On-Site Diagnosis of Cement Job Problems: The Concept of Job Signatures

Abstract

Summary. Since the introduction of free-fall models for primary cementing, several companies have been monitoring cementing operations by continually measuring the rate of returns, the rate of pumping, the density of the fluids pumped, the density of the fluids returned to the surface, and the surface pressure. Measured values of surface pressure and rate of returns have been compared to predicted values obtained with cementing simulators. Discrepancies between measured and simulated behaviors that cannot be explained as normal simulation and job measurement limitations are attributed to problems that develop during the cementing operation. Although problems have been detected by this approach, in some cases it has not been possible to ascertain the exact cause of the misbehavior. Therefore, proper and timely corrective adjustments have not always been made during the job. To improve our ability to diagnose cement job problems while the cement job is in progress, we investigated six potential job problems with a cement job simulator that includes free fall. The job problems were channeling, unsuspected (unmeasured) washout, lost circulation, influx, flow restrictions, and slurry dehydration. Representative graphs show the behavior trends for each of the six potential cement job problems. The graphs show the normal (without trouble) behavior and the departure from the norm for the problem situation. Lists of generalized "job signatures" for each case are also given. Characteristics that are considered the main indicators of each job problem are highlighted. Three field cases illustrate how the concept of job signatures can be used effectively to assist in detecting potential job problems during the actual job. Introduction An earlier version of this papers looked at the information that can be obtained by continually monitoring the rates of pumping and of returns during a cementing operation. It was shown that some job problems can be detected during certain portions of the cementing job, even without having on location a predicted job behavior from a simulator. It was also shown, however, that a more detailed picture of the operation can be obtained by comparing job measured values with predicted data from simulators. The analysis was performed with a generalized materials balance around a well experiencing lost circulation or influx during a cementing operation. The analysis confirmed the value of plotting "calculated" (simulated) vs. "measured" behavior in the same graph during a job. A sizable departure of measured from calculated values during a reasonably long period of time is a clear indication that the cementing job is experiencing a problem. The Concept of Job Signatures The proposed concept of job signatures states that job problems generate distinctive, measurable job signatures that distinguish them from other job problems when compared to a "control" (without trouble) behavior. Job Signatures for Several Cementing Job Problems. Many situations can cause problems during a cementing operation. Channeling, restrictions in the pipe or annulus, lost circulation, etc., can cause the measured behavior to depart from the predicted, calculated, or simulated behavior. To obtain a better understanding of the situation, a cementing simulator was used to develop curves that show the departure of the measured from the calculated free-fall gaps caused by different potential job problems. Curves for other measurable quantities-such as surface pressure and instantaneous rate of returns-were also developed to illustrate the departure from the normal behavior. The graphs show the nominal behavior and the expected departure from the norm for the problem situation for a normal or standard cementing operation. Clearly these graphs cannot be made general enough to be representative of every imaginable cementing operation. Therefore, the knowledge that must be extracted from these simple graphs is the general signature or "fingerprint" of the departure from the norm for each job problem behavior. To try to capture those signatures, a list of generalized characteristics for each case was put together and is included in this paper. The generalized signatures and trend graphs should be viewed as additional tools developed to help the engineer or foreman on location diagnose a cement job problem while the job is in progress. These tools should be used in conjunction with all other available information about the job to arrive at the final diagnosis of the problem. Minor discrepancies between the measured and the calculated job behaviors should be regarded as expected simulation and job measurement limitations and should not be considered indications of problems developing during the cementing operation. The six potential job problems investigated werechanneling,unsuspected (unmeasured) large washouts,lost circulation,influx,restrictions in the fluid flow path, andslurry dehydration. Figs. 1 through 8 each comprise four graphs used to represent the data forsurface pressure,instantaneous rate of returns,well volume "gap," andcumulative rate of returns in each case. All the variables are plotted vs. the cumulative volume pumped at the surface during the job. Strictly speaking, only Graphs a and b are needed for the analysis. Graphs c and d are derived from Graph b. Graph d is essentially the same as Graph c. Some job problems, however, are better detected with the aid of Graphs c and d. Channeling. Channeling was simulated by comparing the results of two cementing operation simulations with different hole diameters. The amount of cement, however, was the same in both cases. The use of a smaller hole size to help simulate channeling is only a crude approximation to the real situation. Unfortunately, it is the only way this simulator can provide us with usable data. In addition, the use of a smaller hole diameter assumes that channeling is taking place from the very start of the operation (even before the cement gets to the shoe). This implies mud channeling within mud during the conditioning period before the cement job. SPEDE P. 374^