This paper presents new correlations to predict Modified Black Oil (MBO) PVT properties for volatile oil and gas condensate reservoirs using commonly available production data instead of PVT reports or complex EOS calculations. The correlations were developed using data from 14 actual reservoir fluid samples representing different fluid compositions and behaviors. MBO properties were generated at 12 separator conditions using commercial PVT software and a statistical approach. The new correlations show reasonable agreement with MBO properties from the software. They were also validated against full compositional simulations and material balance calculations. The correlations explicitly account for separator configuration and conditions, which previous correlations did not. They provide a simpler way to model volatile oil and gas condensate reservoirs using the MBO approach.

1. SPE 164712
Modified Black Oil PVT Properties Correlations for Volatile Oil and Gas
Condensate Reservoirs
Ibrahim S. Nassar, GUPCO, Ahmed H. El-Banbi, and Mohamed H. Sayyouh, Cairo University
Copyright 2013, Society of Petroleum Engineers
This paper was prepared for presentation at the North Africa Technical Conference & Exhibition held in Cairo, Egypt, 15–17 April 2013.
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 have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers,
or members. Electronic reproduction, distribution, or storage of any part of this paper 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 SPE copyright.
Abstract
This work presents new Modified Black Oil (MBO) PVT properties (Rs, Rv, Bo, and Bg) correlations for volatile oil and
gas condensate reservoir fluids. These new correlations do not require the use of fluid samples or EOS calculations. The
correlations have the advantage of taking into consideration the effect of surface separator configuration (two and three
stages) and conditions (separators pressures and temperatures).
The correlations were developed using fourteen actual reservoir fluid samples (7 gas condensates, 3 near critical fluids,
and 4 volatile oils) spanning a wide range of fluid behavior and characteristics. Whitson and Torp method was used to
generate Modified Black Oil (MBO) PVT properties that were used as a data set for correlations development.
The MBO PVT properties data points were generated by extracting the PVT properties of each sample using commercial
PVT software program at twelve different separator conditions spanning a wide range of surface separator configuration and
conditions to generate twelve curves for each sample. A statistical approach using a statistical software program (SPSS) was
used to develop the new correlations models.
The results of the new models show reasonable agreement between Modified Black Oil PVT properties generated from the
new correlations and the MBO properties extracted using Whitson and Torp method. The average absolute error in the
correlations was 8.5% for volatile oils and 17.5% for gas condensates.
These correlations were also validated by comparing the results of modified black oil simulation using MBO PVT
properties generated from these correlations to the results of full equation of state (EOS) compositional simulation. Also, the
generalized material balance equation (GMBE) was used to calculate the initial oil/gas in place (IOIP/GIIP) for many
simulated cases using PVT data generated from the new correlations and data generated from EOS models. The advantage of
the new correlations comes from being the first in the industry (to the best of our knowledge) that explicitly take into
consideration the effects of surface separators configurations (two or three stages) and conditions. Also, all input parameters
in the correlations are readily available from field production data. These correlations do not require elaborate calculation
procedures or PVT reports.
Introduction
It was clear since 1920's that the engineering of oil reservoirs require the knowledge of how much gas was dissolved in the
oil at reservoir conditions and how much the oil would shrink and gas would expand when it was brought to surface. Three
properties (Rs, Bo, and Bg) serve these purposes and constitute the traditional (conventional) black oil PVT formulation7
.
However, it has been known for many years that volatile oil and gas condensate reservoirs cannot be modeled accurately with
conventional black oil technique but require Modified Black Oil (MBO) approach. The MBO approach assumes that the stock
tank liquid can exist in both liquid and gas phases in reservoir condition.
Gas condensate and volatile oil petroleum reservoir fluids are simulated frequently with fully compositional models but
can also be efficiently modeled with a Modified Black Oil (MBO) approach15
. A few authors have addressed the question of
how to best generate the MBO PVT properties including the new function, condensate gas ratio (Rv) which represents the
vaporized oil in gas.

2. 2 SPE 164712
Whitson and Torp3
in 1983 used data derived from CVD experiments to calculate modified black oil PVT properties for
volatile oil and gas condensate reservoirs. Perhaps the most useful application of CVD data is the calculation of liquid
composition, which together with measured vapor composition yield high pressure K-values. At each depletion step,
individual phase compositions (measured or calculated) are flashed using a set of appropriate K-values (ex.: Standing’s K-
values Correlation) through a multistage separator simulator representing field conditions to calculate MBO PVT properties
(Bo, Bg, Rs, Rv).
Coats4
in 1985 developed a different approach from Whitson and Torp to calculate the modified black oil PVT properties
for gas condensate reservoirs only. In his approach, oil-gas ratio (Rv) is obtained by flashing the equilibrium gas at each stage
through the specified surface separator configuration while the remaining parameters are calculated using a material balance
procedure.
McVay2
in 1994 extended Coats’ work to include volatile oil reservoirs. He modified Coats’ procedure in a completely
analogues manner to generate MBO PVT properties for volatile oil reservoirs.
Walsh and Towler5
in 1994 suggested a simple method to compute the black oil PVT properties of gas condensate
reservoirs. The authors used the data available from standard CVD experiments and developed an algorithm to compute the
black-oil PVT properties of gas condensate without the requirement of K-value model or equation of state (EOS) calculations.
The method is rigorous, direct and simple and is ideally suited for spreadsheet applications. However, it depends on how
many pressure steps are taken in the CVD laboratory experiments.
All the methods for generating modified black oil PVT properties presented in the literature need a combination of lab
experiments (PVT reports) and elaborate calculation procedures. Recently, a new oil-gas ratio (Rv) correlation was developed
by Abdel Fattah.9
This correlation doesn’t require the use of fluid samples or elaborate EOS calculations. In practical use of
this new correlation, difficulties were noticed from the use of the surface gas gravity parameter (it is assumed to be
volumetric average between gas gravity in different separators, while gas gravity from low pressure separators may not be
available in many field operations). Therefore, the surface gas gravity used by the correlation14
probably needs advanced
knowledge in PVT to be calculated. Also, the effect of the surface separator configuration and conditions are not explicitly
represented in the correlation. Separator conditions were implicitly represented in the specific gravity term. It was found that
separator conditions would have significant impact on PVT properties for volatile oil and gas condensate reservoirs16
.
In this work, we developed new MBO PVT correlations that combine the advantages of the previously published
correlations9,14
and explicitly use separator configurations and conditions.
Fluid Samples and EOS Modeling
Fourteen reservoir fluid samples are used in this study representing different fluid composition and phase behavior for the
extraction of modified black oil (MBO) PVT properties (7 gas condensates, 3 near critical fluids, and 4 volatile oils).
Table 1 summarizes the major properties for the fluid samples including the fluid type, heptanes plus mole fraction (C7+),
reservoir temperature, saturation pressure and initial gas oil ratio for each sample. Fig. 1 shows a graph for heptanes plus
mole fraction (C7+) for each sample to illustrate that the samples span a large variation of fluids. We can see that a 12.5% C7+
can be considered as a distinguishing value between volatile oil and gas as proposed by McCain6
.
Equation of State (EOS) models were created using commercial EOS PVT Software for each sample and tuned with the
available lab experiments (Constant Composition Expansion, Differential Liberation, and Constant Volume Depletion). The
EOS characterization was conducted following Coats and Smart14
procedure. The EOS Models were developed using the (PR
EOS) or (SRK EOS). We first split the heptanes plus component and then we regressed on OMEGA A and OMEGA B
parameters for the heavy pseudo-components and methane. Also, the Binary Interaction Coefficients (BICs) between methane
and the heavy pseudo-components were used as regression parameters when needed. The regression is conducted by
minimizing an objective function which quantifies the difference between the measured and calculated PVT properties.
Tuned EOS Models were used generate MBO data set that we used to develop the correlations of this work. They were also
used to output EOS parameters in compositional simulation format for validation purposes.
Figs.2 through 6 show the EOS results after tuning with the measured data from the available lab experiments for one of
the samples (Volatile Oil 1) as an example. Fig.2 presents the phase envelop calculated from the tuned EOS for fluid VO1.
Fig. 3 presents the match for Constant Composition Expansion experiment showing an excellent agreement between the
relative volume values calculated from the tuned EOS model and the observed value from the constant composition
expansion experiment. Figs. 4 through 6 present the match for the Constant Volume Depletion experiment observations
(vapor z-factor, liquid dropout, and number of moles produced). EOS models for other fluid samples were developed in a
similar way.

3. SPE 164712 3
Approach
Extracting the modified black oil (MBO) PVT properties for each sample from the tuned equation of state (EOS) using
Whitson and Torp method generated at twelve different separator conditions was performed. The data set included 1,488
points for the 4 PVT curves from volatile oil samples, 1,212 points from near critical samples, and 2,280 points from the gas
condensate samples.
A statistical software program (SPSS) was used to develop the new correlations models by fitting the data sets extracted
above. In selecting the independent parameters for the 4 PVT properties curves, we selected parameters that are readily
available and also have strong correlation with the dependent variables (Rs, Rv, Bo, and Bg). The new correlations do not
require data from experimental fluid analysis (PVT reports), nor elaborate calculations with EOS models, and all the
parameters are easily obtained from field production data.
After developing those correlations, we evaluated them by comparing the results of the modified black oil simulation
using these PVT properties extracted from the new correlations to the results of full Equation of State (EOS) compositional
simulation. Also, the generalized material balance equation was used to calculate the Initial Oil/Gas in Place (IOIP/GIIP).
Models Construction Methodology
The first step in our work to develop the models was to select the independent parameters affecting the MBO properties
which are (reservoir pressure, reservoir temperature, stock tank oil gravity, surface gas gravity, separator configuration,
separator pressure and temperature, and saturation pressure), while the dependent parameters are the Modified Black Oil
(MBO) PVT properties.
A question arises here about dependence of MBO properties on surface gas gravity term which was a point of confusion in
the previous correlation.14
Therefore; in the development of the correlations of this work, we considered the surface gas
gravity of first stage separator and second stage separator as independent model parameters. Surface gas gravity of the stock
tank was not considered because it is usually not easily available in field operation.
The independent parameters with plotted with the dependent parameters to provide us with the initial guess for the model
shape and a trial and error procedure was used to arrive at an appropriate model shape. Non-linear regression was then used to
find the model constants that minimize the difference between observed data points (extracted from the EOS model) and the
calculated points. For models validation, we drew cross-plots between observed and calculated values and calculated the least
mean square error (R2
) and the average absolute error given by the following equation:
Error ∑ .................................................................................(1)
Another step was performed to complete the work for the extraction of MBO PVT properties so those correlations can be
applied for any field case. Because saturation pressure may not be known in some cases, new saturation pressure correlations
were developed. The saturation pressure correlations depend on the same parameters and the calculated saturation pressure
will be used to divide the MBO curves to the saturated and the under-saturated parts. To account for the cases where
saturation pressure may be available from other sources, all correlations were presented for two cases: (1) known saturation
pressure, and (2) unknown saturation pressure. The following presents the new correlations models and their results. First, the
saturation pressure correlations are presented followed by the saturated curve correlations for the 4 PVT parameters, and
finally the under-saturated curves.
Saturation Pressure Models
The most widely used correlations for saturation pressure are probably the ones by Standing6
, Vasquez and Beggs17
, and
Al-Marhoun18
. We tried to modify these correlations to account for three stage separation which is commonly used for
volatile oils and gas condensates and found that the modified “Al-Marhoun” correlation gives the best results with our data
base.
In the following, we present two correlations for saturation pressure: one for volatile oil and the other for gas condensate.
Both correlations have the same form, but for volatile oil it will be function of initial producing solution gas-oil ratio (Rsi)
while for gas condensate, it will be function of initial producing condensate-gas ratio (Rvi).
Volatile Oil Saturation Pressure (Bubble Point) Correlation:

4. 4 SPE 164712
P A ∗ R ∗ X Y ∗ ∗ T .......................................(2)
Gas Condensate Saturation Pressure (Dew Point) Correlation:
P A ∗ R ∗ X Y ∗ ∗ T ......................................(3)
Where, X and Y are given by:
X ..............................................................................................................................(4)
Y ...............................................................................................................................(5)
The above correlation parameters are given in Tables 2 and 3 for two-stage and three-stage separation for volatile oils and
gas condensates, respectively. The average absolute error for the correlations is presented in Table 15.
Saturated Curve Models
The following sections present the new correlations for the 4 PVT parameters (Rs, Rv, Bo, and Bg) for the saturated curves.
Solution Gas-Oil Ratio Model
a) Known Saturation Pressure
The final form of the modified correlation is:
R
∗ ∗ ∗ ∗ ∗
∗ ∗
....................................................................(6)
X ..............................................................................................................................(7)
Y .............................................................................................................................(8)
V ∗ STO ......................................................................................................................(9)
The correlation parameters have been computed by regression and are presented in Table 4 for gas condensate and volatile oil
(for both two-stage and three-stage separators). The average absolute errors and the least mean square errors mentioned
earlier are presented in Table 15.
b) Unknown Saturation Pressure
The final form of the modified correlation is:
R
∗ ∗ ∗ ∗ ∗
∗
.....................................................................(10)
X ..............................................................................................................................(11)
Y ..............................................................................................................................(12)

5. SPE 164712 5
V T ∗ STO .........................................................................................................................(13)
The new correlation parameters are given in Table 5 for gas condensate and volatile oil for both two-stage and three-stage
separators. The average absolute error for this correlation is presented along with other correlations in Table 15.
Oil Formation Volume Factor Model
a) Known Saturation Pressure
The final form of the correlation is:
B
∗ ∗ ^ ∗ ∗ ∗ ∗
..........................................................(14)
X ∗ STO ....................................................................................................................(15)
Y ∗ STO .....................................................................................................................(16)
V T ∗ STO .....................................................................................................................(17)
The oil formation volume factor correlation parameters (when saturation pressure is known) are given in Table 6 for gas
condensate and volatile oil for both two-stage and three-stage separators. The average absolute error for the correlation is
given in Table 15.
b) Unknown Saturation Pressure
The final form of the correlation is:
B A ∗ P A ∗ 10 ^ A ∗ X A ∗ Y ∗ EXP A ∗ V ..................(18)
X ∗ STO ....................................................................................................................(19)
Y ∗ STO .....................................................................................................................(20)
V T ∗ STO .........................................................................................................................(21)
Similarly, the new correlation parameters are given in Table 7 for gas condensate and volatile oil for both two-stage and
three-stage separators. The average absolute error is presented in Table 15.
Condensate-Gas Ratio Model
The initial condensate-gas ratio (Rvi) is one of the independent parameters that have a significant effect on the correlation
accuracy. However; in case of volatile oils, this parameter is not available from production data. For volatile oils, this

6. 6 SPE 164712
parameter is not the reciprocal of the initial producing gas-oil ratio. It is actually the amount of oil (or condensate) vaporized
in the gas coming out of the solution at surface separators. In black oil correlations, the parameter condensate-gas ratio is not
defined as the gas associated with black oil is dry gas19
. Therefore, a new correlation for initial condensate-gas ratio (Rvi) will
need to be used first to compute a value we can use for other correlations in volatile oil cases.
The form of the initial condensate-gas ratio, Rvi, correlation is:
R A ∗ EXP A ∗ X Y A ∗ STO A ∗ STO A A ∗
................................................................................................................................................(22)
X SG ∗ P ...................................................................................................................(23)
Y SG ∗ P ....................................................................................................................(24)
The new correlation parameters are given in Table 8 for volatile oil only for two-stage and three-stage separators. For gas
condensates, the initial Rvi value can be obtained from production data. Now for the rest of the correlations and for both fluid
types (volatile oils and gas condensates) Rvi values will be available. For gas condensates, it will be available from production
data while for volatile oils, it will be calculated from the new correlation.
a) Known Saturation Pressure
The final form of the condensate-gas ration correlation is:
R A ∗ P A ∗ P A ∗ EXP A ∗ X A ∗ Y ∗ EXP A ∗ V ∗ R
................................................................................................................................................(25)
X SG ∗ P ..................................................................................................................(26)
Y SG ∗ P ....................................................................................................................(27)
∗
.....................................................................................................................(28)
During the regression process, we found that it was hard to obtain a good curve fit especially for the tail part of the curve in
the condensate-gas ratio model. This was the main reason to explain the higher error percentage in this correlation for gas
condensates than volatile oils. The new correlation parameters are given in Table 9 for gas condensate and volatile oil for both
the two-stage and three-stage separators. The average absolute error is presented in Table 15.
b) Unknown Saturation Pressure
For the unknown saturation pressure case, the average absolute error percentage was about 16% for volatile oil and 25%
for gas condensate. The new correlation parameters are given in Table 10 for gas condensate and volatile oil for both the two-
stage and three-stage separators. The average absolute error is presented in Table 15.

7. SPE 164712 7
The final form of the Rv correlation is:
R A ∗ P A ∗ P A ∗ EXP A ∗ X A ∗ Y ∗ EXP A ∗ V ∗ R
................................................................................................................................................(29)
X SG ∗ P ...................................................................................................................(30)
Y SG ∗ P ...................................................................................................................(31)
..............................................................................................................................(32)
Gas Formation Volume Factor Model
a) Unknown Saturation Pressure
Knowing that the shape of gas formation volume factor, Bg, curve is monotonic increase below the saturation pressure and
sudden increase to very high values at low pressures (approximately below 1000 psi), we first regressed against the entire
curve followed by regression only against part of the curve at pressure greater than 1000 psi to improve the accuracy of the
correlations.
B A ∗ P ∗ EXP A ∗ X A ∗ Y ∗ EXP A ∗ V ........................(33)
X SG ∗ P ...................................................................................................................(34)
Y SG ∗ P ....................................................................................................................(35)
V STO ∗ T ........................................................................................................................(36)
The new correlation parameters are given in Tables 11 and 12 for gas condensate and volatile oil for two-stage and three-
stage separators. The average absolute error is presented in Table 15. Table 11 is used if we want to calculate Bg values for high
and low pressures. Table 12 is used if we would like to have better accuracy correlation for Bg in the high pressure range (P >
1000 psi).
Under-Saturated Curve Models
The following equations are presented to show how the 4 MBO PVT properties (Rs, Rv, Bo, and Bg) can be calculated for both
volatile oils and gas condensates above the saturation pressure.
Solution Gas Oil Ratio Model
The same model for the saturated curve with the same correlation parameters will be used to calculate the solution gas oil
ratio at the saturation pressure for gas condensate fluids. For volatile oils, it can be obtained from production data as it is
equal to the initial producing gas-oil ratio.
R R ...........................................................................................................................(37)
The average absolute error is presented in Table 15 for both known and unknown Psat.

8. 8 SPE 164712
Oil Formation Volume Factor Model
Under-saturated Bo is frequently calculated using oil compressibility. Several oil compressibility correlations are available
(e.g. Standing, Vasquez and Beggs, and Laster) for black oil fluids. However, these correlations do not take into consideration
the three-stage separators. The following correlation is presented for MBO fluids and it takes surface separator configurations
and conditions into account.
The new correlation form is:
B A ∗ P A ∗ V A ∗ P A ∗ B A ∗ X A ∗ Y ...(38)
∗ ....................................................................................................................(39)
∗ ...................................................................................................................(40)
V T ∗ STO .........................................................................................................................(41)
The new correlation parameters are given in Table 13 for gas condensate and volatile oil for both two-stage and three-
stage separators. The average absolute error is presented in Table 15.
Condensate Gas Ratio Model
The same model for the saturated curve with the same correlation parameters will be used to calculate the initial
condensate-gas ratio (Rvi) at the saturation pressure for volatile oil. The new correlation for Rvi is just used as input for the
saturated curve model. For gas condensates, initial condensate-gas ratio is obtained from production data.
R R .............................................................................................................................(42)
The average absolute error is presented in Table 15 for both known and un-known Psat.
Gas Formation Volume Factor Model
The value of under-saturated gas formation volume factor, Bg, decreases with increasing pressure, regardless of whether
the pressure is above saturation pressure or not. Therefore, the final form of the new correlation is the same as the saturated
curve model. The average absolute error is presented in Table 15. For volatile oils, under-saturated gas formation volume
factor is not defined and therefore, only gas condensate average absolute error is presented here.
Correlations Validation
The accuracies of the new correlations are evaluated firstly by cross plots between actual values and calculated values and
secondly by calculating the average absolute error. Figs. 7 to 10 show example cross plots between observed and calculated
values for the new correlations models.
For further validation and to estimate the effect of the correlation error on the results of the applications these correlations
will be used for, two more procedures were used in validation:
1. The results of the Modified Black Oil simulation using PVT properties generated from the new correlations were
compared to the results of full compositional Equation-of-State (EOS) simulation.
2. The Generalized Material Balance equation was used to calculate the Initial-Oil/Gas-In Place (IOIP/IGIP) for several
simulated cases.

9. SPE 164712 9
In order to examine the effects of MBO PVT Properties, all other potential sources of differences between compositional
and MBO simulation results should be eliminated. First, the same simulator was used for compositional and MBO simulation
runs. Second, the same EOS models that were used for generating MBO PVT properties were also used for compositional
simulation runs.
We used the generalized material balance equation (GMBE) in its straight line form to calculate the Initial-Oil/Gas In
Place using the PVT properties calculated from the new correlations and these values were compared to those calculated from
the compositional simulation models. The procedure to perform this comparison started by running hypothetical
compositional simulation cases to predict reservoir performance for each of the fourteen reservoir fluid samples. These runs
were also used in the simulation comparison between the MBO and compositional models. Then, the GMBE was used in its
straight line form (graphically) to estimate initial oil in place, N, and initial gas in place, G, following Walsh’s approach. Fig.
11 shows an example of the results of GMBE as a straight line using PVT generated from the new correlations. Table 14
compares between the calculated Initial Oil/Gas In-Place using PVT extracted from the new correlations and compared with
the values of the compositional simulation models. The table shows that the errors of material balance calculation using PVT
from the new correlations range from minimum of 3% up to a maximum of 23%, which represent reasonable accuracy.
Finally, we compared the results of the Modified Black Oil simulation using PVT generated from these correlations to the
results of Full Equation of State (EOS) compositional simulation. All simulation runs started from pressure greater than the
saturation pressure and went to pressures significantly below the saturation pressure (no pressure maintenance) up to
abandonment pressure of 500 psi. A commercial simulator program (ECLIPSE) was used for the simulation runs. Figs. 12
and 13 show example comparison results of compositional simulation and MBO simulation. These figures indicate a
reasonable match between reservoir pressure and the producing gas oil ratio calculated from MBO simulation (using PVT
properties calculated from the new correlations) and those calculated from the compositional simulation.
Discussion
The importance of the new correlations comes from the fact that they can generate reasonably accurate PVT properties for
volatile oil and gas condensate fluids without the need for a laboratory report or elaborate EOS calculations. They also take
into consideration the effect of surface separator conditions. Also, all parameters used in the correlations are readily available
especially for the surface gas gravity term that was a point of confusion in previous MBO correlations9,17
.
The developed correlations are expected to have wide application in MBO simulations and volatile oil and gas condensate
material balance applications. To highlight the new MBO PVT correlations applicability, we compared the results from the
new correlations to the results of one of the most widely used black oil correlations (Standing correlation)16
. The results from
Abdel Fattah9,17
correlations were also compared with the new work. We will consider here the MBO PVT properties
extracted with Whitson and Torp1
method as reference for comparison. Two samples (one volatile oil and one gas
condensate) were used for full comparison between the new correlations, Abdel Fattah’s correlations, and Standing
correlations. The two selected samples were not used in developing the new correlations to present unbiased testing. Two sets
of figures (Figs. 14-17 for the new gas condensate sample and Figs. 18-21 for the new volatile oil sample) show the
comparison between the MBO PVT properties calculated by the new correlation, Abdel Fattah’s, and Standing versus the
values extracted from the EOS model. The figures show that the new correlations perform much better than the other
correlations especially the one by Standing (which was developed for black-oil fluids).
All volatile oil samples from this work were then used in similar comparison and the error was calculated for the new
correlations, Abdel Fattah’s, and Standing. Table 16 provides the average absolute error for all the 4 MBO PVT functions
computed with all correlations. The error was calculated referenced to the Whitson and Torp MBO PVT properties
calculation method. Both the comparison figures and summary table show the superior behavior of the new correlations. One
should also notice that the common black-oil PVT correlations will usually perform badly in volatile oil and gas condensate
fluids. Also, the condensate-gas function (Rv) is not defined for commonly used black-oil PVT correlations.
Conclusions
Fluid samples representing different fluid composition and ranging from volatile oils to near critical fluids and up to gas
condensates were characterized using a commercial EOS PVT software program and new MBO PVT properties (Bo, Rs, Bg and
Rv) correlations were developed. Based on work presented in this paper, the following conclusions were made:
1. The new MBO PVT properties correlations do not require lab experiments or EOS model and they take into consideration
the surface separator configuration and conditions. Separate models were developed for volatile oil and gas condensate
fluids.
2. The obtained results show reasonable agreement between MBO PVT properties generated from the new correlations and
those extracted using Whitson and Torp (W&T) method. The average absolute error is 8.5% for volatile oils and 17.5% for
gas condensates.
3. Application of the new correlations in material balance and reservoir simulation was performed for both validation and for

10. 10 SPE 164712
estimation of error in case of applying those new correlations. The error in calculating the initial fluid in place using the
GMBE ranges from 3% to 23%. Reasonable agreement between MBO simulation using PVT from the new correlations and
fully compositional simulation was also obtained.
4. For volatile oil fluids, the new correlations are significantly more accurate than commonly used black-oil PVT correlations.
Acknowledgement
The authors would like to express their gratitude to both Cairo University and GUPCO for making the programs used in
this research work available.
Nomenclature
BIC = Binary Interaction Coefficient
Bg = gas formation volume factor, bbl/SCF
Bgi = initial gas formation volume factor, bbl/SCF
Bo = oil formation volume factor, bbl/STB
Boi = initial oil formation volume factor, bbl/STB
Bosat = Oil formation volume factor at saturation pressure, bbl/STB
CCE = constant composition expansion test
C7+ = Heptanes plus components
CGR = condensate yield, STB/MMscf, equal to Rv at
CVD = constant volume depletion test
DL = differential liberation test
EOS = Equation Of State
GIIP = original gas in-place OGIP, SCF
GC = Gas Condensate
GMBE = General Material balance Equation
IOIP = Initial oil in place, STB
MBE = Material Balance Equation
MBO = Modified Black Oil
PVT = Pressure – Volume - Temperature
PR = Peng-Robison
Psat = Saturation Pressure
Psep1 = Separator Pressure at first stage separator, psi
Psep2 = Separator Pressure at second stage separator, psi
Rs = solution gas-oil ratio, scf/STB
Rsi = solution gas-oil ratio at initial pressure, scf/STB
Rv = vaporized oil-gas ratio, STB/MMscf
Rvi = vaporized oil-gas ratio at initial pressure, STB/MMscf
STO = Stock Tank Oil Gravity, Fraction
SG1 = Gas gravity at first stage separator, Fraction
SG2 = Gas gravity at second stage separator, Fraction
Tr = Reservoir Temperature, F
VO = Volatile Oil
References
1. Whitson, C.H. and Trop, S.B.: “Evaluating Constant Volume Depletion Data,” Paper SPE 10067, SPE, Richardson, TX.
USA, 1983.
2. Schilthuis, R.J.: “Active Oil and Reservoir Energy,” Trans. AIME 1936, 148, pp. 33-52.
3. Walsh, M.P.: “A Generalized Approach to Reservoir Material Balance Calculations,” paper presented at the International
Technical Conference of Petroleum Society of CIM, Calgary, Canada, JCPT, May 9-13, 1994.

11. SPE 164712 11
4. Walsh, M.P., Ansah, J., and Raghavan, R.: “The New, Generalized Material Balance as an Equation of a Straight line: Part 1
– Applications to Under-Saturated and Volumetric Reservoir,” paper SPE 27684 presented at the 1994 SPE Permian Basin
Oil and Gas Recovery Conference, March 16-18, Midland TX.
5. El-Banbi, Ahmed H., Forrest, J.K., Fan, L., and McCain, W.D., Jr.: “Producing Rich-Gas-Condensate Reservoirs--Case
History and Comparison Between Compositional and Modified Black-Oil Approaches,” paper SPE 58988 presented at the
SPE Fourth International Petroleum Conference and Exhibition, Villahermosa, Mexico. Feb. 1-3, 2000.
6. Coats, K.H.: “Simulation of Gas Condensate Reservoir Performance,” Paper SPE 10512, JPT, Oct. 1985, pp. 1870-1886.
7. McVay, D.A.: Generation of PVT Properties for Modified Black Oil Simulation of Volatile Oil and Gas Condensate
Reservoirs, Ph.D. Thesis, Texas A&M University, TX. 1994.
8. Walsh, M.P., and Towler, B.F.: “Method computes PVT properties for Gas Condensate,” OGJ, July 31, 1994, pp. 83-86.
9. Abdel Fattah, Khalid A.: Volatile Oil and Gas Condensate Fluid Behavior for Material Balance Calculations and Reservoir
Simulation, Ph.D. Thesis, Cairo University, 2005.
10. Ibrahim, M, El-Banbi, Ahmed H., El-Tayeb, S., and Sayyouh, H.: “Changing Separator Conditions During Black-Oil and
Modified Black-Oil Simulation Runs,” paper SPE 142462 presented at the SPE Middle East Oil and Gas Show and
Conference, Manama, Bahrain, 6–9 March 2011.
11. Coats, K.H., Smart, G.T.: “Application of a Regression Based EOS PVT Program to Laboratory Data,” SPERE (May 1986)
277-299.
12. McCain, W. Jr.: “Analysis of Black Oil PVT Reports Revisited,” Paper SPE 77386, Oct. 2002.
13. Vasquez, M. and Beggs, D.: ”Correlation for Fluid Physical Property Predictions,” JPT, June 1989.
14. Al-Marhoun, M.A.: “Evaluation of Empirically Derived PVT Properties for Middle East Crude Oils,” Journal of Petroleum
Science and Engineering 42 (2004) pp.209-221.
15. McCain, W.D., Jr.: “Heavy Components Control Reservoir Fluid Behavior,” JPT (September 1994) 746-750.
16. Standing, M. B.: “Volumetric and Phase Behavior of Oil Field Hydrocarbon Systems,” SPE, AIME, 1977.
17. EL-Banbi, Ahmed H., Abdel Fattah, Khalid A., and Sayyouh, M.H.: “New Modified Black Oil Correlations for Gas
Condensate and Volatile Oil Fluids,” Paper SPE 102240 presented at the SPE Annual Technical Conference and Exhibition,
San Antonio, TX. Sept. 24-27, 2006.
Table 1 - Reservoir Fluid Samples Properties
NO. Sample Name Sample Type
C7+
(%)
Tres
(F)
Psat
(PSIG)
IGOR
(Scf/Stb)
1 VO 1 Volatile Oil 19.0 249 4527 1678
2 VO 2 Volatile Oil 16.9 176 4460 N/A
3 VO 3 Volatile Oil 14.9 246 4821 2000
4 VO 4 Volatile Oil 14.2 276 4375 2527
5 NC 1 Gas Condensate 12.7 312 5210 3413
6 NC 2 Gas Condensate 12.2 286 5410 4279
7 NC 3 Gas Condensate 11.7 238 4815 3405
8 GC 1 Gas Condensate 8.2 280 6750 5500
9 GC 2 Gas Condensate 8.2 215 4952 5403
10 GC 3 Gas Condensate 6.9 186 4000 5987
11 GC 4 Gas Condensate 6.5 312 5465 8280
12 GC 5 Gas Condensate 6.4 260 4525 7203
13 GC 6 Gas Condensate 5.9 267 4842 N/A
14 GC 7 Gas Condensate 5.5 240 3360 N/A

12. 12 SPE 164712
Fig. 1 Reservoir Fluid Samples Heptanes-Plus Range
Fig.2 Phase Plot from the tuned EOS for VO1 Fig.3 Comparison between EOS and Observed Relative volume values for VO1
Fig.4 Comparison between EOS and Observed Vapor Z-Factor values for VO1 Fig.5 Comparison between EOS and Observed Liquid Dropout values for VO1
C7+ Range
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
VO 1 VO 2 VO 3 VO 4 NC 1 NC 2 NC 3 GC 1 GC 2 GC 3 GC 4 GC 5 GC 6 GC 7
Sample Name
Volatile Oil
Near Critical
Gas Condensate
0
1000
2000
3000
4000
5000
6000
‐100 0 100 200 300 400 500 600 700 800 900
Temperature, F
Pressure, psi
0
1
2
3
4
5
0 1000 2000 3000 4000 5000 6000 7000 8000
Pressure, psi
Relative Volume
Rel. Vol. (EOS) Rel. Vol. (Obs.)
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
0 1000 2000 3000 4000 5000
Pressure, psi
Vapor Z‐Factor
Vapor Z‐factor (EOS) Vapor Z‐factor (Obs.)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1000 2000 3000 4000 5000
Pressure, psi
Liquid Dropout, fraction
Liq. Sat. (EOS) Liq. Sat. (Obs.)

13. SPE 164712 13
Fig.6 Comparison between EOS and Observed Moles Recovered values for VO1
Table 2- Saturation Pressure Correlation Parameters (Volatile Oils)
Table 3- Saturation Pressure Correlation Parameters (Gas Condensates)
Table 4- Solution Gas-Oil Ratio (Known Psat) Correlation Parameters
Table 5- Solution Gas-Oil Ratio (Unknown Psat) Correlation Parameters
Table 6- Oil Formation Volume Factor (Known Psat) Correlation Parameters
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 500 1000 1500 2000 2500 3000 3500 4000
Pressure, psi
Moles Recovered, fraction
EOS Obs
Fluid Separator Stages A0 A1 A2 A3 A4
VO 2 Stages 0.064 1.00 ‐0.005 ‐5.77 1.73
VO 3 Stages 0.033 1.06 ‐0.004 ‐6.22 1.83
Fluid Separator Stages A0 A1 A2 A3 A4
GC 2 Stages 729 ‐0.06 ‐0.007 0.97 0.38
GC 3 Stages 762 ‐0.06 ‐0.017 0.96 0.38
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages 1.51E‐04 1.06 ‐43.99 ‐6.24 ‐10.10
GC 3 Stages 1.78E‐04 1.13 ‐36.44 ‐774.33 332.73 ‐9.15
VO 2 Stages 4.88E‐04 0.63 407.53 ‐8.32 3.48
VO 3 Stages 4.26E‐04 0.55 356.32 1200.33 ‐524.59 4.00
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages ‐1.5E‐08 6.47E‐04 ‐0.12 ‐4.94 0.0007
GC 3 Stages ‐1.1E‐08 6.94E‐04 ‐0.12 ‐1597.13 698 0.001
VO 2 Stages 1.0E‐07 1.42E‐04 0.08 ‐7.80 0.001
VO 3 Stages 9.2E‐08 1.27E‐04 0.07 988.22 ‐433 0.001
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages 5.55E‐05 0.721 3193 1.6E‐05 0.0022
GC 3 Stages 7.25E‐05 0.690 3396 ‐3.0E‐05 6.6E‐05 0.0020
VO 2 Stages 1.81E‐04 0.294 4382 1.6E‐05 0.0007
VO 3 Stages 1.82E‐04 0.268 4444 ‐1.1E‐05 4.7E‐05 0.0006

14. 14 SPE 164712
Table 7- Oil Formation Volume Factor (Unknown Psat) Correlation Parameters
Table 8- Initial Condensate-Gas Ratio (Rvi) Correlation Parameters
Table 9- Condensate-Gas Ratio (Known Psat) Correlation Parameters
Table 10- Condensate-Gas Ratio (Unknown Psat) Correlation Parameters
Table 11- Gas Formation Volume Factor (Unknown Psat) Correlation 1 Parameters
Table 12- Gas Formation Volume Factor (Unknown Psat) Correlation 2 Parameters
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages ‐2.1E‐08 3.7E‐04 1.00 9.4E‐06 6.5E‐05
GC 3 Stages ‐1.9E‐08 3.6E‐04 1.05 ‐4.0E‐05 7.0E‐05 ‐3.8E‐05
VO 2 Stages 3.8E‐08 7.4E‐05 0.97 1.4E‐05 6.3E‐04
VO 3 Stages 3.8E‐08 6.8E‐05 0.98 ‐1.2E‐05 4.3E‐05 5.8E‐04
Fluid Separator Stages A0 A1 A2 A3 A4 A5
VO 2 Stages ‐1.6E‐23 6.7E‐02 52.94 ‐93.72 40.36 4.7E‐03
VO 3 Stages ‐9.9E‐27 4.8E‐02 63.82 ‐112.41 48.25 5.3E‐03
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages 8.6E‐08 ‐1.9E‐04 0.606 1.2E‐05 ‐5.0E‐02
GC 3 Stages 8.7E‐08 ‐2.0E‐04 0.634 ‐1.2E‐04 2.2E‐04 ‐4.9E‐02
VO 2 Stages 3.1E‐07 ‐9.3E‐04 1.493 ‐3.8E‐04 ‐7.4E‐02
VO 3 Stages 3.0E‐07 ‐9.4E‐04 1.554 ‐1.7E‐03 2.3E‐03 ‐7.2E‐02
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages ‐4.6E‐09 2.5E‐04 2.6E‐01 8.5E‐06 ‐2.4E+02
GC 3 Stages ‐2.0E‐09 2.3E‐04 2.9E‐01 ‐1.4E‐03 2.4E‐03 ‐2.3E+02
VO 2 Stages 4.7E‐07 ‐1.4E‐03 2.3E+00 ‐3.7E‐04 ‐4.6E+02
VO 3 Stages 4.6E‐07 ‐1.4E‐03 2.4E+00 ‐1.7E‐03 2.3E‐03 ‐4.5E+02
Fluid Separator Stages A0 A1 A2 A3 A4
GC 2 Stages 3626 ‐1.07 ‐3.1E‐05 0.0021
GC 3 Stages 3695 ‐1.08 ‐1.4E‐04 1.8E‐04 0.0022
VO 2 Stages 3015 ‐1.07 ‐8.9E‐05 0.0027
VO 3 Stages 2988 ‐1.07 ‐1.7E‐04 1.5E‐04 0.0029
Fluid Separator Stages A0 A1 A2 A3 A4
GC 2 Stages 349 ‐.78 ‐8.9E‐06 0.0020
GC 3 Stages 343 ‐.77 3.3E‐05 ‐7.5E‐05 0.0020
VO 2 Stages 449 ‐.81 ‐2.7E‐05 0.0020
VO 3 Stages 435 ‐.81 9.4E‐07 ‐5.2E‐05 0.0020

15. SPE 164712 15
Table 13- Under-Saturated Oil Formation Volume Factor Correlation Parameters
Fig. 7- Rs Cross Plot for VO known Psat (Three Stage Separator) Fig. 8- B
o
Cross Plot for VO known Psat (Three Stage Separator)
Fig. 9- RV Cross Plot for VO known Psat (Three Stage Separator) Fig. 10- Bg Cross Plot for VO (Entire P. Range) (Three Stage Separator)
Fluid Separator Stages A0 A1 A2 A3 A4 A5
GC 2 Stages ‐1.4E‐04 ‐6.1E‐06 1.4E‐04 0.997 8.1E‐08
GC 3 Stages ‐1.3E‐04 ‐6.1E‐06 1.4E‐04 0.997 8.0E‐07 ‐1.2E‐06
VO 2 Stages ‐7.1E‐05 ‐8.6E‐05 1.0E‐04 0.952 1.3E‐06
VO 3 Stages ‐7.0E‐05 ‐8.8E‐05 1.0E‐04 0.951 ‐5.4E‐06 1.4E‐05
R
2
= 0.9661
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Rs (Obs), Mscf/Stb
Rs (Calc), Mscf/Stb
R2
= 0.915
1
1.5
2
2.5
3
3.5
4
1 1.5 2 2.5 3 3.5 4
Bo (Obs), rbbl/Stb
Bo (Calc), rbbl/Stb
R2
= 0.9726
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Rv (Obs), Stb/Mscf
Rv (Calc), Stb/Mscf
R2
= 0.9724
0.5
0.75
1
1.25
1.5
1.75
2
0.5 0.75 1 1.25 1.5 1.75 2
Bg (Obs), rbbl/Mscf
Bg (Calc), rbbl/Mscf

16. 16 SPE 164712
Fig. 11- (F vs. Eo) for VO 1 (Two Stage Separator)
Table 14- Comparison Between GMBE and Simulation IOIP/GIIP
y = 17447371.084x
R2
= 0.993
000E+0
20E+6
40E+6
60E+6
80E+6
100E+6
120E+6
140E+6
160E+6
180E+6
200E+6
0.00 2.00 4.00 6.00 8.00 10.00 12.00
Eo, bbl/STB
F, bbl
Sample
Name
Sample Type
EOS_STOIIP
(STB)
MBal_STOIIP
(STB)
ERROR
(%)
VO 1 Volatile Oil 15110447 17828640 ‐18
VO 2 Volatile Oil 14361290 15373341 ‐7
VO 3 Volatile Oil 11714572 12684897 ‐8
VO 4 Volatile Oil 12663336 15610385 ‐23
Sample
Name
Sample Type
EOS_GIIP
MSCF
MBal_GIIP
MSCF
ERROR
%
NC 1 Near Critical 38712208 36317597 6
NC 2 Near Critical 37197692 32884012 12
NC 3 Near Critical 39709708 35588587 10
GC 1 Gas Condensate 50497036 44672755 12
GC 2 Gas Condensate 45546884 43311130 5
GC 3 Gas Condensate 48092008 43808088 9
GC 4 Gas Condensate 42919356 41842871 3
GC 5 Gas Condensate 44155820 41464689 6
GC 6 Gas Condensate 46457824 43642928 6
GC 7 Gas Condensate 50783260 46557199 8

17. SPE 164712 17
Fig. 12- Reservoir Pressure for MBO and Comp. Simulation for VO1 Fig. 13- Producing Gas Oil Ratio for MBO and Comp. Simulation for VO1
Table 15- New MBO PVT Correlations Average Absolute Error
0
1000
2000
3000
4000
5000
6000
7000
0 500000 1000000 1500000 2000000 2500000 3000000 3500000
Cum Oil Production, STB
Reservoir Pressure, psi
Pr_Models Pr_W&T
0
5
10
15
20
25
30
35
40
45
50
0 500000 1000000 1500000 2000000 2500000 3000000 3500000
Cum Oil Production, STB
Producing GOR, MScf/Stb
PGOR_Models PGOR_W&T
R Square Avg. Error R Square Avg. Error R Square Avg. Error R Square Avg. Error
Saturation Pressure Correlation 3% 12% 3% 12%
Solution Gas Oil Ratio Correlation
(Known Psat)
96% 12% 88% 21% 97% 11% 89% 19%
Solution Gas Oil Ratio Correlation (Un‐
Known Psat)
97% 11% 78% 28% 97% 11% 79% 26%
Oil Formation Volume Factor
Correlation (Known Psat)
92% 6% 82% 10% 92% 6% 81% 10%
Oil Formation Volume Factor
Correlation (Un‐Known Psat)
96% 4% 70% 11% 96% 4% 70% 11%
Condensate Gas Ratio Correlation
(Known Psat)
96% 15% 85% 22% 97% 15% 85% 22%
Condensate Gas Ratio Correlation (Un‐
Known Psat)
96% 16% 80% 25% 97% 15% 80% 25%
Gas Formation Volume Factor
Correlation (Model 1)
100% 9% 100% 13% 100% 9% 100% 13%
Gas Formation Volume Factor
Correlation (Model 2)
98% 11% 98% 16% 98% 12% 99% 16%
Under‐Saturated Solution Gas Oil
Ratio Correlation (Known Psat)
8% 14% 10% 15%
Under‐Saturated Solution Gas Oil
Ratio Correlation (Un‐Known Psat)
9% 25% 10% 27%
Under‐Saturated Oil Formation
Volume Factor Correlation
1% 1% 1% 1%
Under‐Saturated Condensate Gas
Ratio Correlation (Known Psat)
9% 16% 11% 16%
Under‐Saturated Condensate Gas
Ratio Correlation (Un‐Known Psat)
12% 24% 14% 26%
Under‐Saturated Gas Formation
Volume Factor Correlation
35% 36%
2 Stage Separator 3 Stages Separator
VO GCVO GC

18. 18 SPE 164712
Table 16 – Error Comparison Between This Work, Abdel Fattah’s and Standing Correlations for All Volatile Oil Samples Combined
Method Rs Rv Bo Bg
New Correlation 8 26.8 1.8 0.5
Abdel Fattah
Correlation
33.2 42 5.3 7.6
Standing
Correlation
62.5 N/A 18.9 64
Fig. 14 - Rs Correlations Comparison for Gas Condensate Test Sample Fig. 15 Bo Correlations Comparison for Gas Condensate Test Sample
Fig. 16 – Rv Correlations Comparison for Gas Condensate Test Sample Fig. 17 Bg Correlations Comparison for Gas Condensate Test Sample

19. SPE 164712 19
Fig. 18 - Rs Correlations Comparison for Volatile Oil Test Sample Fig. 19 - Bo Correlations Comparison for Volatile Oil Test Sample
Fig. 20 – Rv Correlations Comparison for Volatile Oil Test Sample Fig. 21 – Bg Correlations Comparison for Volatile Oil Test Sample