[ELI-REVIEW: UG-27] is the paragraph of [ELI-REVIEW: ASME Section VIII, Division 1] that gives the required wall thickness of a cylindrical shell under internal pressure. In plain terms, it says the thinner the shell wall, the higher the hoop stress a given internal pressure creates, and [ELI-REVIEW: UG-27] sets the minimum thickness that keeps that hoop stress within the material’s allowable stress at the design temperature. The formula takes design pressure, shell radius, allowable stress, and weld joint efficiency as inputs and returns a required thickness before corrosion allowance is added.
The formula, in ColdCalcs’ own notation
Written out with each variable defined in plain terms (not reproduced from the code text itself), the [ELI-REVIEW: UG-27] internal- pressure circumferential-stress case is:
| Symbol | Meaning | Typical unit |
|---|---|---|
| t | Required shell thickness (before corrosion allowance) | in |
| P | Internal design pressure at the point being sized | psig |
| R | Inside radius of the shell (or an equivalent radius term on an outside-diameter basis) | in |
| S | Material allowable stress at design temperature | psi |
| E | [ELI-REVIEW: Longitudinal weld joint efficiency (0 to 1)] | — |
The relationship, in ColdCalcs’ own variable names, is [ELI-REVIEW: t = P × R ÷ (S × E − 0.6 × P)] on an inside-radius basis; an outside-diameter basis rearranges the same relationship to solve directly for the corroded inside radius. Corrosion allowance is added on top of this result, not folded into P, R, S, or E. The formula is only valid within a thin-wall regime — [ELI-REVIEW: roughly P ≤ 0.385 × S × E, per the code’s own stated validity limit]— beyond which a thick-wall analysis is required instead.
Why joint efficiency matters so much
E is a discount factor on the weld seam’s strength relative to the base plate: full radiographic examination of the longitudinal seam typically allows the highest E value ([ELI-REVIEW: E = 1.0 for an unwelded, no-seam head]), while spot or no radiography allows progressively lower values, which means a proportionally thicker (and heavier) shell for the same pressure and material. That trade-off — thinner plate with more radiography testing vs. thicker plate with less — is one of the first decisions a vessel design makes, and it also feeds into MAWP; see MAWP vs. design pressure.
Run it
PRELIM runs the full [ELI-REVIEW: UG-27] shell calculation (plus head thickness, MAWP back-check, and MDMT screening) from your inputs and shows the result on screen, free.
Open PRELIM →FAQ
Does [ELI-REVIEW: UG-27] apply to formed heads too?
No. [ELI-REVIEW: UG-27] covers the cylindrical shell specifically. Formed heads (ellipsoidal, hemispherical, torispherical) are sized under a separate set of formulas ([ELI-REVIEW: UG-32] and related paragraphs), which use different geometry terms appropriate to each head shape.
What happens if a design falls outside [ELI-REVIEW: UG-27]'s validity range?
The paragraph's thin-wall assumption stops applying once required thickness gets large relative to the shell radius. Past that point, a thick-wall analysis under a different part of the code is required, and a vessel design tool built only around the thin-wall formula should flag this rather than return an unsafe result.
Is [ELI-REVIEW: UG-27]'s shell formula the only thing that sets shell thickness?
No. External pressure (vacuum or jacketed service), MDMT-driven impact-test requirements, and mandatory full radiography by thickness can all push required or as-built thickness above what the internal-pressure [ELI-REVIEW: UG-27] case alone would require.
Can I reproduce the exact [ELI-REVIEW: UG-27] code text here?
No, and this page does not. ColdCalcs presents the underlying relationship in its own notation with plainly defined variables; the authoritative formula, exceptions, and validity limits live in the current edition of [ELI-REVIEW: ASME Section VIII, Division 1] itself.