Page | 132
poor quality survey in the original well. This is one of the main reasons why relief wells can be severely delayed sw ‘s ch’fo h j coyof h oh w ll. hssof n fus ng nd xp ns p oc ssof nw h considerable public interest and the excuse that we saved money on surveying does not go down well in the press. 6. The combined uncertainty is a specialist calculation that looks at the sources of error affecting both the relief well position and the target well position. In the absence of anything else, if both wells are MWD, it is safest to assume that both error ellipses are full size and geometrically add them together. However, in many cases, when a relief well is called for, the primary cost consideration is not surveying. There is usually justification for multiple gyro runs to minimise the uncertainty on the relief well. (Gyros can be attached to the ranging tools for additional high quality in-runs and outruns with multiple overlaps). As a result, in the calculation of combined uncertainty, the target well uncertainty dominates the error budget and the contribution of the relief well gyros is almost negligible in the sum of the squares of the errors but it cannot be ignored completely. 7. When planning a relief well, if the homing in point is not the same as the kill point, it may be necessary to ‘ ck’ h g w llun lklld ph s ch d. hsusu llym ns h h l fw llc wwllw n o approach the homing in point with a similar inclination and direction as the target well to minimise steering to parallel in order to keep the target well in range. For example if the target well is vertical the bypass may be done at about 15 degrees and a drop planned after the closest approach point to run parallel with the target still in range.
8. In some cases, the kill mud connection will be made by perforation and the accuracy of the intersection need only be a few feet but in many cases, the target well has to be milled. It is likely that the well control engineers will prefer a long milled notch in the target well casing. This means the well plan for the relief well will have to approach the well at an angle of 5 – 10 degrees and may need a casing included before the kill mud is deployed to avoid lost circulation under the kill mud pressure. Again the space for this approach to the kill point has to be adequate to accommodate the final s c onknown s h ‘gunb l’s c on nd hsmus no be close to any other wells. It is also important that this point is in stable rock just above the reservoir to ensure that the action of pumping kill mud does not fracture the formation so badly that another escape route is opened up for the reservoir fluids.
Figure 118: Relief well penetration point.
9. As is clear from the diagram above, the final approach section will be steered and yet it is critical that this steering is of the highest accuracy. It is actually better when aiming a milling tool to aim slightly anticlockwise of the casing centre to ensu h h clockw s m ll ng c on do s no ‘ oll off’ h c s ng on con c . n order to steer accurately when this close to casing, it may be necessary to use GWD on approach to the final casing shoe especially if the inclination is not close to vertical as the azimuth uncertainty could jeopardise the kill run. After the final casing shoe is in place, there is very little steering ability and the success of the kill is dependent on accurate placement of the casing. 10. It is worth noting that a relief well contingency plan will be pointless if the original well is not surveyed with sufficient accuracy AND frequency. When the ranging shots are taken, the entire trajectory is shifted to fit the relative position discovered by ranging. A new relief well plan is generated every time a ranging shot is taken and a new kill point is interpolated in the target well survey. If the original survey is only available every 500ft, it is impossible to predict where the well will be between surveys and the kill could fail sue to unmeasured doglegs creating unforeseen deviations in the well path.
CONTENTS
Made with FlippingBook flipbook maker