Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ASME B31P-2017 — Standard Heat Treatments for Fabrication Processes
🔥 ASME B31P-2017 is a standard within the ASME B31 Code for Pressure Piping devoted specifically to heat treatment during fabrication and installation. It originated from a fairly concrete management problem: before 2011, the individual B31 code books — B31.1 (Power Piping), B31.3 (Process Piping), B31.4 and B31.8 (transportation pipelines) and the rest — differed in their rules for welding preheat and postweld heat treatment (PWHT), so the same weld could be judged differently depending on which code book applied. The B31 Standards Committee concluded that greater consistency was needed and directed the Fabrication & Examination Technical Committee to develop a single standard covering these rules, which became B31P.
The scope needs to be understood precisely: B31P does not itself require that any piping be preheated or given PWHT. Rather, when the applicable code section or the engineering design mandates preheat or PWHT, B31P states the requirements for carrying it out correctly. Because the code sections provide preheat and PWHT rules mainly for ferrous materials, B31P may also offer expanded rules and alternatives for a wider variety of materials, although not every possible material is covered. The standard was approved by ANSI on November 15, 2017, and issued on May 15, 2018.
💡 The key to B31P is to read it as an execution-level heat treatment procedure standard: the code section tells you whether to do it and to what temperature; B31P tells you how to measure, how to control, how to document, and under what conditions an exemption or an alternative is permitted. The two must be used together.
1. Structure and Scope of the Standard 📌
The body of the standard has eight sections, supported by five nonmandatory appendices and two form-based work documents, forming a chain from measurement through documentation, and from routine practice through alternatives:
| Section / Appendix | Content | Engineering Focus |
|---|---|---|
| Sections 1–2 | Introduction, glossary | Unifies terms such as heated band, control zone, gradient control band |
| Section 3 | Temperature measurement | Thermocouple type, attachment, number, representativeness |
| Section 4 | Temperature control | Heating and cooling rates, holding, uniformity |
| Section 5 | Welding preheat (Table 5.1-1) | Preheat by material group and thickness |
| Section 6 | Postweld heat treatment (Tables 6.1.1-1 to 6.1.1-3, 6.2-1) | Holding temperature, time, alternates, exemptions |
| Section 7 | Postforming and material heat treatments | Restoring properties after cold or hot forming |
| Section 8 | Alternatives and exemptions to preheat and PWHT | Legitimate routes to doing less |
| Appendix A | Recommendations for proper use of thermocouples | Dominant measurement error sources |
| Appendix B | Narrow-range local heat treatment control (incl. Form 1, Form 2) | Mandatory for CSEF steels such as P-No. 15E |
| Appendix C | Branch connection examples | Computing control thickness |
| Appendix D | Temper bead method as a PWHT alternative | Where furnace treatment is impossible |
| Appendix E | Fracture mechanics alternative to PWHT | Engineering critical assessment route |
2. Temperature Measurement and Temperature Control 🔧
The first gate in heat treatment quality is whether the measured temperature truly represents the workpiece temperature. Section 3 and Appendix A address this in detail: thermocouple type and wire gauge, the method of attaching the hot junction to the surface (direct contact, capacitor-discharge welding, banding), and thermal insulation between leads and heating elements all materially affect the reading. The most common systematic error in the field is a hot junction separated from the pipe wall by a gap or a layer of insulation, so the reading lags the true wall temperature — especially dangerous during the heating ramp.
Section 4 sets the temperature control requirements. The core parameters are heating rate, holding temperature and time, cooling rate, and temperature uniformity among measurement points during the hold. For hardenability-sensitive materials such as Cr-Mo and CSEF steels, heating and cooling rates above roughly 427 °C (800 °F) must be controlled, commonly scaled by thickness (for example, not exceeding 222 °C/h divided by the maximum thickness in inches, and never exceeding 222 °C/h) to avoid excessive gradients and thermal stress.
⚠️ Heat treatment is not simply “getting to temperature”. Heating too fast creates a large through-thickness gradient in heavy sections and thermal stresses that can exceed yield; cooling too fast, particularly in Cr-Mo steels, may re-harden the material and produce cold cracking. B31P’s rate limits matter as much as its temperature limits.
3. Welding Preheat (Section 5) 🔥
Preheat reduces the temperature gradient at the weld, slows the cooling rate, and encourages hydrogen escape, thereby suppressing cold cracking. Table 5.1-1 states preheat requirements by material group (P-Number) and thickness. The following typical magnitudes are given for orientation only — the actual values must be read from Table 5.1-1, with account taken of carbon equivalent, restraint, and hydrogen level:
| Material Group | Typical Materials | Typical Preheat Magnitude | Notes |
|---|---|---|---|
| P-No. 1 | Carbon steels (A106, A53, A516, etc.) | About 10 °C thin wall; about 100 °C heavy wall | Transition thickness commonly near 25 mm; raise with carbon equivalent |
| P-No. 3 | 1¼Cr-½Mo | About 100–150 °C | Marked hardenability; control cooling rate |
| P-No. 4 | 2¼Cr-1Mo | About 150–200 °C | Often combined with hydrogen bakeout |
| P-No. 5 | 5Cr-½Mo and similar | About 200 °C and above | Insufficient preheat readily causes cold cracking |
| P-No. 15E | CSEF steels (e.g., Grade 91) | About 200 °C and above | Preheat and interpass temperature both tightly controlled |
Beyond the preheat temperature itself, interpass temperature is also controlled: insufficient preheat causes cracking, while excessive interpass temperature can degrade weld metal toughness. Code sections and this standard normally impose an upper interpass limit, especially relevant to thick, highly restrained joints.
4. Postweld Heat Treatment (Section 6) 🌡️
PWHT serves three purposes: relieving welding residual stress, softening hardened heat-affected zone structures, and improving resistance to stress corrosion cracking and brittle fracture. Table 6.1.1-1 gives holding temperature and time by material group; Table 6.1.1-2 gives alternate PWHT requirements for carbon and low-alloy steels (P-Nos. 1 and 3); Table 6.1.1-3 lists exemptions to mandatory PWHT; and Table 6.2-1 gives approximate lower transformation temperatures — critical data, because the PWHT holding temperature must stay below the lower critical temperature or the material’s properties will be destroyed.
| Material Group | Typical Holding Range | Holding Time | Key Limit |
|---|---|---|---|
| P-No. 1 (carbon steel) | About 595–650 °C (1 100–1 200 °F) | About 1 h per 25 mm, 1 h minimum | Must not approach the lower critical temperature |
| P-No. 3 (1¼Cr-½Mo) | About 650–705 °C (1 200–1 300 °F) | As above | Consider temper embrittlement susceptibility |
| P-No. 4 (2¼Cr-1Mo) | About 675–760 °C (1 250–1 400 °F) | As above | Control heating/cooling rate in heavy wall |
| P-No. 15E (CSEF steels) | About 730–775 °C (1 350–1 425 °F) | Per standard, typically longer than carbon steel | Narrow window; Appendix B mandatory |
🚨 CSEF steels (P-No. 15E, such as Grade 91) have a very narrow PWHT window: too low and residual stress relief and softening are incomplete; too high and the material may enter the austenitizing range or lose significant creep strength. The standard therefore makes Appendix B mandatory for local PWHT of these steels — this is a requirement, not a recommendation.
5. Narrow-Range Control for Local Heating (Appendix B) 🎯
Appendix B is the most technically demanding part of B31P. For controlled local PWHT using electric resistance heating pads, it defines the arrangement of the heated band, control zone, gradient control band, and insulation, and it provides thermocouple layouts graded by pipe size:
- Smaller than NPS 6 (DN 150): one control zone, with control and monitoring thermocouples per Figure B-5.2-1.
- NPS 6 to NPS 12 (DN 150–300): two control zones (Figure B-5.2-2).
- NPS 14 to NPS 30 (DN 350–750): four control zones (Figure B-5.2-3).
- Branch nozzle or attachment welds: dedicated monitoring thermocouple layout (Figure B-5.2-4).
Appendix B also covers the case where a heated band from a weld requiring PWHT intersects a weld that does not require PWHT (Figure B-5.3-1), and it supplies Form 1 (standard procedure for local heating) and Form 2 (standard documentation checklist for local heating) so field execution has a verifiable written basis. Table B-5.1-1 summarizes the requirements for each control zone.
💡 The inherent difficulty of local PWHT is the temperature gradient: outside the heated band there must be a transition region, and the gradient in that region creates new thermal stress. The control zone must therefore be wide enough and the insulation long enough that the axial gradient stays within limits. For longitudinal seams, the standard explicitly advises avoiding local PWHT where possible, because longitudinal welds carry higher pressure-induced hoop stress.
6. Alternatives and Engineering Practice 🛠️
Where furnace PWHT is not feasible — a final tie-in weld on a large field-erected line, or equipment already in place and immovable — B31P provides two legitimate alternatives:
- Appendix D, temper bead method: careful control of bead placement lets the thermal cycle of subsequent beads temper the hardened zones of earlier passes and the heat-affected zone, achieving toughness recovery equivalent to full PWHT without it. The method demands tight control of welding parameters, bead overlap, and operator skill.
- Appendix E, fracture mechanics alternative: an engineering critical assessment demonstrates that the joint without PWHT retains adequate fracture resistance under design and test conditions, normally supported by hardness caps, nondestructive examination, and more frequent inspection.
- 🔍 Control thickness (Appendix C): at a branch connection, the “control thickness” is neither the run wall nor the branch wall but is computed from the typical configurations in Figures C-1 and C-2 (including integrally reinforced outlets), because it directly drives holding time.
- 📊 Documentation: time-temperature charts, thermocouple layout drawings, calibration records, operator qualifications, and completed Form 1/Form 2 checklists belong together in one package — the evidentiary basis for acceptance and for any future failure analysis.
- 🔧 Equipment and calibration: confirm power capacity, calibration status of thermocouples and recorders, and spare thermocouple inventory before starting, so an interruption does not leave the hold time short.
- ⚖️ Interface with the code section: B31P applies only where the code section requires preheat or PWHT. If the code does not require it, you cannot add heat treatment “per B31P” — unnecessary heat treatment can itself degrade material properties.
7. Frequently Asked Questions ❓
❓ Q: Does B31P mandate PWHT for every pressure piping weld?
A: No. B31P provides implementation requirements only where the applicable code section or the engineering design mandates preheat or PWHT. Whether it is required, and at what temperature and time, remains a matter for B31.1, B31.3, and the design documents. B31P answers “how to do it correctly”.
A: No. B31P provides implementation requirements only where the applicable code section or the engineering design mandates preheat or PWHT. Whether it is required, and at what temperature and time, remains a matter for B31.1, B31.3, and the design documents. B31P answers “how to do it correctly”.
❓ Q: Is local PWHT equivalent to furnace PWHT?
A: Not fully. Local heating inevitably creates an axial gradient and thus new thermal stress, and temperature uniformity is harder to control. The standard therefore requires adequately wide control zones and properly arranged gradient control bands, and imposes mandatory local-heating procedure requirements for materials such as P-No. 15E. Where conditions permit, furnace treatment remains preferable.
A: Not fully. Local heating inevitably creates an axial gradient and thus new thermal stress, and temperature uniformity is harder to control. The standard therefore requires adequately wide control zones and properly arranged gradient control bands, and imposes mandatory local-heating procedure requirements for materials such as P-No. 15E. Where conditions permit, furnace treatment remains preferable.
❓ Q: Why not simply raise the PWHT temperature “for extra safety”?
A: Because of the lower transformation temperature in Table 6.2-1. Once the holding temperature approaches or exceeds it, the material partially austenitizes and forms hard, brittle structures on cooling, degrading both strength and toughness. Excessive tempering temperature also reduces creep strength. The PWHT temperature is a window, not a target to maximize.
A: Because of the lower transformation temperature in Table 6.2-1. Once the holding temperature approaches or exceeds it, the material partially austenitizes and forms hard, brittle structures on cooling, degrading both strength and toughness. Excessive tempering temperature also reduces creep strength. The PWHT temperature is a window, not a target to maximize.
❓ Q: Why is thermocouple count graded by pipe size rather than weld length?
A: Because the central risk in local heating is circumferential non-uniformity. The larger the diameter, the harder it is to hold a uniform temperature around the circumference, so more independent control zones and monitoring points are needed. Appendix B’s grading — one, two, and four control zones — follows exactly this logic.
A: Because the central risk in local heating is circumferential non-uniformity. The larger the diameter, the harder it is to hold a uniform temperature around the circumference, so more independent control zones and monitoring points are needed. Appendix B’s grading — one, two, and four control zones — follows exactly this logic.
