Jökull - 01.12.1966, Qupperneq 30

GUNNAR BODVARSSON1 and D. J. RYLEY1 2:
The Measurements of the Weight Discharge
from Geothermal Steam Wells
ABSTRACT
Geothermal toells are frequently situated in
remote, inhospitable areas where water may be
scarce ancl communications bad. Measurement
of the mass output of geothermal wells, either
by condensing the efflux or by separation and
measurements of the individual phases may be
impossible or very inconvenient.
This paper reviews the alternative methods
that are available for predicting the flow,
draiving particular attention to the present
states of laboratory and other studies in two-
phase critical discharge from parallel ducts. It
is concluded that no acceptable method of flow
prediction yet exists based on idealised two-
phase flow models, but that the future outlook
in this respect is encouraging.
A new theory of critical flow is presented,
and examined in the light of flow measure-
ments made by R. James.
Menlion is also made of dilution methods
whereon experimental work is currently pro-
ceeding in Iceland.
INTRODUCTION
For the satisfactory exploitation of geo-
thermal wells, it is necessary to know with some
precision the weight of water and steam dis-
charged. The most accurate way of finding the
flow is that of direct measurement either by
separating and metering each of the two phases,
or by condensing and weighing the entire dis-
charge. The flow from a geothermal well is
usually very farge; a typical Icelandic well
having a discharge of 40 to 50 kg/sec., and the
measurement of a flow of this magnitude pre-
sents difficulties wherever the flow is situated.
1) Department of Oceanography, Oregon
State University, Corvallis, Oregon. U.S.A.
2) Department of Mechanical Engineering,
University of Liverpool, England.
In the case of geothermal wells, which are usu-
ally found in inhospitable and often arid
regions both in Iceland and elsewhere, it is
wholly impracticable to attempt to condense
and weigh the efflux. For such a purpose a
condenser would be required with a cooling
water consumption which might itself exceed
the well flow by a factor of 5—10.
A more suitable method for use in the geo-
thermal field is that in which the liquid and
vapour are separated and each phase metered
separately. This method was first employed in
New Zealand and has been subsequently used
in Iceland. The method is not without its pro-
blems. The separating equipment may be large
and not readily portable, and requires level,
even ground on which to stand. The capital
cost of such a plant is high.
Engineers concerned with the measurements
of discharge from geothermal wells have re-
cognized for some time, however, that high
cost and inconvenience might be avoided if it
were possible to calculate the flow from straight-
forward observation of properties made at the
well head and without the introduction of
direct flow-measuring equipment. One such
method was successfully employed by Belin and
Knox (1955) and subsequently improved by
Belin and Bainbridge (1957). These investiga-
tors recognized that with a low-quality-hori-
zontal pipe flow there is gross separation of
vapour and liquid, and that the liquid will
flow through the bottom of a measuring nozzle
in a stream of segmental cross-section. The
thickness of this segment was measured using
beta-ray absorption methods to give the nomin-
al mixture density and this quantity was related
empirically to the mass flow dryness fraction.
An empirical relation between density and up-
stream pressure enabled the mass discharge to
be predicted with fair accuracy from observa-
tion of these quantities.
Banwell (1957) devised a small impact sampl-
184 JÖKULL