SPE 110726
Acoustic Determination of Pump Intake Pressure in Wells Producing Low-API-Gravity
Crude
O.L. Rowlan, SPE, J. N. McCoy, Echometer Company, A. L. Podio, University of Texas, Carlos A. Brunings, PDVSA
Copyright 2007, Society of Petroleum Engineers
This paper was prepared for presentation at the 2007 SPE Annual Technical Conference and
Exhibition held in Anaheim, California, U.S.A., 11–14 November 2007.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.
Abstract
The acoustic determination of the pump intake pressure in
wells producing viscous low API gravity crude is presented in
this paper. These wells have liquid above the pump and the
pump is set above the formation and gas flowing upward
through the gaseous liquid column. The method to determine
the pump intake pressure involves the acoustic measurement
of the liquid level and the casing-pressure buildup rate when
the casing head valve is closed. When these data are used
with this paper's new empirically derived correlation for
viscous crudes, then an accurate gradient of the gaseous liquid
column in the annulus is obtained. This new correlation
provides a more accurate correction of the gaseous gradient of
the fluid above the pump intake for wells producing viscous
crudes.
Results are presented from field testing of numerous wells
where the actual gradients of gaseous liquid columns were
measured is wells having a variety of gas flow rates and an oil
gravity in the range of 10 degrees API. A back pressure valve
is used to increase and stabilize the casing head pressure and
the valve is used to depress and to stabilize the liquid level at
different depths while the well is produced at a constant rate.
Over a period of many days the acoustic measurements to the
stabilized liquid levels are acquired. The gradient of the
gaseous liquid column is then calculated and the pressures are
extrapolated to the pump intake depth. Development of a new
correlation from the analysis of this field data improves the
accuracy of pump intake pressure determination in wells
producing viscous low API gravity crude.
Introduction
This paper addresses the problem of determining the pump
intake pressure in wells producing heavy crude (10-12 API) by
means of pumping systems such as progressing cavity pumps,
plunger pumps and other pumping means, using surface
measurements of pressure and fluid levels. The technology is
well established in industry and has been reported in several
publications 1,2,3,4. It requires determination of the gradient
of the gas-liquid mixture in the annulus above the pump intake
based on knowledge of the oil gravity adjusted by an effective
gradient factor that represents the effect of any gas flowing
through the annular liquid and being produced from the casing
at the surface.
Fig. 1 (left) illustrates the fluid segregation that occurs in a
pumping well that is being operated at stabilized conditions.
Stabilization implies that after pump start-up the following
variables have settled at a relatively constant value:
• The surface producing rate of oil, water and gas
• The produced Water Oil Ratio
• The casinghead pressure
• The annular fluid level
This also implies that the pump has been operating at a
constant speed for a sufficient time to reach stabilization.
At stabilized producing conditions, the oil in the casing
annulus becomes saturated with the gas that is continuously
flowing to the surface. Consequently, if gas is being vented
from the casing annulus at the surface at a constant rate, free
gas is being produced from the formation simultaneously with
the oil. This condition results in a gaseous annular liquid
column. Generally, most oil is produced through the pump
while most free gas is produced up the casing annulus. Pump
intake pressure and bottom hole pressure calculation is
undertaken from a measurement of casinghead pressure,
knowledge of oil and gas properties, and an estimate of the oil
fraction in the annular liquid. The fraction estimate is required
to obtain the gradient of the gas/liquid mixture. This problem
has received considerable attention by numerous authors.
These techniques involve the determination of the gas flow
rate up the annulus and, in turn, the calculation of the amount
of liquid present in the gaseous liquid column by use of such
well conditions as casing/tubing sizes, liquid properties, and
pressure.
A comprehensive field study was performed in 1985 to
determine directly the gradient of gaseous liquid columns3.
The wells tested in that study included casing sizes from 4.5 to
7 in. [11.4 to 18 cm] and oil gravities between 32 and 43°API
[0.86 and 0.81 g/cm3]. Long gaseous liquid columns of more
than 5,000 ft [1525 m] were studied in wells up to 9,000 ft
[2745 m] deep. Annular gas flow rates ranged from 13 to 120
MscfD [368 to 3400 m3/d] and oil fractions ranging from 20%
to 77% were measured. The wells were located in regions of
2 SPE 110726
normal temperature gradients in the range of 0.9 to 1.2°F/100
ft [16.4 to 21.9 mK/m]. Results were converted into what is
known in the industry as the “Echometer S Curve” for
determination of gaseous liquid gradients in pumping wells.
The study reported here addresses the need to extend the
technique to wells producing denser and more viscous crude.
To this purpose a series of field measurements were
undertaken in Eastern Venezuela in wells producing oil in the
range of 10-12 API. In the earlier study the procedure used
for field determination of the gaseous column gradient was
applied, where the liquid level was initially measured and then
depressed by increasing the casing pressure while maintaining
constant production rate. Then, after waiting until stabilized
flow conditions are re-established, the new liquid level was
measured and the process is repeated until the fluid level was
slightly above the pump intake.
Field Setup
Fig. 1 (right) shows a schematic of the installation of a back
pressure regulator at the outlet of the casing in a pumping well
that produces annular gas. This leaves the opposite casinghead
connection available to install a fluid level and pressure
recording instrument.
Fig. 2 is a photograph of the actual field installation used
in acquiring the data reported in this paper. The installation is
more elaborate than would be necessary for the depression
test, since in addition to the pressure regulator a gas flow
meter was installed in the loop to monitor the casing pressure
and gas flow rate using a 24 hour chart recorder. The purpose
of the flow meter was to obtain a gas flow rate independently
of that computed from the acoustic fluid level measurement.
This flow rate is then used in the development of the
correlation for the effective oil fraction.
Subsequent tests were done using a trailer mounted skid
that included the instrumentation and that could be easily
transported from well to well and quickly connected to the
casing and flowline.
Detailed Testing Procedure
With the use of the backpressure valve the casing pressure was
increased to a specific value, then stabilized by allowing
annular gas to vent at its original flow rate. When casing
pressure ceases to increase, liquid from the annulus is no
longer forced into the pump and the pump intake pressure
returns to its original value. After the time when the casing
pressure stopped increasing at the new backpressure the well
is in a stabilized condition and a true gradient can be
calculated.
During the fluid level depression procedure well flow rate
testing of the well was used to monitor that stabilization
occurred. The results showed that a stabilized flowing
condition was accurately indicated by a stabilized liquid level,
obviating the need for this auxiliary equipment in future tests.
Generally it took considerable time (about 24 hours) for
stabilized conditions to be established after an increase in the
casinghead pressure. During this transient period, the pressure
buildup at the surface (caused by the accumulation of gas
flowing into the gas column from the gaseous liquid column)
will result in a pressure increase in the annulus and at the
formation. This casing pressure increase caused some liquid to
be depressed from the casing annulus into the pump, thus
restricting fluid inflow and possibly causing liquid backflow
into the formation. During this time, the liquid level in the
annulus will decrease to a lower level. After gas begins to vent
at the increased stabilized surface pressure, the surface gas
flow rate stabilized at the original value and the liquid level
will stabilize as the producing BHP returns to its original
value.
Typically fluid level measurements were taken twice a day
at approximately the same times. The casing pressure was
monitored on the charts and if everything was considered to be
stabilized, then the back pressure regulator was adjusted to
next higher pressure. The appropriate pressure step to increase
casing pressure was estimated based on the casing’s previous
change in fluid level per unit pressure increase. Correctly sized
pressure steps avoided the danger of depressing the fluid level
to the depth of the pump intake.
A total of 12 wells were tested over a period of one and
one half year beginning in 2005.
Depression Test Results
Fig. 3 is a series of pressure vs. depth diagrams corresponding
to a typical fluid level depression test. Each casing pressure
step is identified as a nearly vertical gas gradientlines from the
surface to the fluid level, while the corresponding fluid level
points are plotted individually. The fluid level data points are
fitted with a least square straight line showing the direct
relationship between casing pressure and depth to the gaseous
liquid column during the 8 day test.
Fig. 4 shows the sequence of acoustic fluid level records
taken during the test. Notice the multiple echoes recorded on
the acoustic traces, due to the relatively shallow initial liquid
level the echoes repeat approximately every 3 seconds. The
first record shows the liquid level at a time of 2.95 seconds,
while the last record taken six days later shows the echo at
3.51 seconds. The echoes are clear and are not affected by
noise produced by the annular gas flow nor the operation of
the pump.
Pump Intake Pressure Determination
In order to determine the pressure at the pump intake, the fluid
level data is converted to pump submergence, which is the
height of the gaseous column relative to the pump intake,
plotted as a function of the pressure at the gas/liquid interface
depth, as shown in Fig. 5.
The slope of the least square best fit line corresponds to the
change in height of the gas/liquid interface per unit increase in
pressure in the casing annulus. The inverse of this slope is a
measure of the average gradient of the gaseous liquid column:
Slope = 2.908 feet per psi
Gradient of gaseous column = 1/ (2.9084) = 0.344 psi/ft
The measured gradient includes both the hydrostatic
(mixture density) and the dynamic (friction and kinetic
energy) components. The linear behavior with depth shows
that the total gradient remains practically constant, even
though the pressure varies by a factor of 3 or 4. This behavior
was observed in all the field tests covered by this study.
SPE 110726 3

Note that, due to the very good linear fit of all the points.
This means that after a relatively short time of only 3 casing
pressure increase steps, then an accurate annular gradient
could have been determined.
The least square equation may be written as:
The pump intake pressure is easily computed from the
measured casing pressure and the gradient of the gaseous
liquid column.
The data points for all the tests have been grouped into a
correlation that allows estimation of the effective gradient of
gaseous liquid columns encountered in heavy oil production.

Submergence (feet) = -2.908 (pressure) + 1140

Where pressure in the equation refers to the pressure at the Acknowledgements

gas/liquid interface computed from the measured casinghead
pressure and adding the weight of the annular gas column.
Setting the submergence above the pump intake equal to
zero and solving for pressure gives the pressure at the pump
intake:
The authors express their appreciation to Victor Sanford of
PDVSA and the engineers of ATS for their dedication in
setting up the field equipment and obtaining the data for this
paper.

Pump Intake Pressure = 1140/ 2.908 = 391 psi

The procedure is simple and yields accurate values of
pump intake pressure. Similar results were obtained in all the
wells tested.
Effective Oil Gradient Correlation
As a continuation to the work presented in reference 3 and
taking advantage of the large number of tests undertaken in
wells producing with different rates of casing gas the data
were grouped in a generalized correlation as a function of the
annular gas flow rate per unit area of annular space.
The measured total gradient of the gaseous oil columns
was expressed in terms of the gas-free oil gradient by
introducing a term defined as the effective oil fraction:
References
1. 1. McCoy, J.N. et al.: "Acoustic Static Bottomhole Pressures,"
paper SPE 13810 presented at the 1985 SPE Production
Operations Symposium, Oklahoma City, March 10-12.
2. Podio, A.L., Weeks, S.G., and McCoy, J.N.: "Low-Cost Wellsite
Determination of Bottomhole Pressures from Acoustic Surveys
in High-Pressure Wells," paper SPE 13254 presented at the 1984
SPE Annual Technical Conference and Exhibition, Houston,
Sept. 16-19.
3. McCoy, J.N., Podio A. L., and K. L. Huddleston: “Acoustic
Producing Bottomhole Pressure” SPE 14254 Annual. Technical
Conference and Exhibition, Las Vegas NV, September 22-25,
1985.
4. "Calculating Subsurface Pressure Via Fluid Level Recorders,"
Energy Resources Conservation Board, Calgary (1978).

Effective Oil Fraction =

Gaseous column gradient / Gradient of Stock Tank Oil

The corresponding annular gas rate, which was measured

in Standard Cubic Feet/day was first converted to actual cubic
feet per day at the average pressure and temperature of the
gaseous column and then divided by the cross-sectional area
of the casing-tubing annulus. Each data point was then plotted
as shown in Fig. 6, where the “Echometer S-curve”3 is also
plotted as the solid black curve.
Note that the data points corresponding to the heavy oil
tests indicate that at the same gas flow rate per unit area, the
effective oil fraction of the mixture is greater than that
observed in oil wells producing crude in the 32 to 43 API

Conclusions
An annular liquid level depression tests were successfully
undertaken in a large number of wells producing heavy crude.
The results for each well at stabilized conditions show a
consistent linear relation between casing pressure increase and
fluid level decrease. The increase pressure and decrease in
fluid level was established without affecting the production
rate of the wells.
The fluid level vs. gas/liquid interface pressure graphs
establish the effective gradient of the gaseous liquid column
by means of the least square fit of the data points.
Table 1
Well Date Test Duration
MFB 456 5/05 4 days
MFB 460 2/05 10 days
MFB 534 6/06 7 days
MFB 613 1/05 5 days
MFB 619 1/05 4 days
MFB 651 1/05 3 days
MFB 653 11/04 4 days
MFB 654 1/05 5 days
MFB 633 4/06 12 days
MFB 652 4/05 8 days
MFB 310 4/02 15 days
MFB 628 12/05 15 days
LS 507 7/02 1 day
MFB 391 2/06 15 days
MFB 587 8/04 9 days

Figure 1 – Segregation of fluids in a stabilized pumping well (left) – Installation of back pressure valve for liquid level depression test (right)
SPE 110726 5

Figure 2 -Liquid Level Depression Test Setup

Back pressure
control and data
acquisition

Acoustic fluid
level
instruments

Figure 3 -Casing Pressure and Liquid Level During Typical Depression Test

MFB652 Depression Test

Pressure, psi
0 50 100 150 200 250 300 350 400 450 500
0

500
y = 4.8386x + 524.62
R2 = 0.9404

1000
Depth, feet

1500

2000

2500
6 SPE 110726

Figure 4 - Sequence of Acoustic Records During a seven day test.

Sec 0 1 2 3 4 5 6 7 8 9 10
316.2 mV

4/5/2006 8:07

4/6/2006 7:29

4/7/2006 10:34

4/10/2006 10:04

4/11/2006 9:24
SPE 110726 7

Figure 5 - Height of Gaseous Column vs. Pressure at the Gas/Liquid Interface and Determination of Effective Gradient. = 1/(2.9084) = 0.344
psi/ft and PIP = (1140/2.9084) PIP = 391 psi

MFB633 - Heavy Oil "S" curve test

900
Height of gaseous column, ft

800
700
600
500
400
y = -2.9084x + 1140
300 2
R = 0.9957
200
100
0
0 100 200 300 400
Pressure at G/L interface, psi

Figure 6 - Heavy Oil Test Points (red line) Plotted on Echometer “S” Curve Graph (black line)

"S"curve for Heavy Oil Compared with the Echometer "West Texas Wells S curve"
100000
Q/A SCFD/in2

10000

1000

100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Effective Oil Fraction