Methodology
How calcnote computes ACI 318-19 column capacity via strain compatibility.
calcnote is a strain-compatibility implementation of ACI 318-19 for rectangular tied and circular (tied or spiral), uniaxially-loaded reinforced concrete columns — short columns by direct analysis, slender non-sway columns by §6.6.4 moment magnification. This page documents the code provisions it implements, the design choices it makes, and the independent verification we have run against it. For a step-by-step worked construction, see how to construct a column interaction diagram; for detailing rules, see column tie spacing requirements.
Independent verification
Verification Report — cross-validated against Wight and StructurePoint spColumn
Fourteen test cases spanning the input envelope (including the §6.6.4 slender witness and the circular tied/spiral cases added 2026-09), cross-validated against three independent authority types: Wight's Reinforced Concrete: Mechanics and Design (7th ed.), StructurePoint's published interaction-diagram examples (the canonical 16×16 column and the 20 in circular spiral column), and independent first-principles ACI 318-19 strain-compatibility derivations.
Directly-tabulated control points for the canonical 16×16 column.
Eccentricity at reference Pn; within published design-chart reading precision.
Independent Python strain-compatibility re-implementation, no calcnote source dependency.
calcnote-verification-report-v1.5.1.pdf
31 pages · 167 KB · full per-case derivations, three-authority cross-checks, ACI clause map.
Prefer HTML? Browse the verification results · the canonical 16×16 case, cross-checked four ways.
Range of applicability
Scope: rectangular and circular columns per ACI 318-19
Inputs outside this envelope are rejected before any calculation runs.
| Parameter | V0 range |
|---|---|
| Section shape | Rectangular tied (b × h); circular tied or spiral (D) |
| Section dimensions | 8 in ≤ b, h ≤ 60 in; 10 in ≤ D ≤ 60 in |
| Concrete strength | 3,000 psi ≤ f'c ≤ 10,000 psi (normal-weight) |
| Steel grade | Gr60, Gr80 |
| Number of longitudinal bars | rect: n ∈ {4, 8, 12, 16, 20}; circ: n ∈ {6, 8, 10, 12, 14, 16, 18, 20} |
| Bar size | #6, #7, #8, #9, #10, #11 |
| Transverse reinforcement | Ties #3, #4 (spiral bar #3, #4; §25.7.3 checks enforced) |
| Bar layout | rect: perimeter two-face / all-face; circ: equally spaced ring, checked at two ring positions |
| Bending axis | Uniaxial (rect: about X or about Y; circ: any direction, see ring orientation) |
| Column length | Short by direct analysis; slender non-sway by §6.6.4 magnification (§6.2.5 gate) |
Not in V0: biaxial bending · sway frames (§6.6.4.6) and columns with transverse loads between supports · non-circular unsymmetric sections · composite (steel + concrete) sections · torsion or shear-friction · fire-rating · seismic special-frame detailing (incl. §18 spiral seismic provisions) · post-tensioned reinforcement. Each will be re-verified in its own future report revision.
Code compliance
ACI 318-19 clauses implemented
Each provision is exercised by at least one of the fourteen verification test cases.
| ACI clause | Provision |
|---|---|
| §22.2.2.1 | Extreme-fiber concrete crushing strain εcu = 0.003 |
| §22.2.2.4.3 | β1 formula (Whitney stress-block factor) |
| §22.4.2.1 | Design cap φPn,max = 0.80·φ·Po tied / 0.85·φ·Po spiral (Table 22.4.2.1). Pn,max explained |
| §22.4.2.2 | Pure-compression nominal capacity Po |
| §21.2.2 | φ formula: 0.65 (tied) or 0.75 (conforming spiral) to 0.90 with linear transition; Grade-dependent εty |
| §6.6.4.5.4 | M2,min = Pu·(0.6 + 0.03·h) (h → D for circular) — advisory on the short branch, binding under §6.6.4 magnification (see design choice 03 below) |
| §10.6.1.1 | Longitudinal-steel ratio limit 0.01 ≤ ρ ≤ 0.08. Reinforcement limits explained |
| §6.2.5 | Non-sway slenderness limit klu/r ≤ min(34 − 12·M1/M2, 40); slender columns proceed via §6.6.4 magnification (added 2026-09). Slenderness limits explained · moment magnification factor explained |
| §20.5.1.3 | Minimum clear cover for columns (1.5 in standard) |
| §25.2.3 | Minimum clear bar spacing |
| §25.7.2.2 | Minimum tie size for given longitudinal bar size |
| §10.7.3.1 | Minimum 6 longitudinal bars in a spiral-confined ring (circular, added 2026-09). Bar count rules |
| §25.7.3.1–.3 | Spiral conformance: 1–3 in clear spacing, ≥ 3/8 in bar, ρs ≥ Eq. (25.7.3.3) — enforced as hard checks because Table 21.2.2's φ = 0.75 is conditioned on them (circular spiral, added 2026-09). Spiral requirements explained |
Design decisions
Design choices documented
calcnote makes three intentional design choices that an engineer reviewing its output should be aware of. All three are aligned with the default behaviors of every commercial RC column-design tool in the same market segment (StructurePoint spColumn, RISA, ENERCALC, SkyCiv).
Positive-(P, M) quadrant polygon truncation
The interaction-diagram polygon is constructed in the positive-(P, M) quadrant only. At high reinforcement ratios where a fixed-εt anchor would otherwise show a small negative-P excursion (axial tension with bending), the affected neutral-axis depth is clamped to cpb (the pure-bending value) and the anchor collapses onto the pure-bending point. This matches the default rendered behavior of all major commercial RC column-design tools. The cap, balanced point, pure-bending capacity, and operating-point ray-scaling result are all unchanged.
Ray-scaling for operating-point capacity
For a user demand (Pu, M2), calcnote reports capacity at the intersection of the interaction polygon with the ray from the origin through the demand point. This is the ACI 318-19 standard interpretation of “capacity at a demand point,” used by every commercial RC column-design tool. Both demand-side D/C ratios collapse to a single scalar 1/t by construction.
Minimum eccentricity (M2,min) per ACI 318-19 §6.6.4.5.4
ACI 318-19 §6.6.4.5.4 defines M2,min = Pu·(0.6 + 0.03·h) within the moment-magnification procedure for slender columns (§6.6.4 is titled "Slenderness Effects"). For short columns this bound is not code-mandated; calcnote retains it there as an advisory check accounting for construction-tolerance eccentricity and unmodeled lateral effects that a first-order short-column analysis may not capture. When user-entered M2 falls below M2,min on a short column, calcnote emits an advisory and proceeds using the user-entered M2; the engineer of record retains the substitution decision. On a slender column (2026-09, §6.6.4 branch) the provision is binding: magnification applies to max(M2, M2,min), with Cm = 1.0 when M2,min governs.
Circular sections
Circular columns: ring orientation
The bars of a circular column sit on a ring, and nothing on site fixes which way the ring faces relative to the direction of bending. Turning the ring moves some bars closer to the compression face and others farther from it, so the capacity along a given demand direction changes with the ring's orientation. The section's radius of gyration r is the same in every direction; its strength varies with ring orientation.
calcnote checks two ring positions: one bar on the compression axis (the arrangement in StructurePoint's worked example, verified as TC-12) and the ring turned half a bar spacing. It builds the φ-reduced capacity polygon for each, checks your demand against both, and reports the larger D/C as the verdict. The result page and the .docx name the ring position that governs, or state that both ring positions agree within 0.05 %, and give the other position's D/C.
Between the two checked positions the ring can sit at other angles. The figures below compare the weaker of the two checked positions with the weakest ring rotation found, separately for 6-bar rings (Class A) and for rings of 8 to 20 bars (Class B).
On exact capacity curves, the weaker of the two checked positions has more capacity than the weakest ring rotation found along the same demand direction by 4.14 %, the largest found by a search over 834 6-bar sections (49 ring rotations and edge sweeps near both checked positions on every section; dense scanning and jump-edge bisection on 375 of them) (2026-09-25); finer sampling finds larger values.
On exact capacity curves, the weaker of the two checked positions has more capacity than the weakest ring rotation found along the same demand direction by 3.13 %, the largest found by a search over 1,149 sections with 8 to 20 bars (49 ring rotations and edge sweeps near both checked positions on every section; dense scanning and jump-edge bisection on 322 of them) (2026-09-25); finer sampling can find larger values.
Sections searched: circular rings of 6, 8, 10, 12, 14, 16, 18 and 20 bars in D = 10, 12, 13, 14, 15 and 16 in with #6, #8, #10 or #11 bars and D = 16.5, 17 and 18 in with #10 or #11 bars; tied (#3 ties, #4 for #11 bars) and spiral (#4 spiral); f′c 5,000 to 10,000 psi (5,000, 8,000 and 10,000), with an f′c 3,000 check on 6-bar sections (D 10 to 16.5 in, clear cover 1.5 and 2.0 in; its largest found value did not exceed the largest found over all f′c 5,000 6-bar sections searched, so it enters no figure on this page); fy 60 and 80 ksi; clear cover 1.5, 2.0, 2.5 and 3.0 in; plus D = 24 in with #9 bars and D = 30 in with #11 bars at f′c 10,000 psi, fy 60 and 80 ksi and clear cover 1.5 and 3.0 in. Combinations that fail the layout, spiral or 1 to 8 % steel-ratio checks are dropped, leaving 834 sections in Class A and 1,149 in Class B. Input values between or beyond the listed ones (D, clear cover and f′c), other bar sizes, and other tie or spiral choices are not sampled by the search. Demand directions (angles in the plane of P [kip] and M [kip-ft]): every 1° between pure compression and pure bending, pure bending itself, and the direction through each polygon corner of both positions except the pure-compression corner. Basis: exact φ-reduced capacity in the plane of bending at any neutral-axis angle, cut at the axial cap. Every strain state is enumerated, including a neutral axis slightly tilted from the bending axis.
| Bars in ring | Largest found, weaker checked position vs weakest ring rotation found |
|---|---|
| Class A · 6-bar rings | |
| 6 | 4.14 % |
| Class B · rings of 8 to 20 bars | |
| 8 | 3.13 % |
| 10 | 2.26 % |
| 12 | 1.22 % |
| 14 | 0.73 % |
| 16 | 1.12 % |
| 18 | 0.47 % |
| 20 | 0.48 % |
Largest found for each bar count by the same search as above: over the 834 Class A sections for the 6-bar row, and over the 1,149 Class B sections, grouped by bar count, for the 8- to 20-bar rows (2026-09-25); finer sampling can find larger values.
Where the dips lie: a dip is a ring rotation where capacity along a demand direction is lower than at both checked positions. The largest dips found lie within a fraction of a degree of the half-spacing position. These largest dips come from the edge of the rectangular stress block (ACI 318-19 §22.2.2.4.1) in the idealized section model the verdict uses: when a bar crosses the edge of that block, the force the model assigns to that bar changes abruptly, so capacity along a demand direction changes abruptly as the ring turns. Smaller dips also occur at other rotations. All of these dips are part of the capacity of the model the verdict uses; the verdict checks only the two ring positions, and the Class A and Class B figures above state how much more capacity the weaker of those two positions has than the weakest rotation found.
These figures and the polygon-accuracy figures in the next section are measured on different bases, so they do not add: exact curves at any neutral-axis angle here, and the conventional curve with the neutral axis parallel to the bending axis there.
The only combined figures
The comparison that relates a reported circular D/C to the weakest rotation found is between the two-position verdict polygon itself and the weakest ring rotation found. On that comparison the polygon has more capacity by 3.45 % for 6-bar rings (Class A), the largest found by a search over 834 sections (49 ring rotations and edge sweeps near both checked positions on every section; dense scanning and jump-edge bisection on 375 of them) (2026-09-25), and by 2.73 % for rings of 8 to 20 bars (Class B), the largest found by a search over 1,149 sections (49 ring rotations and edge sweeps near both checked positions on every section; dense scanning and jump-edge bisection on 322 of them) (2026-09-25), over every demand direction listed above; finer sampling can find larger values. Basis: exact φ-reduced capacity in the plane of bending at any neutral-axis angle. They are the only figures on this page that combine ring orientation with polygon accuracy.
What this means near 1.00: if the governing D/C (shown to at least 4 decimals in the Ring position row of the result page and in the .docx Result Summary) is 0.966 or higher for a 6-bar ring, or 0.973 or higher for a ring of 8 to 20 bars, it can exceed 1.00 at the weakest ring rotation found. 0.966 is 1 / (1 + 3.45 %) and 0.973 is 1 / (1 + 2.73 %), each rounded down to three decimals, on the basis of the combined figures above. The result page does not flag this band.
Polygon accuracy
How close the capacity polygon is to the exact curve
calcnote checks your demand against a polygon through seven anchor points on the φ-reduced interaction curve, not against the curve itself. Between anchors a straight edge can sit outside the curve (more capacity than the curve along that demand direction) or inside it (less). Basis: the conventional exact uniaxial curve, computed by strain compatibility with the neutral axis parallel to the bending axis and dense in neutral-axis depth. It is the curve spColumn, StructurePoint's worked examples and textbook derivations compute, and the curve calcnote's anchors sit on.
| Section | Compared with | Largest outside | Largest inside | Sections |
|---|---|---|---|---|
| Rectangular | the section's conventional curve | 0.53 % | 9.76 % | 43 |
| Circular, Class A (6 bars) | the weaker of the two ring positions' conventional curves | 3.40 % | 34.40 % | 834 |
| Circular, Class B (8 to 20 bars) | the weaker of the two ring positions' conventional curves | 2.65 % | 31.49 % | 1,149 |
Largest measured over the sections shown (2026-09-25), along every 1° demand direction and each polygon-corner direction except the pure-compression corner; the directions listed separately below are not included. Outside and inside are relative to the curve along the same demand direction. The circular rows use the sections searched in Circular columns: ring orientation, split into 6-bar rings (Class A) and rings of 8 to 20 bars (Class B).
Measured separately, because on these directions the cap-intersection and pure-bending root-finding steps also affect the result: directions that meet the axial-cap edge, largest outside 0.01 % and inside 0.01 % (rectangular), largest outside 0.01 % and inside 0.01 % (circular, Class A), largest outside 0.01 % and inside 0.01 % (circular, Class B); pure-bending directions, largest outside 0.04 % and inside 0.05 % (rectangular), largest outside 0.11 % and inside 3.47 % (circular, Class A), largest outside 0.08 % and inside 3.05 % (circular, Class B).
This basis leaves out strain states with the neutral axis slightly tilted from the bending axis. For the mirror-symmetric bar layouts calcnote uses, those states can only add capacity, so leaving them out can keep a D/C the same or make it higher; it does not lower one. That statement is about tilted states only; the outside figures above still stand.
For circular columns, ring rotations between the two checked positions are covered in Circular columns: ring orientation, on a different basis; the two sets of figures do not add, and the combined circular figures are given there.
Market context
Compared to StructurePoint spColumn
calcnote's scope is a strict subset of spColumn's — rectangular tied and circular (tied or spiral) uniaxial only (short and slender non-sway), no biaxial, no sway frames. Within that shared scope, calcnote's cap-point control values match spColumn's tabulated 16×16 example to ≤ 0.03 %, and its circular spiral control points match StructurePoint's spColumn-validated 20 in example to the printed digit on five of the six compared points (pure bending within 0.008 %).
spColumn is a paid, mature desktop application with vastly broader scope; calcnote is a free web tool covering one narrow slice of the same market. Use spColumn when scope demands it. Use calcnote when the narrow slice matches your job and you want an auditable, .docx-formatted calc-note.
FAQ
Frequently asked questions
Which ACI 318-19 clauses are implemented in calcnote V0? Which are not?
V0 implements: §22.4.2.1 Pn,max design cap — 0.80·Po tied / 0.85·Po spiral — with φ per §21.2.2 (0.65 tied / 0.75 conforming spiral); §22.4.2.2 Po pure-compression capacity; §21.2.2 φ formula with Grade-dependent transition; §10.6.1.1 steel ratio (0.01 ≤ ρ ≤ 0.08); §6.2.5 non-sway slenderness limits (r = 0.30·h rectangular, 0.25·D circular); §6.6.4 non-sway moment magnification (Cm, (EI)eff per §6.6.4.4.4(b), Pc, δns with the §6.2.5.3 1.4 limit — added 2026-09); Table 20.5.1.3.1 cover; §25.2.3 clear bar spacing; §25.7.2 tie requirements (size + spacing); §10.7.3.1 spiral minimum bar count and §25.7.3.1–.3 spiral conformance (clear spacing, bar size, ρs — added 2026-09 with circular sections).
Not implemented: biaxial bending (Bresler contour), sway-frame provisions (§6.6.4.6), §18 seismic/confinement detailing, §22.5 shear, §22.7 torsion, §25.4 anchorage/development, fire cover.
See the full ACI clauses table above for the itemized list.
Does calcnote support biaxial bending, sway frames, or circular sections?
calcnote covers: rectangular tied and circular (tied or spiral) sections, uniaxial bending, non-sway conditions — short columns by direct analysis and slender columns by §6.6.4 moment magnification (added 2026-09); circular sections added 2026-09 with the §25.7.3 spiral conformance checks enforced. Biaxial bending (Bresler contour method) and sway frames are deliberately excluded. The scope grows one verified branch at a time.
See the scope section for the exhaustive list.
When is a column “slender” per ACI 318-19?
ACI 318-19 §6.2.5 defines the non-sway short-column threshold as klu/r ≤ min(34 − 12·M1/M2, 40). For rectangular sections calcnote uses r = 0.30·h, and for circular sections r = 0.25·D, per §6.2.5. calcnote evaluates the full formula using the signed M1/M2 ratio (positive for single curvature), so the limit ranges from 22 to 40 — stricter than 34 when the column bends in single curvature. Columns above the threshold are slender: calcnote applies non-sway moment magnification per §6.6.4 (Cm, (EI)eff per §6.6.4.4.4(b), Pc, δns), with hard stops at instability (Pu ≥ 0.75·Pc) and the §6.2.5.3 limit (δns > 1.4). Sway frames remain out of scope, and the magnification assumes no transverse loads between the column's supports (Cm per §6.6.4.5.3(a) from end moments; a laterally loaded column requires Cm = 1.0 per §6.6.4.5.3(b) and is outside calcnote's scope). Worked references: when is a column slender and the moment magnification factor.
Is ACI 318-19 §6.6.4.5.4 M2,min required for short columns?
ACI 318-19 §6.6.4.5.4 defines M2,min = Pu·(0.6 + 0.03·h) (h in inches, result in kip-in). The Code places the provision inside §6.6.4 (Slenderness Effects); its role in the moment-magnification procedure is to prevent under-estimating M2 when computed eccentricity is small. calcnote's posture forks by branch (2026-09): advisory for short columns (where §6.6.4 does not apply), binding under moment magnification — the slender branch magnifies max(M2, M2,min) with Cm = 1.0 when M2,min governs.
calcnote treats M2,min as advisory outside that procedure: the engine emits a non-blocking advisory and proceeds with user-entered M2; the engineer of record retains the substitution decision.
See minimum eccentricity and errata for full rationale, or the full M2,min treatment.
How does calcnote apply the ACI 318-19 §21.2.2 φ factor?
ACI 318-19 §21.2.2.1 defines φ as 0.65 for compression-controlled sections, 0.90 for tension-controlled sections, and a linear transition in between. The tension-controlled limit is Grade-dependent (εt ≥ εty + 0.003), superseding 318-14’s fixed 0.005 threshold. calcnote applies the formula per §21.2.2.1 without approximation: Grade 60 gives εty ≈ 0.00207 and tension-controlled limit ≈ 0.00507; Grade 80 gives εty ≈ 0.00276 and tension-controlled limit ≈ 0.00576. See the full Grade 60 vs Grade 80 comparison.
See φ transitions for the full derivation.
How does calcnote compute the P-M interaction diagram?
calcnote uses strain compatibility: for each trial neutral-axis depth c, the engine computes rebar strains per Bernoulli plane-sections, applies the Whitney stress block to concrete, and integrates force and moment about the plastic centroid to yield (Pn, Mn).
The curve is capped at Pn,max = 0.80·Po tied / 0.85·Po spiral per §22.4.2.1 and traced through compression-controlled, balanced, and tension-controlled anchors to pure bending. φ per §21.2.2 is applied at each point; the resulting φ-reduced envelope is the capacity polygon the demand point is checked against. For circular sections the polygon is built for two ring positions (one bar on the compression axis, and the ring turned half a bar spacing); the demand is checked against both, and the larger D/C is the verdict.
Is there a free alternative to StructurePoint spColumn?
calcnote is free and covers a deliberate subset of what spColumn does: rectangular tied and circular (tied or spiral) columns, uniaxial bending, short and slender (non-sway) columns per ACI 318-19. For that subset, calcnote’s engine has been cross-validated against spColumn within 0.03% on Pn and Mn on the canonical 16×16 case, and against StructurePoint’s spColumn-validated circular spiral example to the printed digit on five of its six compared points (pure bending within 0.008%). spColumn covers additional geometries (irregular sections, biaxial bending) and sway-frame magnification; calcnote covers non-sway moment magnification but not those.
Can I edit the calc-note before submittal?
Yes — calcnote outputs .docx so project labels, headers, and reviewer notes can be added in Word before the engineer of record signs and exports to PDF. The generated file is a working draft, not a locked deliverable.
For the engineer of record
Verification establishes that calcnote's calculations agree with the cited authorities within their respective precision tolerances. It does not constitute a license to use calcnote's outputs without independent engineering review.
calcnote outputs are draft calculation notes intended to assist a licensed structural Professional Engineer. Final designs require independent engineering review and signature/seal by a licensed Engineer of Record. The .docx output is not a licensed deliverable; this is stated as a footer on every generated page.
Errata
Corrections
-
Erratum 1 (Verification Report v1.0): §6.6.4.5.4 M2,min applicability
§6.6.4.5.4 M2,min applicability. The Verification Report v1.0 described ACI 318-19 §6.6.4.5.4 as a code-required check for calcnote's short-column scope, and the engine enforced M2 ≥ M2,min as a hard error. As of 2026-08-05, this is corrected: §6.6.4.5.4 formally scopes the M2,min = Pu·(0.6 + 0.03·h) bound to slender columns under moment magnification (§6.6.4 title "Slenderness Effects"; commentary R6.6.4.5.4 limits it to "when moment magnification is required"). calcnote's V0 scope (short columns only, klu/r within the §6.2.5 limit) is therefore not code-required to enforce it. calcnote now emits a non-blocking advisory when M2 < M2,min and proceeds using the user-entered M2. The Verification Report was regenerated as v1.1 (2026-08-24) and now reflects the correction; see its §5.4 and revision history.
Cross-validation coverage. The 8 verification test cases in v1.0 all use M2 > M2,min, so v1.0 could not cross-validate the calc-proceeds-with-advisory behavior in the M2 < M2,min regime. Report v1.1 verified that the advisory fires with the correct M2,min value and that the calculation proceeds with the user-entered moment (its TC-4); report v1.2 (2026-08-25) closes the gap with operating-point cross-validation inside the regime itself — two new cases (TC-9, TC-10) verified against independent first-principles ACI 318-19 derivations (we found no published textbook or vendor example for this regime).
-
Erratum 2 (circular columns): bar ring checked in one orientation only
- Affected
- Circular-column results produced from 2026-09-12 16:30Z until this correction went live. Results and .docx files produced with the correction show verdict method 2 or later. Rectangular columns are not affected.
- What was wrong
- calcnote checked the bar ring in one orientation only: one bar on the compression axis. A ring turned half a bar spacing can have less capacity along some demand directions, so a reported D/C could be lower than the weaker of the two positions gives. The error was unconservative: the D/C could be under-reported. The effect is largest for 6-bar rings. The correction checks two ring positions. Other rotations can have less capacity; ring orientation gives the largest found differences for 6-bar rings and for rings of 8 to 20 bars.
- How large
- Re-checked with both ring positions, a D/C rose by up to 8.31 %, largest measured over 55 sections (2026-09-25): circular rings of 6, 8, 10, 12, 14, 16, 18 and 20 bars in D = 16 in with #6, #8 or #11 bars, D = 24 in with #9 bars and D = 30 in with #11 bars, tied (#3 ties, #4 for #11 bars) and spiral (#4 spiral), at f′c 5,000 psi, fy 60 ksi and 1.5 in clear cover, run as short columns, along every 1° demand direction between pure compression and pure bending (angles in the plane of P [kip] and M [kip-ft]), pure bending itself, and the direction through each polygon corner of both positions except the pure-compression corner. Combinations that fail the layout, spiral or 1 to 8 % steel-ratio checks are dropped. Example outside the measured sections: StructurePoint's 20 in spiral section (TC-12) at Pu = 500 kip and M2 = 200 kip-ft goes from D/C 0.8002 to 0.8047 and still passes.
- Who was exposed
- The measured window 2026-09-12 16:30Z → 2026-09-13 23:47Z recorded zero non-smoke circular calculation or document events; the rest of the window is being measured. Limits of this count: events are recorded only for sessions with an analytics id; bot-classified user agents are filtered out; a first valid calculation is counted once per session; our own smoke-test runs are identified by an internal tag or by their logged timestamps and excluded.
- What to do
- Re-run any circular-column calculation whose result page or .docx shows no verdict-method line, or shows verdict method 1. Circular documents without a verdict-method line were not produced with the correction. A result page opened before the correction and downloaded after it is re-run under the two-position check at download, so that .docx's D/C can differ from the page you saw; its Verdict method line shows which check produced it.
-
Erratum 3 (Verification Report v1.5): TC-13 end-to-end D/C
- What was wrong
- Verification Report v1.5 printed the TC-13 end-to-end D/C as 0.740 (circular tied column, D = 18 in, Pu = 300 kip, M2 = 100 kip-ft). The engine value for that run was 0.795.
- Effect
- The error was in the report's printed value, not in the calculator's result, which was 0.795 for this run before the ring-orientation correction and is 0.800 with it (see Erratum 2). The run passes at each of these values.
- Corrected in
- Verification Report v1.5.1, which states the run's two-position D/C and records the v1.5 value.
-
Erratum 4 (verdict method 3): b and h hint inversion on the input form
In versions before verdict method 3, the b and h hints on the input form were inverted: b was labeled “Perpendicular to bending axis” and h “Parallel to bending axis”, implying b was width across the bending plane and h was depth in the bending plane. The engine always used h as depth regardless of axis choice.
This version replaces the hints with physical axis labels (b along X, h along Y) and adds a section drawing showing the axis mapping. The engine now uses frame-resolved depth (h for about X, b for about Y).
This is a documentation and axis-model correction, not a change to how the engine computes capacity for the default (about X) case, which still uses h as depth.