With increasing demand of oil and gas production in mature field, there is a growing interest in obtaining formation acoustic properties from cased-hole acoustic logging. The cased-hole acoustic measurement is not a problem when the casing is well-bonded with formation. However, it is problematic when the casing is poorly bonded. The situation is aggravated in the ?free-pipe? condition where there is no cement bond between casing and formation. Because of the need to obtain formation acoustic properties in the poor-bonding or even free-pipe situations, various signal processing techniques are used to separate the formation acoustic signal from the casing signal. It is found that in the free-pipe situation the acoustic data is usually unusable because it is extremely difficult to distinguish the formation signal from the overwhelming casing signal. In this study we investigate signal processing techniques for cased-hole acoustic data processing. We use numerical modeling to identify the formation signal in poor bonding and free-pipe conditions. We demonstrate that the formation signal, when it is present in the data, can be enhanced and extracted from acoustic data even in the presence of the dominant casing signal. The results are then used to provide a means for estimating formation acoustic velocity through casing. We use convincing field data examples to demonstrate the important application.

This content is only available via PDF.
You can access this article if you purchase or spend a download.