SPE 14254 SPE Society of Petroleum Engineers Acoustic Determination of Producing Bottomhole Pressure James N. McCoy, SPE, Echometer Co. Augusto L. Podlo, SPE, U. of Texas Ken L. Huddleston, SPE, Echometer Co. ‘Summary. This paper discusses the acoustic determination of producing bortomhole pressure (BHP). Two different techniques are presented for wells that have liquid above the formation and gas flowing upward through the gaseous liquid column. One technique involves the acoustic measurement of the liquid level and the casing-pressure buildup rate when the casinghead valve ie closed. When these data are used along with an emy ly derived correlation given here, the gradient of the gaseous liquid column in the annulus can be obtained. This technique offers a reasonably accurate procedure for determining the producing BHP fof a well by acoustic means. The second method involves two acoustic measurements. A backpressure valve is used in the casing head to depress and to stabilize the liquid level at two podtions while the well is produced at a constant rate. The gradient of the sateous liquid column is then calculated and extrapolated tothe formation depth. ‘This paper discusses results from the field testing of numerous wells where the actual gradients of gascous liquid columns were ‘measured in 2 variety of casing/tubing sizes, oil gravites, gas flow rates, and pressures. Introduction ‘The producing-rate efficiency ofa well can be determined with the ‘curve of inflow performance relationship, which requires . Textniques for de~ termination of static BHP's by acoustic means have been presented in Refs. 2 through 5. These techniques have proved to be su ciently accurate for most conditions. ‘The producing BHP is the sum of the surface casing pressure plus tie pressure from die column of Quid in the annulus ‘The fluid distribution inthe annulus is function of the producing condition ofthe particular wel. Three situation are generally found in the field: (1) the liquid level is at or near the formation and casinghead gus may or may not produced; (2) the liquid level is fbove the formation und casinghead gas is ot produced; and (3) the liquid level is above the formation and casinghead gas is produced. Fig. 1 illustrates these three cases. For Cases A and B, the pressure distribution is well defined from a measurement of the Dressure a the surface, a knowledge of te propertes of the ful, ‘andthe position ofthe liquid level. Case C, on the other hand, in: volves the uncertainty ofthe gaseous liquid column gradient as a result of the annular gas flow. ‘Liguia Level at Formation (Case A). The casinghead pressure ‘constitutes the major portion of the producing BHP in normal-depth ‘wells hecause the prescare from the gas column i relatively mall Even when gas is being vented, the frictional pressure losses are ‘minimal. BHP calculation is undertaken from a measurement of ‘the casinghead pressure, the knowledge ofthe gas composition, and ‘the temperature distribution as described in Ref. 2. Or the BHP ‘alenlation can he performed by a computer pengram given in Ref 6. This program also includes Cases B and C. ‘The liquid level will always be atthe tubing perforations when a well is being produced wit the casing valves closed and free gas is flowing from the formation. Liguid Level Above Formation Without Free Gas Flow From Reservoir (Case B). At tablized producing conditions, the Hiquid above the tubing perforation is 100% oll. This producing BHP is calculated from measurement ofthe surface casinghead pressure, measurement af the depth tthe liquid level hy an acoustic survey and a knowledge of th oil and ga properties. Deals ofthe calcu: lation are given in Ref. 2. Liquid Level Above Formation With Casinghead Gas Flow (Case ©). This condition results ina gaseous annular liquid column. At iabilized producing comitions, dhe oil im dhe easing annulus becomes saturated with the gas tht is continuously flowing to the surface, Consequently. if gas is being vented atthe surface at a constant rate, free gas is being produced from the formation si- ‘multaneously withthe oil, Generally, most ol is produced through {he pump while most free gas is produced up the easing annulus. BHP calculation is undertaken from a measurement of casinghead pressure. knowledge of oil and gas properties. and an estimate of, the oi! fraction in the annular liquid. The fraction estimate is re- (quired to obtain the gradient ofthe gas/iquid mixture. This problem has receved considerable anenion by numerous authors.” These techniques involve the determination of the gas flow rate up the annulus and, in tur, the calculation ofthe amount of liquid present in the gaseous liquid column by use of such well conditions as casing/tubing sizes, liquid properties, and pressure. All these ‘methods are based on a combination of theoretical ana empirical ‘models and yield different results fora given set of conditions as shown hy Kabir and Hasan. 10 ‘Because ofthe disagreement of these techniques, a comprehensive field study was performed to determine directly the gradient of, ‘gaseous liquid column, ‘The wells tested during this study included casing sizes from 4.5 to Tin. (11.4 wo 18 em] and oll gravties between 32 apd 43° APL [0.86 and 0.81 g/cm). Long gaseous liquid columns of more than '5,000 ft (1525 m) were studied in wells up to 9,000 ft (2745 ml} ‘deep. Annular gas Tlow rates ranged from 13 to 120 Mct/D [368 to 3400 m/d] and oil fractions ranging from 20 to 77% were measured. The wells were located in regions of normal temper- ature gradients in the range of 0.9 to 1.2°F/100 f [16.4 to 21.9 kim}. ‘When a gaseous liguid column exists in the annulus of a well producing at stabilized conditions, the pressure at any depth in the {gateous column is independent ofthe turface pressure. This it Iustrated in Fig. 2, which is a schematic ofa well producing at three different values of casinghead pressure. The producing BHP remains unaffected by the changes in surface pressure and liquid level as Jong as the production rates through the tubing and the casing an- ‘ulus remain constant. ‘The annular pressure for three cases is plotted as a function of, ‘depth in Fig. 3. The gradient ofthe gaseous column (gas/oil mixture) ‘an be obiained from the change in liquid level and in pressure at the gas/liquid interface.ma a os, T Ve | a e { |GRADIENT =0.0962 PSI/FT Z 1 t we Zou! i e a au = ows 3 arene # ed i Bit | oman | 7 7 * a mal tr Fla. 5—Fleld recording of annular fuktloval depth corre. | | Fig. Gaseous quid column halght aaa function of pres ‘ponding to Fig. &- sure atthe gonlqud interface. Sng ™ measured total gradient, and 2 = gradient of gas-free oil. This term was then correlated directly wit the corresponding an- nular pas flow rate and annular area as shown in Pig. 7. Note that implied inthe correlation isthe effect of temperature and gas com- pressibilty factor, because itis derived with field data from regions ‘wih normal temperature gradients and wells producing casinghead {22s with specific gravity corresponding to oils in the 32-to 43°API {0.86- to D.8l-glem?) eeavity vange Direct measurement of annular gas flow in the field isa tedious process that can be avoided by estimating the gas flow from a short ‘asing-pressure-buildup test. The test 1s conducted by closing the casing valve while the well continues to pump. The rate at which casing pressure increases js measured, During the early phase of | this study, it was observed that annular gas flow rates calculated by short casing-pressure-buildup tests generally were less than the ‘ates measured with a critical How prover. This error was a result ‘of considering only the volume of gas above the liquid level and ‘neglecting the gas bubbles present in the gaseous liquid column, ‘The procedure was corrected as follows: once the gradient present ‘was determined, an effective ol fraction was calculated and, in tur, the volume of gas present in the column was estimated and added to the volume of gas above the fuid level, yielding an adjusted depth to the liquid level: Dy=Di+(I=fedls ® which is then used to calculate the annular gas flow rate 0.00008 apAD, q-——, a ar where ‘Ap = change in pressure. ‘At = elapsed time, and ‘A = cross-sectional area of annulus. ‘To facilitate use of the effective-ol-faction correlation, it has been plotted with ite aris representing the terms corresponding tthe ‘asing-pressure-buildup rate times the adjusted depth to the liquid level: )p,. Fig. 7—Echometer gasoous liquld column gradient corection user's cunve CORRECTION FACTOR Fig. 8—Comperison of Gilbert's gradtent-correction-factor curve with data points trom this experimental studTWPEA ‘iguio LeveL, [AT PERFORATIONS Wee 8 LguIo Lever ABOVE FORMATION AXON peaoucine "ANNULAR GAS: Panel Fig. 1—Fluld dletribution In the annulus of @ pumping wel. 2—Pressure relations in a pumping well with a gaseous ‘column. The gaseous oil column gradient is equal to the difference in pressures atthe two fluid-evel depths divided by the distance be- ‘ween them: a where Dr and Dz, are te depts tote gui evel a pressres ga ap respectively, ad pg a Pye Fe the pressures Fl te bette ps cour Eg, | suggest the procedure for fld determination of the gaseous column gradient: the gud level canbe depresed by increasing the casing pressure. ‘Then, ater valting unt sabllzed ow con: tition are re-established, the new iguid level fs meabured ad the fume cnn gratin! i calenlated with Fa Tew generally tke considerable time (more than 24 hours) for stabilized conditions to be extablised when an inerease in the asimghead presse occurs. uring this trasien period, the pressure buildup atte surface (caused by the accumulation of ga Fong inte the gs column from the ger lig clam) result ins presue increas nthe eanlus anda the formation. This increas causes gud to be depresed frm the csing annulus Ino the pom, ths retnctng tad ow and pssbiy causing thd backflow ito te formation During this time, te gud evel in the annul wll drop to lower level. After gus begin fo vent a the increased stabilized srface pressure, the surface gs flow rate wil stabilize atthe orginal value andthe liquid level wil stablze 2 the producing BHP reurm os onginal ve In some stances, this process may take I or more days. ‘Testing Procedure ‘With the use of backpressure valve, the easing pressure was in- creased to a specific value and then stabilized by allowing annular 4225 to vent a its original rate. When casing pressure ceases 1 in- crease, liquid from the annulus ino longer forced into the pump land the producing BHP returns toils original valve. The well in a stabilized condition and a true gradient can be calculated. During earlier tests, BHP sensors, dynamometers, and portahle well esters were used to determine when stabilization occurred, [As experience was gained, the results showed that a stabilized flowing condition was accurately indicated by a stabilized liquid level, obviating the need for this auxiliary equipment Figs, 4 and 5 show the most recent field data, obtained by microprocessorcontroled test system’? that allows data to be taken as frequenily as once a minute, thus yielding an accurate and detailed description ofthe behavior ofthe fluids m the well. The casing pressure is plotted asa function of time in Fig. 4, Measure- ‘monte wore made at three stabilized values of cainghead pressure 480 psi (551 kPa] Points 1 and 2), 180 psi {1.2 MPa] Points 3 and 4), and 250 psi (1.7 MPa} (Ponts 5 and 6). The corresponding uid level behaviors presented in Fig. 5, which shows that for ths par- ticular wel, idle! stabilization occurs in a relatively shor time, 3 to 10 hours. ‘The measured gradient includes both the hydrostatic (mixture density) and the dynamic riction and kinetic energy) components. ‘The linea behavior with depth shows thatthe total gradient remains lly constant, even though the pressure varies by a factor 5 shown in Fig. 6. Thic behavior was observed in all the field tests covered by this study. The measured total gradient was expressed in terms of the gas-tee oil gradient by introducing, 4 term defined as the effective oll fraction: So~tnelbos penne ® > Elou P t ff of oas-.0u0, onstant Bs pee a 0 100 200 300 400 500 PRESSURE ———> Fig. 3—Preseure/aepin traverse in he annulus for the cases. Iietrated in Fig. 2 al ssa ‘CASING PRESSURE, PSI. ww T © + @ Fig. 4—Fleld recording of casinghead pressure during fluld- level stablization procedure.