SPE 25892 Society of Petroleum Engineers

Field Implementation of Proppant Slugs To Avoid Premature

Screen-Out of Hydraulic Fractures With Adequate Proppant
Concentration
M.P. Cleary: Massachusetts Inst. of Technology; D.E. Johnson: Resources Engineering
Systems Inc.; H-H. Kogsbllli and K.A. Owens, * Maersk Olie & Gas A.S.; K.F. Perry, * Gas
Research Inst.; C.J. de Pater, * Delft U. of Technology; Alfred Stachel* and Holger Schmidt, *
RWE-DEA Aktiengesellschaft flir Mineraloel & Chemie; and Mauro Tambini, * AGIP SpA
'SPE Members

Copyright 1993, Society of Petroleum Engineers, Inc.

This paper was prepared for presentation at the SPE Rocky Mountain Regional/Low Permeability Reservoirs Symposium held in Denver, CO, U.S.A., April 12-14, 1993.

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. Permission to copy is restric'ed to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment
of where and by whom the paper is presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A. Telex, 163245 SPEUT.

ABSTRACT INTRODUCTION
Premature screen-outs and/or low proppant concentration are The technology of hydraulic fracturing has been undergoing
the most likely cause of failure in hydraulic fracturing treatments. a drastic re-evaluation of its procedures and appropriate
Although commonly blamed on a variety of presumed applications, especially over the past five years 1- S• This effort has
problems-most typically the treating fluid, or large-scale arisen mainly from two converging causes: (1) the application to
reservoir conditions, such as permeability or stress profile-the more complex environments, and (2) the availability to
true source of most problems has been uncovered only recently operational personnel of on-site instrumentation and real-time
by careful analysis of treatment data. The source is referred to as analysis capabilities 1• The result of this effort has been the dis-
near-wellbore tortuosity, but it can variously arise from deviatoric covery of many new important aspects, which greatly affect the
stress, natural fractures and/or perforation-dominated creation execution of field operations. Some of these issues have been
of complex fracture patterns in the wellbore vicinity. discussed in previous papers2-4, but many of our case studies
Numerous theories have been formulated to deal with near- remain to be published s and review of the literature shows that
wellbore screen-outs and, especially for oriented wellbores from much clarification is still needed to achieve optimal execution.
Arctic or offshore platforms, various perforation strategies have This paper will focus on one such major issue, namely the
been postulated and/or implemented. In contrast to the ideal- problem of near-wellbore tortuosity; this may be the most
izations and costs associated with those theories and strategies, common and most poorly-understood of all aspects in fracturing:
this paper presents simple cheap solutions that are less sensitive failure to understand the cause and effects of tortuosity have
to the wellbore environment This novel strategy involves generated multiple erroneous strategies and conclusions about
injection of proppant slugs into the near-wellbore region and, fracturing procedures, which must now be reversed in order to
when necessary, immediate shut-ins upon small slugs, with three execute more appropriate fracture treatments2-4: most harmful
important results: the response of the near well-bore region can among these, at least for low-permeability thin payzones, has
be measured and characterized; a large part of the near-wellbore been the use of large fluid volumes and low proppant
tortuosity can be removed, by simplifying the near-wellbore concentrations to avoid screen-out. Indeed, the process of
fracture pattern; and the true nature of the large-scale reservoir remedying tortuosity with proppant slugs has led to some other
response can be determined, e.g. from the greatly modified major new discoveries in relation to pressure fall-off behavior in
pressure fall-off obtained after placing slugs near the wellbore. lower-permeability formations 4• These and other such
The paper reports the concept and implementation, in a consequences of our data analysis, and their implications for
number of commercial fracturing environments, in both gas and revised field procedures are discussed elsewhere 1- s. Examples of
oil reservoirs, with both foam and liquid-gel jobs. These show special "problem wells" are presented here to display the
the effective removal of tortuosity varying from 20 to 200 bars diversity of environments and associated tortuosity behavior, for
and associated elevation of allowable proppant concentrations. which we have implemented the proppant-slug methodology.

493
 FIELD IMPLEMENTATION OF PROPPANT SLUGS TO AVOID PREMATURE SCREEN-OUT
2 OF HYDRAUUC FRACTURES WITH ADEQUATE PROPPANT CONCENTRATION SPE 25892

Simple Phenomenological Model of Tortuosity

The phenomenon of tortuosity, in our adopted terminology, is

that of a convoluted pathway connecting the wellbore to the main
!!Pr '" PW-P F'" #(-Z) - _]2n.l
- Q /N
hL r
n I
NE
P wr
,p '" P -P wc (2a)

body of the fracture(s) further away from the wellbore.

Schematics of the concept are shown in Figs. 1 and 2, but these with a resulting opening of the tortuous region given by
represent the process only in a simple conceptual way, which
may (be expected to) have an infinitely-variable form. However, wr~ Pwr Lr / iN; Pwr ~(C1PW + CZPF ) (2b)
the result is a major effect on wellbore pressure during
fracturing z. The causes of near-wellbore tortuosity may also be Note that Pwc is the closure stress at the wellbore, which may
(expected to be) many and variable, as we discuss later (e.g. in (will) not be the same as the far-field closure stress Pc. Of
the context of perforation strategy) but we group them, for course, the geometry hL/ really represents a much more complex
convenience, into two sources: combination of the phenomonology identified earlier in I, II, but
the formula makes a number of key points (See Fig. 3(b»:
I. Multiple fractures competing for opening space in the same (1) The tortuosity pressure drop !:J.pr depends on the flow-rate Q
region. Such parallel or sub-parallel (en echelon) fractures
will each individually have less opening than a single (2c)
fracture with the same pressure and/or flow-rate. Detailed
formulae for this process are provided in Ref. 2 (Eqns. with IX varying from linear to (2n + 2)/n-root as (!!Pr / PF) goes
(2.5», for N fractures, but the essential result is that the from zero to 00. Regardless of such details, the result allows us
individual fracture widths are reduced by some (almost to distinguish between the near-wellbore friction dictated by Eqn.
linear) power of N, relative to a single fracture. (2a), and stagnation pressure at perforations (where IX .. 2), as
II. Curvature (turning) of the pathway leading from the illustrated later in Figs. 6(b), 7(b) and lO(b). This recognition
wellbore to the main fracture(s). Such re-orientation leaves alone may prevent many re-perforationlball-out operations.
a segment near the wellbore which is not supported
(2) The tortuosity !!Pr also depends, even more strongly in Cr.
(mechanically) by/like the main body of the fracture-hence
on the Young's modulus of the rock, E through the formula
it will have less opening in general and may often close
completely when injection ceases, somewhat like a valve. A for "crack opening modulus" E '" E/4(I-v 2) .. E/4.This
simplified formula may be concocted to represent the width has severe implications for tortuosity problems in hard rock.
Ww at the wellbore, in terms of the far-field large-scale width (3) The tortuosity depends on effective fluid viscosity ~, in a
wF of the fracture, using the angle 4> in Fig. 1(a): similar way as to flow-rate «1), with (2n + 2)-root). This, in
tum, has a number of significant sub-implications:
(1)

a. More viscous fluids will show "gel kicks" as they enter

in which the function f goes from 1 (for 4>=0) to 0 at the near-well bore. This has often been confused with the
(q,:4>ait) for zero flowrate Q=O; 4>ait depends on the net behavior in the main fracture, e.g. as expected from
pressure and bias in the stress field, as well as the geometry conventional rheology models,which our review of data
(g) of the tortuous region, but it will be less than 1t/2. has also dis-established z.
b. More viscous gels will tend to produce wider near-
Although we have provided simple "back-of-the-envelope" wellbore fractures, as indicated by Eqn. (2). This may
formulae to allow data-matching for these processes, we have help to explain the established beliefs in using cross-
little hope of practically applying detailed modelling to explain linked gels-which may also display a visco-elastic
our observations: detailed models, which we could readily base effect and decrease the propensity for multiple fracture
on our unique 3D fracture analysis capabilities 6, should be formation, allowing smaller N in Eqn. (2b).
applied only to deterministic processes7 , whereas qualitative
models are more appropriate/adequate for more stochastic events. These main features of tortuosity behavior have been clearly
Such a simple qualitative model, for the flow-induced opening Wr confirmed by laboratory experiments 8• However, that laboratory
of the tortuous region, may be based on the schematic shown in environment also displays substantial frictional effects in the
Fig. 3(a), in which the fluid (effective viscosity3 ~) is flowing fracture, corresponding to essentially clastic behavior of the
through the near-wellbore geometry characterized by N fractures surrounding solid (unlike rock at depth); from careful field-data
with height h and length Lr. under pressure Pw at the wellbore, analysis we have deduced that friction in the fracture is
producing a pressure drop !:J.pr to PF in the main fracture: negligible for many reservoir environments Z, leaving tortuosity as
the (only) major source of rheology-dependence.

494
SPE 25892 M.P. CLEARY, D.E. JOHNSON, H.H. KOGSB0LL, K.A. OWENS, CJ. De PATER 3
K.F. PERRY, A. STACHEL, H. SCHMIDT & M. TAM BINI

Fundamental Mechanics of Near-WeUbore Fractures been advanced for under-balanced 12, vs. over-balanced lS
perforating, but the dynamic breakdown (crack initiation)
Although the details of processes involved are not yet fully should be independent of this: hydraulic follow-through
determined, there are clear fundamental relevant principles which (overbalanced) should bias re-initiation toward fewer
can be enunciated here, in light of results from our laboratory fractures, with shear-induced favorable conductivity (VII).
and field work, supported by careful modelling activities: VI. The dynamic perforation breakdown process should be
I. Multiple underground fractures tend to avoid each other, rather random, i.e. dominated by rock microstructure,
rather than link up, acting only to compete for propagation flaws etc. in the near-wellbore region-including, perhaps,
length; one well-known consequence is the observation of the effects of prior drilling processes. Therefore, the
en echelon fracturing in outcrops. This is in stark contrast cracks available for early (initial) propagation away from
to observations of fractures under tensile conditions, where the perforations should be relatively independent of the
a strong tendency to link-up may exist-especially under stress field around the wellbore-except insofar as those
dynamic conditions. As a result of extensive careful are affected by relevant drilling and perforating damage .
experimentation and analysis" 8, we have concluded that VII. The propagation of cracks away from perforations (and/or
fractures emanating from adjacent perforations will tend to openhole well bores) is dominated by hydraulic
avoid each other, almost regardless of orientation, unless conductivity i.e. the ability for pressurizing fluid to
some other connecting influence is present: our simul- penetrate such cracks over distances comparable to (or
ations' show that very few favorable orientations exist in grater than) the diameter of the relevant hole. The
general, and not enough for reliable fracture linkage9.1O. mechanics of this issue were discussed extensively by the
II. An openhole completion should, in general, introduce the first authorll , emphasizing the dominant role of fluid
"common attractor" required for linking of mUltiple pressure access to the crack(s) being created.
initiation sites: the negative stress concentration is VIII. The role of stresses, in light of V, VI, VII above, may be
removed, since the borehole walls are pressurized, and very different from that conventionally assumed, (e.g. as
fluid has direct access to each available fracture site. reviewed in Ref. 16 and in list of references/contradicted
III. A well-cemented cased borehole acts as a repellent to in/by Ref. 11]: for instance, the cracks created in V and
fractures, discouraging the linking up of fractures VI may have most hydraulic conductivity when oriented
emanating from perforations: the responsible mechanism in the direction of maximum shear (producing offsets
is the higher confining stress induced by presence of the between initially mating surfaces). Although the distri-
wellbore, which is little relieved when the casing is bution of initial crack orientations will be somewhat
pressurized-e.g. as represented by the following formula: random, this feature tends to bias the re-opening process
toward the (cj> =) 45° direction; this is converse to what is
(3)
assumed in ideal rock mechanics theories-which are still
commonly used, e.g. for microfrac tests to determine
for the pressure transmitted to the rock, PR' through a pipe stresses. The effect will be less in openhole completions,
of thickness t in a well bore of radius R. (This is the but may still exist due to natural fractures; in such cases,
maximum value, for transmitting cement.) This effect was the consequence may be an erroneous estimate of closure
clearly shown by experimental and modelling work" 8, stress and fracture orientation (as well as fracture aperture)
although there is some disagreement on this matter [Refs. from wellbore measurements and traces.
12, 13, as discussed partially in Ref. 14]: deviations from IX. Pore-pressure may be expected to play a role in
such behavior may be expected if the cement seal is not determining the near-wellbore fracture trajectory; besides
effective and/or cement blocks are cast around a "rigid" affecting confining stresses 14 , it may enhance the bias
pipe, with Eqn. (3) working in reverse. toward induction of shear-slippage offsets and associated
IY. Fractures should tend to emanate from the ends of dominance of conductivity, thus increasing the tortuosity.
perforations, based simply on the lower confining stresses X. Fractures with a large advantage in hydraulic conductivity
away from the (shadow of the) wellbore. This has been may be able to overwhelm smaller fractures (eventually,
observed in laboratory experiments [e.g. related to Ref. or even before they get established), e.g. as calculated in
14], and would conspire with III (above) to further prevent Ref. 7. This process will decide how many fractures
linking up of fractures from adjacent perforations. emanate from adjacent perforations and what the
V. However, it seems that one or more perforations, at least, separation will be between perforations taking fluid. The
would be broken down during the shooting process: the proppant slug technique exploits this observation of
associated high pressures/stresses suggest this and the hydraulic conductivity dominance. Absence of the slugs
commonly low hydraulic breakdown pressures (relative to may require injection of much larger (pad) volumes to
theoretical values for breakdown) support such a reduce tortuosity and even this may not work consistently,
conclusion (see also Ref. 11). Various arguments have even where allowable «2)B below).

495
 FIELD IMPLEMENTATION OF PROPPANT SLUGS TO AVOID PREMATURE SCREEN-OUT
4 OF HYDRAUliC FRACTURES WIlli ADEQUATE PROPPANT CONCENTRATION SPE 25892

Case Studies of Tortuosity and Its Effects brief placement of the previous slug in the near-wellbore
region, as we discuss later in (1)B, (2)A. Unfortunately, this
Since we have properly understood the nature and role of second slug was completely ftushed-partly because of fears
tortuosity only in the past few years, there is a limited database associated with pressure rise on the first slug displacement.
of adequate data to quantify the various effects just discussed Consequently, the proppant stages of the main fracture
over a broad enough range of wells to constitute a statistical also elicit clear near-wellbore pressure responses (Fig. 4(c»,
sample for each of a reasonable range of rock types. However, which increase with proppant concentration until the pressure
such a correlation study (e.g. of the kind outlined in Refs. 10, limit (i.e. a screen-out) is rapidly reached. However, it is
17-19) has been initiated for data-sets obtained on one of the clear that well-defined-aperture fractures, of the kind
rock-types we discuss here14• Our other examples illustrate the depicted in Fig. 4(a), should have a more dramatic response
variation with rock types and wellbore geometries. to early (18/20) proppant-based on the 20/40 response: we
began to speculate that earlier slugs had changed the
(1) Vertical Gas Wells in Siliceous Formation geometry somewhat, but the concepts and data in this paper
tend to undermine, at least for hard rock, the
We start with case studies for vertical wells, to emphasize "perforation/rock erosion" ideas sometimes used to explain
two points for routine field operations (e.g. onshore U.S.): such observations20.
• Tortuosity is not limited to deviated wells (or even to Unfortunately, none of our other theories were tested at
tectonic and/or naturally-fractured environments) that time; this was understandable, in light of the frustration
• Tortuosity effects more than near-wellbore screen-out: that was experienced on this particular well. Two treatments
especially, it may greatly affect pressure fall-off. had previously been conducted (in the same well) on the
Lower Travis Peak. An erroneous match of (albeit limited)
Although we have observed tortuosity effects on numerous tortuosity as high leak-off in the mini-frac, Fig. 5(a), led to
(nominally) vertical wells in quiet (non-tectonic) sedimentary the use of lOO-mesh sand, to block off natural fractures. The
basins" s, we had not prevailed on operators to (properly) 1OO-mesh gave rise to substantial kicks ahead of the main
implement proppant slugs, i.e. in the manner that we wished to fracture treatment (Fig. 5(b», which was then displaced by
employ them, as against their more incidental use20, which we an injection test (Fig. 5(c» and followed by an attempted
describe first (A below). It was purely fortuitous (including re-frac (Fig. 5(d» at higher rates and gel concentrations.
environmental and economic considerations) that we had our first These latter efforts finally convinced us that many of the
opportunity to properly execute such slugs, without too much conventional views of fracturing were wrong-partly
interference and with support from operating and service because the associated model scenarios needed to match the
company personnel, on overseas jobs. We are now working with data 20 were not realistic: for instance, we came to discredit2. 3
numerous companies, implementing the technique in the U.S. the presumed role of rheology and large pads which led to
the efforts at re-fracturing the lower zone in this well. But
A. East Texas Tight Gas Well Cfravis Peak) we were misled by the role of natural fractures, one of
Our early involvement with proppant slugs was possible which had (remarkably) been found cemented open in core
only because of a conceprl that such slugs could be used to (Vol. 1 of SFE #2, Ref. 20). We credited the leak-off
block downward growth offractures-a concept that has mechanism vs. the associated creation of near-wellbore
been extensively employed by one service company22. Since tortuosity, which we now believe to have dominated those
screen-outs had been observed frequently in this region and jobs and many other screen-outs in that area.
since the environment was that shown in Fig. 4(a), it was This experience with screen-outs has led to many
postulated by the main project participants20 that the problem extremely conservative (and often harmful) practices, as we
was associated with biased growth of the fracture, i.e. have outlined in Ref. 3: there is an unwillingness to pump
injection into the top or bottom narrow comer of a fracture high concentrations (partly motivated by low theoretical
(being created in the Upper or Lower Travis Peak). optima for prop concentration in low-permeability reservoirs)
The response to this "divertafrac" treatment is shown in and use of large pads has ruled field practice, until recently
Fig. 4(b); the problem is clearly near-wellbore, with a at least. Blame for failed treatments is laid on many factors,
pressure kick that eventually disappears as the (20/40 mesh) such as rheology, additives, proppant crushing etc., or often
slug clears the near-wellbore; however, a short shut-in had even ascribed to a "poor producibility" reservoir (kH),
occurred before this (- 20 kJb) slug had fully cleared the depending on the disposition of the advocate. In fact, after
perforations. After clearing this slug, a large (40 klb) slug of careful analysis of commonly-discarded "problem well"
4OnO mesh was pumped, without any pressure kick as it data-sets, the factors responsible are now thought to be much
passed through the perforation/near-wellbore region (note more basic and their resolution allows simplification of job
downhole gauge error): this was ascribed to the smaller schedules, with associated cost reductionsZ-s, 23, 24 and
mesh size at the time20, but it was more likely due to the substantial improvement in well performance.

496
SPE 25892 M.P. CLEARY, D.E. JOHNSON, H.H. KOGSB0LL, K.A. OWENS, CJ. De PATER 5
K.F. PERRY, A. STACHEL,H. SCHMIDT & M. TAM BINI

B. Low Pressure Gas Formation, Onshore Italy (2) Deviated Oil Wells in Siliceous and Carbonate Formations
One of the additional common practices that we have Deviated wells increase the probability and severity of
found is that of ascribing production problems to water- tortuosity problems in at least two possible ways:
sensitivity and associated clean-up problems. Based on this, (i) by increasing the likelihood of generating fractures
many operators have pumped (C0 2 or N 2) foam jobs, with turning into the preferred far-field direction (e.g. Fig. 2);
varying degrees of successlS which our own experience is (ii) the potential for link-up between fractures, emanating
causing us to increasingly question. Such a (NJ job was from various perforations, may also be reduced (but not
recommended by the service company to the operators of greatly, based on results in Refs. 7, 8, in contrast to
this well (see also Ref. 23 for more details). theories like Ref. 9); indeed higher pressures associated
This environment was also found to be tectonic, with with (i) may generate more fractures from more
high associated break-down pressures. The mini-frac testing perforations. Only one good set of (qualitative)
was conducted on the morning of the job, as indicated in correlations have been made, for some "low-perm"
Fig. 6(a). The first and second injections were performed North Sea gas wells 18, but our most detailed data-sets
without proppant slugs-which were not firmly scheduled have been acquired for a number of oil wells·-from
for this treatment, because tortuosity was not expected to be which, of course, the experiences can be (and are being)
a problem at the low bottomhole proppant concentration (of transposed directly to gas wells.
order 5ppg) that would be achieved at the N 2-quality (75%)
designed in the schedule. However, a match of the pressure A. Medium Permeability Oil Wells, Onshore Germany
fall-off data (Fig. 6(a» would have required a permeability A unique opportunity to test our theories in full became
of order micro-darcies, whereas the reservoir was expected available at a site in northern Germany where the operator
to be of order O.lmD. The second (step-rate) injection test (RWE-DEA) had experienced two premature screen-outs on
was used to test for the nature of perforation/near-wellbore a number of wells. This (siliceous) environment was subject
friction, as indicated by the permanent downhole gauge: the to tectonics, generated by salt intrusions and associated
associated plot (Fig. 6(b» clearly showed the tortuosity. warping of the surrounding sedimentary strata: the wells in
To test our associated theory that the fall-off did not question (preetz 26a and 28) were directly adjacent to the
reflect the true fracture behavior, because of a closed/sealed Salzstock (salt-dykes), resulting in extreme distortions of the
near-wellbore region, we chose to place (first 1 ppga, then stress field and extensive natural fracturing. Also, they were
1 & 3 ppga) proppant slugs, as shown for the successive drilled diverted (from adjacent pads, to appropriate reservoir
three injections of Fig. 6(a). A match of the (relatively much locations under a lake), leading to angles of order 10-15° at
faster) fall-off was achieved with (of order) 0.1 mD for the perforation sets. A further adverse factor (typical of old
liquid stages 3 & 4. However, the fall-off after the fifth drilling designs) was the negligible rat-hole below the
(foam) stage is again much slower-mainly due to low leak- perforations, into which non-passing proppant could fall.
off and pressure maintenance characteristics of the foam: this The wells had been on production for about twenty
complicates placement of a proppant pack opposite pay3. years and had become marginally economical for further
Based on the leak-off observed, and recognizing the operation. Matching of production gave a permeability of
high apparent efficiency of the foam, we designed (on-site) order 10 mD over a pay-zone height of order 30 ft., so that
a main fracture treatment about six times as aggressive as an effective fracture radius of order 100 ft. was determined
that originally proposed: our aggressiveness ratio is defined to be adequate for substantial increases in production: this
as the ratio of proppant to (pad) fluid volume pumped. In conclusion was supported by production increases of order
addition, we recommended a decrease in foam quality at the 3-5 on previous jobs (preetz 37 and Plon-Ost 29), which had
end of the job, both to increase the aggressiveness ratio and screened-out prematurely at estimated fracture radii of 25 ft.
to avoid excess uphole concentrations associated with There were, however, two substantial risks that we
increased bottomhole concentrations. We successfully immediately associated with this particular undertaking, one
predicted the job response, as represented by agreement with technical and the other related to cost-tradeoff aspects:
measured pressure (Fig. 6(c». (i) A negligible "rat-hole" was available below the zone to
Various problems developed, after this job, mainly be fractured. Although this situation existed in the
associated with the use of N 2• Although the job was Italian job [(I)B and Ref. 23], due to a sand-plug cover-
successfully designed/executed and production is reasonable, ing open perforations below, we expected much greater
problems (and cost) associated with foam jobs do not seem tortuosity here-with the attendant risk of covering the
to be balanced by benefits of clean-up: very little liquid load perforations if/when proppant (slugs) would fail to pass.
was returned on this job, whereas we have found much more This problem could be resolved with coiled tubing.
rapid clean-up on liquid jobs with good packs-after (ii) The greater risk arose from the question of the fracture
convincing operators to switch from foam to liquid jobs, geometry being created. It had been assumed, as is still
even on low pressure "water-sensitive" reservoirs. all-too-commonly done in the industry, that the fracture

497
 FIELD IMPLEMENTATION OF PROPPANT SLUGS TO AVOID PREMATURE SCREEN-OUT
6 OF HYDRAUUC FRACTURES WIlli ADEQUATE PROPPANT CONCENTRATION SPE 25892

was being contained in the pay-zone. Prior drawdown, 6 ppga slug was pumped as part of the main fracture pad
with associated reduction of closure stress in the pay- (Fig. 7(d», engendering a 650 psi rise in the bottomhole
zone, along with pay depth (8,000 ft.), supported this pressure (BHP, as confirmed later from the memory gauge).
assumption; and data from previous jobs were BHP had to be calculated on-site from the treating pressure,
inadequate to refute or support the theory (although they and it was difficult to determine certainly if the gradual rise
suggested containment). While we recognized the risk in treating pressure was a near-wellbore or pipe effect; in
(e.g. of an uncontained vertical fracture), we agreed to retrospect, both gauges showed a consistent rise, which
proppant volumes and a schedule like that proposed by clearly shows the proppant slug arriving 2-3 mins. ahead of
the service company, with the understanding that on-site the anticipated transit time (- 9 min.), as calculated
evaluation would generate final decisions. Extra (non- precisely by the real-time system. Both aspects confirm the
returnable) stand-by proppant was not ordered, over-displacement of the previous 3 ppga slug, allowing
unfortunately, since cost considerations were substantial tortuosity to re-develop, and leading to a premature screen-
factors in decision-making for this job. out when the 7.5 ppga (900 gil) stage hit the perforations
during the subsequent proppant stages: by this time, we had
All of these risks were, then, exacerbated by the called an early flush (with only 3k1b of proppant left in
extremely short injection times allowable on these jobs, to hoppers), using a high concentration (10 ppga) stage to
avoid perceived water-damage potential. The overall guarantee a near-well bore screen-out, i.e. try to retain
situation was, therefore, extremely unattractive as a characteristics of the prior jobs (preetz 37 and Plon-Ost 29).
commercial proposition-i.e. we foresaw little potential gain After this job, we had an even greater concern about the
over previous jobs-but it was an opportunity for technical location of most of the proppant: our analysis of shut-ins for
progress; this viewpoint was borne out by our experience. increasing volumes (in the minifracs) had strongly suggested
The first well (preetz 26a) was broken down rapidly, vertical uncontained fracture(s) and the pad volumes injected
using residual wellbore fluid, as shown in Fig. 7(a), and the (including mini-fracs) could serve only to provide a "bath",
near-wellbore behavior was measured by the bottomhole down into which proppant might easily convece. Although
gauge, as shown by Fig. 7(b): this clearly shows the strong a longer period of proppant injection (at less than allowable
tortuosity characteristic (much greater than Fig. 6(b), as 7.5 ppga) might have served to "fill up the bath" until
predicted by Eqn. (2». As a result, break-down and re-open- enough proppant was opposite pay, we did not have enough
ing pressures are extremely high-roughly corresponding proppant on-site to do this-and certainly not enough in the
with "lifting" pressures of the formation at that depth; of hoppers-especially in light of the needs for next job (preetz
course, the strata have already been "lifted" by salt intrusions 28). As a result, the job on Preetz 28 (Fig. 8) was similar to
at this depth, so these pressures may be above those required that on Preetz 26a, except for slow displacement of well bore
for opening (lateral sub-horizontal) fractures; however, near- fluid, which merely exacerbated the near-wellbore tortuosity
wellbore fractures may also/alternately be vertical. In any problem and was not greatly remedied by subsequent rapid
case (near-wellbore) closure stresses are estimated around "breakdown" with cross-linked gel.
6,250 psi from the breakdown shut-in of Fig. 7(c). Further Despite having placed more than twice as much
injections did not seem to greatly change this picture until a propp ant (and higher concentrations) in the formation for
1 ppga slug was pumped (around 160 min. from start of those wells (compared to Preetz 37 and PlOn-Ost 29), their
time-base) and shut-in: the major break in the pressure fall- production has not been so good. There are a number of
off had now shifted down to 5,600 psi (approx.), perhaps possible explanations, but the most likely is that most of the
suggesting that we now see the actual closure of the primary propp ant has been placed out of zone, as indicated by Fig.
fracture and that Fig. 7(c) represents near-wellbore closure. 8(b), i.e. not enough was pumped to fill up the "bath".
A second effect of the 1 ppga slug stage, when Multiple fractures on previous wells probably served to trap
re-opened and displaced away from the near-wellbore region, proppant opposite perforations: although we could pump less
is the absence of a kick during the following 1 ppga and 3 proppant and accept a screen-out, as on Preetz 37 and Plon-
ppga stage (just after 270 mins.), with the consequence that Ost 29, our final conclusion was to pump more proppant,
those may have been over-displaced-as suggested by the after fixing the tortuosity problem, until "the bath is filled.·
associated fall-off, where only the closure around (above) The lesson of Fig. 8(b), for all very limited payzones, is that
6,250 psi is apparent; indeed, the near-wellbore (tortuosity) tortuosity must be overcome while limiting fluid volumes
pressure drop at shut-in of this stage had reduced to about injected to avoid convection 3-all of which requires careful
500 psi, suggesting that multiple fractures had been removed realistic on-site analysis and re-design to achieve any
and/or a better pathway had been established. reasonably optimum economic result on marginal wells. This
Because of this concern about return of tortuosity, if the may be contrasted to other situations 10, 17-19, where proppant
3 ppga had been over-displaced, and also to avoid a full- placement opposite pay is not a real problem and efforts can
wellbore screen-out (since proppant was in short supply), a focus on tortuosity, as we illustrate next.

498
SPE 25892 M.P. CLEARY, D.E. JOHNSON, H.H. KOGSB0I.L, K.A. OWENS, CJ. De PATER 7
K.F. PERRY, A. SfACHEL, H. SCHMIDT & M. TAM BINI

B. Low Permeability Oil Wells, Offshore Denmark perforation friction but a near-wellbore obstruction. For
A converse situation, to that in (2)A, existed in the Dan further interpretation of the pressure record, near-wellbore
Field, operated by M~rsk26-28. Previous fracturing efforts-
on "vertical" wells reaching from platforms-bad proved less
J
friction pressure drop was taken proportional to flowrate .
Observed net pressure was determined by correcting bottom-
than optimal in stimulating production from the (estimated
hole pressure for near-wellbore friction at shut-in, recog-
1 mO) chalk formation, of order 100 m thickness (Fig. 9).
nizing that true net pressure cannot change instantaneously.
Althougb it had been suggested theoretically21, the concept
At the first shut-in, bottomhole pressure fell instant-
of fracturing horizontal wells was first executed by M~rsk.
aneously by 1,800 psi. The minimum stress in the field was
The major problem encountered by M~rsk was that
4,350 psi (approx.): the ISIP of 5,680 psi was higher than
tortuosity, and associated premature screen-outs, increased
reasonable closure stresses, indicating that the fracture closed
dramatically as the azimuth of the wellbores varied from the
near the wellbore. The true near-wellbore friction was there-
direction of alignment with the far-field fracture (surface)
fore higher than 1,800 psi, but the level was obscured by the
orientation. This made it difficult to place the required
early closure. 1\\'0 sand injection tests were performed using
amount of proppant in each of these fractures and required
cross-linked gel with sand slugs having peak concentration
urgent solution. The three main strategies adopted were:
of 0.6 and 1.6 lb/gal, respectively. The pressure decreased
• Matrix acidizing of the region around the wellbore.
due to the cross-linked fluid, which seemed to remove some
• High-energy large perforations (HELP)
obstruction. When the sand slug hit, the pressure increased
shot close together over short intervals.
and then decreased again; a similar response was seen in the
• Injection of proppant slugs.
third injection test; both were analogous to observations in
As in (l)A above, the slugs used by M~rsk have been Figs. 4(b), 5(b). After this test, the ISIP was much lower and
quite large, with the concept of both screening-out (all but a closure was seen between 4,250 and 4,280 psi, as shown
one of the) multiple fractures created, and also of possibly in Fig. 10(d). The friction was about 20 MPa (i.e. 2,900 psi,
eroding the near-wellbore rock material (vs. eroding the sum of the pressure drop of 1,800 psi at shut-in and the
perforations)-a reasonable possibility in soft rock like difference between observed ISIP and true net pressure).
chalk. These proppant slugs have been quite effective in In the main treatment (Fig. 100c», the pressure
reducing the near-wellbore screen-out problems. Although decreased further upon entry of the proppant into the
the (HELP) perforation strategy may be somewhat akin to formation. The designed proppant volume was placed and
the slotting approach employed in other wells 20 , it also seems the increase of net pressure indicated that the fracture was
to be more effective in the chalk environment. effectively packed off. At the end of treatment the near-
Sample (out of ten zone) treatments on Well MFA-24 in wellbore friction had decreased to 2 MPa (300 psi). The
Dan Field (Fig. 9), are shown in Fig. 10, to illustrate the steady decrease in observed net pressure during the first
many aspects of tortuosity behavior exhibited by various proppant stages was probably due to erosion of near-
wells (and even different fracture treatments within a well). wellbore obstructions. The peaking of pressure upon entry of
These are taken from the hundreds of fracture treatments that the proppant slugs during the injection tests may be
M~rsk has conducted, including treatments with the largest explained by screening out branches of the fracture and/or
proppant mass for a (horizontal well) fracture treatment and passage through the tortuous near-well bore region.
also the largest overall proppant mass injected in a well. Primary conclusions, from observations in Fig. 10 and
Various special interpretations of these obselVations have from many other such jobs conducted by M~rsk, are:
been analyzed: some are based on the concepts presented in • The apparent breakdown pressure may be very high in the
this paper14; others use idealized initiation formulae, despite presence of near-well bore tortuosity (although this was not
Ref. 11. However, the present authors have concentrated on obselVed consistently in all treatments). In cases with
the operational aspects of identifying the tortuosity, severe near-wellbore obstruction, the pressure decline may
minimizing it and then executing a successful fracturing not reflect the true leak-off behavior of the formation.
treatment. In some cases, it was found that the tortuosity • Flow rate changes are a readily-available diagnostic tool
reduced dramatically as (allowable) pad volume increased, for detecting the nature of near-wellbore problems.
something we have observed on other wells in numerous • Both plugging of multiple fractures and rock erosion seem
environments, but adequate (high) concentrations will to aid in removing near-wellbore obstruction, certainly in
generally require other remedial action, like proppant slugs. the special case of soft chalk of the Dan Field.
The first injection test consisted of 20k gals. of linear However, it is extremely important to note that many of
gel. The flow rate changes aided in determining the nature the experiences in this reselVoir may not apply to hard rock
obselVed near-wellbore restriction. The friction pressure drop environments, even when interpreted correctly. Indeed, this
can be related to flowrate (Fig. 10(b» following Eqn. (2c), case study is presented here in apposition to many of our
with oc < 1, indicating that the friction was not due to other experiences and associated strategies.

499
 FIELD IMPLEMENTATION OF PROPPANT SLUGS TO AVOID PREMATURE SCREEN-OUT
8 OF HYDRAUUC FRACTURES WIlli ADEQUATE PROPPANT CONCENTRATION SPE 25892

Conclusions and Implications for Field Operations (9) Tortuosity may be the only reasonable explanation of
observations that pumping stoppages during proppant can
The occurrence and effects of near-wellbore tortuosity imply a cause screen-out: unless the stoppages are long and/or the
much needed re-evaluation of wellbore drilling and completions "rat-hole" is minimal (which still would require a poor
operations. By combining the ideas and analyses presented in the transport ftuid), it is not easy to conceive of other sudden
previous sections--and based on many field observations, screen-outs mechanisms--and we have certainly
including those discussed in the previous section-we have experienced some such stoppages without problems.
deduced the following general conclusions and guidelines for ~Thrtu~~~~~m~oo~rt~a~~
dealing with the reality of tortuosity in well fracturing: rat-hole) for using viscous ftuids and/or high ftow rates:
the other reasonable use of high ftow-rate, for high leak-
(1) Tortuosity caused by multiple fractures can be eliminated off, is not so critical [e.g. in light of Ref. 3]-and should
by placement of proppant slugs to bias growth toward the certainly not change the job results dramatically. However,
minimum number of fractures (hopefully one). This may very low-rate breakdown (especially with inviscid ftuid)
be at least as effective as the best perforation strategy does exacerbate the tortuosity problem.
(HELP), or they may be combined, if appropriate. (11) An openbole completion-possibly including use of
(2) Most perforation strategies may be expected to leave a uncemented casing or slotted liners-should be optimum,
residual, ifnot exacerbated, tortuosity problem-including purely from the viewpoint of tortuosity minimization. Of
strategies based on oriented perforations. Only one course, this may not be allowed by wellbore stability
procedure seems reasonably sure to minimize tortuosity, considerations and/or control of production; it may also
namely high-energy large perforations (HELP) shot close present problems in failing to fracture effectively along
together (i.e. the equivalent of notching): these may leave extensive horizontal sections, especially if there are limits
the curvature problem of re-orientation to the preferred on vertical growth due to nearby undesirable ftuid zones
far-field fracture orientation (Fig. 1(a», but the latter may (e.g. water, gas-caps etc.). However, a reasonable com-
be unavoidable--even with 45° (or less) phasing of HELP. promise may be an open-hole segment at the end of a
Accurate alignment (of HELP) with the far-field direction longer hmizontal wellbore section, especially if tortuosity
may effectively eliminate the tortuosity, but hardly seems problems are severe (e.g. due to wellbores perpendicular
worth the expense, in light of (1). Although more to the preferred fracture direction). Something equivalent
numerous perforations may increase the possibility of to openbole is thought by some to be achieved by
some cracks initiating in the "right" direction, they also extensive acidizing in soft carbonate formations; related
increase the likely number of cracks (N) and conductivity strategies might involve degradable (or poor) cement jobs
may dominate over minimum stress considerations. opposite the zones to be fractured: neither of these
(3) Tortuosity is especially important in oriented well bores, approaches is supported by the Ma:rsk experience.
but is not limited to them: we have found important
tortuosity effects in nominally-vertical wellbores, even in Two broad implications for field operations may be drawn:
the absence of tectonics. A Tortuosity has been a major source of trouble and
(4) Tortuosity may be expected especially in tectonic confusion in fracturing operations. It may, in fact, be
environments, where a strong bias exists between the responsible for the routine limitation of proppant
stresses, especially in planes perpendicular to the wellbore. concentration to excessively low poundage (e.g. contrary
(5) Tortuosity should be exacerbated by the presence of to requirements in Ref. 3, even for low-perm
natural fractures, even for openbole completions. reservoirs). A related, but even more detrimental
(6) Tortuosity problems may be more severe in hard rock consequence, has been the tendency toward large
formations, especially when natural fractures exist: deeper pads--even where small/negligible pads are required2, 3.
structures may exhibit less effects of such fractures, B. Tortuosity may be strongly inftuenced by "upstream"
because of fracture closure by high confining stresses. decisions and operations in drilling and completions.
(7) Tortuosity, if it cannot be removed, limits the proppant However, possible remedial actions-such as (openbole)
concentration that may be placed, hence the ability to completions and/or fracturing while drilling or
adequately place the proppant opposite pay, e.g. resisting cementing-may experience the same skepticism and/or
convection3: the maximum concentration is inversely delayed implementation as many previous concepts
related (in a complex way) to Apr in Eqn. 2. proposed in the past by us [e.g. Ref. 21].
(8) Tortuosity can lead to rapid near-wellbore closure after Nevertheless, our experience with the proppant slug technique-
injections, creating erroneous interpretations of pressure especially unftushed shut-in, which was considered a "risky
fall-off data and poor fracture designs, especially in low- concept" until the present authors all agreed to try it in the field
permeability reservoirs (see also Ref. 2). -provides substantial hope for future innovation and risk-taking.

500
SPE 25892 M.P. CLEARY, D.E. JOHNSON, H.H. KOGSB0LL, K.A. OWENS, CJ. De PATER 9
K.F. PERRY, A. STACHEL, H. SCHMIDT & M. TAM BINI

Acknowledgements 13. Hallam, S.D. and Last, N.C., "Geometry of Hydraulic

The first author (Cleary) is especially grateful to the Gas Fractures from Modestly Deviated Wells," SPE Paper No.
Research Institute for funding his efforts over the past decade 20656, September 1990.
and to the many companies who have undertaken to help gain a 14. Weijers, L., et aI., "Geometry of Hydraulic Fractures
reasonable understanding of the physics involved in underground Induced from Horizontal Wellbores", SPE Paper No. 24994,
(fracturing) operations. The authors express gratitude to the November 1992.
management of Agip S.p.A, Edison Gas, RWE-DEA and Maersk 15. Branagan, P.T. and Wilmer, RH., "Breakdown Procedures
Olie og Gas AS, the operator for Danish Underground Designed to Minimize Naturally Fractured Reservoir
Consortium (AP. M!Illler, Shell and Texaco) for permission to Damage," SPE Paper 17716, June 1988. (See also, J. Dees,
publish the data. They are also grateful for the co-operation of private communication (Oryx Patent».
the particular (Halliburton and Schlumberger) service company 16. Economides, J.J. and Nolte, KG., "Reservoir Stimulation",
field personnel involved in the various new approaches required 2nd Edition, published by Schlumberger Educational
for field operations. Services, Houston, TX 77023; 1989.
17. Martins, J.P., et aI., "Deviated Well Fracturing and Proppant
References Production Control in the Prudhoe Bay Field", SPE Paper
1. Cleary, M.P., "The Engineering of Hydraulic Fractures- No. 24858, October 1992.
State of the Art and Technology, Journal of Petroleum 18. Ovens, J., "The Performance of Hydraulically-Fractured
Technology, 13-21, January 1988. Stimulated Wells in Tight Gas Sands: a Southern North Sea
2. Cleary, M.P., Wright, C.A and Wright, T.B., "Experimental Sample", SPE Paper No. 20974, October 1990.
and Modelling Evidence for Major Changes in Hydraulic 19. Pospisil, G., et aI., "Results of a Large-Scale Refracture
Fracturing Design and Field Procedures", SPE Paper No. Stimulation Program, Kuparuk River Unit, Alaska," SPE
21494, 1991. Paper No. 24857, October 1992.
3. Cleary, M.P., and Fonseca, A Jr., "Proppant Convection and 20. Gas Research Institute Staged Field Experiment (SFE) #2,
Encapsulation in Hydraulic Fracturing: Practical Implications Report No. GRI-90/0093, April 1990.
of Computer and Laboratory Simulations", SPE Paper No. 21. Cleary, M.P., "Mechanisms and Procedures for Producing
24825, October 1992. Favorable Shapes of Hydraulic Fractures", SPE Paper No.
4. Wright, T.B., Johnson, D.E. and Cleary, M.P., "Real- 9260, 1980.
Data/On-Site Analysis of Hydraulic Fracturing and 22. Mukherjee, H., et aI., "Successful Control of Fracture Height
Procedures for Design Optimization", November 1992. Growth by Placement of Artificial Barriers," SPE Paper No.
5. Cleary, M.P., et aI., "Documented Case-Studies and 25917, Joint Rocky Mountain Regional Meeting and Low-
Procedures for Proper Design and Execution of Hydraulic Permeability Reservoirs Symposium, Denver, CO, 1993.
Fracturing", in preparation for publication. 23. Johnson, D.E., et aI., "Real-Data On-Site Analysis of
6. Cleary, M.P., Barr, D.T. and Willis, R.M., "Enhancement of Hydraulic Fracturing Generates Optimum Procedures for Job
Real-Time Hydraulic Fracturing Models With Full 3-D Design and Execution", SPE Paper No. 25920, Joint Rocky
Simulation", SPE Paper No. 17713, 1988. Mountain Regional Meeting and Low-Permeability
7. Narendran, V.M. and aeary, M.P., "Analysis of Growth and Reservoirs Symposium, Denver, CO, April 1993.
Interaction of Multiple Hydraulic Fractures", 1983. 24. Johnson, D.E., et aI., "On-Site Real-Time Analysis Allows
8. Johnson, D.E. and aeary, M.P., "Implications of Recent Optimal Propped Fracture Stimulation of a Complex Gas
Laboratory Experimental Results for Hydraulic Fractures", Reservoir," SPE Paper No. 25414, SPE Production
SPE Paper No. 21846, 1991. (See also Papadopoulos, J.M., Operations Symposium, Oklahoma City, OK, March 1993.
Narendran, V.M. and aeary, M.P., "Laboratory Studies of 25. Juranek, T.A, et aI., "Mini fracture Analyses and Stimulation
(Multiple) Hydraulic Fracture Growth and Interaction", SPE Treatment Results for CO 2-Energized Fracturing Fluids in
Paper No. 11618, 1983.) South Texas Gas Reservoirs," SPE Paper No. 20706, 1990.
9. Yew, C.H., Schmidt, J.H., and Yi, L., "On Fracture Design 26. Andersen, S.A, et aI., "Exploiting Reservoirs with
of Deviated Wells," SPE Paper No. 19722, October, 1989. Horizontal Wells: The Mrersk Experience," Oilfield Review,
10. Pearson, C.M., et aI., "Results of Stress-Oriented and pps. 11-21, July 1990.
Aligned Perforating in Fracturing Deviated Wells," SPE 27. Conlin, J.M., et aI., "Multiple-Fracture Horizontal Wells:
Paper No. 22846, October 1991. Performance and Numerical Simulation," SPE Paper No.
11. Cleary, M.P., "Rate and Structure Sensitivity in Hydraulic 20960, October 1990.
Fracturing of Fluid-Saturated Porous Formations," 19th U.S. 28. Owens, KA, Pitts, M.J., Klampferer, H.J. and Krueger,
Symposium in Rock Mechanics, June 1979. S.B., "Practical Considerations of Horizontal Well Fracturing
12. Behrmann, L.A, Elbel, J.L., "Effect of Perforations on in the "Danish Chalk", SPE Paper No. 25058, European
Fracture Initiation," SPE Paper No. 20661, September 1990. Petroleum Conference, Cannes, France, November 1992.

501
 .....
o

/NlrrATfON t INTER.ACTIVE I MASSIVE C::-RAcrUR£

REORIENrATlON REGIME EXTENSION ,=<EGfM£

Ffw,Q

I SUCCOSSM:
FRACTURe ...-"
............ PERIMETERS

#f:LLBOR£

Figure l(a) Schematic of fracture re-orientation from

well bore into far-field preference.

Figure l(b) Schematic of fracture departure from well bore

01 trace, even for (nominally) vertical wells.
oI\)
Figure 2 Schematic of multiple fractures emanating
from perforations sets in various rock types.
'--.-

Figure l(b) Figure 2

w..- width

A~f
Tortuosity

---
~

Q
---t LT
t AP'
~c::p'

Figure 3(a) Simplified geometry to represent main tortuosity parameters,

~ -p- -
w
Ap'
T
t
AQ1 AQz AQ a
AQ

Figure 3(b) Contrast between pressure response of tortuosity vs. perforations.

 I ---T"'' i----- .------------------
I

. - - - \:;:'

I -

"'t\;:':;
1- -
:. :

~~-

~i- . JIII[[_
Figure 4(a) Schematic of Fracture Growth Postulated by Participants in Gas Reserach Institute (GRI) Staged Field Experiment (SFE) #2, Travis Peak, East Texas [Ref. 20]

SFE #2 Upper SFE #2 Upper

Divertafrac Treatment Main Fracture Treatment

- ~eGeIVISC
_ . - Slurry Flowral. _ . - Slur!'Y F1owrat. - - - Propp.,t Cone
- Btrntiae Presaure 250.0 - 6trnliole Pressure - - CGng Preaaure
150.0 15.00 15.00
150.0

I~
U500 4000 5000
\JOOO

,/L ) \
120.0
8700

11_ - \ ! I
fll~ :--- r-
. . .
/!
i
12.00
3200
200.0
120.0
8000
I ! -- f-v-- -- --- -, ./ --
12.00
4000

110.0 \f- . , ~ 1\
•.00
150.0 L /....",
•.00
7IlOO I~ 1/ • I 2400
110.0 r- /' 3000
I \i
\ \
7000 ~

1\
'-J
80.0
7100
)
~
A
l J ~
h ~\
~' II
'- ,....
6.00
1800
100.0
80.0
8000
)" ~.-- --.- f-.. .. - I-~
" ---- 6.00
2000

I ~ !i 1/ i
30.0
8300
,- f-'J..~
I
1 j-' --r'-' r" ;..
'-..J
_.J
I
! f\ "-- 3.00
800
SO.O
30.0
5000
ii
, r ..J
\
\
,j
3.00
1000

, ,
0.0
j "
I I
!I I I
\ j rI 0.00
0.0
0.0
Ii I
I
J- -, 0.00
5500 0.0 60.0 120.0 UIO.O 240.0 300.0 0 0.0 30.0 80.0 110.0 120.0 150.0 0
4000
Tim. (mlns) Tim. (minI)

Figure 4(b) Pressure, Flow & Proppant: Divertafrac Treatment, Upper Travis Peak Figure 4(c) Pressure, Flow & Proppant: Main Frac Treatment, Upper Travis Peak ....
....
 SFE #2 Lower SFE #2 Lower
Minifrac #2 First Propped Fracture Treatment

-- Slu~Row -- SI~Row - - - Sand Cone

150.0
- CER Bmhol. PIe.. - - Casing Pless - CER Dnhol. PIe - - Tr. .llngPle
100.0 I 5.00
4000
uooo 10000

1 -r.~.
5000

120.0
8200
r ... (' f- ......,.. ~.... - 32CO BO.O
uooo
n 1 ~I' \~jl I~L
~~ .t........... "' I 2.00

..... / I I \-..1 ..... )

4000

f r"\ r1 ;J ----~ " I "f1

:
00.0 BO.O 0.00
2400

/ I
7400 '-- 8000 3000
\.
I f"- ...... ,...,
--
"
Ir '--- I / ,- ~
~
I

--
00.0 ".0 •.00
1000
8600

If '----
\. '-- 7000
1/
...... ....... '\ 2000

II
~

~ \
If '---
30.0 20.0
~(j ~1-' 3.00
800
j
~
I

--' -- "--_."'\ II I
5000 8000 1000
r-' I ~~
L __ Ir---,
0.0
5000 0.0
~r i
50.0 100.0 150.0 200.0
l
250.0 0
0.0
SOOO 0.0 ".0
I
00.0
I
120.0
,
I
I

1600
I \
200.0
0.00
0
Time (min:s) Time (mins)

Figure 5(a) Pressure and Flowrate Data: Mini-Frac #2, Lower Travis Peak Figure 5(b) Pressure, Flow & Proppant: Main Frac Treatment, Lower Travis Peak

SFE #2 Lower SFE #2 Lower

Injection Test Second Propped Fracture Treatment

- - Slurry: Row - - - Sand Cone

100.0
- - TubingPr:s - CERDnholePnJ - - Treating Prs '1
100.0 15.00
5000

r . . . . h IL
5000 10000 0000

BO.O r.... ~ 4000 BO.O 12.00

W~_ r'"
uooo _/
---
4000 5000
/ .... 1\ -./
'--
....f'"
BO.O
p ' - .....
t- "'_ 00.0
r II / 0.00
3000
3000

\..
8000 jI'-.r
rv--- ......
- ~
./
~
4000

r-- r- 1'0
-----
".0
) .A ".0 •.00
2000
2000
7000 3000
\
20.0
,J
. -..,.".
(;., I-----,
I.,..
1 1000
20.0
r~
h r r-----·h. -
.
_...... _ Jr --
'",
.....\.,
3.00

....-- ~~ ( LJ ij-J
1000 8000 I 2000
i I
0.0
a
JI
0.0
r- 30.0 00.0
i
00.0 120.0 150.0
0
0.0
5000 0.0 ".0
,1 ,
BO.O
IJ
r-

120.0
'r 1,
180.0
-- 200.0
0.00
1000
Time (mins) Torn. (mlnoj

Figure 5(c) Recorded Treatment Data for the Injection Test, Lower Travis Peak Figure 5(d) Pressure, Flow & Proppant: Second Treatment, Lower Travis Peak
SPE 25892 M.P. CLEARY, D.E. JOHNSON, H.H. KOGSB0LL, K.A. OWENS, Cl. De PATER 13
K.F. PERRY, A. STACHEL, H. SCHMIDT & M. TAMBINI

AGIPS.p.A
Net Pressure Match

A 8tm Siry Rate (bpm) 8tm Prop Cone (ppg)

o Observed Net (psi) Model Net Pressure (psi) 20.00
40.00
700.0 700.0

32.00 16.00
560.0 560.0

~
24.00 12.00
420.0
~ rr,
420.0

% If" \
16.00 ~ ~ 8.00

\
280.0 280.0

\ 1\ 1\

8.00 \ 1\ 1\ \ 4.00

\ ~\ \ \
140.0 140.0
~ .....

- ~- -
j ~ .\ - ~ ~
0.00
0.0 0.0 110.0 220.0
TIme (mins)
330.0 -- 440.0 550.0
0.00
0.0

All Minifracs Job Date: April 14, 1992

Figure 6(a) Matching of net pressures measured in minifrac tests, showing tortuosity effects on pressure fall-off.

10\
AGIP S.p.A
Net Pressure Match

A P,op Cone (ppg) o SkmyR ... (bpm)

SO 00 0 Observed Nel (psi) a Model Net P,essufe (psi)
"'00
""'0 0000

4000 2000

1/
6400 0<00

/ :>l00 f \\ ,SOO
~\
4000 <000

~~
\lb..
2000 r-. 10.00
3200 c.a-' ~ 3200
~
-fi. I r
~ '"'''IF
1000
'600
Ir rI SOO
1600
I
r
IOJ.
100
~. _____ 1 _ _ _ _ _ _ .J ____ ,_

10'
___________ , __ ~_'~~_~~_L~L' J 000
00 0
1 6200
.n -6(l( -'1)( -,,,Cll 1!Ilo
000
00
Time (mifl~

Flnw R~lc (hrlll) Main Fracture JoI> Dale: April 14. 1992

Figure 6(b) Relationship between near-well bore friction and tlowrate. Figure 6(c) Predicted VS. observed pressures in main frac.

505
 Preetz 26a
Database Data

- - Surf _ (Tbg) (pal - - _hale Pr_ (psi)

Preen 260 Friction versus Flow Rate
0000

- -- -
,\

'.
-
1600
r-_
~oJ
n '- """
...., ~

- -
1.«Xl
II, ./

- V--
- ----
U' 1(\
'- U I'-' -...
""" 1200

,"" UI "'00
., lCOO
.,-AJUl 80.'
,, 2«, 3'" <00, A ..., V
Time (mins)
~
If /
'-.
600
- Slurry Row Rat. (bpm) - - Density ~b1gol)
,.00

- J -. .- ~
.- 11',_, ....
'.00
.«Xl
/
2000

'''Xl
''''
000

000 200
7
'.00 .00
/
10 15 20
.00
I .00
I Flowa_(bpm)
II.
'00 oc 80.0 , 0 >«, »<0 <000
.00

Time (mins)

Minifrac Job Date: June 30, 1992

01 Figure 7(8) Preetz 26a minifrac database data, Figure 7(b) Near-wellbore friction vs, flow rate for Preetz 26a breakdown test.
o
0>

Preetz 26a Preetz 26a

Square-Root lime Pressure Data

--
- - Surf _ (Tbg) (psi) - - _hale Pr_ (psi)
- - BoItanhoi. Prees (psi)

~
"000

8800

-
/
/'
A
-
-...
•1000

-
-/
~ "'" ~ It'
.....

""""" 00

-·-TB
--
34. ..
Tim. (mlno)
.
--
... .. 00

f Sklrry Row Ret. (bpm)

""""" ""-
.000 .. 00
3000
.,00 .., .oJ
I
...
11.:10

.... J
"", ""
'000

.eo
eooo
"~ ........,, . ,
MOO ~.~ 51." 57.~ ea.~
N
75.~
o
.00 00 34.
II
.
Tim. (mino)
. ... ..00
.00

Time (mini)
Minifrac - First Shut-in Job Date: June 30, 1992 Main Fracture Job Date: July 1, 1992
Figure 7(c) Preetz 26a breakdown pressure decline during shut-in, Figure 7(d) Database data for the Preetz 26a main fracture treatment.
 Preetz 28
Database Data

- - Surf_ 1'rHsu.. 11
Preetz 28 Main Fracture
.»-..
--
80S)

r
8:>,.

-
\I
If
U ...
..,
- - Slu"" Row (bpm)
.1-----'
~

rome (min.)
- -
r

''"'
- 0ensi1y Oblgal)
.....
-..,

g
8110

""
00." 1 8150

'" I~
, ... 40 ",.
II
»" »00
""
11.20
.,,,
'/""1' .-_n_ It- r- -t' l- I- .- .J
" :I't --1' fT-
.eo

Time (mins)
''"' ...,
Minifrac 113,114 and Main Frae Jcb Date: July 3, 1992 Ooeu.ser_f,..

Figure 8(a) Minifrac #3, #4, and main frac database data. Figure 8(b) Proppant-concentration profile: Preetz 28 main fracture treatment.

(J'l
o
.....
1000 1500 2000 2500 3000 3500 1000 1500 seee
L-_LI--~-L~__J_~I_ _~_L_ _L _ _ L_ _IL_-L~_ _J_~_ _~I_ _~.~~._J_ _J_~_ _~_L_ _L _ _ L_ _L _ _ L_ _L__L~_ _J - - J_ _J-~ __J _ _ L_ _~_L__L_~ '"'"
lD
If)
,
U1
()).
.
<D '"
'",
'"co GDC '"
"-
'"
If)

:c CS>
CS>
CS>
,
lD

CS>
~TOP CHALK
';'rOP 02
TOP MIA
~OP HI
<r

-----------~~~------------------
t.O
, 'TOP H2
O'l CS>
en CS>
IS> lD
.tS>
en
())
lD
,
Cil
IS> '"
10\HI 1500 2000 25e0 3000 3500 100e 1500 seee '"'",
lD

11 F R .... 2 if , VERT I CRL SECT I ON RLni~G TRRJECTORY

IIRK. JANUARY 27, 1992

Figure 9 Well layout, with 3,165 ft. horizontal section through the payzone; azimuth was approximately 280°: well path was nearly perpendicular to preferred fracture plane. ....VI
 8oo0~ ____________________________ ~

MFA24 zone 2
Measured Data
7500
- - !!Iurry........
- (pol) - - - Sluny dona (ppg)
-!ItIIp (poI) - - Tbh 1"'_ (pSI)
100.0
10000
20.00
10000
7000 o
o o
6500
10.0
IlOO
,. 11J!tJ
8000
~
.e 6000
10.0 I ~ I.. 15.20
~
o
7eOO
I r r'l '" ~ r"'-.. q .....
8000
a. 5500
V 'I ~ II
...... --;:;.,
. ~.I
40.0

"'"
"

,.
,
12.60
8400
l-1/ I 4000 5000

i,
j I
""- I

" --- "-t,

I
i i
--
I
20.0 I 10..40 4500
..............
J --+
S200 I 2000
f-
l"

I ~ ,l
0.0
4000 0.0
...-'"-
r-i---~
30.0.
,..''1L
.... ~
60.0.
-- 110.0. 120.0
L

H 150.0
aoo
0 10 20 30 40 50
Tame (mina)
Flow rote (Bpm)
Figure 10(8) Treatment data: injection test of well MFA-24, Zone 2. Figure 10(b) Friction pressure vs. flow-rate for the injection test of well
(J1
MFA-24, Zone 2.
o
CO

MFA24 zone 2 MFA24 zone 2

pressure match Third injection test

- Ob ....od Net (Psi) - - Net Pressure (psi) ,,.. - Btf1 pressure (Psi)
"'" 4400
':100
"\1

I
'I
... ~
J
,/
'\ l\
.,., 4350

,..
-- '". --
yo

.~
~.

...
Time (mins)
V
-- ''".
300

4300
------ ---........... -......
........
...........
,..
:1000
- . - N e ' - Frie(psi)
- Btm Prop Cone (ppg)
---- Freet,,," V"" (cp)
- - Sluny Rate (bpm) ....'". I'--.
....
....,... ......• <250 .......

.... '".
......
':100
..,
......
r
300.•

:100.
»... <200 "'" """'"
. "'"
100.0
'r-_ '0"
'". TIm. (min")
... ,,. ,'". o.
0 ..
4150
72.0 75.6 78.2 82.8 86.4 90.0
Tme (mine)

Figure 10(c) Observed and modelled net pressure for the injection tests and Figure 10(d) Pressure decline after third injection of MFA-24, Zone 2.
main treatment of well MFA-24, Zone 2. Oosure could either be picked at 4,250 or 4,280 psi.