-diameter arrange- can only appear when a horizontal permeable region
ment) and W~ (the flow through the single-diameter is reached; this is because the generation of steam would
arrangement): result in a large increase in the specific volume, which
in an approximately parallel-sided column would bring
the flow to a halt. On entering a permeable horizon,
Wl [ dl / (14) however, there would be no such restriction as the
water of the column can now expand radially outwards
It is clear that it is preferable to increase the well dia- without restriction. Hence the temperature of the wat-
meter over the whole of the depth down to the per- er column may largely be influenced by the presence

1512
of a geologically permeable layer with a depth which displaced downwards by 600 ft (as at Ahuachapan inE1
can be evaluated from equation (15) if the temperature Salvador). For comparison purposes, curve (C) is drawn
of the water column is known and if the B.P.D. curve with the B.P.D. relationship displaced 1800 ft down-
intel:cepts the ground [urface. At Wairakei, the temper- wards. A selection of Wairakei bocehole test results
ature of the water column is now estimated to be are shown plotted, which are considered to be only
257°C and using this in the equation above, the depth roughly accurate because it has, in the past, been found
at which boiling commences is 1750 ft which is very difficult to be sure of the source water temperature
close to the depth of the major permeable region at supplying a borehole at M.D.P. Because of this diffi-
1800 ft. Recent results from the Broadlands geothermal culty, it is probable that the curves of Figure 10 will
be used to determine the source temperature of bores
rather than the other way around and particularly af-
ter the vertical displacement of the B.P.D. curve has
been established by early investigation holes. Deep holes
o /// penetrating below the reservoir water surface should
result in an estimate of the rising water column tem-
perature and equation (15) will then give the depth t o
the lower permeable horizon.
For dry steam fields, curve (D) is given which can
also apply to pressurised hot water systems when bores
are bleeding (severely throttled condition) and in which
the depth of the borehole is filled with steam or gas.
5OO For bores which are drilled to shallower depths
9~ ~mtJ
over which the B.P.D. curve operates, Figure 10 will
give temperature of the water supplying the bore but
t" .~_ ~ I I I this will be less than that of the rising water column.
Such shallow bores are liable to fill with steam and
gas when closed as they are often tapping water which
is closed to the boiling point and is not pressurised to
any marked extent.
*" , I As an example of the use of Figure 10, the case
of an actual shallow Wairakei borehole is considered
which had an M.D.P. of 220 psig when discharging
a steam-water mixture. Complete closure of the well-
head valve led to steam and gas filling of the well so
that over a time period of 1 day the wellhead pressure
rose from 220 psig to a final figure of 325 psig. Plotting
of these pressures on Figure 10 indicate that the source
Water Teml~rotUm water temperature is 222°C and that curve (A) is ap-
o I
100 150 2OO 250 3~
plicable showing that the B.P.D. curve starts at groun d
level. Later tests confirmed these results. If the B.P.D.
Flo. 10. -- Maximum discharge pressure (MDP) as a ]unction curve was downwards by, say, 600 ft, then curve (B)
o/ reservoir water temperature.
would have applied, and for the same source water
temperature of 222 °C the wellhead pressure would have
been 190 psig at M.D.P. with an unchanged value of
region, which is about 18 miles from Wairakei, indicate 525 psig for the final closed-in pressure.
that the rising water column temperature there is at As an example of a bore penetrating deep into the
abouth 300 °C which, when used in equation (15), gives pressurised hot water, we may take the case of Ahua-
a depth of 3600 ft to boiling at the permeable layer. chapan No. 1 of which measurements of the bottom hole
Evidence from the Broadlands investigation pores in- pressure and temperature indicate has a B.P.D. curve
dicate that the permeable horizon is close to that depth, displaced downwards by 600 ft. With a M.D.P. of 202
for the deep reservoir. psig, the rising water column temperature is 226°C
Employing equations (5) and (15) with H = L2, from Figure 10. Equation (15) gives H = 950 ft which
a curve can be plotted which correlates the maximum with the addition of the 600 ft displacement, results
discharge pressure (M.D.P.) of a borehole with the tem- in a depth to the permeable horizon of 1550 ft at its
perature of the supply water at the bottom of the well. highest point. Actual depth to the top very permeable
This is drawn on Figure 10 as curve (A). Also drawn is layer has been given in the U.N. Progress Report (1968)
curve (B) where the B.P.D. relationship is considered as as from 1640-1800 ft which is reasonably good agree-

1515
ment.. It should perhaps be mentioned that the max- where:
imum discharge of a bore is only slightly reduced by
O = borehole flow, cma/sec;
drilling deeper into a hot water permeable aquifer of
constant temperature. This because increasing the depth k = rock permeability, darcies;
L as given in Figure 8 results in a nearly similar in- L = thickness of permeable horizon, cm;
crease in L, and only a small increase in L~. Doubling
= viscosity of fluid, centipoise;
the depth L can result in a decrease in flow of only
about 5%. Po and Pb = reservoir and bottom hole pressure, atm
With successes in predicting the temperature of R and r,, = radii of the drainage boundary and the well bore
the rising deep water column from chemical analyses
of geyser fluids, it appears that the approximate depth Test results on Wairakei bores which have flows
to permeable horizons may be estimated before actual up to a million lb/h indicate that there is no meas-
drilling commences. If the maximum displacement down- surable draw-down, i.e. ( P o - Pb) is probably less than
wards of the B.P.D. curve is assumed as not likely to about 4 psi; it is also believed from geological evidence
exceed 1000 ft, then drilling to, say, 5000 ft (about that the wells do not dra,w from porous-permeable
1.5 kin) should locate any available permeable horizons media but instead tap fissures of generous proportions
in hot water reservoirs with temperatures up to 307 °C. which radially supply the pressurised hot water exist-
Of course, there is no guarantee that such horizons ing at depth. Recent tests on Broadlands bores indicate
will be successfully pieced as they may present an that flows are less than those at Wairakei and values
elusive target, but they should exist somewhere within of drawn-down attain hundreds of psi. But the rela-
that depth range at a level where the imposed hydro- tionship established by tests between the flow Q, and
static pressure matches the vapour pressure correspond- draw-down (Po B Pb) is not a linear one as in equa-
ing to the water temperature, and where the permea- tion (19), and O is in fact found to be approximately
bility provides room for the steam expansion. proportional to ( P o - Pb)°L This strongly suggests that
For mathematical and extrapolation purposes, the fissure flow is taking place because an equation suitable
respective equations for curves (A), (B) and (D) are for boreholes and based upon radial flow through fis-
given below (curve (C) has been neglected as outside sures is:
the normal range expected).
r ,.,.,, lO. ,,
W = 7500dt [ V,,,J (20)
Curve (A), C = 46.8 pm0"~ (16)
where:
Curve (B), C = 63.4 Pm°'24 (17)
w -- borehole flow, lb/h
Curve (D), C -- 51.3 Pm°'2¢Is (18)
d = diameter of well bore, inches
t -- thickness of fissure, inches
It should be noted that for convenience, the pres- AP = draw-down, psi, (= difference between reservoir and
sures of Figure 10 are given in psig (gauge pressure), bottom hole pressure)
but in the above equations, the maximum discharge V., = specific volume of fluid, ftVlb.
pressure is given in psia (absolute pressure). C is the
water temperature in °C. For New Zealand wells of about 8 inches diameter,
drawing on hot water with a specific volume of appro-
ximately 0.02 ft3/lb, equation (20) simplifies to:
F l o w f r o m h o t w a t e r reservoir to borehole
W = 424000 t(AP) °'5 (21)
T h e relationship between the flow of a borehole
and the pressure gradient surrounding it can be calcu-
lated from Darcy's law of linear filtration according to Applying equation (21) to Wairakei bores indi-
which the flow is proportional to the pressure-gradient cates that the fissures there are about 1 inch in width,
and inversely proportional to the viscosity of the fluid. while for the Broadlands bores, they would be more
The equation derived from this relationship and com- than an order of magnitude smaller (i.e. less than about
monly applied to the radial flow of homogenous fluid 0.1 inches). It would appear, therefore, that the con-
into a borehole is: cept of flow through porous-permeable media must be
rejected for these fields, and probably elsewhere, and
problems involving the identification of reservoir per-
= 2~kL (Po Pb) meability in equation (19) need not be pursued. This
(R)
~

is substantiated by two facts; (a) a deep hole has been

I~ log~ ~ (19)
drilled at each of the Wairakei and Broadlands fields,

1514
neither of which produced any flow from thousands steam. Where the fissure thickness is much less than
of ft of uncased hole. It appears then, that unless a this, the fissure itself increasingly controls the flow to
fissure is penetrated, flows are virtually non-existent, a point where the discharge is completely dependant
hence it can be concluded that the rock is grossly im- on the fissure thickness. When this occurs the flow-rate
permeable from the viewpoint of geothermal explora- is proportional to d.
tion. (b) Although the top 2000 ft of the Wairakei field From either point of view, improved flows are ob-
has declined in pressure over the last decade under the tained with larger diameter bores.
combined draw-off of 60 production boreholes, the field
output enthalpy has not changed in any significant de-
gree, hence there has been very little heat contributed REFERENCES
by the rock in the regions of falling pressure. This ASME 1950 - - Fluid meters, their theory and application.
is in accordance with the concept of flow through fis- Amer, Soc. mech. Engrs. New York.
sures, with minimum fluid contact with the rock. Con- KEF.NAN J. H., KEYES F. G., HILL P. G., MOORE J. G. 1969 - -
Steam Tables. John Wiley & Sons.
ductive heat flow through rock is very slow and re-
LAPPLE C. E. 1943 - - Isothermal and adiabatic flow of com-
quires a large and intimate surface contact to make pressible fluids. Trans. amer. Inst. chem. Engrs., 39, 385.
a substantial contribution, and this has apparently not MAnes W. A. J. 1966 - - A method for determining the en-
happened. thalpy of a steam-water mixture discharged from a geo-
thermal drillhole. N. Z. ]. Sci., 9, 791.
]AMES R. 1962 - - Steam-water critical flow through pipes.
Effect o f bore s i z e o n f l o w f r o m fissures Prec. Inst. mech. Engrs., 741.
As fissures appear to dominate the mode of flow JAMES R. 1964 - - Maximum steam flow through pipes to the
atmosphere. Prec. Inst. mech. Engrs., 173, 473.
in geothermal reservoirs, it is worth noting that flow-
]AM~.S R. 1965 - - Metering of steam-water two-phase flow by
rate is directly proportional to the bore diameter when sharp-edged orifices. Prec. Enst. mech. Engrs., 180, 549.
a fissure is intercepted, as given by equation (20). For JAMES R. 1966 - - Measurement of steam-water mixtures dis-
fissures of a width of the order of 1 inch however, or charging at the speed of sound to the atmosphere. N. Z.
greater, the geometry of the bore controls the flow (be- Engng., 21,437.
cause draw-down is negligible) and as has been men- P~RRY R. H. 1963 - - Chemical Engineers Handbook. McGraw-
Hill.
tioned earlier, the discharge is proportional to dT M for 1968 - - El Salvador. Survey of geothermal resources. United
pressurised hot water, and proportional to d262 for dry Nations, Progress Rep..

1515