This paper describes and quantifies the importance of some geometrical parameters on the main output of the cement bond log, i.e. the attenuation of a sonic wave propagating along a casing to cement interface. These geometrical parameters are the cement thickness, the casing stand-off, the percentage of cemented area and the shape of the noncemented channels. Also the nature of the fluid in the noncemented channel has been investigated.
These experiments were performed with a laboratory cement bond tool of variable spacing with between 0.5 and 3.0 feet, coupled to an oscilloscope and a data acquisition unit.
It is observed that cement sheath thickness is an important parameter than can significantly affect the output of the tool up to a thickness of more than 2 inches. As a first approximation, the percentage of cemented area is confirmed to be in linear relationship with the attenuation rate. Casing stand-off can also change the log output by a factor of up to 30%. Conversely the shape of the channel is found to have no significant influence on the result provided the percentage of cemented area is the same. The nature, from gas to liquids, density and gel strength been found to be negligible factors.
Some of the conclusions are illustrated through the interpretation of several field logs of cased hole or of open hole sections where both the cement and the caliper logs were available.
Since the introduction of the Cement Bond Log (CBL) in the late 50's little experimental work has been performed in order to explain the output of the log in relation with the downhole annular configuration. The effect of vertical channels was studied and a linear relationship was observed between the attenuation rate and the percentage of circumference bonded for a centralized pipe. Later experimental and theoretical studies quantified the influence on the CBL amplitude of casing diameter, casing thickness, transmitter to receiver spacing and cement compressive strength leading to the construction of a nomograph well known as CBL interpretation chart. This last chart which has been modified to take foamed-cements into account suffers from some limitations with regards to tool and casing centralization, cement thickness, well history, etc…
Later on, it was also found that the bond log amplitude was greatly influenced by the acoustic properties of the fluid in the casing and by downhole conditions affecting the transducer response, leading to the elaboration of a borehole-compensated cement bond tool.
However not all the parameters have been considered and some field logs still cannot be explained with existing CBL interpretation techniques, resulting in totally unsuccessful squeeze cement jobs in wells showing poor bond logs.
In the present study, we investigate the influence of critical borehole geometrical parameters which were not considered, or not completely covered in the past. These include casing stand-off, cement thickness, percentage of bonded area, shape of the channel and nature of fluid.