RECORDS SET, DRILLING CALCULATION REFINED IN DUTCH PIPELINE CROSSINGS

Jan Spiekhout, Jan Gorter, Willem H. Hartlief
N.V. Nederlandse Gasunie Groningen, The Netherlands

Horizontal drilling for pipeline installation made significant progress in The Netherlands in 1991 with establishment of two diameter records and development of a simplified design calculation method.

The records were established in two natural gas pipeline projects by N.V. Nederlandse Gasunie, Groningen.

In companion waterway-crossing projects, installation of two 48-in. gas lines by directional drilling set a record for diameter-to-length ratio.

Also, directional drilling for a 36-in. pipeline crossing established a record for entry and exit angles for such large-diameter pipe. Precise location was particularly important in this second project because a pipeline bundle (12-in. insulated oil pipeline, 10-in. chloride pipeline, 12-in. nitrogen pipeline) was installed to run parallel over part of the distance.

The major data on Gasunie's crossings are given in Table 1.

The calculation method used in horizontal drilling was modified to correct for overconservatism and to provide an approach which better approximates actual pipe and soil behavior.

The simplified method is supported by finite-element computer calculations and has been incorporated into the new Dutch code on steel pipeline systems (NEN 3650).

DUTCH PROGRESS

Horizontal directional drilling has become increasingly popular since its introduction into The Netherlands in 1984.

The maxi-rig has been followed by the midi-rig and the mini-rig, and measurement techniques employed have improved both in locating the drilling head and in establishing the as-built position of the product pipe.12 Work has also been done on design and calculation procedures.35

A great deal of pioneering work has been conducted in the field of soil mechanics, covering such areas as borehole stability and allowable mud pressures in connection with the risk of fracturing.6

Notwithstanding all these developments, success of such projects still depends heavily on the human factor, as well as the ever-present uncertainties of geology.

In the first 48-in. crossing, a pipeline crossing was installed on the Noord Hollands Canal near Julianadorp't Zand, The Netherlands. The project also involved installing a parallel 8-in. pipeline at a spacing of 10 M.

A 48-in. crossing by horizontal directional drilling had never before been accomplished.

To reduce the risk of damage to the 48-in. pipe, it had to be installed after the 8-in. pipe. Also, the position of the pilot hole had to be verified with a gyro compass to provide the most accurate possible information on the location.

The choice lay between drilling with a gyro compass or drilling with position checks at three points through the washover pipe. The contractor, Horizontal Drilling Inc., Paris, chose the latter option, i.e., checking the location at three points with a gyro compass.

These three points are just beyond the shore, half-way across the canal, and beyond the exit point.

The contractor also chose to drill the two pilot holes first, then to install the product pipes (after reaming).

An "instrument sub" was used to establish the position of the drilling head during drilling of the pilot holes. The sub incorporates an inclinometer (to measure orientation with respect to the gravitational field) and a magnetometer (to measure orientation with respect to the earth's magnetic field).

The TruTracker method1 was also used on both banks of the canal. This method involves laying a cable at ground level above the planned route of the drill in an accurately measured polygonal pattern, through which a welding transformer feeds a dc current of approximately 100 amp.

This generates a magnetic field whose strength is measured by the magnetometer at the drilling-head position. This method improves location accuracy and reduces the effect on the earth's magnetic field of stationary or moving ferromagnetic objects, high-voltage lines telephone lines, and so forth.

Because the washover pipe provides a magnetic screen, the magnetometer cannot be used when the pilot pipe is drawn back into the washover pipe unless the washover pipe is also withdrawn.

Regular measurement of the profile each time a given length is drilled is therefore labor-intensive.

PROBE MEASUREMENTS

Measuring the as-built position of the washover pipe at the Noord Hollands Canal crossing was carried out with a completely separate method: a probe incorporating an inclinometer and a gyroscope.

For interim measurements, this probe was moved through the washover pipe by attaching it to the pilot drillstring and pushing or pulling it through the washover pipe. On completion of drilling, the as-built position was measured by pulling this instrument sub to and fro through the washover pipe on a cable.

Fig. 1 gives the results of the as-built measurement of the profile of the pilot hole for the 48-in. pipeline in the horizontal plane for approximately one third of the distance and for the entire distance. The interval between these two measurements was 2 days.

The minor differences between the two sets of measured values indicate a high degree of reproducibility of the as-built measurement procedure.

Fig. 1 also compares the measured values recorded during drilling (inclinometer + magnetometer) with the as-built measurements, once again in the horizontal plane.

The shapes of these two independently measured profiles are virtually identical, but the average heading differs because an incorrect value was used as input for the starting heading (azimuth), which was established by surveying techniques. The two profiles practically coincide if correction is made for this error.

Fig. 2 gives the results for the pilot hole for the 8-in. pipeline in the horizontal plane.

Both figures clearly show the effect of drilling-head control.

The relatively sharp correction applied after about 500 m, particularly for the 8-in. hole, may occur because the measurement pipe came within the detection range of the TruTracker system.

Once the hole was reamed and the product pipe was pulled through, the final profile of the product pipe was much smoother than shown in Figs. 1 and 2. This could be demonstrated by measuring the profile of the product pipe.

Fig. 3 shows the profile of the pilot hole in the vertical plane, both as measured during drilling and after completion. The measured values are virtually identical.

Experience has shown that measuring the profile in the vertical plane is generally more accurate than in the horizontal plane.

This may derive from navigation in the vertical plane being achieved mainly by determining direction with respect to gravity (measuring inclination) and in the horizontal plane by determining direction with respect to the earth's magnetic field (measuring azimuth), which is much more susceptible to disturbance.

This problem does not arise if a gyroscope is used.

Lastly, Fig. 4 shows the final entry and exit points in the horizontal plane, as measured during drilling and after completion. The discrepancies between the lines connecting these points appear large, but appearances can deceive because, at an exit angle of around 10 with the horizontal, a small discrepancy in this angle gives a relatively large displacement in the horizontal plane.

The other project was a 48-in. crossing of the Kolenbrandersbos and Dedemsvaart waterway. Primary contractor was Visser & Smit Hanab, Papendrecht, The Netherlands.

In this case, positioning was far less critical than with the 48-in. Noord Hollands Canal crossing.

ANGLE SEVERITY

The 36-in. crossing project involved a pipeline crossing of the Oude Maas waterway near Pernis, The Netherlands. Primary contractor was Land & Marine (overseas), Bromborough, U.K. It established a record for severity of entry angle (12) surpassing the normal angle of 8.

The tolerances on the positioning of the pilot hole were: horizontal, 0.5 m; vertical, 1 m.

To arrive at the correct point and remain within the limits set by the Rotterdam municipal authorities, the pipe had to follow a curved horizontal route. After the installation of the product pipe, a pipe bundle was to be laid running in parallel over part of the distance.

The distance between centers at the exit point was 8 m. The entry and exit angles were higher (12) than normal (8) for this diameter because of the constraints on the length available for inserting the pipe.

The procedure used for measuring the location of the pilot hole was the same as for the Noord Hollands Canal crossing.

Fig. 5 gives the coordinates in the horizontal plane for the entry and exit points for both the 36-in. pipeline and the three-pipe bundle. The corresponding entry and exit points are connected by a straight line and a dotted line to show the curved route followed by the 36-in. pipeline.

Fig. 6 shows the distance between centers of the two pilot holes, both the horizontal distance and the absolute distance. Fig. 6 also shows that the distance is nowhere less than the permitted minimum distance (8 m) at the exit points, which are the critical points.

Although both twin pipeline projects discussed here progressed successfully, not least in terms of drill-head navigation, it should not be concluded that it is sufficient in every case to measure position with an inclinometer and magnetometer (combined with the TruTracker system) alone.

In many instances it is advisable to measure the actual position of the pilot hole. Measurement of the as-built profile of the product pipe is advisable if it is necessary (for example in critical areas) to establish (or map) the location of the product pipe with a high degree of accuracy (for example, within 1 m).

ABOUT RECORDS

Records are constantly being improved upon, and there is no theoretical reason why it should not be possible to install larger diameters over greater lengths at higher entry and exit angles.

Nothing appears to preclude reaming holes to larger sizes than have been attempted so far and temporarily preventing their collapse. The drilling rigs are in principle capable of handling larger diameters.

In the projects described here, the forces needed to pull the product pipe through were relatively low. The force needed to install the 48-in. pipeline across the Noord Hollands Canal was less than 195 tons.

Nevertheless, the present achievements may represent more or less the maximum achievable. Experience has shown that, not only on these projects but also on other projects involving smaller diameters and longer lengths, drilling equipment can encounter difficulties during the installation procedure.

Although the equipment may be essentially sound and well maintained, the dynamic loads occurring in practice occasionally lead to fatigue cracks. So far, however, such problems have not caused any of Gasunie's drilling operations to fail.

CALCULATION PROCEDURE

The calculation procedure currently in use in The Netherlands gives large wall thicknesses and virtually precludes use of plastic pipe because, in analogy with buried pipelines lying on rigid supports, it assumes that the maximum limit values for (passive) earth pressures would be reached.

In consultation with authorities of the Department of Public Works, Province of Zuid Holland, the calculation procedure for pipelines installed by means of directional horizontal drilling was modified in 1991. These authorities became involved because they must check design and strength calculations for crossings and agree on the calculation method used.

The modification was based on results of finite-element (FE) calculations carried out for three cases.

The cases taken were two installed crossings. For Case 3, the geometry and soil mechanics data from Case 2 were used, but the material was plastic instead of steel.

These three cases are described in Table 2.

Fig. 7 plots the soil reactions for the three cases as calculated with Expert Design Systems' PLE (pipeline engineering) software. This is an FE package specially designed for buried pipelines supported on springs to represent the surrounding soil. The curved profile of the hole was entered in the calculation as the settlement line.

The calculations took account of the second-order effect. The parameters for the start and end points were specified as "free" and "open." The calculation results are summarized in Table 3.

For the curved sections, the variation in soil reaction indicates a beam loaded by a bending moment at its extremities. For Case 1, there is a straight section in the center (the "floor pipe"), but this is not the case with Case 2.

According to this detailed calculation procedure, the maximum soil reactions are significantly lower than the limit values indicated by the former calculation procedure. For the steel pipelines, it appears that the maximum soil reaction can be accurately calculated with the model of an elastically supported beam loaded by a moment.4

The loading pattern also corresponds closely to that given by the computer calculation. Agreement is not as good for the plastic pipe, but the method is usable in practice and is on the conservative side.

The results obtained with the simplified calculation method for the three crossings are given in Table 4. Wall thicknesses with this calculation are approximately 25% less than previously called for.

CALCULATIONS

For service conditions at the time of installation, the pipeline design can be calculated with Equations 1 and 2 found in the accompanying box.

For the soil, the indirectly transferred load is found with Equation 3 and the directly transferred load (at the deepest point in the hole) with Equation 4.

For the bottom of the pipe, Equation 5 applies.

In case of heavy soil cover (15 m), the neutral earth pressure can be reduced.7

This calculation method for horizontal drilling has been incorporated into the new Dutch code on steel pipeline systems (NEN 3650).

ACKNOWLEDGMENT

Valuable contributions were made to the discussions and design phase of the projects by A. M. de Koning, K. E. van der Woude, P. M. Wesselius, and M. H. W. Kuperus, all N.V. Nederlandse Gasunie, Groningen. The authors would also like to thank H. J. M. Hergarden, Delft Geotechnics, Delft, and R. A. J. de Kock, Province of Zuid Holland.

REFERENCES

1. Gorter, J., "Verschillende meet-methoden voor bestuurbare horizontale boringen-Techniek vertelt exact waar de boorkop opduikt" ("Location method for horizontal directional drilling-technique determines drill bit location"), Land + Water, No. 1/2, January 1992 (in Dutch).

2. Spiekhout, J., "Developments in Directional Drilling," Pipeline & Gas Journal, April 1991.

3. Pijpleidingcode 1990 (Version 6).

4. Hetenyi, M., Beams of unlimited length (Ann Arbor; University of Michigan Press, 1983).

5. NEN 3650: "Requirements for steel pipelines."

6. Luger, H.J., and Hergarden, H.J.A.M., "Directional drilling in soft soil: influence of mud pressure," No-Dig 88.

7. Terzaghi, Soil Mechanics (John Wiley and Sons, New York, 1966), pp. 194-202.

Copyright 1993 Oil & Gas Journal. All Rights Reserved.