Not Applicable.
Not applicable.
This disclosure is related to the field of pulsed neutron well logging instruments. More specifically, the disclosure relates to pulsed neutron well logging instruments having neutron burst and measurement timing controlled by measurements made by a detector in the instrument.
Pulsed neutron well logging instruments known in the art include instruments that have gamma ray detectors operated to detect gamma rays emitted as a result of thermal neutron capture (“capture gamma rays”) by selected elemental nuclei in subsurface formations having high neutron capture cross section. The most common of such chemical elements is chlorine, and measurements from such instruments are commonly used as a proxy for brine content in the formations. Brine content may be related to the fractional volume of pore space (porosity) in the formation and the fractional volume of the pore space that is occupied by brine (water saturation).
Measurements made by such instruments may be first used to calculate a parameter referred to as the thermal neutron decay time (tau). The value of tau calculated may then be converted to a value of the thermal neutron capture cross section (sigma) of the formation by the expression:
Σ=4550/τ
SPE paper no. 2252, Sep. 19, 1968, revised manuscript received Oct. 8, 1970 published by SPE International, Richardson, Tex. explains in detail the advantages of the “Complete Scale-Factor Method”, and explains many of the technical aspects of thermal neutron decay time well log measurements.
An “automatic tau loop” data acquisition technique, where the neutron burst duration and the detector acquisition timing gates (both starting time and duration) are all adjusted according to the thermal decay time of the formation, is fully described in U.S. Pat. No. 3,662,179 issued to Frentrop et al., and is defined as “the Complete Scale Factor Method.”
The “Dual Burst” data acquisition technique is fully described in U.S. Pat. No. 4,721,853 issued to Wraight. The dual burst data acquisition technique, as described in the foregoing patent, is used in a fixed timing instrument where the short burst duration is always about 20 microseconds and the long burst is about 150 microseconds.
The sonde 14 may be disposed similarly in a pressure resistant housing 114 configured to traverse a wellbore and to couple to the cartridge housing 112. The sonde housing 114 may contain therein the pulsed neutron generator 14F (PNG), the high voltage power supply 14E for the PNG 14F, the gamma-ray detectors 14C, 14D and some electronic circuits 14G configured for driving the PNG 14F. The sonde 14 will be described in more detail with respect to
The cartridge 12 may include any form of electrical/mechanical connector 8A at its upper end for coupling the cartridge 12 to a cable head or another instrument above the instrument 10 between the cable head (not shown) and the instrument. The sonde 14 may include an electrical/mechanical connector 8B at its lower end for coupling to another well logging instrument, or such connector 8B may be a termination or “bull plug” if no instruments are to be connected below the pulsed neutron well logging instrument 10.
The FPGA 54 controls the above described measurement loop for the sonde (obtaining detector measurement data and setting the duration of the pulsed neutron burst). The FPGA 54 may receive specific commands from the surface for safety reasons. Thus both the FPGA 54 and the safety microcontroller 52 may be configured to detect a specific sequence to start operation of the PNG (14F in
In
Counts detected in gates N1 and N2 are acquired starting after a delay of 0.1 AFTDL delay following the end of the short burst SB and are used to determine an Apparent Borehole Tau (ABT). A long duration neutron burst LB may begin at a time of 1 AFTDL after the beginning of the short bursts. The duration of the long burst LB may be equal to 1 AFTDL. Counting gate N3 may start after a time delay of 1 AFTDL following the end of the long neutron burst LB. Counting gate N3 may have a duration equal to the duration of the long burst LB and may be followed by contiguous detection timing gates N4, N5, N6 each having a duration equal to the duration of the long burst LB. There may then be time delay of 1 AFTDL, after which contiguous counting gates N7 and N8 may occur. At the time at which gate N7 begins, the thermal neutron capture count rate may have decreased to essentially zero and during gates N7 and N8 a long term activation count rate that builds in the near detector may be measured. The gamma ray detection measurements made in gates N7 and N8 may be referred to as the “background” radiation measurement.
The downhole tau loop, which regulates the overall neutron burst timing and detection counting gate timing may be controlled by counting rate data from the near detector, in a manner very similar to that described in U.S. Pat. No. 3,662,179. However, because the background is only collected for 2 AFTDL times rather than 3 AFTDL times, as described in the foregoing patent, the count rate equation that needs to be balanced becomes:
4*(N5+N6)−2*N4−3*(N7+N8)=0
The controller adjusts the duration of the neutron burst timing until the above equation condition is met. The overall measurement cycle lasts 10 AFTDL times, the long neutron burst LB may be 1 AFTDL and the measurement counting gates are either 0.1 AFTDL or 1 AFTDL in duration, as explained above.
The downhole tau loop, which in the present disclosure may be called “Adaptive Timing” provides an Apparent Formation Tau (AFTDL) which is quite accurate and precise, but an improved result may be obtained by calculating an “Apparent Formation Tau Calculated” (AFTC) from the near detector timing gates (N3+N4), (N5+N6) and (N7+N8). By using the counts from gate N3 the statistical precision of the measurement may be improved and the rate at which AFTC may change is not limited by the downhole tau loop regulation time. Even if the downhole loop (or Adaptive Timing) is not exactly locked in to the changing Apparent Formation Tau, the AFTC will be correct.
The following operations may be performed on the counts in specific counting gates in order to determine AFTC.
The Adaptive Timing loop operation may be programmed into the controller and may balance the equation
4*(N5+N6)−2*N4−3*(N7+N8)=0
In one example counts in the foregoing gates transmitted to the surface may be averaged over a 1 second sample period. So the following count rates may be transmitted to the surface:
N0, N1, N2, N3, N4, (N5+N6), (N7+N8)
F0, F3, F4, (F5+F6), (F7+F8)
(N5+N6), (N7+N8), (F5+F6) and (F7+F8) may be transmitted to the surface using the telemetry as sums because the individual count rates in the foregoing individual gates are not needed and by combining them saves bandwidth in the telemetry.
First, all the count rates may be expressed as instantaneous count rates before the dead time correction, i.e.,
N0′=100*N0
N1′=100*N1
N2′=100*N2
N3′=10*N3
N4′=10*N4
(N5+N6)′=5*(N5+N6)
(N7+N8)′=5*(N7+N8)
F0′≦100*F0
F3′=10*F3
F4′=10*F4
(F5+F6)′=5*(F5+F6)
(F7+F8)′=5*(F7+F8)
Next, the counts in each of the time windows may be corrected for detector dead time:
N0″=N0′(1−N0′*K)
where in the present example, K is the dead time per pulse and in the present example K=0.000001. Other methods for correcting detector counts for dead time are known in the art.
N1″=N1′/(1−N1′*K)
N2″=N2′/(1−N2′*K)
N3″=N3′/(1−N3′*K)
N4″=N4′/(1−N4′*K)
(N5+N6)″=(N5+N6)′/(1−(N5+N6)′*K)
(N7+N8)″=(N7+N8)′/(1−(N7+N8)′*K)
F0″=F0′/(1−F0′*K)
F3″=F3′/(1−F3′*K)
F4″=F4′/(1−F4′*K)
(F5+F6)″=(F5+F6)′/(1−(F5+F6)′*K)
(F7+F8)″=(F7+F8)′/(1−(F7+F8)′*K)
The result is a set of instantaneous, dead time corrected count rates.
One may then perform a background subtraction on all the count rates in gates other than N7, N8, F7, F8. The background may be averaged over a selected time interval, in the present example at least 21 seconds to smooth the background count rate before subtraction. First the average background may be calculated from the counting rates in gates N7 and N8, and F7 and F8 to subtract from all the windows:
BKG
—
N=(Σi−ni+n(N7+N8)″)/(2n+1)
BKG
—
F=(Σi−ni+n(F7+F8)″)/(2n+1)
wherein n represents the number of acquisition cycles. If n=10 then the background will be averaged over 10 acquisition intervals of 1 second before the i-th level and 10 intervals after, i.e., it is a balanced 21 second filter. Now the background subtraction may be performed for all the counting gates other than the background gates (N7, N8, F7, F8).
N0′″(i)=N0″(i)−((BKG—N)/2)
N1′″(i)=N1″(i)−((BKG—N)/2)
N2′″(i)=N2″(i)−((BKG—N)/2)
N3′″(i)=N3″(i)−((BKG—N)/2)
N4′″(i)=N4″(i)−((BKG—N)/2)
The reason why the background count rate is divided by 2 for the foregoing measurement gates is because the BKG_N is calculated over two gate times (N7+N8). The background counts during one gate time interval is thus half of that.
(N5+N6)′″(i)=(N5+N6)″(i)−BKG—N
No division by two is needed for the foregoing gate measurement because there are two timing gates in the represented value. Similarly for the fare detector (14C in
F0′″(i)=F0″(i)−((BKG—F)/2)
F3′″(i)=F3″(i)−((BKG—F)/2)
F4′″(i)=F4″(i)−((BKG—F)/2)
(F5+F6)′″(i)=(F5+F6)″(i)−BKG—F
It is then possible to calculate the outputs at each i-th level. First one may generate an output corresponding to the Apparent Formation Sigma derived from the Downhole tau Loop (AFSDL), and this is 4550/AFTDL. For example, if the tau time of the i th measurement is 180 microseconds then AFSDL=4550/180 which equals 25.28 capture units. This output may be used as a quality control indicator.
Next, determine the Apparent Formation Tau Calculated from the transmitted, dead time corrected count rates:
Next determine an Apparent Formation Sigma Calculated (AFSC)(i), 4550/AFTC(i). The output of AFSC may be averaged over 5 seconds for presentation on a well log:
AFSC(log output)=(Σi−ni+nAFSC(i))/(2n+1), where n=2
At this time one may also average over 21 levels the count rates N1′″ and N2′″. These averaged count rates may be used to calculate an Apparent Borehole Tau Calculated (ABTC) and they need to be averaged before taking logarithms. Also average AFTC over the same 21 levels.
N1′″(averaged)=(Σi−ni+nN1′″)/(2n+1), where n=10
N2′″(averaged)=(Σi−ni+nN2′″)/(2n+1, where n=10
AFTC(averaged)=(Σi−ni+nAFTC)/(2n+1), where n=10
Next calculate the averaged ABTC from the following equation:
Next calculate an apparent borehole sigma value:
The output to be displayed on a well log for the foregoing parameter will be ABSC(averaged). Next one may calculate a porosity indicator ratio:
RatPor(i)=(N3′″(i)+N4′″(i)+(N5+N6)′″)(i)/(F3′″(i)+F4′″(i)+(F5+F6)′″(i))
Log Output of RatPor=(Σi−ni+nRatPor(i))/(2n+1) where n=2
It may be observed how RatPor varies with limestone porosity by examining the graph in
As can be observed in
Log Output of Apparent Porosity Indicator=Log Output of RatPor̂2/2.8
The foregoing output may be an adequate indicator of varying porosity that can be calibrated in situ with a known open hole porosity value (e.g., from a well log) if such data are available. A very significant amount of response characterization and Monte Carlo modeling would be needed to have a characterized porosity response. To show the response of RatPor in limestone, sandstone and dolomite, in and 8 inch diameter wellbore having therein a 20 pound weight per foot length, 7 inch external diameter casing, one may observe such results in
Next one may use the inelastic count rate ratio (IRAT) as an Apparent Gas Indicator. The inelastic ratio may be calculated as:
IRAT=N0′″/F0′″
IRAT may be displayed as a raw on a scale of 0 to 20 entitled, “Apparent Gas Indicator.” There may be an indicator on the log (e.g., a darkened or other coded scale line) at a value of 10 and an indicator that shows an Apparent gas Indicator value of less than 10 there is a high probability of gas being present either in the borehole or the surrounding formation. As IRAT moves higher than 10 there is a decreasing probability that there is gas present.
While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims.