Since the early 1940s casing centralization has been identified as being key to efficient mud removal and therefore to successful primary cementing. Prior to a job it is common for field engineers to spend time optimizing casing centralization using commercially available or proprietary software, in particular for highly deviated wells. However predicted casing centralizations suffer several limitations due a number of possible assumptions. For example, the mechanical response of centralizers is usually characterized in a single hole size, and different rules are used to extrapolate their behavior in different hole sizes. Needless to say, there are not too many holes that are perfectly in gauge and therefore centralization calculations in out-of-gauge sections can be questioned. With the recent introduction of a new generation of cement evaluation tools making use of flexural waves propagating vertically along casing and radiating into the annular space, it may be possible to measure the time of arrival of the cement/formation interface echo and therefore to measure casing eccentering. Therefore, one can now compare the calculated casing centralization to the one measured downhole. The objective of this paper is to review some of the current limitations of casing centralizer calculations as they are being performed in the field on a day to day basis. The analysis is supported by a comparison between measured and predicted casing centralizations on a number of field cases. Examples showing the effects of the observed discrepancies on cement placement using a displacement simulator are discussed. Recommendations to improve centralization predictions are proposed. Introduction Pipe centralization has long been identified as a key parameter for effective mud displacement and cement placementFurthermore, good cementing practices always include recommendations regarding pipe eccentricityHowever, it is only in the early 90s that more quantitative rules were introduced to account for the effect of pipe eccentricity on fluid placementAt the end of the century full 2-D simulators were made available to field engineers. Maps showing fluid concentration and the risk of mud on the wall could then be predicted on at least the most critical jobs, if not the majority in certain geographical areasSimulators show that mud removal and cement placement can be very sensitive to pipe centralization, and allow engineers to define the minimum centralization needed to achieve 100% cement coverage all around the pipe and across the zone(s) of interest. The type of centralizers as well as their distribution along the pipe can then be determined accordingly using another model predicting pipe centralization.

In a challenging well in a mature gas field offshore Germany, four different size liners with annular clearance of less than 1-in. with the open hole or previous casing were cemented. Successful cementing of these liners was critical in order to drill and complete the well before putting it into production. To avoid lost circulation during cementing, we established loss-free mud circulation rates before the cementing operation and designed cement slurries with long thickening times for placement at low rates. Additionally, we used fibers in the cement slurry, and pumped treated micronized barite drilling fluid ahead of the cement slurry to lower friction pressures and optimize mud removal. We validated cement slurry designs through a rigorous laboratory testing procedure, including go/no-go procedures in the high temperature/high pressure (HPHT) consistometer. Long thickening times were needed, especially on a 600 m long 6-in. expandable liner cement job. We tested the setting times of cement accurately with ultrasonic cement analysis and static gel strength analysis devices. Additives to reduce the risk of gas migration were included in all cement slurries. In order to optimize mud removal and reduce the chance of cement slurry getting contaminated, custom made centralizers were strategically placed in highly critical areas. We optimized spacers through a full laboratory analysis of compatibility between spacer and the low-toxic oil based mud, as well as performing reverse emulsion and grid tests. We validated the mud removal design with 2-D simulations to predict the effectiveness of mud displacement from the wellbore. Cement slurry was placed without losses on all liners. In all cases, we had positive indications that cement slurry reached up to the liner hanger by observing cement slurry in the returns back to surface. All formation integrity tests on the liner shoe were successful and allowed drilling of the consecutive section. The final production liner was logged using a radial bond tool and advanced cement evaluation processing was run on the data in order to better evaluate cement quality. Good cement bond quality was observed behind the production liner.