PVTi and E100 problem HELP !!!

[ivan_aljeburi]

i'm having huge a problem with PVT file for E100... E100 is giving me some errors and warnings due to NEGATIVE COMPRESSIBILITY.... can anyone help with this one?

Thanx

[Shakespear]

First check would be to look at the PVT Tables you have and make sure the curves are smooth (the data points you have make nice smooth lines). None smooth data can make E100 extrapolate badly and cause this problem. That is the first place to look.

I looked on the NET but nothing obvious poped out. Here is a translation of a Chinese site that mentions this problem and OTHERS !!

[link Point to another website Only the registered members can access]

[ivan_aljeburi]

Thank you bro..

The same thing is written in E100 manual but no use of it.. The .PVO file i use is an export from PVTi so i suppose it is smooth enough...i can do that manualy but i don't think it is a good solution...
Thanks for your effort

[Shakespear]

Next thing to look at is when the problem happens in the simulation and what changes in the run at that moment, as you might have a big pressure transient and this may cause e100 to try extrapolate PVT data where you have none in your PVT table.

I would still inspect the curves from PVTi in OFFICE or PVTi. Were they from correlations ?

[ivan_aljeburi]

i absolutly agree with you.. here is the situation... my PMAX is 700 bars but the PVT data go to the last highest pressure in the experiment table (PVTi)... i need it to go to 700 bars but i can't do that in PVT.. i really think that the problem lies in exporting .PVO from PVTi to PMAX.. but i can't do it, i don't know how.

[Shakespear]

Forgot another thing you might try. Add more data points to make those curves better defined and smoother.

[Shakespear]

If you look in your DATA file in the section with you PVT data, you could manually adjust those tables. Not sure how complicated you model is but this manual adjustment could be done for instance in EXCEL. A lot of people do it this way

Does you simulation when you run it ACTUALLY go to 700 bar?

Is you simulation running OK and then after lets say 2 months you see this problem, or does it start from the very start?

[ivan_aljeburi]

Hahahahah... first i wanted the PVT file to be done in PVTi becasue i have the lab data but now i will use the best fitting correlation and make it .... stupid PVTi...
But the main question remains ... how can i make PVTi extrapolate to a value i want?
that problem is killing me....!!!!!!!

[Shakespear]

Ivan I am not a Pvti expert, but there must be in the package an option that tells it to output or generate tables over a given range of pressure.

Search on keyword "range" in the PVTi manual.

When choosing the values of the max/min temperature and pressure consider the following:
a The default values of Pmax=1000psia, Pmin=50psia, Tmax=400F and Tmin=50F are considered to be reasonable variations of conditions within a reservoir.
b Make sure you alter these values to suit your particular reservoir. Try to make the region that is defined by your values as small as possible. For example if you think that the maximum temperature in your reservoir is going to be 300F then do not leave the default value in the panel (400F).

Read this on page 22

Can Black Oil tables be extended above the liquid
bubble point in PVTi? Yes. You should make sure that the DL/CVD .....

[LOST]

I had the same problem during simulation with petrel.
it is just due to bad license or -----. try it with new version of licensed software

[Avers]

i'm having huge a problem with PVT file for E100... E100 is giving me some errors and warnings due to NEGATIVE COMPRESSIBILITY.... can anyone help with this one?

Thanx

I am familiar how to solve the problem. If it is still actual, just write me to alknyaz@yahoo.com
I will send you excell spreadsheet to fix it.

[Avers]

i'm having huge a problem with PVT file for E100... E100 is giving me some errors and warnings due to NEGATIVE COMPRESSIBILITY.... can anyone help with this one?

Thanx

No reasons to share it with one person. I tink it is very common problem. So,

Sometimes ECLIPSE will calculate negative total compressibility values even if the PVT tables are not extrapolated. This issue seems to occur due to either PVT tables created by PVT packages that do not have total compressibility checks as in PVTi, or user errors when tables are exported. Whatever the reason, it is possible to fix the total compressibility by slightly changing Bo values as shown in the attached tables. These small changes should improve simulation convergence.

Gas and oil PVT properties are defined in the ECLIPSE deck using PVDG and PVTO
(or PVCO). ECLIPSE then issues the following message:

@--WARNING AT TIME 0.0 DAYS ( 1-JAN-1990):
@ NEGATIVE COMPRESSIBILITY FOUND IN GAS
@ PRESSURE TABLE 1 AND OIL PRESSURE
@ TABLE 1 AT A SAMPLE PRESSURE VALUE
@ 5911.25172 . ADJUST SATURATED FLUID
@ PROPERTY VALUES AT THIS PRESSURE.
@ NEGATIVE COMPRESSIBILITIES OCCUR
@ FOR GAS SATURATIONS LESS THAN 0.01461

Why does ECLIPSE issue this message? What does it mean? How do I fix it?

Answer Details:
To ensure the total volume of the oil-gas mixture reduces as the pressure
increases, ECLIPSE checks that the total compressibility of the hydrocarbon
system along the "saturated line" is positive.

The total compressibility of the oil-gas mixture is defined as:

Ct = Sg*Ct,g + So*Ct,o --- (1)

where,
Ct,g = [-dBg/dP + dRv/dP * (Bo-Rs*Bg) / (1-Rs*Rv)]/Bg --- (2)
Ct,o = [-dBo/dP + dRs/dP * (Bg-Rv*Bo) / (1-Rs*Rv)]/Bo --- (3)

For each PVT region, ECLIPSE selects a pressure range which spans the complete
range of the pressure data in the corresponding oil and gas PVT tables. This
pressure range is then subdivided into 30 equally spaced pressure nodes for the
evaluation of the total hydrocarbon compressibility.

At each pressure node, ECLIPSE checks two limiting total compressibilities:

Ct = Ct,o at Sg = 0, So = 1 --- (4)
Ct = Ct,g at Sg = 1, So = 0 --- (5)

If either or both limiting values are found to be negative at a pressure node,
ECLIPSE will issue a warning message indicating the offending pressure and the
gas saturation range for which negative total compressibility occurs:

If Ct < 0 --- (6)
Then Sg < - Ct,o / (Ct,g - Ct,o) --- (7)

The effect of a negative total compressibility is usually to cause the simulator
to experience numerical difficulties (convergence failures and/or erratic
changes in the solution).

A complete table of total compressibility evaluation can be written to the .DBG
file by activating switch 21 of the DEBUG keyword in the PROPS section.

If the range of sample pressures extends above the maximum bubble point entered
in the PVTO table or above the maximum dew point entered in the PVTG table, then
ECLIPSE will be forced to extrapolate above the highest entered Rs or Rv. This
extrapolation is linear in Rs, 1/Bo, 1/(Bo*Visco), Rv, 1/Bg, and 1/(Bg*Viscg).
In this case, it is not unlikely that negative compressibilities could occur as
a result of extrapolation.

It is recommended that the highest bubble point node in the PVTO table is
constructed so as to avoid extrapolations above the highest entered Rs in
the table. Similarly, in a run with vaporized oil present, it is recommended
that the highest dew point node in the PVTG table is constructed so as to avoid
extrapolation above the maximum entered Rv during the simulation.

Three examples of PVT tables are provided below for a model with dry gas (Rv=0)
and live oil (Rs>0):

__________________________________________________ _____________________________

Example 1. PVDG and PVTO are defined with a single "undersaturated line".
ECLIPSE extrapolates Bo and Rs from 4014.7 to 9014.7 psi along the
saturated line. This may cause negative total compressibilities to be
calculated.

PVDG
-- PG BG VISG
14.7 177.506 0.0086
264.7 11.218 0.0098
514.7 5.943 0.0111
1014.7 3.108 0.0135
2014.7 1.614 0.0182
2514.7 1.306 0.0205
3014.7 1.098 0.0228
4014.7 0.835 0.0272
5014.7 0.675 0.0314
6014.7 0.568 0.0355
7014.7 0.490 0.0394
8014.7 0.432 0.0432
9014.7 0.386 0.0468
/

PVTO
-- RS (PBUB)O BO VISO
0.0061 14.7 1.117 0.994 / -- saturated
0.0944 264.7 1.160 0.953 / -- saturated
0.1776 514.7 1.202 0.913 / -- saturated
0.3384 1014.7 1.284 0.838 / -- saturated
0.6492 2014.7 1.435 0.706 / -- saturated
0.8014 2514.7 1.505 0.648 / -- saturated
0.9520 3014.7 1.571 0.595 / -- saturated
1.2497 4014.7 1.691 0.502 -- Psat
5014.7 1.668 0.549
6014.7 1.646 0.597
7014.7 1.623 0.645
8014.7 1.601 0.692
9014.7 1.578 0.740 /
/

__________________________________________________ _____________________________

Example 2. PVDG is the same as that in the Example 1 but PVTO has multiple
"undersaturated lines". Since the saturated PVT data exists for
the complete pressure range (14.7 to 9014.7 psi), ECLIPSE does not
extrapolate Bo and Rs along the saturated line.

PVTO
-- RS (PBUB)O BO VISO
0.0061 14.7 1.117 0.994 /
0.0944 264.7 1.160 0.953 /
0.1776 514.7 1.202 0.913 /
0.3384 1014.7 1.284 0.838 /
0.6492 2014.7 1.435 0.706 /
0.8014 2514.7 1.505 0.648 /
0.9520 3014.7 1.571 0.595 /
1.2497 4014.7 1.691 0.502 -- Psat
5014.7 1.668 0.549
6014.7 1.646 0.597
7014.7 1.623 0.645
8014.7 1.601 0.692
9014.7 1.578 0.740 /
1.5436 5014.7 1.796 0.423 -- saturated
6014.7 1.772 0.463
7014.7 1.748 0.503
8014.7 1.724 0.543
9014.7 1.700 0.583 /
1.8346 6014.7 1.886 0.356 -- saturated
7014.7 1.861 0.390
8014.7 1.835 0.424
9014.7 1.810 0.458 /
2.1232 7014.7 1.960 0.300 -- saturated
8014.7 1.934 0.329
9014.7 1.908 0.357 /
2.4097 8014.7 2.019 0.253 -- saturated
9014.7 1.992 0.277 /
2.6945 9014.7 2.063 0.213 -- saturated
10014.7 2.035 0.234 /
/

__________________________________________________ _____________________________

Example 3. PVDG is the same as that in the Example 1 but PVCO has the multiple
compressibilities (CO) and viscosibities (CV) of the undersaturated
oil. Since the saturated PVT data exists for the complete pressure
range (14.7 to 9014.7 psi), ECLIPSE does not extrapolate Bo and Rs
along the saturated line.

PVCO
-- (PBUB)O RS BO VISO CO CV
14.7 0.0061 1.117 0.994 1.334E-05 9.501E-05 -- saturated
264.7 0.0944 1.160 0.953 1.334E-05 9.501E-05 -- saturated
514.7 0.1776 1.202 0.913 1.334E-05 9.501E-05 -- saturated
1014.7 0.3384 1.284 0.838 1.334E-05 9.501E-05 -- saturated
2014.7 0.6492 1.435 0.706 1.334E-05 9.501E-05 -- saturated
2514.7 0.8014 1.505 0.648 1.334E-05 9.501E-05 -- saturated
3014.7 0.9520 1.571 0.595 1.334E-05 9.501E-05 -- saturated
4014.7 1.2497 1.691 0.502 1.334E-05 9.501E-05 -- saturated
5014.7 1.5436 1.796 0.423 1.334E-05 9.501E-05 -- saturated
6014.7 1.8346 1.886 0.356 1.334E-05 9.501E-05 -- saturated
7014.7 2.1232 1.960 0.300 1.334E-05 9.501E-05 -- saturated
8014.7 2.4097 2.019 0.253 1.334E-05 9.501E-05 -- saturated
9014.7 2.6945 2.063 0.213 1.334E-05 9.501E-05 -- saturated
/

Attached excel sheet illustrates how a small change in specific Bo values can remove negative total compressibilities. Just turn on option 21 in DEBUG keyword to generate the check table. JUST RENAME the attached *.doc file in to *.xls

Regards,
Alex