Producing Wells on Casing Flow - An Analysis of Flowing Pressure Gradients

Abstract

Published in Petroleum Transactions, AIME, Volume 213, 1958, pages 202–206. Abstract The Cretaceous limestone wells of the Mara/Maracaibo Dist. of Venezuela are extremely prolific producers. To maintain production on cessation of natural flow, large scale gas-lift operations were commenced involving high production rates on casing flow. For the design of these gas-lift installations it was essential to have some knowledge of the pressure gradients involved in casing flow, so that the required injection pressures optimum gas injection rates, etc., could be forecast. This paper presents a method of calculating these annular pressure gradients. Basically the method makes use of an energy balance equation coupled with an empirical energy lab factor derived from field data. The calculations have been put in a form suitable for "punch card-type" calculating machines. A set of gradient curves for Mara/Maracaibo conditions covering flow rates up to 12,000 B/D, and GOR's of 500 to 2,000 cu ft/bbl is presented. From these (curves the flowing BHP can be predicted from surface data and the vertical flow performance under varying conditions of gas injection can be estimated. The method has been applied to La Paz and Mara gas-lift operations for over three years and has given consistently accurate results. Introduction The Cretaceous limestone fields of the Mara/Maracaibo area in Venezuela are characterized by the presence Of widely distributed fissure systems. Under these conditions individual wells frequently have extremely high potentials and such wells are generally flowed on the casing-tubing annulus in order to reduce the pressure loss in vertical flow. To maintain production on cessation of natural flow, an extensive gas-lift system was planned. As a first approach to this gas-lift design problem, an attempt was made to construct a set of empirical flowing gradient curves from the available subsurface pressure data. Such a method has been used successfully for tubing-flow conditions in California by Gilbert. Unfortunately, the empirical construction of casing-flow gradient curves presents additional difficulties, as pressure bombs cannot be run in the casing-tubing annulus against high rates of flow. Thus, the full flowing gradients cannot be defined by actual measurement and the best that can be done is to obtain spot pressures at the tubing shoe. Under these conditions a vast accumulation of BHP data is required to construct a set of gradients.