Patent Application
20100066560
The present device relates to logging radar devices functionally attached to (or integral with) the casing. More particularly, the present invention relates to radar systems that use electromagnetic wave propagation to locate and identify geometry aspects of proppant packed fractures created in geological structures located in the ground, wherein at least a portion of the operation occurs in a borehole (also known as “down-hole” as referred to in the art).
Oil exploration and developing oil wells often pose great financial risks because the costs are substantial. To mitigate some of the financial risks, logging has become essential in nearly every phase of exploration as well as drilling, completing, and producing the well. Logging techniques provide information on the depth of formations, the presence of oil, the bottom-hole or formation temperature as well as data associated to the success of completion techniques, initial formation/reservoir pressures and various data related to stimulation treatments that are often applied to increase production rates.
Often the key to attaining an acceptable production rate and its associated financial results lies in the well's response to stimulation techniques (in particular hydraulic fracturing). The technique referred to as hydraulic fracturing describes a process in which a fluid (either thin or viscous) is pumped into the targeted formation at a rate in excess of what can be dissipated through the natural permeability of the formation rock. This results in a pressure build up until such pressure exceeds the strength of the formation rock. When this occurs, the formation rock fails and a so-called “fracture” is initiated. With continued pumping, the fracture grows in length, width and height. At a predetermined time in the pumping process, solid particulate is added to the fluid that is being pumped. This particulate is carried down the well, out of the wellbore and deposited in the created fracture. It is the purpose of this specially designed particulate to keep the fracture from “healing” to its initial position (after pumping has ceased). The particulate is said to be propping open the fracture and is therefore designated as “proppant”. The fracture, which is generated by the application of this stimulation technique, creates a conductive path to the wellbore for the hydrocarbon. Critical to the process of optimizing the design of a hydraulic fracturing treatment, is the determination of the created fracture geometry (in particular fracture length).
Currently there are logging techniques that give limited information on fracture height, but virtually no technique that gives any reliable data connected to propped fracture length.
The lack of an accurate assessment of propped fracture length is due to a combination of factors. First and foremost is the fact that propped fractures can extend for hundreds of feet away from the wellbore. Prior to the development of the technique of the present invention, which utilizes penetrating radar waves, there was no proven technology available that could determine this substantial propped length aspect (with a reasonable degree of accuracy). Secondly, the down-hole conditions (in particular temperature and pressure) encountered by logging equipment limited the electronic equipment that could be used, types of signals that could be generated and the type of data gathered by this type equipment. It is not uncommon for logging equipment to be subjected to temperatures in excess of 200° C. and pressures up to and exceeding 10,000 psi.
Thus, via logging and other technologies such as pressure analysis and production history matching, the potential productivity of a given well can be more accurately evaluated. However, current logging devices do not address all critical data requirements and more sophisticated equipment may not stand up well to the environmental conditions of a borehole. For example, temperatures may exceed 200° C. down-hole, and this type of heat limits the electronic sensors and circuits that can be used in a logging device.
Ideally, fracture location and orientation, and its dimensions width, height, and length would be known values. However, as mentioned previously, there is limited data available on fracture height and virtually no method available to accurately measure an extended propped fracture length. Therefore, there has been a long-felt need in the art/industry for a logging device that can be used to generate this critical element of fracture geometry while being subjected to the elevated values of temperature and pressure (for example about 200° C., or greater, and 10,000 p.s.i.) associated with down-hole wellbore conditions. There is also a need in the art for a system that can be arranged to operate with existing wellbores that have already been perforated and fracture stimulated and newly drilled wells that may be completed according to the present invention to simplify the measurement process or to enhance its ability to describe the propped geometry generated from a fracturing treatment.
The invention provides a radar logging device/tool, system and method for determination of propped fracture length, height and azimuth (direction from the wellbore). The present invention addresses the industry need for accurate measurement of these important aspects of fracture geometry. The invention accomplishes this goal using a design expressly suited to operate under adverse conditions associated with a wellbore, as it penetrates the producing formations and its associated elevated formation temperatures and pressures. If you know propped length and height, you can calculate the third fracture dimension (propped width).
The present invention provides a radar logging system, apparatus, and method that includes above ground instrumentation and a down-hole hybrid transceiver. Microwave signals, which are generated above ground, are used to drive an intensity modulated (IM) laser. The laser output travels along a fiber optic cable down into the wellbore.
The radar logging system includes a transceiver that may include a photodiode or has a photodiode separately connected thereto. In one particular embodiment, the laser output travels down a fiber optic cable arranged inside the well casing and will be converted by the photodiode back into an electromagnetic signal, including but in no way limited to a radar signal. Depending upon the chosen frequency, the radar signal can be transmitted through holes/perforations (or possibly a slit) in the casing and based on the reflection from the boundaries of the proppant and formation walls contained in a fracture, the difference between the reflected signal and the transmitted signal in at least one characteristic (for example, frequency) can be fed to a mixer to send an audio signal (the difference in frequency between the transmitted signal and reflected signal, as both typically are about 1 GHz (this number can vary greatly) will tend to have a difference in the KHz region. The audio frequency is generated without using active components that would be affected by the extreme heat and pressure down-hole. Fracture length may be calculated based on the elapsed time from when the signal is transmitted by the antenna to when it returns. The fracture length is calculated by taking the time increment and multiplying it by the velocity of the wave through a proppant pack.
Generally the proppant is not reflective. However, in some embodiments a reflective material may be put into proppant at the distant ends of the fracture or as a thin coating along the propped fracture sidewalls. Also, in some embodiments based on a change in at least one characteristic of the transmitted signal, data may be generated that can be used to describe the propped fracture length and height. This technology can be used to describe all fractures connected to the wellbore and differentiate between the dimensions of the two vertical wings of a propped fracture. Commonly accepted theory says that most fractures are orientated in the vertical direction (the width of the propped fracture is largest at the wellbore) and that vertical fractures have two wings orientated approximately 180° from each other. In most instances (due to the inability to accurately measure fracture length) the two propped fracture wings are assumed to have similar propped geometry (length, width and height). This technology will allow actual measuring of the geometry of both wings.
Inside the wellbore, the transceiver, containing passive components that can withstand the high temperatures, such as a photodiode, converts the IM laser signal back to a microwave signal. The signal is split between an antenna and a mixer, where the output from the antenna is transmitted out into the fracture. The fracture containing proppant serves as a wave guide for the radar signal and inconsistencies in the fracture, including the fracture termination reflect the radar signal to form a reflected wave. This reflected wave is mixed to generate a beat frequency used to determine the dimensions (e.g., length) of the fracture. Advantageously, the transceiver is operable at low down-hole temperatures or high (e.g., about 200° C., 220° C. or 300° C.) down-hole temperatures without any cooling apparatus.
The invention includes a plurality of ways to bring the signal down-hole and generate data based on the audio signal. For example, to eliminate the issues relating to transmitting the signal beyond the casing, the transceiver, twisted pair and/or fiberoptic cable can be arranged along an outside of the casing. This configuration also permits monitoring the fracture geometry during a fracturing treatment (in so called real time).
Moreover, the arrangement of some of the other hardware that comprises the present invention can all be arranged external to the casing, and arranged along an exterior of the casing. For example, a series of antennas can be arranged circumferentially around the outside of the casing to examine 360° around the wellbore.
In another embodiment the twisted pair may be replaced by a series of transmitter/receivers to relay a signal up to the surface.
In another embodiment, a portion of the pipe (casing) is pre-designed to contain perforations large enough to permit a series of antennas to transmit through the perforations. The pre-designed casing can be a shorter than normal length of pipe.
In some embodiments, the invention permits accurate radar logging measurements using only passive components down-hole (no amplification down-hole). The more temperature sensitive active components are above-ground and away from the high temperatures, pressures and potentially corrosive environment often associated with down-hole well conditions. However, active components, for example an amplifier, which can operate in downhole environments may be employed if desired.
Also, the invention advantageously achieves very low loss signal transport mechanisms. For example, an embodiment has a fiber microwave feed exhibiting at or less than a 1.2 dB RF/electrical loss per kilometer. This is dramatically lower than would be possible with an RF transmission line and is a reason for using fiber optic. Also, the audio frequency output signal can be transported for kilometers over a twisted pair of wires with minimal loss.
A notch or slot antenna may be employed to radiate the signal outside the casing. The vertical slot is created in the wellbore casing by a high-speed saw/cutting device or cutting laser that may or may not be a part of the radar logging device or by a shaped charge separate from the radar logging device. The slot is created in the location that is determined by the tool as oriented toward the fracture. This slot location is determined by probing the casing perforations with a RF signal (whose wavelength is short enough to pass through a single perforation) to locate those perforations shown to be in communication with the propped fracture. Typically, the slot is approximately 5 mm in width by 20 cm in length. The length is typically greater than or equal to one half the wavelength of the signal being used. The tool orients its vertical slot antenna with the casing slot. Contact of the antenna with the casing is not required. The slot antenna of the tool and casing slot create a strong electrical coupling allowing the tool's electromagnetic signal to enter and penetrate the propped fracture.
The logging device may be positioned down-hole in conjunction with a gyroscope. The gyroscope has a dual purpose of being useful to pinpoint the location of perforations that are found to be in communication with the propped fracture and it also provides useful information that contributes to the invention's ability to determine the azimuth/direction of the propped fracture as it leaves the proximity of the wellbore. However, it should be understood that a gyroscope is not a requirement of practicing the invention.
The gyroscope may also be used to position the above-mentioned slot that is used as an exit point for the device's RF signal.
While the slots in the casing can be made down-hole with a cutting tool or saw/cutting device, it is also within the spirit and scope of the invention that, in an embodiment, wellbore casings are manufactured with prefabricated slots that are selected/customized according to the specifics of the wellbore. It is also within the spirit and scope of the invention that, in an embodiment, wellbore casings may be made of fiberglass (which is transparent to the signal).
For purposes of illustration and not intended to limit the scope of the invention in any way, the aforementioned and other characteristics of the invention will be clear from the following description of a preferred form of the embodiments, given as non-restrictive examples, with reference to the attached drawings.
It is understood by a person of ordinary skill in the art that the drawings are presented for purposes of illustration and not for limitation. The embodiments shown and described herein do not encompass all possible variations of the arrangement of structure, and an artisan appreciates that many modifications can be made within the spirit of the invention and the scope of the appended claims. In the following description, well-known functions or constructions are not described in detail so as not to obscure the description of the invention. For example, power sources, bias voltages, and their respective connections are not shown in the drawings so that the subject matter emphasized in the description is not obscured with unnecessary detail. However, an artisan understands and appreciates that any such items not shown may be advantageous and/or required for operability.
There are a number of ways that the present invention may be practiced. While newer casings can be designed with the present invention in mind, older casings/wellbores can be modified so that an antenna located down-hole will be able to transmit and receive a signal through a series of slots that are cut into the casing.
Referring to
Microwave signals, generated above ground by the microwave signal source 221 (e.g. radar signal source), are used to drive a laser intensity modulated (IM) by an intensity modulator 226. The laser output travels along the fiber optic cable 300 down into the wellbore.
Inside the wellbore, the transceiver 510 containing passive components that can withstand the high temperatures, such as the photodiode 505, which converts the IM laser signal back to a microwave signal (e.g. radar signal). The radar signal is split between an antenna 512 (
The second frequency (e.g. about 0.5 to 2 GHz, typically about 1 GHz) is used for transmission via a notch or slot antenna to determine the fracture dimensions, (e.g. fracture length and/or height). This would be the frequency of choice if the fracture being examined is >30 feet and it is desired to measure the propped fracture from wellbore to tip. It is expected that this frequency can be used to measure propped fracture lengths up to and exceeding 500 feet. This frequency is used with one or more exit slots. The reference numerals shown in
Still referring to
In addition, in an embodiment the multiple antennas 702 may not transmit at the same time, so it would be easier to detect the angle of the reflected signal. The transmission may be in sequential order, or random, or follow some pattern spaced apart by, for example, 270°, 180°, 90°, 45°, etc. The pattern could be spaced apart as desired, to reduce problems with interference. As mentioned above, typically, multiple antennas and transmitting 360 degrees employ RF transparent material, such as fiberglass pipe, or multiple slots or exit ports.
If desired, the casing may have perforations sufficient to transmit an approximate 1 GHz signal (plus or minus about 30 to 50%). In an internal arrangement of antennas, higher frequencies can exit through smaller perforations. For example, it is possible to use a signal near 10 GHz. Moreover, the data from a 10 GHz signal has significant value in that it can determine fracture length on the order of 10-30 feet. This is a significant advantage over other technology can only see the first 1-2 feet of propped fracture length. In other words a 10 GHz signal can tell you that a propped fracture length >20 feet is connected to a particular perforation.
In addition, the logging device can be arranged (at least partly) down-hole in wellbores that would require modification of existing casings, or customized casings can be constructed to permit an antenna to transmit the signal toward the fracture. Some of the installation steps and the modifications to the structure of the logging device that assist in this process are described below.
As shown in
To determine the location for placing the slot in the casing, the device 405 transmits a signal to locate casing perforations connected to a proppant packed fracture (of at least several meters). Once such perforations are found, the saw or cutting device 410 (which may comprise a laser) cuts a vertical slot or slot 99 in the casing 100 (dissecting the perforation). The slot 99 is designed to be sufficient in size to allow a lower frequency signal (suitable to make the trip down the fracture and back to the transmitter/receiver) to travel from within the casing to the propped fracture. During the cutting, the gyroscope 430 assists in positioning the cutting device 410 (e.g. high-speed saw or laser) for cutting the slot or notch.
In accordance with
(a) As shown in
(b)
(c) As shown in
(d) As perforations are located, data from the return of an approximate 10 GHz signal is gathered and analyzed (data being transmitted to the well surface through the twisted pair). From the aforementioned data, it is determined which perforations are connected to a propped fracture of at least minimal length.
(e) As shown in
(f) As shown in
(g) After the fracture geometry is measured, the tool 405 can then be raised out of the wellbore, as shown in
Using a similar procedure to steps (e) and (f) above, the fracture height can also be established. In this case, the saw/cutting device 410 would be positioned to cut narrow slots above and/or below the first or last perforation to be identified as being in communication with the propped fracture connected to the fracture. These narrow slots (above or below the communicating perforation) can be used to establish if the propped fracture extends past the perforation in question.
In addition, the use of this technology/logging device 405 is simplified when all or at least portions of the casing are made of composite material which is relatively transparent to the signals. This procedure, in which a signal can be transmitted through composite casing material, is a variation upon the embodiment discussed above and would be particularly applicable to new wells. Old wells are typically completed without these composite sections. However, a customer designing a new well completion can decide to include composite casing material to facilitate employing the present invention to determine the created fracture geometry of the well.
For example, new casings being arranged down-hole can be made to include portions, referred to as pup joints or “pups” 402 (
These composite pup joints (or subs) would be substantially transparent to the RF signal being directed at the fracture area and could simplify locating the position of the fracture (and measuring its extent). The composite pup joints 402 permit the signal from the logging device 405 to penetrate through the composite pup joints 402 (
An advantage of arranging the composite pup-joints 402 in the casing 400 is that the composite material makes it easier to position the logging device 405. It also readily allows 360 degrees to be examined.
Also, an advantage of arranging the composite pup joints 402 in the casing 400 is that the logging device 405 could be fully functional for sending a signal through the composite pups with only a source having an approximately 1 GHz signal, e.g., 1 GHz±0.1 GHz. Thus, the approximately 10 GHz signal, e.g., 10 GHz±1 GHz, and the cutting saw/cutting device 410 would not be used when transmitting a signal through the composite pups 402. However, a transceiver for the approximately 10 GHz signal and the cutting saw/cutting device 410 may be included, even if being used with a casing 400 including composite pups 402, in the event the logging device 405 is also to be used to send a signal through parts of the wellbore casing 400 not made of composite pups 402.
In an embodiment of the invention, for maximum definition of the fracture height and length, the production interval can be completed using only composite pipe (instead of the combination of composite subs and conventional casing). In such an example it would be possible to scan the entire production interval without interruption.
Typical composites are fiber glass reinforced-cured epoxy resins. To maximize the geometry data obtained through the use of this invention, all or a substantial portion of the casing, positioned adjacent to the producing interval, may be made of composite material. An example of such composite material is RED BOX fiberglass reinforced aromatic amine cured epoxy resin casing and tubing available from Future Pipe Industries. Such casing and tubing is designed for downhole service of medium to high pressure at depths as great as 13,000 feet.
Additionally, the use of composite pup joints or joints 402 permits one to check above and below the zone of interest to see if the top and bottom of the propped fracture have been located, to provide accurate height determination.
In contrast, employing the method of the present invention on casings made without the composite pup-joints, such as existing casings already installed down-hole that are cemented across the treating interval, normally includes an additional step of using the cutting saw/cutting device 410 to enlarge the perforations 105 already in the casing 100 (as shown in
Although an artisan will understand and appreciate that other values can be substituted, the invention has been designed using as many “off the shelf” components as possible to aid in manufacturing ease and reduce costs. For example, the optical fiber selected was a high temperature polyimide coated mode 1550 nm fiber and can withstand the 210° C. and above temperatures. In addition, intensity modulators that operate in the 1550 nm wavelength range are available. There are also many lasers that operate with sufficient power at the 1550 nm wavelength range. Moreover, there are Erbium-doped fiber amplifiers available that can provide further optical amplification of 1550 nm signals should more optical power be required.
With regard to
It should be noted that virtually any type of signal propagation can be compatible with the present invention. For example, a pulsed wave sent in bursts, or a continuous-wave (such as in a Doppler system), and/or the signal may comprise a chirp that increases or decreases in frequency, linearly or geometrically. An artisan appreciates the use of these and/or other known systems as being within the spirit of the invention.
TABLE 1 is an overview of the operation of an embodiment of the invention along with a discussion of the methods steps. In particular, TABLE 1 identifies as steps some of the functions that are performed by the logging system, and notes whether these functions occur above ground, down-hole or both.
The use of the beat frequency to determine the range is a principle similar to the principle of many modern police radars, and even garage door opening systems. The greatest recorded range with a significant return corresponds to the length of the propped fracture being measured.
An artisan will appreciate other methods of coding the microwave signal, such as direct sequence coding, can be employed within both the spirit of the invention and the scope of the appended claims. For example, in the case of direct sequence coding, the outgoing signal is modulated by a digital code with tightly controlled auto-correlation properties. After the signal is propagated out to a target and is reflected back, the mixer will correlate the original signal with the delayed received signal. The source signal can then be controlled to give a correlation peak at only one particular range (delay). Thus, range gated radar measurements can be made with direct coding in lieu of using a chirp. The aforementioned are two of several ways to measure the fracture.
An example of typical diodes D1A, D1B for use in the mixer 520 include Skyworks Semiconductor SMS7630-006 diodes which are low bias diodes and provide high conversion efficiency (12 dB conversion loss) when operating these diodes D1A, D1B with zero DC bias. T1 and T2 are used to provide a DC ground to the diodes D1A, D1B, and R1, T3 and C1 provide a match to the mixer output port. Still referring to
Circuit topology could change if the frequency is changed. For example, the transmission line that could be employed with a 10 GHz system could be replaced with direct inductors in a 1 GHz system. Also, the hybrid coupler employed in a 10 GHz system could be replaced by having the mixer itself couple part of the signal to the antenna. In other words, while operation at different frequencies is substantially the same, different components may be used at different frequencies that are better suited to the particular frequencies.
The invention measures aspects of propped fracture geometry. Propped fractures provide a conductive pathway for the flow of hydrocarbon and they are designed to be stable in their environment.
Typical suitable proppants include sand, ceramics and resin coated sand and ceramics to prop fractures.
It is possible to mix additive particles/filler with proppant and a variety of additives and/or fillers can be used for determining the geometry of the fracture. The additives and/or fillers (hereinafter additives and/or fillers will be termed “particles”) can be electrically conducting, semi-conducting or electrically non-conducting. However, the particle size of the additive particles/filler should be selected to not be significantly smaller in particle size than the proppant. The use of an overly fine particle (as part of a mixture with a standard proppant) may result in a loss in fracture conductivity.
Electrically conducting particles can be used, for example at the tips of the fracture, for reflecting the electromagnetic radiation signals. Semi-conducting and non-conducting particles can be used to absorb the electromagnetic radiation signals or to propagate them during radar operations and/or imaging operations. The particles and/or proppants can be either electrically conducting, semi-conducting or non-conducting if desired. In an exemplary embodiment, the particles and/or proppants are electrically conducting and can reflect the electromagnetic radiation that it incident upon them. The electrically conducting particles facilitate the transmission of incident and reflected electromagnetic radiation. This would scatter a radar signal giving a radar return along the full fracture length. The scatter would stop at the end of fracture. In another exemplary embodiment, the particles have a high dielectric constant and can facilitate the wave-guiding of the radiation signal.
In one embodiment, the semi-conducting and/or non-conducting particles are transparent to the electromagnetic radiation signals, i.e., they permit the electromagnetic radiation signals to pass through without any substantial attenuation. In another embodiment, the semi-conducting and/or non-conducting particles are opaque to the electromagnetic radiation signals, i.e., they completely absorb the electromagnetic radiation signals.
In one embodiment, a combination of semi-conducting, conducting and non-conducting particles and/or proppants may be introduced into the fracture to facilitate the process of developing an image of the fracture. Combinations of different types of particles and/or proppants can be used to improve imaging capabilities of the process. For example, it may be desirable to screen certain sections of the fracture from the electromagnetic radiation signals to facilitate imaging of other sections. Different types of particles and/or proppants can be introduced into the fracture either sequentially or simultaneously.
Examples of electrically conducting particles are metallic particles, non-conducting particles with metallic coatings, carbonaceous particles, electrically conducting metal oxides, electrically conducting polymer particles, or the like, or a combination comprising at least one of the foregoing particles.
Examples of non-conducting particles that can be coated with metals (in order to render them electrically conducting) are polymers such as thermoplastic polymers, thermosetting polymers, ionomers, dendrimers, or the like, or a combination comprising at least one of the foregoing polymers. The polymers are generally electrically insulating but can be made electrically conducting by coating them with a layer of electrically conducting metals. In an exemplary embodiment, the conducting particles and the non-conducting particles with metallic coatings can be magnetic or magnetizable.
When non-conducting particles are coated with metals (e.g. by disposing a metallic coating upon a polymeric substrate), it is generally desirable for the coated particles to have a bulk density of about 0.5 to about 4.0 grams per cubic centimeter (g/cm3). It is desirable for the polymeric substrate to withstand down-hole temperatures.
For example, it is desirable for the polymeric substrate to withstand temperatures of up to 200° C. In an extreme case, proppants such as sintered bauxite or coated bauxite may be used at temperatures as high as about 260° C.
Examples of carbonaceous particles are carbon black, coke, graphitic particles, fullerenes, carbon nanotubes such as single wall carbon nanotubes, double wall carbon nanotubes, multiwall carbon nanotubes, or the like, or a combination comprising at least one of the foregoing carbonaceous particles. Various types of conductive carbon fibers may also be used to reflect the electromagnetic radiation.
In one embodiment, the proppants or particles may comprise ceramic substrates or polymeric substrates that are coated with an electrically conducting coating that comprises polymers, carbon nanotubes and/or carbon black. The electrically conducting coating generally has a bulk resistivity of less than or equal to about 105 ohm-cm. In another embodiment, the electrically conducting coating generally has a bulk resistivity of less than or equal to about 103 ohm-cm.
It is desirable for the conducting particles and/or proppants to have an electrical resistivity less than or equal to about 1012 ohm-cm. In one embodiment, the conducting particles and/or proppants have an electrical resistivity less than or equal to about 108 ohm-cm. In another embodiment, the conducting particles and/or proppants have an electrical resistivity less than or equal to about 105 ohm-cm. In yet another embodiment, the conducting particles and/or proppants have an electrical resistivity less than or equal to about 103 ohm-cm.
The semi-conducting particles can comprise silicon, gallium-arsenide, cadmium selenide, cadmium sulfide, zinc sulfide, lead sulfide, indium arsenide, indium antimonide, or the like, or a combination comprising at least one of the foregoing semiconducting particles.
Non-conducting particles and/or proppants include insulating polymers such as those listed above. The non-conducting particles and/or proppants and the semi-conducting particles and/or proppants referred to herein are all at least electrically non-conducting or semi-conducting. Non-conducting particles are also termed dielectric particles. Non-conducting particles or also include inorganic oxides, inorganic carbides, inorganic nitrides, inorganic hydroxides, inorganic oxides having hydroxide coatings, inorganic carbonitrides, inorganic oxynitrides, inorganic borides, inorganic borocarbides, or the like, or a combination comprising at least one of the foregoing inorganic materials.
Non-conducting particles and proppants also include electrically conducting metallic substrates or non-metallic inorganic substrates that are coated with electrically non-conducting polymeric coatings or electrically non-conducting ceramic coatings.
One exemplary class of non-conducting particles and/or proppants includes high dielectric constant particles and/or proppants. One class of non-conducting particles and/or proppants comprises non-conducting polymeric substrates that have filler dispersed in the particle. The non-conducting filler can comprise non-metallic inorganic particles, naturally occurring organic particles such as ground or crushed nut shells, ground or crushed seed shells, ground or crushed fruit pits, processed wood, ground or crushed animal bones; synthetically prepared organic particles, or the like, or a combination comprising at least one of the naturally occurring particles.
Another class of non-conducting particles is granules comprising a porous glass or ceramics that can absorb electromagnetic radiation incident upon them. Suitable granules can comprise a ferrite such as nickel-zinc or barium-ferrite, wherein the mass of carbon to ferrite is greater than 0.225. Examples of such materials are described in patent/patent application WO 02/13311. These granules have an average particle diameter of 0.2 to 4.0 millimeters. The total porosity is about 70 to about 80 volume percent. The bulk density if about 0.5 to about 0.8 grams per cubic centimeter.
The particles can have any desirable geometry and any desirable particle size distribution. The particle geometry can be platelet like, spherical, spheroidal, cuboid, conical, cylindrical, tubular, polygonal, or the like, or a combination comprising at least one of the foregoing geometries. The particles can have aspect ratios of greater than or equal to about 1. The aspect ratio as defined herein is the ratio of the largest dimension to the smallest dimension of the particle. In one embodiment, it is desirable to have an aspect ratio of greater than or equal to about 5. In another embodiment, it is desirable to have an aspect ratio of greater than or equal to about 50. In yet another, embodiment it is desirable to have an aspect ratio of greater than or equal to about 100.
The particles and/or proppants can be modified after being introduced into the fracture. For example, electrically non-conducting particles and/or proppants can be reacted after introduction into the fracture to form electrically conducting or semi-conducting particles and/or proppants. In one embodiment, the electrically non-conducting particles can be disposed upon a support prior to introduction into the fracture. The support can be a proppant, a porous inorganic substrate, an organic substrate, a fiber, or the like. In one embodiment, the electrically non-conducting particles can be coated onto the support and can exist in the form of a continuous coating upon the support. In another embodiment, the electrically non-conducting particles can form discrete particles on the support. After introduction into the fracture, the reaction converts the electrically non-conducting particles into electrically conducting or semi-conducting particles.
Examples of electrically non-conducting particles are metal salts such as metal sulfates, metal nitrates, metal chlorides, metal chlorates, metal fluorides, metal hydroxides, metal iodides, metal hydroxides, metal carbonates, metal acetates, metal bromides, or the like. The electrically non-conducting particles can be reacted with a gaseous or liquid reactant to form an electrically conducting particle. The reactants can be contained in the fracturing fluid or can be added to the fracture independent of the fracture fluid to facilitate the reaction. The fracture temperature, which is about 100 to about 250° C., can facilitate the reaction. Examples of suitable metal salts are aluminum nitrate, copper sulfate, copper nitrate, or the like, or a combination comprising at least one of the foregoing.
It is desirable for the smallest dimension of the particle to be on the order of 0.1 nanometers or greater. In another embodiment, the smallest dimension of the particle can be on the order of 10 nanometers or greater. In yet another embodiment, the smallest dimension of the particle can be on the order of 100 nanometers or greater. In yet another embodiment, the smallest dimension of the particle can be on the order of 1000 nanometers or greater.
If desired, particles having specific predetermined reflecting or absorbing characteristics different from other proppant may be restricted to being the first proppant pumped. This should ensure their deposition near the tip/end of the propped fracture (point of greatest distance from the wellbore). Thus, a first portion of proppant can be injected through the wellbore into the subterranean formation and subsequently a second portion of proppant can be injected through the wellbore into the subterranean formation such that the first portion of proppant travels to ends of fractures of the subterranean formation distal to the wellbore, wherein the first portion of proppant contains particles which reflect or absorb the source radar signal and the second portion of proppant has an absence of such particles.
If desired, the logging device could be employed in a method comprising injecting a first portion of proppant through the wellbore into the subterranean formation and subsequently injecting a second portion of proppant into the subterranean formation such that the first portion of proppant travels to ends of fractures of the subterranean formation distal to the wellbore, wherein the first portion of proppant contains particles which are nonlinear and create new frequencies, frequency distortions, or frequency disruptions from the source radar signal and the second portion of proppant has an absence of such nonlinear particles of the first portion. Use of nonlinear particles can help differentiate a reflection off a wall or turn from a signal/reflection which results from a signal from the transmitter reaching the end of the propped fracture.
Typical nonlinear materials function as a rectifier or a piezoelectric material or create intermodulation. Examples of non-linear components include lithium niobate, nickel oxide, iron oxide (ferric oxide or ferrous oxide), or copper oxide (cuprous oxide or cupric oxide).
For example, in detecting the end of a fracture using intermodulation noise the end of a fracture may be tagged using materials with a non-linear relationship between impedance and the voltage (or current) to which the tagging material is exposed. Such a non-linear relationship creates new frequencies that arise from the sum and difference of the frequencies that the material is exposed to and is called “Intermodulation”. In particular, materials that create intermodulation contain a non-linearity (e.g., a non-linear bond) having impedance that varies according to the magnitude of the voltage or current to which it is exposed. When signals at two different frequencies (f1 and f2) pass through a non-linearity they create signals at their sum and difference frequencies (f1−f2 and f1+f2). These are known as “intermodulation products”. When three signals pass through a non-linearity they create signals at the sum and difference frequencies of each pair of frequencies, plus frequencies corresponding to a number of other sum and difference relationships between them to achieve typically 6 intermodulation products in total.
Thus, in an advantageous method of the present invention such a tagging material with a non-linear relationship between impedance and the voltage (or current) to which the tagging material is exposed is sent into a fracture with a first portion of proppant to be preferentially sent to fracture tips distal to the wellbore. Subsequent portions of proppant sent into the wellbore would not include the selected tagging material of the first portion. Then, by simultaneously applying two or more different frequencies downhole to these fractures, intermodulation products can be created to create a distinctive signal or signals which can help differentiate a reflection off a wall or turn from a signal/reflection which results from reaching the end of the propped fracture.
In an alternative embodiment the tagging material is in all the proppant sent into fractures.
Ferrous metals, ferrite materials, metal salts, intermetallic species of copper, aluminum, iron, carbon and silicon are examples of materials which exhibit such a non-linear relationship between impedance and applied voltage to create intermodulation products.
1) Presence of ferrous metals in the region of high RF fields: The hysteresis associated with permeable materials and a non-linear V-I curve produce intermodulation. Typical materials are steels, nickel alloys, and nickel iron alloys, for example INVAR (also called FeNi36, is an alloy of iron (64 wt. %) and nickel (wt. 36%) with some carbon and chromium) or variations such as FeNi42. “Super-Invar” (31 wt. % Ni-5 wt. % Co-Balance Iron, possibly with some carbon and chromium) or nickel-cobalt ferrous alloys, such as KOVAR (29 wt. % nickel, 17 wt, % cobalt, 0.2 wt, % silicon, 0.3 wt. % manganese, and 53.5 wt. % iron).
2) Metals in contact: This can form an inefficient rectifier: Cuprous oxide is a p-type semi-conductor. “Tunneling” through a thin oxide layer between similar metals is another mechanism.
3) Microarcing: Non-touching surfaces in close proximity can microarc above a certain potential, especially at high temperature and altitude.
4) Electrostriction of dielectrics and magnetostriction of ferrite material: The physical stress in the material alters the material's physical dimensions.
5) Excitation of spin-wave modes in Below Resonance Ferrite Devices: By controlling the microscopic properties, the bulk properties are modified.
6) Proximity to Ferromagnetic Resonance in Above Resonance Ferrite Devices: The mechanism which causes intermodulation generation also causes the non-reciprocity in the ferrite medium.
Embodiments with Components Outside the Casing
The present invention, instead of locating the transceiver inside the casing 205 (
The driver and instrumentation 222 is connected above ground to a source 215 of cables (cable bundles) 217A, 217B. Typically, the source 215 provides respective rolls of cables 217A, 217B and the cables 217A, 217B are fed down-hole with the casing 205. Each cable 217 communicates with one or more downhole signal sending and receiving apparatus 202 located outside the well casing 205. Also, each cable 217A, 217B communicates with one or more sending and receiving apparatus 202A located outside the well casing 205. By providing the driver and instrumentation 222 above-ground the signal sending and receiving apparatus 202, 202A only has passive components capable of operation under downhole conditions. Each cable 217A, 217B would be run on the outside of the casing 205. The array of antennas would be pre-wired to the location where the cable bundle terminates to the particular piece of casing 205.
The antenna of the downhole sending and receiving apparatus 202 is functionally attached to the casing 205. This includes having the downhole sending and receiving apparatus 202 (of which the antenna is a part) directly or indirectly attached to or integral with the casing 205 or having the antenna itself directly or indirectly attached to or integral with the casing 205. For example, a border (perimeter) of the downhole sending and receiving apparatus 202 could be welded, soldered, bolted or glued with adhesive, e.g., epoxy, to the casing 205.
Each cable 217A, 217B has an outer rigid sheath 216 (
However, if desired a single downhole sending and receiving apparatus 202B may be provided with multiple antennas 512 arranged on a collar 512B (see
A variation of this is to have an antenna configuration located on the external casing surface utilizing a downhole signal sending and receiving apparatus 202A having a flexible antenna strip 512A that would cover 360° of the casing 205 external surface. This approach would also allow for a “phased array” which translates to signals looking out in 360° around the outside of the casing.
The pair of wires 302 in the cable bundle 217A, 217B is used to return data as an audio signal above ground (see
For each cable 217A, 217B, respective above ground instrumentation 222 is provided (one shown). Each above ground instrumentation 222 includes a microwave signal source (microwave frequency generator) 221, a laser driver (laser transmitter) 225, an intensity modulator 226, an audio amplifier/filter 230, mixer 250, microwave frequency generator 240 and RF spectrum analyzer 260 as shown in
This embodiment may be employed with a series of shorter (subs, also known as pups) 205a pieces of casing 205 that are not meant to be perforated on which the antenna array is positioned. This eliminates the possibility of damaging the circuitry (that is positioned on the external walls of the casing) during the perforating process. Another option, for eliminating possible damage created by the perforating process, is to pre-position the location of the perforations to insure the externally mounted antennas are not damaged during the perforating process.
Also, it is possible to use this downhole signal sending and receiving apparatus 202, 202A (
In this embodiment an initial radar signal is used to drive an optical (laser) modulator to transmit an optical signal downhole. This optical signal is converted downhole by a photodiode back into a radar signal for transmission by the multiple antennas (or antenna array). The reflected signal is provided to a mixer that also receives the initial radar signal to provide a beat frequency. From about 1 GHz to about 10 GHz is a typical frequency range for the signal to drive the modulator. The reflected signals are received back by the same multiple antennas used to transmit. However, this is not the only method that may be used. For example, if desired one antenna may transmit and another may receive.
In this embodiment the signal at the surface is still converted to optical and travels down the fiber optical cable. However, rather than sending a beat frequency on twisted pair, the reflected signal is going to be relayed back up. Repeater spacing depends on the technique and frequencies used. However, in general spacing of the repeaters will be on the order of 1 to 10 meters apart. Generally the reflected signal is being relayed back to above ground without changing its characteristics along the way, namely it is amplified, but its audio frequency is unchanged.
As shown in
An embodiment of the invention was tested with the components that would be down-hole tested in an oven at temperatures from 20 to 210° C.
The test apparatus had the components shown in
The laser transmitter 225 employed in the example encompassed all of the components required to generate a laser signal in fiber, and modulate the signal with an RF carrier. Since in use at an oil well the laser transmitter would reside above ground and would not be subject to any unusual temperatures, in these tests the laser transmitter was not in the oven. The laser transmitter included a custom driver and temperature controller board, an external intensity modulator, and a power RF amplifier/driver. An operation wavelength of 1550 nm was chosen for these tests for the following reasons:
The availability of high temperature polyimide coated single mode 1550 nm fiber.
The availability of high-RF bandwidth 1550 nm intensity modulators.
The availability of lasers with sufficient power at 1550 nm.
The availability of photodiodes with sufficient bandwidth and good responsivity at 1550 nm.
The capability to provide further optical amplification of 1550 nm signals with Erbium doped fiber amplifiers.
The RF power delivered over an intensity-modulated link can be derived from fundamental principles. The RF signal power, Srf delivered to the photodiode at peak AC photocurrent) is given by equation (1),
S
rf
=I
2
R
Lβ/2=P02ηRLβ/2 (1)
where I is the peak AC photocurrent, RL is the load impedance, Po is the average optical power, η, is the quantum efficiency of the photodiode, and β is the optical modulation index. Taking β=1 and η=0.8, results in calculating that an average optical power of 7 mW (8.4 dBm) is required at the photodiode to obtain a 0 dBm RF power output into RL=50Ω.
The typical insertion loss into a Mach Zehnder intensity modulator is just over 3 dB (optical). The loss expected in the fiber is 0.6 dB/km. Hence, in the setup for these examples, with 1 km of fiber, the loss will be 0.6 dB. Hence the laser itself must deliver at least 8.4 dBm+3.6 dB=12 dBm (16 mW) of optical power.
In particular the apparatus of these examples included a JDS Uniphase CQF 935.708 telecommunications laser. This laser is rated for up to 40 mW of optical power. This telecommunications laser includes a built in thermoelectric cooler and thermistor.
A regulated current source supply and a temperature control system were also provided to be used with the laser. The thermal control circuit used Analog Devices ADN8830, a microchip designed specifically for laser temperature control. This is a pulsed current driver for the thermoelectric cooler with PID feedback compensation. The regulated current source uses a pair of Analog Devices ADN2830 chips. This device is designed specifically for driving laser diodes.
The laser diode, driver, and, cooling controller were designed as a single board as shown in
The transceiver 510 included the components shown in above-discussed
The photodiode 505 converted optical power into approximately 0 dBm (1 mW) of RF power at 10 GHz. As shown in
The photodiode 505 of this example was a Discovery Semiconductor DSC 40S photodiode. This device is designed for 10 Gb/s telecommunications applications, and has sufficient bandwidth to receive our 10 GHz intensity modulated signal. The diode is not rated for high temperature operation but, as seen below, the data resulting from this example did not indicate any degradation at our test temperatures of up to 210° C.
The antenna 512 of this example was a wideband bowtie antenna design. The bowtie antenna was modeled as a transmission line terminated in a radiation resistance. The characteristic impedance is a function of the bowtie angle and the resistance seen at the terminal is a function of both the antenna length and the bowtie angle. The antenna was fabricated on ROGERS 4350 laminate, due to the high frequency characteristics and high temperature thermal stability of this material.
The antenna 512 was a bowtie antenna with a full width antenna angle of 90 degrees with a characteristic impedance of 198 Ohms as measured by time domain reflectometry. The best-measured match to the antenna was at 9.9 GHz where a feed-point impedance of 53-2 j Ohms at the bowtie feed-point was measured. Antennas other than a bowtie can be used, but the bowtie antenna is preferred for approximately 10 GHz operation.
The antenna dimensions are shown in
The antenna 512 was fed through a one wavelength transmission line having a characteristic impedance of 147 Ohms. The feed-line was designed to be one wavelength long at 10 GHz to avoid transforming the antenna feed-point impedance. As will be discussed below, the inventors achieved better results with a signal on the order of 1 GHz rather than 10 GHz.
The hybrid coupler of this example was a standard 180 degree circular coupler, designed to operate at 10 GHz. A photograph of the coupler is shown in
RF power sent to the LO and RF mixer ports is split between the “I” and “Q” ports on the (square) 90 degree splitter. The “I” and “Q” ports are each populated with zero bias RF mixing diodes, forming a single balanced mixer configuration. The output of the diodes is an audio frequency difference signal that is sent to the surface through twisted pair. Note that the input port contains a quarter wavelength stub; this provides a DC ground for the photodiode.
The following design, prototyping, and testing of each component, the radar system of this example was tested at down-hole temperatures of up to 210° C. The experimental setup is shown in
The continuous 1550 nm laser signal from the transmitter board was intensity modulated with a 10 GHz CW microwave tone. The modulated laser signal was routed into an oven through 1 km of high temperature single mode polyimide coated fiber. The optical power loss of the fiber was measured as 0.6 dB/km. This corresponds to only 1.2 dBe/km of electrical/RF loss. The RF power delivered to the photodiode was up to −2 dBm with a DC photocurrent of 6 mA and an optical power of 7.4 mW.
The photodiode and custom transceiver were placed inside the oven. The oven contained a small (18 cm×28 cm) window, through which the 10 GHz signal could be radiated and received. The audio frequency signal output from the transceiver was fed back out of the oven through TEFLON coated twisted pair wires. For this experiment the photodiode bias was delivered through a separate pair of TEFLON coated twisted pair wires, although in the field, this DC bias voltage would be sent down-hole on the audio twisted pair.
Since this example was not performing tests in an anechoic chamber or the actual down-hole environment, the tests were performed with a moving test target to distinguish the target signature from background clutter such as workbenches, test equipment, and building structural components. Moreover, signal processing development was not included in this example and the moving target allowed the performance of the example without additional signal processing.
The target was a 30 cm diameter fan with metal blades. The target was placed just outside of the oven window. The metal cage was removed from the front of the fan such that the RF would reflect from the moving fan blades. A picture of the setup is shown in
The oven temperature was then increased from 20° C. to 210° C. A plot of the measured return signal power vs. oven temperature is shown in
The data in
When the oven temperature exceeds 150° C., the slope became steeper. However, the solder connecting the photodiode to the receiver board was visibly melting and the SMA connector (between photodiode and transceiver) was coming out of the transceiver board under its own weight. This drop in signal was accompanied by a reduction in photocurrent. When cooled back down to room temperature, the measured signal remained degraded. After the solder connection was repaired, the signal level returned to within 1 dB of its previous level at room temperature. These results are all consistent with the melting solder that was observed.
To further verify that the melting solder near the photodiode was the only mode of failure, the photodiode was removed and a 10 GHz signal was injected directly into the transceiver port. Individual components including the antenna and diode mixer were individually heated with a 260° C. heat pencil and no performance degradation was observed.
The photodiode was then placed in the oven alone, and the RF power from the photodiode over temperature was measured. Less than 0.5 dB change in power was observed over the temperature test range of 22° C. to 180° C. These results all point to melting solder as the only mode of failure. Thus, this can be remedied by using a higher temperature solder.
It is expected that if the solder used here is replaced with a higher temperature solder, then the 0.05 dB/° C. trend will return and the output signal will be reduced by approximately 10 dB as the temperature is increased from 20° C. to 210° C. Increasing the laser drive power and/or employing good signal processing design in the receiver can counteract this reduction in power.
The system of Example 1 was designed to operate at 10 GHz with a free space wavelength of 3 cm. This short wavelength was chosen to facilitate a 24 mm form factor on the down-hole receiver and antenna.
However, it was desired to learn about the electrical properties of various proppant that could be introduced into a subterranean fracture. The real and imaginary dielectric constant of such proppants influences the electromagnetic guiding properties of the fracture. This information, as well as the size of the casing, assists in choosing an optimum operating frequency.
Thus, after completing the 10 GHz prototype design and testing, the propagation loss in a few sample proppant materials was measured (understanding the optimal frequency may not be exactly 10 GHz).
The propagation experiment example was set up as follows. Two known lengths of 5 cm inside diameter Polyvinyl Chloride (PVC) pipe were capped at both ends after being filled with proppant. Slits were cut in each end of the pipe, allowing our broadband bowtie antenna to be inserted at each end. A photograph of the setup is shown in
Transmission loss, return loss, and delay between antennas were measured between 5 GHz and 15 GHz using a 20 GHz network analyzer. Table II summarizes the GHz network analyzer measurements:
Referring to Table II, comparing the results from the air filled PVC to the proppant filled PVC, indicates the presence of the proppants has a dramatic effect on the transmission loss (in dB).
Measurement of transmission loss through the air filled PVC tube shows a much lower loss than one would get from purely free space transmission, indicating that even the air filled PVC tube provides dielectric guiding. The measured 10 GHz propagation losses (L) in PVC tubes filled with air, sand, and CERAMAX curable resin coated ceramic proppant, available from HEXION Specialty Chemicals, Houston, Tex. are estimated by the following equations (2)-(5):
L (dB/m)=[S21(l1)−S21(l2)]/(l2−l1) (2)
L
air (dB/m)=[−48 dB−(−53 dB)/(2.47 m−0.42 m)]=2.44 dB/m (3)
L
sand (dB/m)=[−39 dB−(−44 dB)/(2.47 m−0.42 m)]=2.44 dB/m (4)
L
CERAMAX (dB/m)=[−45 dB−(−76 dB)/(2.47 m−0.42 m)]=15.12 dB/m (5)
The CERAMAX proppant material, with 15 dB/m, has far too large an attenuation to be of practical use in a radar system. The air filled tube and sand filled tube both exhibited 2.44 dB/m attenuation. This ratio indicates that the PVC tube itself can be a significant source of loss when it is the only guiding dielectric material. However, note that sand filled tube produces a much longer propagation delay, indicating that the sand, not the PVC, is the dominant guiding material when the tube is sand filled.
TABLE III summarizes measurements of the propagation delay that were also made in order to estimate the effective dielectric constant of the proppant. The measured delay does include the delay required to radiate and receive the signal, therefore, the propagation velocity is best estimated from the longer, 247 cm measurement. The delay can be compared to the calculated free space delay to determine the relative velocity factor and index of refraction as shown in TABLE III. The air filled PVC tube alone has a velocity of 0.8×c. The addition of sand reduces the velocity to only 0.55×c. The loss in the CERAMAX proppant was too high to make an accurate measurement of the delay and velocity factor.
Measurements of the antenna return loss were also made to diagnose any detuning of the antenna due to dielectric loading. In the presence of both proppants, the antenna resonance is shifted below 10 GHz and the return loss becomes greater than our acceptable value of −15 dB.
Table IV summarizes results from measurements made at a frequency of 2 GHz. This experiment was done to evaluate the benefits of operating at lower frequencies. The 2 GHz measurements were made with the same antenna designed to operate at 10 GHz; therefore the antenna was electrically short and mismatched at 2 GHz. Furthermore the receiving antenna in the short 42 cm tube is not in the far field at 2 GHz. Hence, these results cannot be used to make a very accurate measurement of attenuation per unit length.
Nonetheless, it is noted that the overall transmission loss over 247 cm is significantly better than in the 10 GHz case in spite of the large antenna mismatch. Hence, the inventors conclude a reduction in frequency of operation to well below 10 GHz will significantly reduce overall transmission loss. This is consistent with theory that suggests approximately a 10 dB reduction in absorptive losses per order of magnitude reduction in frequency.
These results point toward the conclusion that the 10 GHz operational frequency results in unacceptably high transmission losses (if the targeted propped length to be measured is >30 feet), and that the losses can be improved by reducing the frequency. The choice of operating frequency will be based upon a trade-off between antenna form factor efficiency, and frequency dependent transmission loss in the proppant and the distance that can be measured.
The 10 GHz frequency is not preferred to measure a substantial propped fracture length (>20 ft) but has benefits if the use is to identify when a casing perforation is connected to a propped fracture of substantial length.
TABLE IV shows the success with some of these items:
The new down-hole radar-logging device overcomes two important challenges. Namely, the radar can operate at high temperatures, and the microwave signals can be sent through several kilometers of fiber to reach the down-hole transceiver with minimal loss.
The invention solved the propagation problem by employing a high temperature low loss (about 1.2 dBe/km) fiber optic signal feed. The high temperature problem is solved by using a passive electronic design for the down-hole transceiver. The unique design uses passive electronic parts down-hole, including a photodiode and mixing diode. Both components performed well at high temperatures in this example.
Tests were preformed demonstrating the invention at temperatures as high as 210° C. This demonstration included all down-hole electronics and optics. For the first demonstration we used a moving test target. It was then determined that additional signal processing would be required to view a static target in the cluttered lab environment.
Temperature tests showed some degradation in signal level at high temperatures. Reduction of signal above 150° C. was found to be due to melting solder, a problem that will be easily remedied by use of a higher temperature solder in the next design. Additionally, a signal reduction of 0.05 dB/° C. was observed. This drop is acceptable since it can be countered by increasing the optical drive power.
There were some propagation measurements performed to determine the expected propagation loss in some sample proppants. It was determined that at 10 GHz, signal attenuation is too high to provide a practical range measurement over more than a few meters in the CERAMAX proppant. The sand proppant may be usable over a larger range at 10 GHz with additional signal processing.
Preliminary propagation measurements made at 2 GHz showed the propagation loss is lower (better) than at 10 GHz.
This example used a simulated fracture sandstone model to perform propagation tests with proppants to determine the optimal operational frequencies to be used in fracture length detection. The test frequency range used was from 250 MHz to 3 GHz. Sand and CERAMAX proppants were used in the sandstone test.
The network analyzer 620 generates a signal in the frequency range from 250 MHz to 3 GHz. This signal is amplified and sent to the transmitting antenna (element 601). The received signal is measured from the receiving antenna (element 600). Before the measurement, the system is calibrated to account for loss in the cables and any internal losses in the instrument.
The antennas 600 and 601 are mounted parallel to the sandstone sides. The antennas are linearly polarized, and the azimuthal radiation pattern is approximately ominidirectional around the neck of the antenna.
The fracture model 651 simulates different environments. With different kinds of proppants, we have the following testing scenario matrix as listed in TABLE V:
The testing media “air” connotes a simulated fracture that is not filled with any proppant. Tests were done with wet and dry proppants, because water may be present in the actual porosity of the fracture. The sandstone was saturated with water for three days prior to the test in order to simulate water content in the surrounding formation.
The propagation test is to measure signal loss from 250 MHz to 3 GHz in the proppant media under different conditions. In theory, the power received by the fixed receiving antenna can be related to the location of the transmitting antenna by equation (6):
P(z2) [dBm]=P(z1) [dBm]−α [dB/m]×(z2 [mm]−z1 [m]) (6)
where P(z1) is the received power when the separation between transmitting and receiving antennas is z1, P(z2) is the received power when the separation between transmitting and receiving antennas is z2, and where P(z1) is the received power when the separation between transmitting and receiving antennas is z1, P(z2) is the received power when the separation between transmitting and receiving antennas is z2, and α is the attenuation in Decibels per meter.
The S parameter measured by the network analyzer measures the power loss (in dB) between the transmission and reception ports, such that equation (1) can be rewritten as equation (7):
S
21(z2) [dB]=S21(z1) [dB]−α [dB/m]×(z2 [m]−z1 [m]) (7)
Using equation (7) and measuring the S21 at z1 and z2, the attenuation per unit length, a can be determined. Equation (7) is a good estimate only when the transmission mode is a guided wave and when z1 and z2 are large enough to ensure that measurements are made in the far-field.
Using the above-described simulated fracture sandstone model, tests were performed for a specialized down hole radar. In the field, the radar signal will be propagated through a down hole fracture, which will be reinforced/filled with proppant. The model was developed to obtain electromagnetic propagation characteristics in this environment.
It was a goal of the test to determine the attenuation characteristics of the electromagnetic modes that would be supported in a fracture with proppant.
As described above, the simulated fracture sandstone model is outfitted with slots allowing the insertion of wideband antennas 600 and 601 into the simulated fracture at discrete locations. Network parameter (S21) measurements were made at each of these locations to determine propagation characteristics.
The measurements were performed under both wet and dry conditions. The wet condition simulates water filling the porosity of the proppant pack.
Propagation through the fracture model will occur in two types of modes. First, a portion of the launched power will travel as a guided mode as described by equation (7).
However, a portion of the launched power may also travel as an unguided mode, expanding in a spherically shaped wave front as it traverses the fracture. However, propagation loss of this unguided mode is expected to decrease very quickly according to the following equation. Therefore at large distances, the guided mode described by equation (8) is more important.
L [dB]=20×log(r2/r1) (8)
where L represents the signal path loss in dB between radius r2 and radius r1 when the transmission antenna is placed at the origin.
A receiving antenna 600 was embedded at the origin of the model (as shown in
The receiving antenna 600 was a wideband high pass antenna (1.1 GHz 3 dB roll-off). This antenna 600 was connected to Port 2 (element 640) of the network analyzer 620. Two additional transmission antennas 601 were then embedded in two of the three available antenna slots. The transmission antennas 601 were identical wideband high pass antennas (606 MHz 3-dB roll-off). The transmission antenna 601 was connected to Port 1 (element 630) of the analyzer 620 (
Two of the following three S parameters were then measured over frequency: S21-A, S21-B, S21-C. Here S21-A implies excitation of the transmission antenna (601) at Slot A and reception by the receiving antenna (600) placed at the origin.
Before performing any measurements, the network analyzer 620 and antenna feed lines were calibrated by use of a 50 Ohm load, short circuit, and open circuit termination. The calibration removes the effects of frequency dependent loss in the transmission line and internal to the analyzer.
After embedding the antennas 600, 601 into the model, the S21 parameters were measured with an empty fracture (air). The fracture was then filled with proppant and the measurements were repeated. The proppant was soaked with water and the measurements were again repeated. Finally, the foam surrounding the fracture was soaked with water and the measurements were repeated.
Measurements were performed at two sets of antenna slots. In the first experiment, S21-A, and S21-B were measured.
In the second set of experiments S21-B, and S21-C were measured. By comparing the measured levels at either pair of locations, attenuation per unit length was estimated according to (7).
Choice of measurement location involved some trade-offs. In making the S21-A measurement, Slot A is only 0.12 meters from the source. This implies that at our lowest frequencies (<800 MHz), the antenna at Slot A and the receiving antenna will exhibit near field coupling. Therefore, the S21-A measurements are more meaningful at higher frequencies. Results for frequencies below 800 MHz were not recorded in the S21-A data TABLES VI and VII.
Similarly, in making the S21-C measurements, the signal loss between the origin and Slot C was high enough to attenuate the highest frequencies below the noise floor of the analyzer 620. Thus, the S21-C measurements at the lower frequencies provide more meaningful data. The measured signal level was at least 15 dB above the measured (post calibration) noise floor for the data to be recorded in the tables of results. For these reasons the data recorded in the tables for each of the experiments is at different set of frequencies.
The measured data for S21-A and S21-B are presented in Table VI. A sample network analyzer screen photograph from this experiment is shown in
The measured data for or S21-B and S21-C are presented in Table VII. As discussed above, the most meaningful measurements from Table VI are those made at the highest frequencies, while the most meaningful measurements in Table VII are those made at lower frequencies. The reduction in attenuation with increased frequency in Table VI is consistent with near field effects occurring at the lower frequencies.
The highest frequency measurements from TABLE VI, shows an attenuation of about 5 dB/m in the 2-3 GHz range.
The lowest frequency measurements from TABLE VII for both (dry) air and PR600 proppant, shows a loss of 4-6 dB/m in the 250 MHz to 1 GHz range. The Ceramic proppant is somewhat more lossy with a 6-12 dB/m loss depending on the frequency and conditions (wet or dry). The air measurement with wet foam also presents us with a higher loss of 7-8 dB/meter.
The present invention provides a significant advantage over attempted radar logging devices of the prior art. The high temperature problems associated with the non-operation of active components, and/or possible attempts to cool these components have been solved by using just passive components down-hole. The transceiver comprises a photodiode, a diode mixer, and a hybrid coupler (as well as the antenna). No amplification is required of the reflected signal as it will be mixed and travels back along a pair of wires as a beat frequency comprising an audio signal that is a fraction of the original microwave frequency.
However, if desired an amplifier may be added to amplify the radar signal sent into the fracture and/or the signal returned to the surface.
It is also understood that an artisan will appreciate that another advantage of the present invention is that the spectrum analyzation can be performed above ground without having the down-hole constraints to overcome.
It is also clear that, although the invention has been described with reference to a specific example, a person of skill will certainly be able to achieve many other equivalent forms, all of which will come within the field and scope of the invention.
For example, while the generation of the modulated light signal occurs above the wellbore, it is possible that this signal could also be generated down-hole. To elaborate, the laser transmitter 225 and the modulator 226 (see
In this example two test wells, labeled as F1 and F2 were drilled. A custom made steel pipe was put into each of these wells. The steel pipe had a dimension of 7 inches outside diameter (od), and 6.5 inches inside diameter (id) and had two 8-foot long and 0.5-inch wide slots cut on opposite sides 8 inches from the bottom of the pipe. For the 8-foot slot, metallic joints were installed to hold the slot every 2 feet along the slot cut.
The steel pipes were put about 32 feet down into the wells and they were put about 100 ft apart. Hydraulic fracture jobs were performed on each well and the dimensions of the fractures were mapped out by a method other than employed in the present invention.
Although the wells were supposed to be dry, it was found that substantial water remained in the wells. In well F1, the water level was found to be more than 13 feet. In well F2, the water level was found to be more than 6 feet. Instead of pumping the water out from the wells to make the wells dry, it was decided to make the tested tool waterproof Waterproofing materials were put around the joints of the PVC pipes and around the composite pipe so that no water could get into the radar system. These waterproof protections shielded the electronics from being immersed in water. Also, as the majority of the readings collected were with the tool totally immersed inside the water, this proved that even with the tool operating in a water saturated environment, it was able to collect effective RF signals for determining the fracture information.
The RF system was enclosed in a composite pipe. At the test site, several PVC pipes were connected to the composite pipe housing the electronics of the tool 600 to permit raising and lowering of the tool 600. The whole tool unit (including the tool and PVC pipe) was about 40 feet long. The PVC pipe was attached to the composite (holding the antenna and other components) so that we could control the height of the tool in the casing. The PVC was marked at regular intervals so that it was easy to determine (by referencing the marking) where the antenna was located (as far as vertically) within the well. The PVC pipe was also marked to show the direction that the antenna was facing.
A scissor lift truck was employed to stand up and lower the tool into the wells to avoid connecting or disconnecting the pipes. The tool was first lowered all the way down into the well, and then the tool adjusted to the specific height to collect data.
For the angular measurement in the direction tests, the casing for well F1 was labeled as shown in
For the angular measurement in the direction tests of well F2, the casing was labeled as shown in
The testing equipment was setup as shown in
The test equipment included an 81110A pattern generator 628 having a first channel 629A and a second channel 629B as well as a first E4438C vector signal generator 632 and a second E4438C vector signal generator 634. The test equipment also included a computer with sound input 636, an audio amplifier 638 and a laser driver 640. The first and second vector signal generators 632, 634 each had both I and RF ports. The laser driver 640 had a DC port, two RF ports and a laser port.
The first channel 629A was connected to the I port of the first signal generator 632 and the second channel 629B was connected to the I port of the second signal generator 634. The RF port of the first signal generator 632 was connected to an RF port of the laser generator 640. The RF port of the second signal generator 634 was connected to another RF port of the laser generator 640. The test equipment also included a down-hole radar transceiver 646 connected to a slot antenna 648. The DC port of the laser driver 640 was connected by twisted pair of wires to the down-hole radar transceiver 646. The laser port of the laser driver 640 was connected by SMF-28 fiber optic cable to the down-hole radar transceiver 646.
The test equipment 630 also included a computer with sound input 636 connected to an audio amplifier 638 which was connected to the down-hole radar transceiver 646 by a twisted pair of wires.
While fixing the range gate setting, the IF amplitude measurements were made at the following frequencies: 1 GHz, 1.01 GHz, 1.02 GHz, 1.03 GHz and 1.04 GHz.
The average IF amplitudes at these frequencies were recorded as the corresponding radar return. The range gate setting was then changed and the measurement at the four frequencies was repeated. The data was then plotted as amplitude vs. range.
The range gate was set by adjusting the bitwise offset between the two sequences. Hence, the range granularity corresponds to the period of one bit. Two bit periods were used in these experiments, 10 ns (100 Mb/s) and 12.048 ns (83 Mb/s). The delay setting maps to range according to the following equation (9):
R [m]=(n [bits]×T [ns]+D [ns])×f×0.15 (9)
Where “n” is the bitwise offset between sequences, “T” is the bit period, “D” is the residual offset between the codes when the bitwise delay is set to zero, and “f” is the velocity factor.
The velocity factor “f” is included to account for a reduction in the propagation velocity within the fracture when compared to free space. The velocity factor is a function of the dielectric constant of the proppant, the fracture cross section size, and the dielectric constant of the material surrounding the fracture. The velocity factor is estimated to be within the range of 0.5 to 1.0.
It is also useful to know the range resolution that can be obtained with a particular bit period setting. The range resolution corresponds to the minimum lateral feature size that can easily be resolved. The range resolution is estimated according to the following equation (10):
ΔR [m]=f×T [ns]×0.075 (10)
With a velocity factor of f=1 and a bit period of 10 ns, the range resolution is 0.75 m.
The range gated radar return from the well F1 was measured by placing the transceiver 646 of the tool 600 near the center of the fracture at a depth of 8.8 meters (29 ft.) The pattern bit rate was set to 100 Mb/s.
As shown in
Signal data acquisition for this fracture continued through 10 bits of offset (about 15 m), and no additional signatures were observed within 10 bits.
Well F1 was measured again, but this time using a 83 MHz pattern bit rate.
To be certain that the M-sequence was not contributing to observed features the cross correlation of the sequence was checked on the same computer being used to load the pattern into the pattern generator. At an offset of zero bits the pattern correlates to 211−1=2047. At offsets other than zero (the range of 1 to +16 bits were tested), the pattern correlates to −1 as expected. Hence, the pattern peak to side lobe ratio, S, is given by the equation (11):
S [dB]=20×log(MAX/MIN)=20×log(2047/1)=67 dB (11)
The actual contrast will be smaller (<60 dB) due to limited carrier suppression in the signal generator modulation.
To verify the observed signatures at about 8 meters in
As shown in
The range gated radar return from well F2 was measured by placing the transceiver 646 near the center of the fracture in well F2 at a depth of 10 meters. The pattern bit rate was set to 100 Mb/s.
A signature also appears at about 8 meters, although it is less distinct than in the first fracture. It is believed this fracture has the same 90 degree aspect rotation at a range similar to well F1.
Well F2 was measured again, but this time the transceiver 646 was raised up by about 0.3 meters such that the transceiver 646 was still in communication with the fracture, but at a different lateral position.
As shown
As described in above, the fractures in the test wells were created through a steel casing with two 8-foot slots cut in the opposite direction 8 inches from the bottom. Because of the geometry of the slots, the directionality of the signatures was relatively weak to be detected.
To calibrate the tool against one known fracture length one slot on one side was “shorted” by conductive materials to improve its directionality and the antenna aligned to the slot on the opposite side, and then the tool was rotated 180°. To “short” one slot on one side, a steel wool blanket and wires were wrapped around the tool 600 as shown in
The steel wool wrapped tool was put in well F2 at two different depths: 26 feet-10 inches and 28 ft. At 26 feet-10 inches, the radar signals was measured in two different directions: T11 and T20.5. At 28 ft the radar signals was measured in only the T11 direction.
Only one signature, at about 25 feet (8 meters), was detected in the direction T11 at a 28 foot tool depth. However, 3 signatures, about 25 foot (8 meters), about 45 foot (18 meters) and about 60 foot (21 meters), were detected in either directions T11 and T20.5 at a 26 feet-10 inches tool depth. This may have occurred because the steel wool wrapped tool 600 was first put in at the depth 26 feet-10 inches. The steel wool may not have been as “electrically tight” during the first set of readings being collected at depth 26 foot-10 inches as it was during the second set of readings at a 28 foot tool depth. In other words, there might have been small gaps along the slot where the steel wool was supposed to “short”. As the tool 600 was moved up from the well F2 to rotate it and put it back into the well for the second set of readings at a 28 foot tool depth, the steel wool 650 might have tightened up to give a better electrical contact with the slot.
The spectrogram results from the 28 ft readings indicated the shorter fracture, about 25 feet (8 meters), was in the direction T11 and the longer fracture signatures, about 45 feet (18 meters) and about 60 feet (21 meters), were on the other side of T11.
The present invention provided information of the estimated fracture length and directionality. For the fracture of longer length, based on the readings gathered according to the present invention, there was also a fracture structure which produced the about 60 foot (21 meter) signature. Furthermore, this fracture structure was formed after the fracturing job and was like a branch out from the fracture of length of approximately about 45 feet (18 meters) on the opposite side of the fracture of shorter length, about 25 feet (3 meters). With the wool wrapped tool it was determined that the propped fracture corresponding to 25 feet was on the side of the F-2 facing F-1. This meant the longer fracture measurements were on the opposite side. Resistivity measurements established limits to the propped fracture length that were achieved during the fracturing treatments. The only way to obtain a length of 60 feet was for a wave to be traveling on a diagonal from the bottom of the propped fracture (at the wellbore) to the tip of the pancaked fracture located nearer the surface (on the side looking away from F-1).
Based on the data collected, the wells were profiled according to their characteristics.
The above-exemplified tests on wells F1 and F2 show equipment could be set up on the site to perform downhole tests. Even though the configurations and the environments of the test wells were different than originally planned, the tool was sufficiently flexible that it could be modified to make changes to the equipment and setup to overcome these challenges. Among all adjustments, an important one was the waterproof protection for the electronics. This adjustment, permitted putting the tool down into the water remaining inside the wells, and collecting RF signals through such a water saturated environment. After days of signal data acquisition and analysis, we were able to match the radar signals we collected to the characteristics of the fractures of the test wells.
In the first well, well F1, using the tool according to the present invention permitted matching its radar signals to the fracture structures. The fracture structures were planned to be vertically oriented of plane like fractures of length of 35 feet. The radar signals from the present invention, however, found that there were RF signatures of structures of around 25 feet (8 meters). It was later found out from another method that these signatures matched a ‘pancake’ like fracture. This “pancake” like structure was not planned. In fact, it was produced because the geology of the ground and the fracturing process was too close to the surface. The discovery of such “pancake” fracture structure from the present system is highly significant. The positive test results from the matching fracture structures not only validate the whole system architecture in a downhole environment, but also validate the usability of the signals measured downhole for determining fractures.
The tests in well F2, also found positive results for the fracture measurement. These tests found 3 signatures: about 25 feet (8 meters), about 45 feet (18 meters) and about 60 feet (21 meters). However, because of the way the slots were cut in the casing, directionality of the fracture signals could not be cleanly determined. To overcome this issue steel wool was used to wrap around the tool so that a slot on one side could be “electrically shorted” when collecting the signals using the tool. This version of the tool showed directionality with the measurement at the tool depth 28 feet for well F2. The readings from this tool indicated the signature at about 25 feet (8 meters) was not in the same direction as the signatures at about 45 feet (18 meters) and about 60 feet (21 meters). Another important fact was that the invention provided fracture information in terms of length and directionality.
Converting the to be directional was key to getting confirmation that the tool could accurately measure a propped fracture length because it enabled us to focus on one signature that had been measured by other means. Coupling one signal to a verified length also allows the velocity of the signal traveling through the proppant pack to be calculated.
This example repeats experiments of the Example 4 in-house fracture model but used a National Instruments signal generation setup employing PXI-based signal processing hardware and measures a known length fracture. The optical chain of the device for this example was made up of the above ground optical equipment such as a laser, modulator, ED amplifier, and a downhole RF transceiver.
A laboratory model of a down-hole fracture was set up. The laboratory model is constructed from ABS cylindrical plastic pipe 668, 18 cm in diameter 4.2 m in length and having a longitudinal line along line “z”.
The pipe 668 was filled with HEXION ceramic proppant suitable for use in a down-hole environment. The pipe 668 contained several access ports for inserting a round metal target with a diameter of 15 cm into the pipes. The ports are located at A, B, and C as shown in
The transceiver 670 and its antenna (about 8.5 cm outside diameter copper cylinder) were placed in front of the plastic pipe 668. The orientation of the slot in the transceiver antenna is indicated by a white dot in
A laser driver and PXI-based signal processing hardware was utilized. The transceiver 670 was positioned as shown in
This example employed an antenna 680 having a diameter of about 3.0 inches. The overall length of the antenna was 15 inches and it had a RF slot having a length of 5.9 inches. The antenna 680 was fabricated out of brass.
It should be apparent that embodiments other than those specifically described above come within the spirit and scope of the present invention. Thus, the present invention is not limited by the above description but rather is defined by the claims appended hereto.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/US2007/019815 | 9/12/2007 | WO | 00 | 12/2/2009 |
