Travelling cylinder plots: what they show, and why two can disagree

A normal-plane travelling cylinder plot shows other wells (the offsets) relative to a reference trajectory. In the original convention, the reference plan forms the centre of a polar plot. Offset positions appear by distance and direction in a plane perpendicular to that plan at the reference measured depth. On the original plan-centred diagram, the drilled reference well can also appear as a separate trace. Some software also allows an as-drilled trajectory as the reference, so check which trajectory forms the centre.

How the plot is built

Picture a disk set at right angles to the reference (the normal plane), with the well through its centre, and push the disk down the well. Thorogood and Sawaryn describe a polystyrene disk with the offsets as hot wires, each burning a trace as the disk moves. A near-parallel well burns one large hole. A fast-converging well burns a line that moves quickly across the disk.

The plot is the record of those traces:

• The centre is the reference trajectory.
• The distance from the centre is the distance to the offset, in the plane.
• The angle is the direction of the offset.
• Depth labels identify the reference measured depth (MD) at which the normal plane is constructed. The figure labels selected points.

View the normal-plane plot looking down the reference well. Angles increase clockwise.

For the original diagram, SPE 19989 specifies MD labels from the reference plan. True vertical depth (TVD) is ambiguous because it varies across a tilted normal plane.

With regular sampling, closely spaced points along one trace often indicate nearly parallel well directions. Wider spacing can indicate faster convergence or divergence. Point spacing also depends on the sampling interval and scan direction. Check the depth labels and sampling interval before interpreting convergence. Points moving towards the centre indicate decreasing plotted separation.

Figure 1 opens on the Kick-off beside a vertical well scenario, with the normal plane and TC North up. Buttons on the figure change the scenario, the method and which way is up. Move the depth slider, or press Play, to watch the plane travel down the reference in 3-D. The trace shows the full scan. Rings and the table show results at the selected MD. The wells are computed by the minimum curvature method (Sawaryn and Thorogood, 2005).

Thorogood and Sawaryn of BP published the principles in 1990 and 1991, in SPE 19989. They noted that wellsite staff sometimes had real trouble reading separations and convergence from plan and section views, and that this had apparently contributed to some subsurface collisions. By their account, the original hand-drawn version was made for the THUMS offshore island wells in the East Wilmington field.

The first difference: how each point is chosen

Thorogood and Sawaryn (1990) warned that readers could misinterpret the geometry if they did not understand the differences between diagram variants.

The ISCWSA's Current Common Practice document lists the three methods that software typically supports. The figure has all three:

In this figure, the normal-plane trace scans from the offsets to the reference. The other two methods scan from the reference to the offsets. The scan label beside the controls identifies the direction.

The normal-plane convention matches SPE 19989 and the ISCWSA Collision Avoidance Lexicon. The 1990 paper discouraged horizontal-plane and closest-approach versions of the diagram for collision avoidance. That warning concerned the display; 3-D minimum distance remains a common input to numerical clearance calculations. The ISCWSA calculation guide also identifies limitations of horizontal-plane scanning.

In this figure's closest-approach view, the radius is the full 3-D distance. The angle comes from the separation vector's projection onto the reference normal plane. That convention loses the along-hole direction. Other closest-approach displays can use different conventions.

The Lateral crossing above scenario illustrates the effects of these choices. An offset lateral drills north and crosses 3 m above the eastbound reference lateral at reference MD 2,800 m. The normal plane draws a straight line across the plot, every point at 2,800 m. This figure's closest-approach view puts every point at the reference's high side, the distance falling to 3 m and rising again. With Up set to High side, that looks like a well that came down from above and went back up. The horizontal plane finds no points, because the offset never reaches the reference's TVD.

Scan direction matters too. A plane stepped down the reference can miss a well that crosses at right angles, because that well meets the plane at one exact depth. Move the slider off 2,800 m and the current-depth rings disappear. The full trace remains visible because the figure scans from the offsets to the reference. SPE 19989 addresses this failure by scanning in that direction, finding the nearest reference point for each offset point. ISCWSA common practice describes this reverse scan for travelling cylinder clearance reports.

Reverse scanning addresses the perpendicular-crossing failure. The software must also check scan intervals, depth coverage, endpoints, and wells approaching from ahead. A missing plotted point does not demonstrate clearance.

The second difference: which way is up

High side up puts the high side of the reference (the top of the hole, in an inclined well) at the top. The ISCWSA e-book says this is easier to understand. It also matches steering by toolface. But a vertical hole has no high side. Some software switches its angular reference from north to high side at low inclination, which can make offsets jump on the plot. This figure uses a 0.5° switch threshold; other implementations can use different conventions. In low-angle hole, changes in reference azimuth can rotate the high-side view. The e-book shows a well that turned away from a vertical well but, on a high-side plot, looks as if it drilled straight through it.

TC North, also called north up or "TFO+AZI", adds the reference azimuth to the offset's angle from high side. The azimuth uses the well's selected north reference, such as grid or true north. In inclined hole, this plotted angle is not a compass bearing. In the Lateral crossing above scenario, an offset 3 m above an eastbound lateral plots 3 m out on the E side. The convention avoids the high-side reference jump through vertical hole. High-side arrows identify the reference's high-side direction on the plot. The horizontal-plane view uses ordinary horizontal bearings instead.

In the Kick-off beside a vertical well scenario, the vertical well is 10 m north and the reference builds towards 225°. North up, the vertical well turns smoothly from 0° to about 31° by 640 m, where it leaves the plot. Switch Up to High side and move the slider from 455 m to 460 m. The vertical well jumps from 0° to about 135° as the inclination passes 0.5°, where the figure stops measuring from north.

SPE 19989 recommends north up and advises against high side as the reference direction. The slides for SPE/IADC 184730 (Sawaryn, 2017) also show a north-up, normal-plane diagram. At a 2024 ISCWSA meeting, some practitioners argued that north up reads poorly in long laterals. The display convention explains part of this concern. Adding the reference azimuth introduces a rotation in the TC North display. Both views can change when the relative geometry changes. The meeting notes also record that the committee has never recommended one style, so check which one a plot uses.

Use the operator's clearance rules

ISCWSA common practice describes the travelling cylinder plane as typically used for plotting. Numerical clearance calculations commonly use 3-D minimum distance. Operators can specify several clearance rules; a separation factor is one possible measure. Rules that use position uncertainty must account for the relevant uncertainty of both wells. Calculating a separation factor in the travelling cylinder plane can change its result.

Clearance decisions require the operator's specified calculation methods, limits, and procedures. A plot alone does not establish clearance. Figure 1 illustrates geometry and does not calculate clearance.

SPE 19989 draws tolerance lines on the diagram that the well being drilled should not cross. Today the same idea is the allowable deviation from plan: how far the well can stray from its plan before it comes closer to an offset than the minimum allowable separation of the current separation rule (Sawaryn and co-authors, 2019).

SPE 19989 also suggests a plot radius large enough that survey error over the interval is no more than 10% of that radius. This historical guidance concerns diagram presentation; it is not a clearance threshold. Select a view radius that shows the relevant wells and their clearance margins.

Before you compare two plots

If two plots of the same wells disagree, check how each was built before you decide that a well has moved:

1. Point method: normal plane, closest approach or horizontal plane?
2. Scan direction: down the reference, or from the offsets back?
3. Which way is up: high side, TC North, or horizontal bearing, and which north reference?
4. Centre: the plan or the drilled well?
5. Depth labels: reference MD, offset MD or TVD?
6. Plot radius: does it show the relevant wells and clearance margins?
7. Survey data: the same survey versions, coordinate and depth references, and units?
8. Scan coverage: the same scan intervals, scanned depth ranges, and included offsets?
9. Clearance overlays: the same uncertainty settings and clearance rules, where tolerance lines or uncertainty are shown?

References

• Thorogood, J.L. and Sawaryn, S.J. (1990). The Travelling Cylinder: A Practical Tool for Collision Avoidance. IADC/SPE Drilling Conference, Houston. IADC/SPE 19989. doi:10.2118/19989-MS
• Thorogood, J.L. and Sawaryn, S.J. (1991). The Traveling-Cylinder Diagram: A Practical Tool for Collision Avoidance. SPE Drilling Engineering 6(1), 31-36. SPE 19989. doi:10.2118/19989-PA
• Sawaryn, S.J. and Thorogood, J.L. (2005). A Compendium of Directional Calculations Based on the Minimum Curvature Method. SPE Drilling & Completion 20(1), 24-36. SPE 84246. doi:10.2118/84246-PA
• ISCWSA Collision Avoidance Work Group (2013). Collision Avoidance Calculations - Current Common Practice. iscwsa.net
• Sawaryn, S.J. (2017). Well Collision Avoidance Management and Principles. SPE/IADC Drilling Conference, The Hague. SPE/IADC 184730. Presentation slides, iscwsa.net. Journal version: Sawaryn, S.J. et al. (2018), SPE Drilling & Completion 33(4), 335-350. doi:10.2118/184730-PA
• ISCWSA Collision Avoidance Work Group (2017). Collision Avoidance Lexicon. iscwsa.net
• Jamieson, A. (2017). Introduction to Wellbore Positioning, V09.10.17 (ISCWSA / UHI e-book), sections 20.12-20.13. iscwsa.net
• Sawaryn, S.J. et al. (2019). Well-Collision-Avoidance Separation Rule. SPE Drilling & Completion 34(1), 1-15. SPE 187073. doi:10.2118/187073-PA
• ISCWSA Error Model Maintenance / Collision Avoidance Joint Sub-committee (2024). Meeting notes, 60th ISCWSA meeting, 25 September 2024 (travelling cylinder discussion). iscwsa.net