Views: 0 Author: J-VALVES Publish Time: 2026-10-06 Origin: Site
On offshore platform and pipeline purchase requisitions, "extended weld end" is frequently ticked as a safety margin without further discussion. The reasoning is consistent and correct as far as it goes: welding heat damages seats, so extending a short piece of pipe from each valve end keeps that heat away from the sealing components.
The reasoning is not wrong. The problem starts when it hardens into an assumption that a longer pup piece is a safer pup piece — and pup piece lengths begin to be copied from the previous revision of the drawing rather than calculated.
This article addresses three specific questions: how long a pup piece actually needs to be, at what point additional length starts creating costs of its own, and when a weld end valve should not be specified in the first place.
Search API 6D for a table stating that an extended weld end must be a given number of millimetres and you will not find one. Extension length falls under industry practice and project specification, not a single mandatory international standard.
The prevailing industry practice today:
Small bore conventional range: 100–150 mm, with 150 mm the single most widely adopted value
Large bore grading principle: NPS 2–8 commonly 200 mm; NPS 10–20 a minimum of one nominal diameter (1D) or a maximum of 500 mm; NPS 22 and above 800 mm
An alternative tier table common on large projects: 2–6 in at 400 mm, 8–14 in at 600 mm, 16–24 in at 800 mm, 26–48 in at 1200 mm
Worth noting: IOGP S-562, the supplementary requirements to API 6D for ball valves, requires in its weld end clause that butt-welded end valves be supplied with pup pieces, and provides a length table. This means on high-end oil and gas projects, pup piece length can become a contractual requirement — but the basis is the project specification and supplementary requirement, not API 6D itself.
So when someone states that "the standard requires a 150 mm extended weld end," that claim does not hold technically. 150 mm is a practice consensus, not a standard clause.
The sole function of a pup piece is to move the heat affected zone (HAZ) of the field girth weld away from the seat sealing elements. The real basis for the length value is therefore the combination of two variables: the temperature limit of the seat material and the heat input of the field weld.
Approximate temperatures at which seat materials begin irreversible deformation:
Seat material | Onset of irreversible deformation |
PTFE (polytetrafluoroethylene) | approx. 200 °C |
PEEK (polyetheretherketone) | approx. 260 °C |
That appears to leave margin. The complication is that heat flow during field welding is not one-dimensional. The welder works on the pipe side; heat conducts through the valve body wall toward the cavity. Where pipe and body wall thickness differ significantly, heat flux concentrates at the transition. Industry control targets are therefore set well below material limits:
Valve size | Seat temperature control limit |
NPS 2–6 | < 120 °C |
NPS 8–14 | < 150 °C |
NPS 16–24 | < 180 °C |
NPS 26–48 | < 200 °C |
This table explains why pup piece length increases with bore size. A larger bore means greater body thermal mass and thicker walls; heat has a wider path into the cavity and requires a longer transition distance to attenuate the gradient. It also explains why a 150 mm pup piece proven on small bore does not transfer to a 24 inch valve.
The correct approach is not to look up a number but to require the manufacturer to supply a welding heat management procedure with defined thermocouple locations, then verify it on site against actual heat input. Pup piece length is one variable in that procedure, not the whole of it.
Extending the pup piece solves the distance problem. If the bevel at the pup piece end is wrong, field welding returns the problem immediately — incomplete root fusion, misalignment beyond tolerance, and repair gouging all drive heat back toward the seat.
ASME B16.25 governs the bevel and is the starting point of that chain:
Wall thickness ≤ 3 mm: end may be square cut or given a small chamfer
Wall thickness > 3 mm to 22 mm: single angle bevel 37.5° ± 2.5°, root face 1.6 mm ± 0.8
Wall thickness > 22 mm: compound bevel — inner zone 37.5°, outer zone transitioning to approximately 10°
The compound bevel is not a cosmetic choice. Compared with a single 37.5° bevel, a compound bevel can reduce field weld metal fill volume by roughly 20%. For a weld end valve that figure maps directly to welding man hours and heat input: 20% less fill metal means the welder spends proportionally less time adjacent to the valve, and cumulative thermal exposure at the seat drops with it.
Inside diameter tolerance matters equally. ASME B16.25 specifies a bore tolerance for dimension C of +0.25 / −1.02 mm (scaled for larger sizes). The bias toward the negative side means the pipe bore can end up smaller than the valve bore, creating an internal step. A step becomes a turbulence and erosion initiation point, which under sand-laden or high-velocity service is not merely an efficiency concern.
Note also that ASME B16.25-2022 introduced changes: inch and metric tables were consolidated, figures and tables renumbered, the former mandatory Appendix I deleted, and referenced standards updated. Where a project specification cites an earlier edition, drawing review should confirm edition consistency, since deletion of an appendix can remove the basis for certain special bevel forms.
An easily overlooked point: once welded to the valve, the pup piece is treated contractually and technically as part of the valve itself.
Unless the purchaser supplies pipe, the pup piece must be furnished as a forging of a grade not lower than the higher of the body and pipeline material grades
The pup piece must be welded (and heat treated as required) before valve trim is installed
The transition taper slope must not be steeper than 1:4 (consistent with ASME B31.3 Fig. 328.4.3, ASME B31.8 Fig. I5 and ISO 13847 clause 7.7)
Valve body wall thickness to pipe wall thickness ratio not to exceed 1.5:1
Pipe SMYS to valve body material yield strength ratio not to exceed 1.5:1
Weld end valves must be supplied complete by the manufacturer or its designated subcontractor, to ensure welding compatibility and body/seal performance
Where a subcontractor welds, the manufacturer must inform them in writing of the maximum permitted body temperature and required PWHT conditions; WPS and PQR must be approved
Heat treated delivery condition must be marked on the pup piece with a low stress stamp
The 1:4 taper slope and 1.5:1 thickness ratio clauses sit at the root of many field problems. Exceeding either ratio moves the stress concentration point from the weld into the transition zone — which frequently falls outside NDT coverage.
Material grade matching contains a real contradiction. A common ASTM A216 WCB carbon steel body carries a carbon equivalent (CE) ceiling of 0.50, and where the actual value exceeds 0.45 preheat to 100 °C is required to prevent hydrogen induced cold cracking. The mating API 5L X70/X80 line pipe sits at a CE of roughly 0.38. The two are not equivalent in strength or hardenability. When a WCB body is welded into X80 line pipe, ER70S-6 class filler is typically selected to balance the strength differential — meaning the weld metal may be lower strength than either parent material. Design should acknowledge that rather than design around it.
A pup piece extends the heat path but does not cancel the trigger conditions for PWHT. Where pup piece wall or weld thickness reaches the threshold, PWHT remains mandatory.
Under ASME B31.3, PWHT is mandatory for carbon steel piping above 19 mm wall thickness. Typical parameters:
Temperature range: 595–650 °C
Holding time: 1 hour per 25 mm of wall thickness (commonly stated as 1–2 hours at 595–705 °C)
PWHT does three things: hydrogen removal, residual stress relief, and HAZ microstructure improvement. Of these, hydrogen removal is a safety item for high strength steel welding, while residual stress relief is an SSC resistance item for sour service — and the latter is often the genuine limiter on design life.
One comparison worth putting in front of the procurement decision: welds without controlled PWHT show a stress corrosion cracking (SCC) incidence of roughly 8%–12% within 3 to 5 years, while controlled welds remain at zero SCC over 8 years. The difference is not welder skill; it is whether the heat treatment procedure was enforced.
The practical difficulty is that pup piece welding must be completed before trim installation. That means the furnace or local heating temperature during PWHT conducts through the complete valve body assembly. If a manufacturer installs trim first to save schedule, the sealing elements will already have experienced one over-temperature excursion even if PWHT is performed later.
Pup piece length and bevel geometry are variables controllable before dispatch. Internal leakage more often originates on site — in construction management rather than design.
Transport and lifting damage — a long pup piece lengthens the moment arm, significantly raising bending stress on the valve end under impact
Inadequate site protection — sand and debris entering the cavity through the open pup piece end
Loss of welding heat control — preheat, interpass and welding temperatures not held to the manufacturer's seat temperature limits, damaging seal rings and sealing surfaces
Valve not in the full open position during welding — spatter adhering to the ball surface
Slag not removed — scratching sealing surfaces during reassembly or pressure testing
Hydrostatic strength test with the valve full open at excessive pressure — seat subjected to differential pressure beyond design conditions
Items 3 and 4 account for the majority of field leakage causes. Their common feature: neither is a valve quality problem — both occur because the welder does not know the valve has a temperature limit. That in turn says something about specification practice: printing the maximum permitted body temperature in the installation manual is not sufficient; it has to be confirmed as received at the working level.
There is a further structural problem that typically surfaces only after mechanical completion: once welded, valve orientation cannot be adjusted the way a flanged valve can rotate. Handle or actuator orientation is fixed at the factory, and the field must complete pipe routing to that orientation exactly. If interference with adjacent piping or platform structure is discovered during construction, the only remedy is cutting out the weld — which means repeating the PWHT.
The most consistently underestimated part of the selection discussion is construction-side cost. For a Class 300, NPS 6 ball valve:
Item | Butt weld end (extended) | Flanged |
Field welding + NDE + heat treatment | approx. 12–20 man hours | approx. 1.5–2.5 man hours |
Productivity ratio | — | weld end is roughly 1/5 to 1/13 of flanged |
Crew impact | A 6-welder crew switching to weld end valves loses 40%–60% productivity | — |
One real case: a 2024 ethylene plant turnaround scheduled 18 days for six NPS 8 weld end ball valves; actual duration was 41 days, a 128% overrun. The overrun came primarily from serial waiting on PWHT scheduling, repairs and follow-up NDT re-inspection — not from the welding itself.
Everything above points to one conclusion: the extended weld end trades construction complexity for sealing reliability. It therefore has a definable applicability boundary.
Pipe sections requiring frequent removal for maintenance — removing a weld end valve is a hot work operation every time
Installation points where orientation is uncertain or may change later — valve orientation is locked at the factory and cannot be adjusted in the field
Inlet points to equipment subject to in-service replacement — strainers and flow meters, for example, with high maintenance frequency
Sites without PWHT execution capability or verifiable furnace temperature records — in sour service this is not a process preference but a safety floor
Platform interstitial spaces too confined to stage welding equipment and heat treatment gear
Buried or subsea pipelines where no flanged connection point can be provided
Critical isolation points where the cost of leakage far exceeds the cost of installation
Projects whose specification explicitly invokes supplementary requirements such as IOGP S-562, where welding and heat treatment procedures are fully controllable
Cases where the pipeline-to-body material combination has been assessed for compatibility by the designer and WPS/PQR are approved
In subsea and platform applications, pup piece length has a wider impact than onshore, because three additional variables enter.
Subsea valves receive protection through an anode system, and a longer pup piece increases both the protected surface area and the electrical distance involved. CP design under the NORSOK M-503 framework needs to account for the pup piece. NORSOK U-001 additionally requires verification of electrical continuity for the CP system — resistance between pup piece and valve body is a measured item, not an assumed one.
NORSOK M-001 restricts Alloy 625 (UNS N06625) in structural applications to ambient temperature, with Table 10 giving a maximum seawater operating temperature of 30 °C. The same standard identifies Alloy 625 and stainless steels with PRE ≥ 40 as "critical materials" where cathodic protection is absent, and states they shall not be used in mechanical connections where material temperature exceeds ambient seawater temperature — threaded connections being especially susceptible to crevice corrosion.
The relevance to extended weld ends: the connection between pup piece and body cannot be selected on strength alone; it must also be checked against whether it falls inside the prohibited zone for critical materials. Super duplex stainless steel, which does not require overlay in these services, is a common alternative route.
NORSOK U-001 requires UNS N06625 (Alloy 625) overlay on all critical areas, including seat pockets and surfaces between mating parts where crevices can form. Seat and seal spring materials are restricted to UNS R30003, R30035, Alloy 625, C276 or NiAl bronze; subsea valve bodies and bonnets must be forgings, rolled products or hot isostatically pressed parts.
For the subsea ball valve body itself, a further set of features must be coordinated with the pup piece design: metal-to-metal seat sealing (tungsten carbide coating 0.2–0.4 mm thick), non-elastomeric sealing combinations (ELGILOY + PEEK + PTFE lip seals), anti-blowout stems, double piston effect (DPE) or single piston effect (SPE) design, DBB / DIB-1 / DIB-2 sealing configuration, CRA (Inconel 625) overlay, and ROV or diver interface (per API 17H Class 7 torque tool interface). Safety integrity level is typically assessed to IEC 61508, with SIL 3 required for critical isolation valves.
The cumulative effect: pup piece length on an offshore project cannot be determined by a single dimension. It is simultaneously constrained by welding heat management, CP design, material temperature limits and overlay requirements. Neglecting any one can force rework at FAT or during commissioning.
Compressed into an executable checklist, the technical specification for a weld end valve should state at minimum:
Pup piece length, and the basis for that length (heat management calculation, specification clause, or tier table reference) — "same as previous revision" is not a basis
Pup piece material grade, heat treatment condition, and low stress stamp marking requirement
Maximum permitted body temperature, with confirmation that the value has been incorporated into site construction procedures
The ASME B16.25 edition year governing bevel form and dimensional tolerance
PWHT trigger conditions, temperature range, holding time, and record submission requirements
Confirmation that body/pipe wall thickness ratio and SMYS-to-yield ratio satisfy the 1.5:1 limits
Confirmation that pup piece welding and heat treatment precede trim installation, with process records
For offshore projects additionally: measured CP electrical continuity values, overlay extent and material, and a material temperature limit compliance statement
An extended weld end is a design decision that must be justified, not a default setting that can be inherited. Its value is moving welding risk away from the seat. Its cost shows up in bending moments, man hours, locked orientation and heat treatment procedure. Until both sides are written into the specification, the selection is not finished.
No global mandatory standard exists. API 6D itself does not specify extension length; 150 mm is the most widely adopted industry practice value, with large bore sizes typically tiered at 200 / 500 / 800 mm. IOGP S-562 requires butt-welded end valves to be supplied with pup pieces and provides a length table, so on projects invoking that supplementary requirement the length can become contractual.
No. The direct benefit is pushing the welding HAZ away from the seat, but it brings four countervailing costs: a longer moment arm at the valve end, higher deformation risk in transport and lifting, greater distance for subsea support and cathodic protection potential distribution, and more difficult field alignment and mismatch control. Length should be derived backwards from seat temperature limits and actual welding heat input; anything beyond that adds cost and risk without benefit.
PTFE undergoes irreversible deformation above approximately 200 °C, PEEK above approximately 260 °C. Actual control targets sit well below material limits: industry practice requires seat temperature below 120 °C for NPS 2–6, below 150 °C for NPS 8–14, below 180 °C for NPS 16–24, and below 200 °C for NPS 26–48.
Under ASME B31.3, PWHT is mandatory for carbon steel piping above 19 mm wall thickness. Typical practice is 595–650 °C with holding time of 1 hour per 25 mm of wall thickness. The purpose is hydrogen removal, residual stress relief and HAZ microstructure improvement. In sour service, welds without controlled PWHT show an SCC incidence of approximately 8%–12% within 3 to 5 years.
The difference lies mainly on the construction side rather than purchase price. For Class 300, NPS 6, field welding plus NDE plus heat treatment on a weld end valve takes approximately 12–20 man hours against 1.5–2.5 man hours for flanged — a productivity gap of roughly 5 to 13 times. A 6-welder crew switching to weld end valves loses 40%–60% productivity overall.
No. Handle or actuator orientation on a weld end valve is fixed at the factory and the field must complete pipe routing to that orientation. Unlike a flanged valve it cannot be rotated. If interference with adjacent structure is discovered during construction, the remedy is cutting out the weld, accompanied by a repeat PWHT.
It should be completed by the valve manufacturer or its designated subcontractor before trim installation. IOGP S-562 requires weld end valves to be supplied complete by the manufacturer or designated subcontractor to ensure welding compatibility and body/seal performance; the manufacturer must also inform the subcontractor in writing of maximum permitted body temperature and required PWHT conditions, and approve the WPS and PQR.
Three main constraints are added. NORSOK U-001 requires verification of cathodic protection electrical continuity, with pup piece to body resistance measured rather than assumed. NORSOK M-001 sets a maximum seawater operating temperature of 30 °C for Alloy 625 in structural applications, and identifies Alloy 625 and PRE ≥ 40 stainless steels as critical materials that shall not be used in mechanical connections where material temperature exceeds ambient seawater temperature without cathodic protection. Critical areas including seat pockets and crevice-forming mating surfaces require UNS N06625 overlay.
