10.7.2 Continuous mode

Attitude data derived using continuous gyro survey systems have a tendency to drift exponentially with time. In many gyro systems, it is common practice to compensate for this effect by forming estimates of the drift at regular intervals during the survey. This is achieved by holding the tool stationary for short periods of time, and subtracting the estimated components of Earth’s rate so which the tool is subjected a the current location. The drift estimates generated by this process are the accumulated effect of all physical errors at the given interval, and are used to implement a real-time re-calibration of the tool. The quality and effectiveness of this re-calibration are dependent on many factors that are difficult to predict, including the change in tool-face. It is almost impossible to keep track of what happens to the different physical sources of error when drift compensation is applied.

Studies of field data, where comparisons of in-run and out-run surveys have been made, indicated that accumulated azimuth error in most continuous surveys can be estimated using four simple empirical parameters. The four empirical parameters are:

• the error of the initial reference
• a term proportional to measured time (gyro drift)
• a term proportional to the square root of measured time (random walk)
• a random error which is irrelevant for position error calculations.

Whilst the use of empirical error sources is a departure from the usual form of error model linked to physical uncertainties, it does allow realistic uncertainty estimates to be produced. Gyrodata has developed a new method for final calculation of continuous surveys called Continuous Drift Correction (CDC). It is logically equivalent to averaging in-run and out-run surveys, but adopts a more complex approach which facilitates the estimation of error model terms, including the linear drift and random walk components.

CDC also checks for tool misalignments provided that the in-run and out-run high-side tool-faces are not the same throughout the survey. The misalignment correction compensates for systematic misalignment for the whole survey and provides additional quality control to the data. The use of CDC provides several benefits compared to a simple in-run/out-run comparison. It corrects for linear drift and systematic misalignment and allows QC checks to be implemented by providing estimates of residual errors that can be checked against the tool error model.