GB2257785A

GB2257785A - Method and apparatus for obtaining borehole information downhole

Info

Publication number: GB2257785A
Application number: GB9115543A
Authority: GB
Inventors: Benjamin Peter Jeffryes
Original assignee: Anadrill International SA
Current assignee: Anadrill International SA
Priority date: 1991-07-18
Filing date: 1991-07-18
Publication date: 1993-01-20
Anticipated expiration: 2011-07-18
Also published as: GB9115543D0; GB2257785B

Abstract

A method of, and apparatus for, obtaining downhole well bore information from a well being drilled with a drill string, a drilling fluid being circulated through the drill string and the well bore, the method comprising: a) generating an acoustic signal at a known position 20 inside the drill string; b) determining a parameter of the signal in the drill string; c) detecting the signal at a downhole position 22, 28 in the borehole outside the drill string; d) determining said parameter of the signal detected outside the drill string; e) comparing values of said parameter determined inside and outside the drill string so as to determine a difference therebetween; and f) relating the difference to one or more downhole characteristics of the well bore in order to determine factors affecting the signal due to the formation being drilled at any one time. The parameter may be the speed and/or the amplitude of the signal. <IMAGE>

Description

METHOD AND APPARATUS FOR OBTAINING BOREHOLE INFORMATION DOWNHOLE The present invention relates to a method and apparatus which can be used to obtain information about the downhole part of a borehole and is particularly, though not exclusively related to, a method and apparatus which can be used while the well is being drilled.

In the rotary drilling of wells, a drill string having a bit mounted on the end thereof is rotated in the well. The drill string is formed from tubular members or "pipes" linked end to end and a drilling fluid is circulated in the string, through the bit and back to the surface in the annulus formed by the drill string and the wall of the well bore. The fluid or "mud" serves various purposes, including cooling and lubrication of the bit, transport of cuttings to the surface and maintaining a hydrostatic pressure on the drilled formation to prevent collapse of the borehole or an influx of fluids, especially gas, from the formation being drilled.

In order to determine what is taking place at the region of the drill bit, which may be several thousand feet from the surface, a technique known as measurement while drilling ("MWD") has been developed. In order to perform downhole measurements, sensors are mounted in the drill string near the bit which operate to determine parameters such as natural formation radioactivity (a well-known indication of rock type); rock resistivity (showing the presence of oil or gas); temperature near the bottom of the borehole; weight and torque on the drill bit; inclination and direction of the borehole.The sensors are powered by a turbine generator mounted in the mud flow in the drill string and information is transmitted to the surface by a rotor-stator mechanism which generates a series of modulated pressure pulses in the mud which travel up the drill string to the surface where they are detected and decoded.

US 4 730 233 proposes the use of an MWD signal in the detection of fluid influxes. The infusion of fluid from the formation being drilled into a borehole is detected in an MWD system by modulating the drilling fluid stream in the drill pipe and detecting pressure variations commensurate with the modulation at the surface in the annulus between the drill pipe and wall of the well. Modulation to generate the pressure variations in the annulus is accomplished by the same pressure generating element used to generate pressure signals in the mud column for MWD measurements. The detected pressure variations are compared in phase and/or amplitude with their own near term past history or with the drilling fluid pressure variations in the drill pipe resulting from the modulation.Variations in phase or amplitude which cannot be attributed to changes in the drilling operation are suggested as being indicative of fluid invasion. However, the signal detected in the annulus will have been affected by the whole of the bore and it is not possible to extract only information about the formation that is being drilled at a given time by this technique.

The present invention resides in the understanding that a pressure wave signal generator such as that found in MWD produces not only an upward travelling pulse but also a downward travelling pulse which is reflected by the bottom of the hole both back up the drill string and upwardly in the annulus where the signal will be affected by interaction with the borehole wall. The invention proposes detecting the pulse downhole in the annulus in an attempt to determine factors affecting this signal due to the formation being drilled at any time.

Accordingly, the present invention provides a method of obtaining downhole well bore information from a well being drilled with a drill string, a drilling fluid being circulated through the drill string and the well bore,the method comprising: a) generating an acoustic signal at a known position inside the drill string; b) determining a parameter of the signal in the drill string; c) detecting the signal at a downhole position in the borehole outside the drill string; d) determining said parameter of the signal detected outside the drill string; e) comparing values of said parameter determined inside and outside the drill string so as to determine a difference therebetween; and f) relating the difference to one or more downhole characteristics of the well bore.

It will be readily appreciated that any influence in the annulus which can change the parameter can typically be the speed of the signal or the amplitude of the signal. In a further embodiment, both speed and amplitude are determined. Such a parameter of the signal can potentially be detected and quantified. Typical characteristics affecting the signal in the annulus are shear modulus of the formation, wellbore geometry, influxes of gas into the wellbore, permeability of the formation, etc.

Where the parameter to be measured is the speed of the signal, it is necessary to know the time at which the signal is generated and the time of arrival of the signal at the position downhole outside the drillstring.

The parameter of interest can be determined either by direct measurement or by calculation based on measurement of other parameters.

The method can be achieved using an MWD siren to generate the signal and one or more pressure transducers on the outside of the drill string to detect the signal in the annulus and optionally one or more transducers inside the drill string to provide a reference measurement for a signal that is unaffected by interaction with the formation.

As well as detecting the time of arrival of the signal, the amplitude of the signal in the annulus can be measured and optionally can be compared to the strength of the corresponding signal in the drill string.

As well as the speed of the signal in the annulus, the attenuation can also give information which can be used to identify formation effects.

The present invention also provides apparatus for obtaining downhole wellbore information in a well being drilled with a drill string, the apparatus comprising: a) means for generating an acoustic signal inside the drill string; b) means for determining a parameter of the signal in the drill string; c) means for detecting the signal at a downhole position in the borehole outside the drill string; d) means for determining said parameter of the signal detected outside the drill string; and e) means for comparing values of said parameter determined inside and outside the drill string so as to determine a difference therebetween.

The invention will now be described in relation to Figure 1 which shows a schematic view of an apparatus for performing the method defined above.

Referring to Figure 1, there is shown therein the bottom of a borehole being drilled in a formation 10. The drilling is being conducted using a drill string 12 and a bit 14, drilling mud 16 being circulated down through the string 12 and back to the surface via the annulus 18. An MWD siren 20 is located in the drill string and pressure transducers 22, 24, 26, 28 are mounted on the outside and inside of the string 12. The siren and pressure transducers are connected to suitable timing, measuring and encoding means to provide signals for transmission to the surface in the normal manner. Electrical power is generated by a conventional turbine generator (not shown).

In the MWD telemetry system, the siren 20 produces a continuous wave variation in mud pressure, which is modulated to convey information up to the surface. As well as the upgoing wave (a) a downgoing wave (b) is produced that is transmitted down to the bit 14 and cutting face 30 and then reflected back up the pipe (bp) and also upwardly in the annulus (ba).

The pressure sensor 22 in the annulus is mounted at about the level of the mud siren 20 although this is not essential and monitors the travel time of the signal (b-ba) from the siren and the amplitude of the transmitted wave. Additional pressure sensors 24, 26 inside the drillstem, either above the siren, below the siren or both above and below, will allow measurement of source strength, acoustic travel time to the bit and back (b-bp) and the reflection coefficient of the bit. This gives greater flexibility and allows self calibration of the measurements.

If only a single annular sensor 22 is present then travel time measurement is still possible. Given knowledge of the bit stream modulating the siren signal and a computed model of the signal generated then the delay that best matches the measured signal to the calculated signal can be found. This delay will be the time the signal takes to reach the bit plus the annular travel time. To extract the annular component from this the bulk modulus and density of the mud must be known, but even without this information relative changes may be measurable, given sufficient accuracy in the delay estimate. It will be appreciated that any changes in the total travel time will be due to changes in the annular region rather than in the drill string as the factors affecting the signal in the drill string remain substantially constant or vary predictably.For a sensor and siren twenty meters above the bit the total travel time will be about 27ms, this corresponds to 1170 of phase at 12Hz, 2440 at 25Hz. The rocks detailed in Table 1 below will introduce delays of between 4% and 25% of this.

With one annular sensor measurement of transmitted signal strength will be possible. Although this cannot be calibrated against the signal inside the string the flow rate may be continuously calculated by measuring the turbine voltage, and hence some comparison with theoretical or experimentally determined values may be possible. Even without this, for constant flow rate the variation in signal strength will convey information about the change in rock properties below the bit More becomes possible with pressure transducers inside the string. In the first case, a transducer 24 is located above the siren. If it is possible to measure the signal produced by the siren without also measuring reflected signal (bp) the task of calculating the travel time to the annulus can be achieved by either a cross-correlation or best-fit delay method.However, the signal above the siren is composed of the initial upgoing signal (a) plus multiply reflected contributions from the bit (bp) and the siren itself and possibly other components of the BHA. If the upgoing signal (a) can be separated from the reflected parts then the possibility exists to measure the acoustic travel time to the bit and back. Subtracting this from the total travel time to the annular sensor 22 gives the desired travel time from the bit to the annular sensor. In addition measuring the travel time to the bit and back to the sensor 24 enables the speed of sound in the fluid to be calculated

and thus the measurement of p to be calibrated.The main difficulty in separating the direct signal (a) from the reflected parts (bp) above the siren is that the interaction of the up going reflected wave with the siren is non-linear. The travel time to the bit and back can be expected not to vary substantially if the flow rate is constant and the mud unchanged, so a very-long period correlation could produce a peak at the required time, however the "noise" produced by non-linear effects may result in a systematic bias or hide the peak in the background. A comparison of the transmitted signal with a computed model of the signal allows the direct component to be found, and then cross-correlating the remainder allows the travel time to the bit and back to be found. By estimating the magnitude of the direct component, the annular measurement of transmitted signal can be calibrated.

For some combinations of bit nozzles and flow rates there may be little or no reflection from the bit. In this case measuring the direct signal amplitude is easier, but a computed travel time to the bit must be used in calculating annular transit times.

The use of two pressure transducers inside the string, one 24 above the siren and one 26 below, allows the easier separation of direct and reflected signals. As the siren is a vertical dipole source, the pressure wave travelling down has opposite sign to that travelling up. The wave reflected from the bit on the other hand has the same sign whether measured above or below the siren. Ignoring re-reflection from the siren, when the two sensors are the same distance from, and close to, the siren, then summing their output gives the signal reflected from the bit and, differencing them, the direct signal from the siren. The same is true if they are differing distances from the siren but phase compensation is required.It is also possible to measure the reflection coefficient from the bit and while this varies little with lithology, it may be useful for checking consistency amongst other measurements.

Apparatus for the implementation of this method requires pressure transducers with a frequency range up to about 100Hz, and on the provision of some computing power downhole. A number of pressure transducers at the same vertical position inside and outside the drillcollar are better than one, since an average can be used and this provides for redundancy against transducer failure so reducing this cause of total failure.

It is also possible to use transducers at different vertical positions. This is useful if the same transducers are used for other functions such as discrimination of reflected downgoing waves and allowing the signal travel time between sensors to be determined, so improving resolution.

The downhole computer processing required comprises mainly simple signal processing operations, with some "intelligent" supervision to select only sensible numbers. This is well within the possibilities of current microprocessors.

The two measurements that provide most information are the travel time of acoustic waves in the annulus from the bit to a pressure transducer on the outside of the drill string, and the transmission coefficient for acoustic waves at the bit. This will affect the amplitude of the wave detected in the annulus and so provide some indication of the hardness of the formation being drilled.

The speed of sound in the annulus (CA) is given by

where p is the mud density, k the bulk modulus of the mud, pL is the shear modulus of the surrounding rock (assumed uniform), d is the borehole radius and b is the drillcollar external radius.

The total travel time for the wave in the annulus is Tann = UcA [2] where L is the distance from the bit to the pressure sensor.

There are three effects that can alter the travel time Tann; changes in rock properties Q, changes in fluid properties (p, k), and changes in hole geometry d. The properties of rock can have a very large effect on wave speed, Table 1 shows CA for various rocks and a RHA occupying half the borehole (b2 = 2d2).

Table 1

rock W (Pa) cA (ms-l) shale 800 1.5e9 750 shale 1000 2.3e9 870 shale 1200 3.3e9 980 sandstone 6.e9 1130 sandstone 12.e9 1280 limestone 25.e9 1380 rigid oo 1500 For the slowest shale the annular travel time will be double that for a rigid formation. For non-uniform rock the total travel time is approximately proportional to the integral of the inverse of the wave-speeds between the bit and the sensor. As the hole is drilled from one rock to another the annular travel time will gradually change.

The fluid properties can be changed either by the addition of cuttings, or the addition of gas. Cuttings alter significantly only the density of the fluid, and the change can be allowed for in the calculations of u from Tann by direct measurement at the surface, or based on ROP. Gas will reduce the fluid's bulk modulus k by an amount that depends on the hole depth as well as the amount of gas. As the gas moves up the annulus at the mud speed or faster the travel time will quickly increase. For small downhole gas volumes the bulk modulus is given approximately by

where ko is the bulk modulus without gas, z is the depth, g is the gravitational constant, m is the surface gas concentration measured as volume gas/volume liquid and Pg is atmospheric pressure.Small quantities of gas at large depths will give a similar increase in transit time to a change of rock type, however the rate of change of the transit time should enable the two effects to be distinguished. Gas which occupies a significant proportion of the hole bottom will have a large effect on acoustic propagation. Studies have shown an increase in amplitude in the annulus, measured at the surface, as the gas is admitted to the well followed by substantial attenuation.

The permeability of the formation can also affect wave speed. With normal mudcake the effect is likely to be small but if the wellbore is under-pressurised then the effect will be larger.

Changes in hole diameter d can be mistaken for changes in formation shear modulus , however a simultaneous caliper log enables a correction to be made. The measured travel time will be slightly lower in a flowing fluid than a static one (due to a Mach number effect), this will be a small effect is easily compensated for since the fluid velocity is known.

The transmission coefficient of the acoustic wave from the drill string to the annulus will be affected by the properties of the rock face. The transmission coefficient is calculated as

where Ac if the fluid area inside the BHA, Aa is the fluid area in the annulus, Ab is the total cross-sectional area of the borehole, cc is the speed of sound of the mud in the BHA, or is a rock impedance, and X is a dimensionless constant that for a given bit is proportional to flow rate.

The annular area is normally much larger than the fluid area inside the BHA, and the borehole area is bigger again. This means that T may be significantly affected by moving from one rock type to another. Although the zr calculated from [4] may not correspond to a more usual measure of rock impedance it should be related to poroelastic properties of the rock, and increase with rock hardness. Because it is a poroelastic property it should be affected by rock properties ahead of the bit, thus possibly allowing some anticipation of changing rock type.

It is also possible to perform the method of the present invention when not drilling but circulating off-bottom such as when circulating bottoms-up. In this case, the siren signal can be devoted to providing a signal that is optimised for the acoustic logging technique described herein. Much clearer measurements will be possible as the noise created when drilling is absent and it will be possible to measure a transfer function (amplitude or phase v frequency) to the annular sensors.

Claims

Claims

1A method of obtaining downhole well bore information from a well being drilled with a drill string, a drilling fluid being circulated through the drill string and the well bore,the method comprising: a) generating an acoustic signal at a known position inside the drill string; b) determining a parameter of the signal in the drill string;

c) detecting the signal at a downhole position in the borehole outside the drill string; d) determining said parameter of the signal detected outside the drill string; e) comparing values of said parameter determined inside and outside the drill string so as to determine a difference therebetween; and f) relating the difference to one or more downhole characteristics of the well bore.
2 A method as claimed in claim 1, wherein said parameter is the speed of the signal and/or the amplitude of the signal.
3 A method as claimed in claim 2, wherein the parameter is the speed of the signal, the method comprising: a) generating the signal and determining the time of generation; b) detecting the time of arrival of the signal at said downhole position in the borehole outside the drill string;

c) determining the speed of the signal inside the drill string; d) determining the speed of the signal outside the drill string; e) comparing the speeds and relating the difference to a characteristic of the well bore.
4 A method as claimed in claim 3, further comprising detecting the arrival of the signal at a position inside the drill string and determining the time of arrival after generation.
5 A method as claimed in claim 2, wherein the parameter is the amplitude of the signal.
6 A method as claimed in claim 5, wherein the amplitude of the signal in the drill string is determined either by direct measurement or by calculation based on other measurements.
7 A method as claimed in any preceding claim, wherein the signal is generated downhole.
8 A method as claimed in any preceding claim, wherein the signal is detected above and/or below the position of generation in the drill string
9 A method as claimed in claim 8, wherein the signal is detected at a plurality of vertically spaced positions inside the drill string.
10 A method as claimed in any preceding claim, wherein the signal is detected at a plurality of vertically spaced positions outside the drill string.
11 A method as claimed in any preceding claim, wherein the signal is generated by the siren of an MWD sub, the signal detected outside the drill string being the downgoing signal reflected by the bottom of the well.
12 A method as claimed in any preceding claim, wherein the characteristic of the wellbore is the shear modulus of the formation.
13 A method as claimed in any of claims 1 - 11, wherein the characteristic of the wellbore is an influx of gas into the well from the formation.
14 A method as claimed in any of claims 1 - 11, wherein the characteristic of the wellbore is the wellbore geometry.
15 A method as claimed in any of claims 1 - 11, wherein the characteristic of the wellbore comprises one or more properties of the drilling fluid under downhole conditions.
16 Apparatus for obtaining downhole wellbore information in a well being drilled with a drill string, the apparatus comprising: a) means for generating an acoustic signal inside the drill string; b) means for determining a parameter of the signal in the drill string;

c) means for detecting the signal at a downhole position in the borehole outside the drill string; d) means for determining said parameter of the signal detected outside the drill string; and e) means for comparing values of said parameter determined inside and outside the drill string so as to determine a difference therebetween.
17 Apparatus as claimed in claim 16, wherein the means for detecting the signal at a downhole position in the borehole outside the drill string comprises a pressure transducer.
18 Apparatus as claimed in claim 17, wherein one or more pressure transducers are also provided inside the drill string above and /or below the signal generation means.
19 Apparatus as claimed in claim 18, wherein a plurality of spaced transducers are provided inside and outside the drill string.
20 Apparatus as claimed in claim 19 comprising vertically spaced transducers.
21 Apparatus as claimed in any of claims 16 - 20, wherein the signal generation means comprises an MWD siren.