Annular Pressure and Temperature Measurements Diagnose Cementing Operations

Abstract

Summary Pressure and temperature measurements were made in the Pressure and temperature measurements were made in the annulus of wells with sensors placed on the casings as they were run into the wells. The primary purpose of these measurements was to study the phenomenon of pressure reduction in the cement as it cures. This aspect of the measurements was reported in Ref. 1. Other phenomena were observed during these measurements, however, which are important to the cementing of a well and to casing design. This paper discusses three such phenomena: temperature in the annulus during cementing, loss of returns during cementing, and long-term pressure decline in a mud column above cement. The industry has made several studies over the years to improve the prediction of temperature of cement during pumping. The predictions of bottomhole circulating temperature (BHCT) from an extensive API study agree well with measured data in three of the five wells discussed in this paper. However, better predictions of cementing temperature in all five wells were obtained with a numerical method. The phenomenon of loss of returns during cementing can be very complex. Pressures in the cement during pumping may be greater than expected because the pumping may be greater than expected because the cement is rising more than anticipated (because of bypassing of mud) or for other reasons. The resistance of the wellbore to hydraulic fracturing and consequent loss of returns is difficult to predict accurately. The behavior in two wells is described. Casing design requires assumptions about the pressure behind the casing to calculate internal burst pressure. The data available from two wells indicate that hydrostatic pressure exerted by mud left in the annulus above cement pressure exerted by mud left in the annulus above cement decreases with time, and original mud weights should not be used as the backup pressure in casing burst design. Introduction Results of pressure and temperature measurements made in the annulus of wells during and after cementing operations were reported in Ref. 1. That paper focused on the importance of pressure decline in the cement column in the time interval after the cement is pumped and before the cement has cured. Observations regarding remedial cementing and acidizing treatments were reported also. The procedures and equipment used in those measurements were described and are not repeated in detail in this paper. This paper provides comparisons of measured and predicted BHCT's during cementing. The predicted predicted BHCT's during cementing. The predicted temperatures were developed from two sources: an API publication and a numerical calculation. An API task publication and a numerical calculation. An API task group provided very valuable data for predicting BHCT. The API correlation is now a part of Ref. 3. (The same data were provided in the preceding API document, RP 10B, which was superseded.) These data were collected from 78 wells in 10 states, at depths of 4,000 to 23,000 ft [1219 to 7010 m]. Numerical procedures for predicting downhole temperature during cementing also have predicting downhole temperature during cementing also have been developed over the years, the most recent being that by Wooley. This paper provides references to many earlier papers. Here, we compare measured temperatures to those predicted by the numerical method described by Sump and Williams. This numerical model was developed by fitting 28 unsteady-state circulating temperature measurements made in nine wells to derive heat transfer coefficients and effective formation thermal conductivities. The second topic concerns cementing of lost return zones (or lost circulation zones), a long-standing problem. Loss of returns often will occur during problem. Loss of returns often will occur during cementing even though the problem did not occur during drilling. Usual steps to solve the problem include the use of granular or fibrous materials in the cement or preflushes of material designed to form a precipitate and plug the channels or the fracture where returns are lost. The mechanism by which loss of returns occurs in some formations is not well understood, but in the examples in this paper the mechanism is believed to be hydraulic fracturing. A fracture is initiated when fluid pressures at the wellbore surface are greater than the tensile strength of the rock plus the compressive earth stress at the wellbore at some depth. The fluid pressure required to initiate a fracture is usually higher than the pressure required to open a preexisting natural or induced fracture or to propagate a fracture after it is formed. propagate a fracture after it is formed. The third topic discussed in this paper is the long-term decrease in hydrostatic pressure exerted by the mud column above the cement. JPT P. 2181