Fracture Treating Pressure

Behavior
 Purpose of Fracture Treating
Pressure Analysis
Provides diagnostic information about
the completion
May indicate general trends in fracture
geometry development
Determines whether the job can be
pumped to completion
May suggest reasons for post-frac well
performance

Copyright B&A 2004

 Observed Surface Treating Pressure
is Influenced by Many Factors

Pwh  Pclose  Phyd  Pff  Ppf  Pnw  Ptf

Pwh = Observed Wellhead Treating Pressure
Pclose = Bottomhole Fracture Closure Pressure
Phyd = Hydrostatic Fluid Head
Pff = Fracture Friction Pressure Drop
Ppf = Perforation Friction Pressure Drop
Pnw = Near Wellbore Tortuosity Losses
Ptf = Tubing Friction Pressure Drop
Note that all frictional components approach zero
during shut-in.

Copyright B&A 2004

 Hydrostatic Head Changes
Continuously with Ramped Proppant
 Total head is the sum of the head loss for all
proppant stages in the tubing.
P   0.433 h
hyd f s
where
f = the density (g/cc) of the slurry stage
hs = the height (ft) of the slurry stage in the tubing

 f  Cv  s  1  Cv   l
Ppg
01199
. s
Cv 

1  01199
. Ppg
s 
Copyright B&A 2004
 Pipe Friction Losses Can Be
Estimated for Each Fluid Type
1000
A Tubular Friction Losses
for 40 lb/Mgal HPG Fluid
B
Friction Pressure, psi/100 ft

C
A — 1.25 in tubing
100 B — 1.5 in tubing
D
C — 2.375 in tubing
E
D — 2.875 in tubing
E — 3.5 in tubing
10 F
F — 4.5 in casing
G G — 5.5 in casing
H H — 7.0 in casing
1
1 10 100
Flow Rate, bpm
Copyright B&A 2004
 Pipe Friction Losses Are Affected
by Proppant Concentration
120
10 bpm
15 bpm
100
20 bpm
% Friction Pressure Increase

80 30 bpm

Increase in pipe friction

60 over clean fluid with the
addition of 20/40 sand
40

20

0
0 5 10 15
Pounds Per Gallon Sand
Copyright B&A 2004
 Sand Size May Also Affect
Pipe Friction Losses
90
10/20 mesh
80
20/40 mesh
70
% Friction Pressure Increase

60

50 Frictional pressure
increase at 30 bpm
40
for two sand sizes.
30

20

10

0
0 5 10 15
Pounds Per Gallon Sand
Copyright B&A 2004
 Tubing Friction Estimates with
Laminar-Turbulent Transition

 
Ptf
o
 0.48d 4.8 q 1.7 L for base solvent (psi), d (inches), q(bpm),
L (ft)

if CHPG  0, for gel and slurry, CHPG(lb/Mgal), Cs (ppga)

 
0.45 Ptf Exp 0.029Cs 
P   o

 
0.3 ln  C HPG 
atan  q 0.06d 1.7 CHPG 
tf s 0.9
 

Copyright B&A 2004

 Perforation Restriction Causes a
Large Pressure Drop

. q f
1975 2

Ppf  2 2 4
; psi
C N d
D p p

where
q = the total pump rate, bpm
f = the slurry density, g/cc
CD = the perforation coefficient
Np = the number of open perforations
dp = the perf diameter, inches

Copyright B&A 2004

 Near Wellbore Tortuosity May
Also Cause Pressure Losses
2000
Ppf
1800
Pnw
 Near wellbore tortuosity
1600
may be caused by:
1400
• turning fractures
Pressure Drop, psi

1200 • multiple fractures

1000 • poor cement
800
• perforation orientation
• deviated wellbore
600
• perforation damage
400

200
 It cannot usually be
0
0 5 10 15 20 25
predicted, but can be
Rate, bpm measured.

Copyright B&A 2004

 Pressure Change at Shut-In
Measures All Frictional Components
DPISI = Pff + Ppf + Pnw + Ptf

5500

5000
Surface Pressure, psi

4500

4000

3500 DPISI2
3000 DPISI1
2500

2000

1500

1000
0 20 40 60 80 100 120

Pumping Time, min

Copyright B&A 2004

 Accounting for Frictional Losses
During SRT
Wellbore Pressure Calculations

12000 12

10000 10

8000 8
dPfr(psi)
Phydr
Pressure, psi

Psurf
Psurf(meas)
6000 6
BHP
BHP+dPsf
Q(bpm)
PPG@surf
4000 4

2000 2

0 0
0 2 4 6 8 10 12 14 16 18
Time, minutes
Copyright B&A 2004
 BHP Screenout Indicated by Incorrect
Accounting for Slurry Friction
10000 35

9000
30

8000

25
7000

6000 dPfr(psi)
20
Pressure, psi

Phydr
BHP
5000
Psurf(meas)
Q(bpm)
15
4000 PPG@surf

3000
10

2000

5
1000

0 0
36.5 46.5 56.5 66.5 76.5 86.5
Time, minutes
Copyright B&A 2004
 Correct Accounting for Friction
Shows no Screenout Behavior
10000 35

9000
30

8000

25
7000

6000 dPfr(psi)
20
Pressure, psi

Phydr
BHP
5000
Psurf(meas)
Q(bpm)
15
4000 PPG@surf

3000
10

2000

5
1000

0 0
36.5 46.5 56.5 66.5 76.5 86.5
Time, minutes
Copyright B&A 2004
 Use of the Inverse Hydrostatic
Head Curve
PUSER001 (psi) A Slurry Rate (bpm) B
Proppant Concentration (lb/gal) C Calc'd BH Pressure (psi) A
A B Bottomhole Proppant Conc (lb/gal) C Hydrostatic Pressure (psi) D C
20000 30 6 -5000

18000
-2500
25 5
16000
0
14000
20 4
2500
12000

10000 15 3 5000

8000
7500
10 2
6000
10000
4000
5 1
12500
2000

0 0 0 15000
01:40 01:45 01:50 01:55 02:00 02:05 02:10
3/9/2004 3/9/2004
Time
Copyright B&A 2004
 What’s Wrong with This
Picture?
Casing Pressure (psi) A Calc'd BH Pressure (psi) A
Slurry Rate (bpm) B Surface Proppant Conc (lb/gal) C
A Bottomhole Proppant Conc (lb/gal) C B C
12000 40 7

11000 6

10000 30
5

9000
4
8000 20
3
7000

2
6000 10

5000 1

4000 0 0
13:20 13:30 13:40 13:50 14:00 14:10 14:20 14:30 14:40 14:50
6/6/2002 6/6/2002
Time Copyright B&A 2004
 Net Fracture Treating Pressures
Follow a Predictable Pattern
1000
m=1:
m=0: Restricted
m=1/8 to 1/4: height and m>1:
Net Pressure, psi

Stable height Near well

Contained height length
growth restriction
100

m<0:
Rapid or unstable
height growth

10
1 10 100
Pumping Time, min
Copyright B&A 2004
 Nolte-Smith Log-Log Diagnostic
Plot
Y-axis:
 Log of BH treating pressure minus closure pressure
X-axis:
 Log of elapsed pump time or CumVol/Avg Rate
Must know accurate closure pressure
Curve shapes are non-unique and affected by
other factors
 Perf (and near-well) erosion
 Fluid changes (friction, density, viscosity)
 Changes in leakoff (fissures, etc.)
 Perf (and near-well) bridging or packing
Copyright B&A 2004
 Net Pressure Analysis Provides
Information about Fracture Growth

10000
Pressure, psi

1000

Pnet = PBH – Pclose

100
1 10 100

Pumping Time, min

Copyright B&A 2004
 Pressure Derivative Analysis Clearly
Identifies Growth Behavior
10000

1000
Pressure, psi

100

 Pnet
Derivative  t
10
t

Net Pressure
Derivative

1
1 10 100

Pumping Time, min

Copyright B&A 2004

 Fracture Growth through
a Thin Barrier Indicated by Derivative
10000

1000
Pressure, psi

100

10

Net Press
Derivative

1
1 10 100

Pumping Time, min

Copyright B&A 2004
 Conclusions
Treating pressure behavior can provide
good general diagnostic information
about fracture development
Pressure used should represent the
actual treating pressure
Frequently wellbore, near-well, and
perforation events dominate the
measured pressure
These events can lead to
misinterpretations of fracture geometry
Copyright B&A 2004