SWBPIPE · The open manual

SWBPIPE

The Open Manual

The fundamentals, the method, the evidence and the source of a program for piping flexibility and stress analysis

Published openly, 2026 Price $0.00 · MIT licence Chirality AI Ltd · Calgary, Alberta
Forewordi

Foreword

This manual teaches piping flexibility analysis and shows, step by step, how SWBPIPE performs it: the fundamentals a stress engineer works from, the mechanics the solver implements, how to model a line, how a design passes to the program you validate in, the cases the solver is checked against with their hand calculations, and the source itself. It is written to be read, checked and taught from.

For more than fifty years the programs engineers use to analyse piping have been proprietary. Their vendors document and verify them, but the source that produces a result cannot be read or run by anyone else. This manual goes the other way. The mechanics behind any result can be followed through Part III and read in the source, and every verification case can be run again on your own machine.

The program is young. The status of this edition lists where it falls short of the method described here, and every such place in the text is marked Today. Nothing here is a design basis. SWBPIPE does not judge compliance with any code and does not vouch for its results; work that needs validation is validated in the program your industry accepts.

Contentsii

Contents

  1. Status of this Editioniii
    1. A. Results that are wrong today
    2. B. Models that cannot yet be built or solved
    3. C. Not yet in the program
  2. Part I. IntroductionI
    1. A. What flexibility analysis is for
    2. B. Where the beam model stops
    3. C. How the method grew
    4. D. The name: Shipman, Watts and Burrows
  3. Part II. Fundamentals of Piping FlexibilityII
    1. A. Thermal expansion and restraint
    2. B. Primary and secondary loads; pressure
    3. C. Load cases
    4. D. The displacement stress range
    5. E. Bends, SIFs and branch connections
    6. F. The elastic centre and the force method
    7. G. Cold spring; hot and cold reactions
    8. H. Supports and restraints in practice
    9. I. Spring and constant hangers
    10. J. Equipment nozzles and imposed movements
    11. K. Occasional loads; how results feed a code check
  4. Part III. The Method in SWBPIPEIII
    1. A. The 3D beam model
    2. B. Loads
    3. C. Supports
    4. D. Load cases and combinations
    5. E. Stress recovery
  5. Part IV. Modelling GuideIV
    1. A. Building a model
    2. B. Components
    3. C. Supports, hangers and nozzles
    4. D. Load cases for a typical job
    5. E. Code data you enter
    6. F. Checking a model
  6. Part V. Validation ElsewhereV
    1. A. Design here, validate there
    2. B. The export today
    3. C. What changes when you validate
  7. Part VI. VerificationVI
    1. A. System benchmarks · 1 case
    2. B. Elements and loads · 14 cases
    3. C. Stress recovery · 15 cases
    4. D. Nonlinear supports · 6 cases
    5. E. Algorithm and software checks · 26 cases
    6. W. Withdrawn · 2 cases
  8. Part VII. Theoryin preparation
    1. The derivation of every equation in the solver, published chapter by chapter
  9. GlossaryG
    1. Practitioner’s terms, with their equivalents in other programs
  10. SourceMIT
    1. The solver, the program and the schemas
From Part VI · VerificationVI–A1
A plane L-shaped pipe between anchor A at the bottom and anchor B at the right: a 3 metre vertical leg, a quarter-circle elbow of 0.5 metre radius, and a 4 metre horizontal leg, under a uniform temperature rise of 150 kelvin. ANCHOR A ANCHOR B T1 T2 L1 = 3.0 m L2 = 4.0 m R = 0.5 m X Y ΔT = +150 K UNIFORM, NO PRESSURE, NO WEIGHT ELBOW FLEXIBILITY × k
Fig. 3.—Plane L-bend between two anchors

An expansion loop, solved twice.

A pipe between two anchors, bent through a quarter circle, warmed by 150 K. The reactions are worked by hand with the force method, as the 1941 manual would, then computed by the solver’s bend element, and the two agree to seven significant figures for flexibility factors from 1 to 20.

Every case in Part VI is set out the same way: the problem, the inputs, the hand calculation, the solver’s result, what the comparison does and does not show, and the command to run it yourself.

Read the case →

SWBPIPE is named for W. H. Shipman, G. W. Watts and W. R. Burrows, credited in the 1941 Kellogg manual with the work its general method rests on.

The program: swbpipe.com · Made by Chirality AI Ltd · MIT licence