Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
Abstract A 500 ft vertical well was used to study slip velocity of air in mud and pressure gradients through a 2.93" annulus (5.43 - 2.50") during continuous two phase flow in flowing liquids and in stagnant liquid columns. The well was instrumented to measure liquid- and air flow rate, surface back pressure and annular pressure gradients. Tests were undertaken with a broad range of air and liquid rates, different liquid properties, and with the injection of air properties, and with the injection of air slugs at different rate combinations. It was possible to detect these slugs as they passed the pressure transducers in the passed the pressure transducers in the annulus. Results were applied to determine gas rise velocity. Correlation has been developed for gas rise velocity, which was used to estimate gas and liquid hold up. The in situ gas velocity and terminal settling velocity were determined for both dispersed bubbly flow and slug flow. The resulting pressure gradients have been compared to estimates from 8 different empirical correlations. The best results were obtained by using the Zuber and Findley correlation for holdup estimation with a gas holdup of 0.6 to distinguish the boundary between bubble and slug flow. This high transition value was mainly caused by the geometry of the well (tool joints) and partly by the rheology of the mud. A very good agreement between recorded and estimated downhole pressure was achieved, with a mean error of approximately 1% and a standard derivation of 2.9%. Introduction During drilling into a shallow gas sand or during circulating out a gas kick it is important to know the bottom hole pressure. If not known, it is difficult to bring the well under control. Few two-phase studies have been performed in large scale annular geometries. Most have involved tubing with diameters of 2" or less and short distances. The annulus of the experimental model in the present work has a hydraulic diameter of 3", which is equivalent to a 5" OD drill pipe in a 9 5/8" casing or a 8 1/2" hole and has a total length of 550 feet. The drill pipe is mounted with tool joints similar to those in the field. A problem of experimental two-phase studies is the determination of liquid hold up. Quick operating valves to shut off a representative part of the flowing mixture have been commonly applied, while local capacitance measurements have been used in more recent studies. Hold up was determined in this study through the bubble rise velocity derived from material balance considerations: Hg = (1) in which vgs is the superficial gas velocity (qg/A), and vg is the absolute gas bubble velocity in the flowing mixture, which was determined experimentally. P. 527
- Well Drilling > Drilling Operations (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring > Production logging (1.00)
ABSTRAC T The transition from stratified to slug flow in gas-liquid horizontal flow was studied in 6.35 cm and 18.4 cm ID pipes for a wide range of air and water flow rates by video-recording the flow field and by ultrasonic measurement of the location of the air-water interface. The ultrasonic imaging techniques provided an accurate means of monitoring the liquid depth and wave characteristics in stratified flows. The experimental results were checked with theoretical models of flow regime transitions, which showed that the Taitel-Dukler model of the stratified to intermittent flow transition gave a reasonable prediction, particularly in the small pipe. In the large pipe, stratified flow occurred for all conditions tested. INTRODUCTION We have been investigating the application of ultrasonic velocimetry to the measurement of volumetric flow rates in gas-liquid flow from deep, subsea wells. In this application, it would be advantageous to locate the measurement system in the subsea pipeline near the wellhead. Our preliminary results show that the efficacy of ultrasonic velocimetry in two-phase flow depends on the flow regime, with stratified flow being a preferred measurement environment compared with intermittent flow. Thus, we are interested in gas-liquid flow regime behavior in large pipes near the pipe entrance. This paper presents the results of a laboratory investigation of the conditions leading to the transition from stratified to slug flow in gas-liquid, horizontal flow. The results were tested with the theoretical model of Taitel and Dukler and the empirical flow regime map of Mandhane et al. For the 6.35 cm ID pipe, reasonable predictions of the stratified to intermittent flow transition were obtained with both of these methods. For the 18.4 cm ID pipe, both methods predicted intermittent flow to occur for some of the conditions tested, while stratified flow was observed in all cases. Previous work has demonstrated the applicability of ultrasonic imaging and velocimetry to the metering of upwards two-phase flow in vertical or inclined pipe.3 In this study we extended these techniques to horizontal gas-liquid two phase flow. It was found that the ultrasonic imaging techniques provide a means of accurately measuring the distribution of the phases in horizontal gas/liquid flow. A clear picture of the water-air interface and many of the wave characteristics, such as wave amplitude and wave frequency, can be easily obtained from the ultrasound signals. By measuring the wave velocity from the video-recordings of the flows, the wave length can also be obtained. EXPERIMENTAL APPARATUS Flow Loop Experiments were performed using a flow loop consisting of a pipeline of 6.35 cm ID and a pipeline of 18.4 cm ID (Fig. 1). Both pipes are 12.8 m long and are clear, so flow can be observed visually. Water and air are conducted to the flow loop separately through two hoses. By changing the connections between the hoses and the pipelines, either or both of the fluids can be conducted into one pipe and be exhausted from the other. In this way, flows with different flowing directions can be studied.
- Research Report > New Finding (0.54)
- Research Report > Experimental Study (0.34)
Abstract A laboratory investigation was undertaken to evaluate and compare the performance of various microprocessor-based gas flow meters. A total of 45 tests were performed with both steady- and unsteady-state flow regimes to simulate field conditions. In general, the differences in cumulative volume obtained with the seven types of metering systems studied were in the range of 2 to 7%, with the mode being between 3 and 4%. In particular, the study revealed that under unsteady-state conditions the turbine meters agreed reasonably well, i.e., less than 3% percent deviation in cumulative volume, with the orifice plate devices. Data sampling rates were examined and found to have a pronounced effect on unsteady-state flow measurements. pronounced effect on unsteady-state flow measurements. also, bearing in mind the generally high Cost of pressure sensing elements, instantaneous values of flowing and differential pressure across the orifice were affected with randomly generated errors in order to investigate the effect on cumulative flow, and thus quantity the degree of transducer accuracy necessary to measure flow under a given set of conditions. Introduction The objective of this research was the evaluation of the performance of four different gas flow measurement systems performance of four different gas flow measurement systems when installed in series in a metering run by measuring the flow of compressed air at variable rates from 35 to 905 mscf/d. The flow meters tested consisted of two gas turbine meters and two standard orifice plate meters which were connected to various data recording devices including commercially available gas flow computers, a standard three pen chart recorder, and a general purpose data acquisition pen chart recorder, and a general purpose data acquisition system. Real field conditions were simulated by allowing a wide spectrum of flow conditions to occur. Results from the various meters were compared mutually as well as with results integrated from recorder charts, which were used as a common reference. Further insight was gained in terms of the effects that higher sampling rates and larger errors have on the calculated total volume. Experimental System The laboratory system used for this study is represented schematically in Fig. 1. The Ingersol Rand compressor was generally operated at a discharge pressure between 90 and 100 psig for variable flow rates. Flow was controlled at the inlet metering run which consists of a 2-inch Camco orifice plate holder and metering tubes outfitted with a temperature sensor, a static pressure transducer, and stacked differential pressure transducers which are fed to the inputs of gas flow computer FC#1. Regulation of flow rate was accomplished with manually operated flow control valves located downstream the metering run. The majority of the tests were undertaken by flowing directly from this metering run into the production separator (4 × 10) and from there to the test metering run. The back pressure in the separator and the test metering run was pressure in the separator and the test metering run was controlled by the regulator at the outlet of the metering system and was varied over the range of 5 to 90 psig in order to cover the broadest possible range of turbine meter TFC#2 allowed by the compressor system. Also included in this experimental setup -but not shown in Fig.1- is a fully completed 500 ft deep test well which was used to simulate multi-rate flow tests, i.e., unsteady-state flow conditions. P. 291
- Reservoir Description and Dynamics > Formation Evaluation & Management > Drillstem/well testing (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring > Production logging (1.00)
- Information Technology > Hardware (0.85)
- Information Technology > Sensing and Signal Processing (0.54)
- Information Technology > Communications > Networks > Sensor Networks (0.54)
Abstract Measurement of return mud flow is the single most diagnostic indicator of potential well control and lost circulation problems during drilling operations. To date, except for electromagnetic induction flowmeters, which can only be applied for conductive drilling fluids, it has not been possible to have access to accurate measurements of return mud flow rate over a wide range of drilling fluid properties and adverse conditions such as drilling through gumbo shale, gas cut mud and other non ideal conditions. This paper describes the design, development and field testing of a full bore mass flowmeter based on the measurement of forces generated due to change in momentum of the return flow stream. This is accomplished by attaching a short J-shaped extension to the return flow line just before the shale shaker. This extension is instrumented with force sensors that measure forces caused by the flowing fluid. The first generation prototype has been extensively tested in the laboratory and in the field over a range of flow rates up to 1700 GPM and over wide ranges of mud weights. Results indicate that the flowmeter has performed successfully under adverse field conditions for extended periods of time without the need of extensive modification of the existing mud flow system. Introduction It is essential for the safety of the drilling operation that any influx of gas or liquid from the formations below or losses of drilling fluid are detected promptly and recorded accurately. Up to this date, however, flowmeters have not been able to accurately record the flow rate of the returning mud. The main reason for this difficulty is that mud is a dirty, sticky and non-linear-viscous liquid, and measured parameters such as mud level inside the return pipe, surface velocity of the mud, drag forces on immersed bodies, etc. are not directly convertible to flow rate over the practical range of mud properties. Presently there are two exceptions; one being the electromagnetic flowmeter and the second an ultrasonic doppler meter. Two important disadvantages severely limit application of the electromagnetic flowmeter: it does not work with oil based mud because the fluid is non-conductive, and since one of the requirements for correct operation is a liquid-filled pipe, the return flow line needs to be of a U-tube shape. P. 435^