Chapter 4 of 10 · 10 min read
Positions, spools and field welds
See how rolled and fixed pipe positions change a weld's difficulty and time, move welds into the spool shop, plan field welds, and price the difference between a shop joint and one made in a trench.
A pipe joint that can be turned is a different job from one that cannot. With the pipe rolling on stands, the welder stays at the top and lays every bead in the easiest position, downhand, at a comfortable current, for as long as the joint takes. With the pipe fixed, the welder must work round it, from the flat position at the top through vertical on the sides to overhead underneath, changing angle and current as the pool’s behaviour changes, often lying on the ground or reaching round a rack. The same joint takes longer, needs a more skilled welder, and is more likely to need repair when it is fixed.
Willis described the answer in 1917 for gas mains: lay the lengths on the ground, weld them while they could be turned, roll the long string into the trench, and weld in the trench only where there was no other way, from a pit dug under the joint. Modern piping does the same thing on a larger scale, by building spools in a shop and making as few welds as possible in the field. This chapter compares the positions, works out what a shop weld saves, and covers planning and making field welds.
The positions
Pipe welding positions are named by the pipe’s axis and whether it turns.
| Pipe axis | Pipe turns? | What the welder does | |
|---|---|---|---|
| Rolled | horizontal | yes | welds at the top in the flat position while the pipe is turned |
| Horizontal | vertical | no | welds round the pipe in the horizontal position |
| Fixed horizontal | horizontal | no | welds from flat at the top, through vertical, to overhead underneath |
| Fixed inclined | inclined at about 45° | no | every position at once, with the joint itself on a slope |
The fixed horizontal position is the common field weld, on a pipe run lying horizontally that cannot be turned. The fixed inclined position is the hardest, and a welder’s qualification test in it is generally accepted as covering a wide range of other positions, which is why experienced pipe welders take it. The rolled position is the easiest and fastest, and it is what the spool shop aims for.
Rolled and fixed by the hour
The difference in time is large, and it is the basis of most piping plans.
A joint welded rolled in the shop takes about a third of the time of the same joint welded in a pit. Some of the difference is the position, and some is everything around it: carrying tools and gas to the work, rigging access, waiting for permits, working in heat or rain, and the slower pace of a welder who is uncomfortable. Willis’s own figures for the number of joints a competent welder could make in an hour fell steeply with pipe size, and the manual added the condition that the welder should not be delayed too much in moving from one joint to the next. That condition is the field’s whole difficulty.
Spools
A spool is a section of piping, pipe and fittings welded together, built in a shop to the drawing and shipped to site to be bolted or welded into place. The spool shop can roll each spool on positioners so that most welds are made in the flat position, fit up on benches with jigs and good light, use mechanised and higher-deposition processes, inspect easily, and work regardless of the weather or the state of the site. Field welds are left only where spools join each other or the equipment, or where a spool would be too large to transport.
- Joints
- 640
- Shop share
- 40% in a site-built plan; 75% in a spool plan
- Hours
- shop about 1.1; field on a rack about 2.4
- joints moved = (0.75 − 0.40) × 640 ≈ 224
- hours saved ≈ 224 × (2.4 − 1.1) ≈ 291 welder hours
- saving ≈ 291 × 135 ≈ A$39,000, before the better quality and shorter site programme
The saving is not only in hours. Shop welds have lower repair rates, because they are easier to make well, and site time is usually the most expensive part of a project, so every week of site welding avoided shortens the programme and the site overheads that go with it. Against the saving must be set the shop’s own costs, transport of spools, and the risk of spools not fitting on site.
Designing spools to fit
The risk of spools that do not fit is real. Buildings are not built exactly to their drawings, equipment nozzles end up a few millimetres from where they were shown, and a spool made exactly to the drawing can be short or long, or point slightly the wrong way. The usual answer is the field-fit weld: one end of a spool, or one joint in a run, is left with extra length, so that the fitter on site can measure, cut to suit and weld it. The spool designer chooses where those welds go, putting them where access is reasonable and the position not too difficult, and keeps their number down. Surveying the building and equipment as built before fabrication, with modern measuring tools, reduces them further.
Breaking a line into spools
Spool size is limited by transport, by lifting on site, by the access through the building to where the spool must go, and by how the spool will be galvanised, painted or pickled if that is needed. A spool too large to get through a doorway is a spool cut in half on site. The spool designer also groups spools by system and installation sequence, so that the site crew receives them in the order it needs them, and marks each spool with its number, its line and the field welds that will join it.
Field welds
Field welds are made in the positions the installed pipe dictates, which means fixed positions and often difficult access. The welder needs room to work round the whole joint, which on a rack means space between pipes and on the ground means space under the pipe. Willis’s pit dug under a joint in the trench is the same requirement. The contractor’s spool designer allows clearance for field welds, and the installation planner provides access: scaffolds, platforms, and in trenches, pits dug wide enough and safe enough to work in.
Field welding brings weather. Wind blows away gas shielding, so gas-shielded processes in the field need screens or shelters; rain and dew wet the joint and cause porosity and, in hardenable steels, hydrogen cracking, so joints are dried and sometimes preheated; heat and sun affect the welders. Self-shielded flux-cored wires and covered electrodes are less sensitive to wind, which is much of why they are used in the field.
Progression: uphill and downhill
On a fixed pipe, the welder can travel from the bottom up each side, uphill, or from the top down, downhill. Uphill welding, slower and with a larger pool, gives thorough fusion and is the usual progression for process piping and heavy walls. Downhill welding, fast with a small pool and electrodes or wires designed for it, is used on cross-country pipelines, where a crew of welders works along the line each laying one pass on every joint in turn. The procedure states the progression, and a welder qualified uphill is not automatically qualified downhill.
Working in trenches and pits
A trench or pit for a tie-in weld is an excavation, and its walls can collapse. Excavations deeper than a modest depth need shoring, benching or battering, and access and egress, under the work health and safety laws, and a pit with a welder in it is also a space where fume and shielding gas collect, because both are heavier than air or are trapped by the walls. Safe Work Australia’s model code for welding processes covers the welding risks; excavation work has its own code and regulations. The contractor’s pipeline crews plan each tie-in pit as a piece of excavation work, not as a hole to stand in.
Mechanised pipe welding
On long pipelines and in large spool shops, pipe welding is increasingly mechanised. Orbital heads that clamp to the pipe and travel round it make fixed-position welds with a consistency no manual welder can sustain for long; in the shop, positioners and mechanised heads weld rolled spools at high rates. The Davis-Bournonville course of 1919 described an early form of machine pipe welding, which heated the whole joint at once with a ring of flame and pushed the ends together. Modern orbital welding is quite different, an arc carried round the joint, but the purpose is the same: to take the hardest, most repeated part of the job out of the welder’s hands. Chapter 7 returns to orbital welding for stainless tube.
Field welding in a running plant
Tie-ins to running plant are the slowest field welds of all. The line being joined must be isolated, drained, purged and proved safe; hot work permits and fire watches are needed; the area must be screened from production; and the work often has to fit a short shutdown. Each tie-in is planned as a small project of its own, with its materials, spools and welders ready before the shutdown begins, and the joints welded at the shutdown are kept to the minimum by doing everything possible beforehand.
Welders and positions
Position also decides who can do the work. A welder who makes excellent rolled joints in the shop may not yet be qualified, or skilled enough, to make the same joint fixed on a rack, and the best fixed-position pipe welders are always in demand. The spool plan therefore changes the crew as well as the hours: the shop can use welders qualified for rolled positions and mechanised heads, while the field needs fewer welders, but the most skilled ones. Planning the crew from the split of shop and field joints, rather than from the total, avoids finding in the middle of the job that the field welds are queued behind the two people qualified to make them.
Inspection access
Field welds must be inspected as well as made, and the access that a welder needs is usually what the inspector needs too. Radiography of a field weld on a rack needs the area cleared for radiation safety, which on a busy site may mean night work; ultrasonic testing needs room for the probe on both sides of the weld; a visual inspection of the cap underneath a pipe needs a mirror or a camera. Shop welds can be inspected on the bench. The spool plan saves inspection time as well as welding time, and makes the remaining field inspection easier to schedule.
The running case: the spool plan
This is an illustration. The contractor’s first estimate for the dairy assumed most welding on site, with about 40% of joints made in a temporary site shop. The supervisor proposed a spool plan instead: the building and equipment would be surveyed as built, spools fabricated in the contractor’s own shop with positioners and a mechanised head for the larger sizes, shipped in installation sequence, and joined on site with field welds planned in accessible positions. The shop share rose to about 75%. On the figures in this chapter the plan saved roughly 290 welder hours, and the site welding programme shortened by about five weeks.
A failure: the spool that did not fit
This is an illustration. On an earlier project, a long stainless spool for a product line was fabricated exactly to the drawing, with no field-fit allowance, because the drawing was believed to be accurate. On site, the tank nozzle it joined was 40 mm higher than shown, and the spool could not be fitted. It had to be cut in two, a new section fabricated, and two extra orbital field welds made, delaying the commissioning of the line by a week.
A survey of the installed tank, or one field-fit end on the spool, would have avoided the problem at almost no cost. The contractor’s spool designs now carry at least one field-fit allowance on every spool that joins equipment, and the survey of as-built equipment is a step in the fabrication programme.
Exercise: the water main tie-ins
The raw water main is about 4 km of DN 200 pipe in 12 m lengths. If lengths are welded into strings of ten beside the trench at about 1.6 hours per joint, rolled where possible, and the strings are joined in trench pits at about 3.6 hours per joint, how many joints of each kind are there, and how many welder hours? What would it cost at A$135 an hour if all the joints were made in the trench?
The main has about 4,000 ÷ 12 ≈ 333 lengths and about 332 joints. In strings of ten there are about 333 ÷ 10 ≈ 34 strings, so about 33 trench joints and about 332 − 33 = 299 string joints.
The hours are about 299 × 1.6 + 33 × 3.6 ≈ 478 + 119 ≈ 597 welder hours. All in the trench, they would be about 332 × 3.6 ≈ 1,195 hours, roughly double, costing about 1,195 × 135 ≈ A$161,000 against about A$81,000, which is the arithmetic behind Willis’s practice of welding on the bank and rolling the strings in.
