It is well known that the MWD sensors in themselves are extremely accurate but the weak link in the system is the accuracy of the magnetic field in which the azimuth readings are taken. We discussed earlier the effects of local crustal anomalies on the accuracy of our background reference vector but in this section we will look at the effects of the drillstring magnetisation which is always present to some degree. It should be noted however that since the sensor pack is always set inside non-magnetic material, the major interference component is likely to be either from below or above the sensor in the drillstring and therefore the z axis interference is usually the major influence on azimuth accuracy.
In the 1990s, Dr Robin Hartman of Shell International developed SUCOP (Survey Correction Program) to implement a technique for measuring and removing the magnetic influence of the drillstring.
If a magnetometer is placed in a magnetic field it will measure the component of the field along its own axis. A good analogy is if a tube with a small flow meter was inserted in stream, the flow through the tube will be the component of flow along the axis of the tube. Clearly if the tube is held perpendicular to the flow there will be no flow in the tube and if it is in line with the flow it will experience the full flow rate of the stream. At any other angle it will experience the flow vector x cosine of the angle of incidence. This value is often referred to as the vector dot product. This can be defined as the product of two vector lengths x cosine of the angle between them.
In below figure the magnetic field is represented by the green arrows and the magnetometers will only measure the component of that field along their own axis.
Now imagine two magnetometers facing in opposite directions at some angle to the magnetic field. Clearly they will read the same magnitude of field but with opposite signs. Hartman realized that if an MWD sensor pack was rotated around the z axis, the x and y magnetometer readings would follow a sine wave which should have an average of zero. If the average was anything else, there must be a component of the observation which is permanent. This will be some combination of a sensor bias or a magnetic field component which is rotating with the sensors and is never going away.
In this example, the magnetometer is clearly carrying a positive magnetic field value that is not going away.
This rotational data can then be used to correct the x and y magnetometer values for this bias. This still leaves the potential for a further magnetic field bias on the z axis. In practice the z axis is often the most significant direction of influence since the significant magnetic material in the drill string is always above or below the MWD sensors. Clearly it is not possible to flip the z axis so the clean z axis is derived from the following formula:
In this formula, Bt means the total background magnetic field strength and Bx By and Bz mean the sensor axes readings from the magnetometer. This formula relies on the accuracy of Bt and ideally should have an IFR survey to accurately measure Bt. This simple technique allows the surveyor to calculate bias values for all three axes and remove the majority of the magnetic interference from the subsequent observations.
One caveat that should be kept in mind is the stability of the calculation. The value for Bz is very sensitive to azimuth and inclination.
The formula is identical to a Pythagoras formula for calculating one side of a right angled triangle where Bt would be the hypotenuse and the base would be the cross axial field as follows.
If the value of Bz is very small, which will happen when drilling at high angle heading East or West, the resolution of clean Bz will be extremely sensitive to the accuracy of the other two sides. The slightest error in Bt can produce a very exaggerated effect on Bz and, in many cases, produce a value in Bz more erroneous than the magnetic error. In other words, this technique should not be used in such geometries in case the correction is more erroneous than the original reading.
For this reason and because the Bz value can be positive or negative it is not recommended that this technique be employed in the following ‘no go zones’ – above 70 degrees inclination within 20 degrees of East / West (magnetic).
Feedback
Post your comment on this topic.