Refracs: What can we decipher? - Mike Vincent

$: Refracs: What can we decipher? - Mike Vincent$
1

Mike Vincent

Insight Consulting

[email protected]

303 568 0695

Refracs –

What can we decipher?

SPE 134330 & 136757

Fracwell

LLC

Outline

Introduction

• Non-unique solutions, uncertainty, value of field data

Refrac Field Results

• Refrac definition

• History of restimulation

• Why do refracs work?

– Mechanisms; cause/effect

– Field examples

• Why do refracs fail?

Interpretations and Recommendations

2

Can I reinforce my misconceptions?

SPE 106151 Fig 13 – Production can be matched with a variety of fracture and reservoir parameters6

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 100 200 300 400 500 600

Production Days

Sta

ge

Pro

du

ctio

n (

mcfd

)

0

20

40

60

80

100

120

140

160

180

200

Cu

mu

lative

Pro

du

ctio

n (

MM

scf)

Actual production data

Long Frac, Low Conductivity

Medium Frac, Low Conductivity

Short Frac, High Conductivity, Reservoir Boundaries

500' Xf, 20 md-ft, 0.5 uD perm, 23 Acres 4:1 aspect ratio

100' Xf, 20 md-ft, 5 uD perm, 11 Acres 4:1 aspect ratio

50' Xf, 6000 md-ft, 10 uD perm, 7 Acres 4:1 aspect ratio

• History matching of production is surprisingly non-unique.

• Too many “knobs” available to tweak

• We can always blame it on the geology

If I don’t uniquely match this, can

I trust my refrac interpretation?

Removing the Uncertainty

• If we require a production match of two different frac designs, we remove many degrees of freedom

– lock in all the “reservoir knobs”!

– The difference in production must be explained with the difference in the FRAC descriptions, not the reservoir description, right?

7

3

Recent Recent attempts to history-match field production:

• SPE 119143 – summary of 200 published field trials in

which initial frac designs were altered

• SPE 134330 – summary of 143 published reports of the

outcome of refrac attempts (worldwide)

• SPE 136757 – evaluation of ~ 100 Bakken refracs

– Direct search - minimizes “publication bias”

– First field to allow evaluation of proppant durability on refrac success

• There are many surprising results that are difficult to

explain with traditional models

• “It’s awfully dark down there”

• Fracs are not as efficient/durable as we anticipate

• Keep an open mind, dig, challenge.

• None of us have it figured out. Consider outlandish ideas!

Refracs

My Definition of Refrac

• A second propped fracture treatment (with production data between!)

• Also tri-fracs, quads-, and “cinco-de-fraco” restimulations

• I have excluded a large dataset where propped fracs

improve previous acidized or unpropped initial treatments

– What can we learn about durability of unpropped portions of the

frac?

Goals

• Identify restimulation opportunities

• Improve initial designs

• Understand fracture and refrac mechanisms

• Improve production models and frac models

4

History1953

Observations:

May recover or exceed initial

productivity!

Decline rate also reduced!

1955 – Refrac

and Tri-Frac

Adapted from Garland, 1957

Sallee&Rugg,1953

Does this outcome fit your

intuition or model?

– Through 1970

• 35% of the 500,000 stimulation treatments were refracs

– 1996

• Only 2-3% of current stimulation activity was refracs

– Why?• Did we improve design, implementation, durability of initial frac

treatments?

• Did technological evolution cease?

• Did economic parameters change?

• Are we no longer staffed to analyze refrac opportunities?

• Did refracturing fall from vogue and become lost art?

• Are refracs not as effective as initially believed?

– Current: Desire to understand refracturing

History

5

Where Have Refracs Worked?Well Types

Oil Wells (under primary depletion, waterflood, or EOR)

Condensate Wells

Gas Wells

Gas Storage Wells

Water Production Wells

Water Injection Wells

Steam Injection Wells

Huff-n-Puff , cyclic injection/withdrawal Wells

Disposal Wells

Formation Types

Carbonates, limestones, dolomites, chalks, evaporites

Sandstones, cherts, siliceous diatomites

Coal (CBM), immature ductile shales, brittle shales

Conglomerates, unconsolidated formations, siltstones

16-page Appendix is attached to

SPE 134330 describing all field

examples

– Immediately post-frac

• Unsuccessful implementation

• To achieve new entry points, diversion, reorientation

– Somewhat later

• Most common published examples are 1-10 years later

– Much Later

• >20 years, Clinton Sand, high perm oil, Ohio

• >30 years, Mesaverde tight gas sand, Colorado

• >30 years, Rangely Field, high perm oil, water + CO2

flood, Colorado

• >40 years, Pembina mature waterflood, Alberta

• >30 years, Medicine Hat, Milk River shallow gas

– Ideas available on the theoretically “optimal” time to refrac,

but we need to discuss mechanisms first!

When have Refracs been Performed?

6

Refrac after 25 years of Depletion?15 refracs (shallow gas – Medicine Hat & Milk River)

All 15 responded to restimulation

Average increase 600-900%

6 of 15 achieved higher production than peak 25+ years prior

4 of 15 achieved rates within 25% of peak IP

Rate prior to Refrac

First 3 months, Initial Frac

3 month average, Post Refrac

Adapted from Gutor, 2003

Observation:

Excellent

success

despite

decades of

production.

Why Do Refracs Work? (mechanisms)

Refrac success (worldwide) has been attributed to:

• Enlarged frac (more reservoir contact)

– Improved pay coverage (add pay in vertical wells)

– Better lateral coverage (horizontal wells)

• Increased frac conductivity

– Restore conductivity lost – frac degradation

– Address unpropped/poorly propped portions

• Improve wellbore-to-frac connection/conductivity

• Reorientation

• Use of more suitable frac fluids

• Re-energizing natural fissures

• Other mechanisms

7

Improved Reservoir Contact

• Bypassed Pay in Vertical Wells

– Where radioactive tracers or detectable proppants indicated missed

opportunities, refracs are frequently successful

Adapted from Hecker, 1995

Refrac contacted initially bypassed pay

Improved Reservoir Contact• Bypassed Pay in Vertical Wells

– 50 discrete sands in Pinedale [Huckabee 2005]

• Can we touch it all with 15-20 stages?

– ~26 sand intervals in Piceance

• 28% of those stages fail to produce measurable gas? [Esphahanian 1997]

• Limited entry didn’t induce effective diversion? [Craig & Odegard 2008]

– 30-40 sand intervals in Jonah?

• 35 to 40% fail to contribute meaningfully [Eberhard 2003]

– Almond, Cotton Valley, Delaware, Red Fork

• In 40% of wells, at least one targeted interval failed to accept any RA tracer

[Fisher 1995]

• Touching Pay is Job #1– If kv/kh is tiny, you MUST create conductive, durable, vertical breach of

these horizontal laminations

– Despite what a simple model will predict, not all gas molecules are

hydraulically connected to a perforation

– Is bypassed pay a primary refrac opportunity?

8


• Improved Coverage of Horizontal Laterals

– Where radioactive tracers or detectable proppants indicated missed

opportunities, refracs are frequently successful

Adapted from Lantz, 2007

• Montana Bakken, cemented laterals believed drilled for longitudinal growth

• Initial fracs ~300,000 lb 20/40 RCS

• RA tracer showed incomplete coverage

• Add perfs, refrac 600,000 lb 20/40 sand, 10 ppg slugs, ball sealers


Adapted from Lantz, 2007

Refrac reduced GOR,

increased oil rate and reserves

• Lower breakdown pressures, but 50% greater net pressure

• We should revisit this issue!

• Oil rates up, GORs down

• Refracs diverted into undrained areas of the reservoir

• EUR increased 1,300,000 bbls in 16 refractured wellbores

• 1 mechanical problem, but 16 of 17 successful (94%)

9

Refrac on a Schedule?

Pagano, 2006

– Gas Condensate wells in DJ Basin – up to 5 restimulations

– Rangely oilfield – 1700 refracs 1947-1989. Most wells have

received 3-4 refracs yet remain viable restimulation candidates.

– Pembina oilfield – Conductivity was understood to degrade over

time, with production falling to unstimulated rates in 6-7 years.

Does Conductivity Degrade?

0

0.2

0.4

0.6

0.8

1

0 15 30 45 60 75

Perm

eab

ilit

y R

ati

o

Days at Constant Stress

IDC at 10,000 psi (69 MPa)

LWC at 10,000 psi (69 MPa)

Sand at 5000 psi (35 MPa)

Cobb, 14133

200F, 5000/10,000 psi [93C, 35/69 MPa]

All non-corrodible surfaces, prop in

Teflon tube, continuous flowing 2% KCl

McDaniel , 15067

275F, 8000 psi [135C, 55MPa]

showed importance of using silica

saturated, deoxygenated brine

10

100

1000

10000

0 10 20 30 40 50

Co

nd

ucti

vit

y (

md

-ft)

Days at Constant Stress, 8500 psi

IDC - Intermediate Density Ceramic

Proflow - Precursor to LWC

RCS - Resin Coated Sand

Ottawa Sand

0

20

40

60

80

100

0 30 60 90 120 150 180 210 240 270

% O

rig

inal

Co

nd

ucti

vit

y

Days at Constant Stress, 5000 psi

20/40 Sand at 75F

10/20 Sand at 250F

Hahn, SPE Drilling 1986

300F, 8500 psi [149C, 59 MPa]

Teflon tube, continuous flowing 2% KCl,

Non-silica saturated

Montgomery (12616) 1984

75/250F, 5000 psi [23C/121C, 34 MPa]

API “short term” cell: Metal plates,

continuous flowing 2% KCl,

Non-silica saturated

25

Does Conductivity Degrade?

Previous testingHandren 110451

250F, 6000 psi [121C, 41MPa]

Modern conductivity cell, Ohio Sandstone

Deoxygenated, Silica Saturated 2% KCl,

Does Conductivity

Degrade?

All

proppants

degrade, but

at different

rates.

Is replacing

degraded

proppant a

major factor

in refrac

success?

11

Perhaps this makes sense?

Pagano, 200627

Provocative Statements• Refrac DJ with minimal data

– >5000 refracs performed

– Few wells individually metered, even fewer with production log

or single-zone test (comingle J Sand, Codell and Niobrara)

– >8 papers written – arguing different mechanisms that account

for success

– We don’t know where, why, or how these refracs work

• No correlation between reorientation and production? (Wolhart)

• Which do you agree with?

– “Engineering malpractice”

– “Statistically validated approach to blanket refrac”

– “Honestly? Do this 5000 times without actually trying to

investigate whether we can address durability?”

– “Financially prudent approach”

12

Proppant Durability Affect Refracs?

• SPE 136757

– Examines refrac success of ~100 horizontal wells in

Bakken

– Large number of wells completed with:

• Sand

• RCS

• Ceramic

– Large number of refracs attempted on each type

• ~90% success restimulating Sand or RCS completions

• <50% success restimulating wells initially completed with

Ceramic

– More durable proppants appear to delay or avoid the

need to restimulate this field

Increase Conductivity in Refracs?Dozens of examples in literature

Shaefer, 2006 – 17 years later,

tight gas

0

500

1000

1500

2000

2500

3000

3500

Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01

Ga

s R

ate

, M

CF

D

0

50

100

150

200

250

300

350

400

450

500

Wa

ter

Ra

te,

BW

PD

Gas

Water

Initial Frac in

1989:

48,000 lb 40/70

sand + 466,000

lb 12/20 sand

May 1999 Frac:

300,000 lb 20/40

LWC

May 1995 Frac:

5,000 lb 100 mesh

+ 24,000 lb 20/40

Sand

Vincent, 2002 – 9 years later,

CBM

0

500

1000

1500

2000

2500

3000

3500

4000

May-84 May-86 May-88 May-90 May-92 May-94 May-96 May-98 May-00

Date

Pro

du

cti

on

fro

m F

rac

ture

(b

fpd

) Original Fracture (20/40 Sand)

Phase I refrac (20/40 Sand)

Phase III refrac (16/20 LWC)

Incremental

Oil Exceeds

1,000,000

barrels

Incremental

Oil exceeds

650,000

barrels

First

Refrac

Second

Refrac

Pospisil, 1992 – 6 years later,

20 mD oil

0

500

1000

1500

2000

2500

Sta

biliz

ed

Ra

te (

MS

CF

D)

Pre Frac 10,000 gal

3% acid +

10,000 lb

glass beads

80,000 gal +

100,000 lb

20/40 sand

75,000 gal +

120,000 lb

20/40 ISP

Ennis, 1989 – sequential

refracs, tight gas

0

20

40

60

80

100

120

Well A Well B Well C Well D Well E

Pro

du

ctio

n R

ate

(to

nn

es/d

ay)

..

Initial Frac

Refrac

Dedurin, 2008, Volga-Urals

oil

30

13

Refrac Reorientation

Barnett - Cipolla, 2005

• Refrac Reorientation Documented in:

– Lost Hills Diatomite

– Austin Chalk

– M. Bakken Dolomite

– Codell, DJ Basin

– Barnett Shale

– Van oilfield

– Ansai oilfield

– Daqing oilfield

– Xin Zhan field

• Diversion/reorientation evidence:

– Mounds cuttings disposal site

– Black Warrior CBM

– Undisclosed locations

31

Cross Sectional View

0 100 200

Xf, ft

Core

Projections

2400

2500

2600

2700

2800

2900

-50 0 50 100 150 200 250 300 350

GR & Hor. Dist., ft

De

pth

Core#

4

Core#1

Fracture

Intersections

Core#3

GR and Hor. Dist.,

ft

Mounds Wilcox Interval Tiltmeter & µS Mapping

No hydraulic

fractures present

in Sidetrack #3,

confirming

vertical

containment.

-100

-75

-50

-25

0

25

50

0 25 50 75 100 125 150 175

Easting (feet)

No

rth

ing

(fe

et)

Wellheads

Perforation Locations

Wellbores

Core

Frac Intersections

Mapped Fractures

Average azimuth

N 71° W ± 11°

Sidetrack #1

Sidetrack #4

Note: Dips of vertical fracs are not

properly represented on this plan view.

Plan View

- 5 diagnostic injections and 17 cuttings slurry

injections

- Core #1 detected ~20 hydraulically induced fractures

- Core #3 detected no hydraulically induced fractures

- Core #4 detected ~9 hydraulically induced fractures

Sidetrack 4 ABOVE

fracture center

Case C-01 See also: SPE 63032

14

Mounds

Cuttings

Injection

Reference: 77553, after

Moschovidis, SPE 59115,

63032, 48987

• Core through verified multiple fracs in recovered core

• Core above Wilcox confirmed containment in mapped interval

Why Do Refracs Work? (mechanisms)

Refrac success (worldwide) has been attributed to:

• Enlarged frac (more reservoir contact)

– Improved pay coverage (add pay in vertical wells)

– Better lateral coverage (horizontal wells)

• Increased frac conductivity

– Restore conductivity lost – frac degradation

– Address unpropped/poorly propped portions

• Improve wellbore-to-frac connection/conductivity

• Reorientation

• Use of more suitable frac fluids

• Re-energizing natural fissures

• Other mechanisms

– Fracturing past condensate block, inducing complexity, rearranging

existing proppant pack, better containment of refrac, etc

15

Why Do Refracs Fail? (poorer candidates)

Refrac failure (worldwide) often attributed to:

• Low pressure

– depleted wells (limited gas reserves)

– poor recovery of frac fluids

• Inadequate reservoir quality

• Diagnostics indicated drainage to reservoir boundaries

• Undesirable existing perforations

• Poor mechanical integrity

• Failure to clean out well prior to restimulation

• Inadequate procedures to select refrac candidates

– Poor wells often make poorest refrac candidates

– Unless there was a failure in initial frac design or implementation

35

Additional Examples

Horizontal Well Refracs

16

Bakken Refracs (Lantz, 2007: 108117)• Montana, cemented laterals believed drilled for longitudinal growth

• Initial fracs ~300,000 lb 20/40 RCS

• RA tracer showed incomplete coverage

• Add perfs, refrac 600,000 lb 20/40 sand, 10 ppg slugs, ball sealersupper track of Figure 4.

The effect of depletion was observed in the treating records, as breakdown pressures averaged 669 psi lower than initial

treatments. However, net pressure generated during the refracs was approximately 50% greater than during the initial fracs.

Figure 4 – Tracer logs in horizontal well indicate coverage was increased by refrac. [Figure adapted from Lantz, 2007]

Tracer from original frac▲

▼Tracer from refrac

• Lower breakdown pressures, but 50% greater net pressure

• Oil rates up, GORs down

• Refracs diverted into undrained areas of the reservoir

• EUR increased 1,300,000 bbls in 16 refractured wellbores

• 1 mechanical problem, but 16 of 17 successful (94%)

Observations:

Can refrac into new rock from

cemented wellbores

37

Bakken Refracs (Dunek, 2009: 115826)• Uncemented liner. N-S oriented lateral. Surface tiltmeter mapping

• Unintentional termination of frac job (wellhead isolation tool failure)

– 3892 bbl crosslinked fluid and 296,000 lb proppant at 48 bpm

• 2nd stimulation treatment 6 weeks later

– 6533 bbl slickwater and 193,000 lb proppant at 61 bpm

Transverse: 45% heel

Longitudinal: 30%

Horizontal: 10%

Oblique: 15%

Transverse: 45% toe, mid

Longitudinal: 35%

Horizontal: 20%

Observations:

Diversion/ Initiation at new locations

Both Transverse and Longitudinal Growth

17

Bakken Refracs (Eberhard, 2008: WBPC)

Cemented laterals initially treated with sand/RCS

• Reperf/jet new entry points

• Diversion slugs

• 100% success rate– 30 days prior – 46 bopd

– 30 days post refrac – 144 bopd

Uncemented Liners initially treated with sand/RCS

• Retreat existing perfs

• 87% success rate (27 wells)– 30 days prior – 64 bopd

– 30 days post refrac – 122 bopd

Observations:

Adding entry points helps.

Initial cemented laterals failed to drain

all rock

Bakken wells frac’ed with sand usually

benefit from refrac (~90% of time)

39

Bakken Refracs (Besler 2007/2008,

SPE 110679 + 2008 WBPC)

Bakken wells initially stimulated with ceramic

• Only ~50% success rate with refracs using ceramic– Refrac success appears to correlate with wells in which RA tracer indicated

poor diversion

– In pre-refrac cleanouts, only minimal quantities of ceramic proppant

recovered from wellbore

Observations:

Less need to refrac if wells treated with ceramic?

Perhaps due to reduced proppant flowback or better durability?

40

18

Bakken Refracs (Phillips 2007 SPE

108045)

Bakken wells initially stimulated with ceramic and

XLG

• Showed 1st month benefit refracturing with sand/slickwater

41

Bakken Refracs (Operator B)

• 3 Wells initially stimulated with 20/40 ceramic in XLG

• Restimulated 13-39 months later with >200,000 lb 30/50 or

40/70 Ottawa (much information unavailable via public data)

• 2 of 3 wells subsequently TA, but did give short term benefit.

0

20000

40000

60000

80000

100000

120000

0 12 24 36 48

Cu

mu

lati

ve O

il P

rod

uct

ion

, Bar

rels

Months on Production

Well B-1

Well B-2

Well B-3

Refrac

Refrac

Abandon

Trifrac

Abandon

Refrac

Sand refracs of initial ceramic

completions have been

successful, but 2 of 3

abandoned?

19

0

10000

20000

30000

40000

50000

60000

0 10 20

Cu

mu

lati

ve O

il P

rod

uct

ion

, Bar

rels


Well C-1

Well C-2

Refrac

Refrac

Bakken Refracs (Operator C)

• 2 wells initially stimulated with sand > 1,000,000 lbs (staged)

• Restimulated 8-11 months later, slickwater with 100,000 lb

30/50 lightweight ceramic (bullhead; overflushed)

Limited data, but:

100,000 lb LDC refracs in

slickwater diagnostic plots

similar/superior to initial fracs

0.00001

0.00010

0.00100

0.01000

0 0.5 1 1.5 2 2.5 3 3.5

1/B

OP

M

Square Root of Months on Production

Well C-1

Well C-1 Refrac

Well C-2

Well C-2 Refrac

Saskatchewan Bakken Refracs

Vincent, 2010 - 136757

• 9 refracs identified in Viewfield area

• 8 initially completed barefoot with 7-8 stages coiled tubing

• 8 recompleted uncemented, ball activated sliding sleeves

• 1 recompleted-cemented liner, abrasive jet, annular refrac

• All 9 show oil increase. Watercut down in 7 of 9 refracs!

0

200

400

600

800

1000

1200

1400

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50

GO

R (m

cfd

/bb

l)

BO

PD

an

d W

ater

cut

%


BOPD

BOPD normalized for downtime

Watercut

GORRefrac

Observations on this well• Oil increased from ~25 to

~140 bopd

• Water decreased from ~250

to <100 bwpd!

• GOR temporarily dropped

44

20

Restimulation of Gas WellsDick Leonard, ProTechnics

Expanding Shale Plays – 2009. PNR Energy

Forum, Brookhaven College May 28, 2009

Barnett Shale

Refracture Treatment

• Min/Max stress are very similar in the Barnett.

• There are now over 4000 traced HZ wells in the Barnett,

about 50% show anomalies (unstimulated perfs, cement integrity,

casing problems)

• “Refracs are, and will become, critical to these shale plays”.

It appears that we are not draining very far past the

fractures.

Refracturing of BarnettDick Leonard

Tri

21


Barnett Refrac

SPE 131783 Curry, Maloney (Range Resources)

• Denbury Winston A3 well:

– Original Completion:

– 3000 ft cemented lateral

– Original perfs spaced at 450 ft

– Restimulated 2007

– Added intermediate perforations as shown on next page

– Two stage refrac, placing 855klb proppant

– Stage 1: 240 klb 40/70 + 290 klb 20/40

– Stage 2: 220 klb 40/70 + 105 klb 20/40

– Slickwater at 102 bpm

– RA tracer shows diversion of treatment into new areas

– Incremental EUR 1.0 BCF

53

22


Green bars

denote new

perf clusters

evenly

spaced

between old

existing

perfs. Note

that on this

stage, the

new perfs

took most of

the refrac

Apparent

stress

diversion.

This is one

of the best

cases

This well

was

identified as

the Denbury

Winston A3

well in SPE

131783


A packer

was set to

isolate the

lower (red)

stage.

It appears

the middle

green new

perfs took

proppant,

but

upper/lower

did not?

This well

was

identified as

the Denbury

Winston A3

well in SPE

131783

23


Marcellus Refrac - Public


• Range Nancy Stewart 4H well:

– Original Completion:

– 3000 ft cemented lateral

– 100 ft perf spacing

– 8 stages, placing 1.9 mmlb 100 mesh + 1.1 mmlb 30/50 sand

– Crosslinked gel & slickwater used at average of 54 bpm

– Restimulated Aug 2008

– Added “intermediate perforations”

– Single stage refrac, placing 880klb 100 mesh + 390klb 30/50

– Slickwater at 131 bpm

– RA tracer shows diversion of treatment into new areas


57

24

Marcellus Refrac


• Range Nancy Stewart 4H well:


58

New perfs appeared to preferentially

receive RAT in refrac

Refracturing of

Marcellus

Buddy Woodruff, ProTechnics

SPE ATW, Banff, May 3 2011

Original FG = 0.99 psi/ft

Refrac FG 1.05 psi/ft

Orig EUR 1.7 BCF

Post ReFrac EUR 2.5 BCF

25

– I am aware of 7 refracs performed in EF

• Swift

• Pioneer

• Results are not yet disclosed

• Also in Eagle Ford, hesitation fracs have been shown by

some operators to have induced diversion via

microseismic mapping

Eagle Ford


• Relax-a-frac? SPE 136873 Eagle Ford

Eagle Ford Relax-a-Frac

26


• Relax-a-frac? SPE 136873 Eagle Ford

Eagle Ford Relax-a-Frac

Successful refracs have been

performed in Barnett, Woodford,

Eagle Ford, Bakken, Marcellus,

Haynesville, Niobrara, Spraberry,

Wolfcamp…

What did we miss the first time?

27

– Standard models fail to predict refrac performance

• Multiple mechanisms

• Models fail to recognize compartmentalization, degradation, realistic

conductivity, etc.

– Artificial Intelligence, Neural Nets

• Excellent success in some fields (Shelley, 1999)

• Dismal failure at predicting success in others

• Some refracs have worked when all diagnostics said they shouldn’t!

– Reorientation

• Beneficial when it occurs (except in some waterfloods)

• Is not necessary for refrac success in many fields

– Refracs have worked in most reservoir types

• And there are examples of failures in most, also!

– Well design

• We need wells that can be cost-effectively isolated and restimulated

Comments

65

– No “magic bullet” was apparent. However, most successful refracs have

incorporated:

• Larger proppant mass

• Improved proppant quality

• Higher proppant concentrations (if placed with crosslinked fluids)

– Adding perfs (if pay was missed) is valuable. Diversion!

– Proppant durability is an issue

– Some refracs clearly touch new rock

• Some refracs do not, yet are still economic

– Good wells are often the most economic refrac candidates!

• Refrac “dogs” if design or implementation failure on initial frac

– Even the best wells can be improved! (corroborates SPE 119143)

• Non-optimal initial fracs, proppant degradation, or reorientation “gift”

– Refrac candidate selection takes effort

• Not as “routine” as drill & complete

• Outstanding economic opportunities available

Some Interpretations 134330 & 136757

66

28

Human beings, who are almost unique in

having the ability to learn from others,

are also remarkable for their apparent

disinclination to do so.

Pertinent Quotation

Douglas Adams (1952-2001)

English writer

Refracs: What can we decipher? - Mike Vincent

News & Politics

Transcript of Refracs: What can we decipher? - Mike Vincent