# Errata in ISO 5136

> One page of the phonometry documentation, as Markdown. Index: https://docs.jmrp.io/phonometry/llms.txt

Canonical: https://docs.jmrp.io/phonometry/reference/errata/iso-5136/
Language: en
Alternate: https://docs.jmrp.io/phonometry/es/reference/errata/iso-5136/index.md
Version: phonometry 4.0.0
Install: pip install "phonometry[full]"
Author: José Manuel Requena Plens (https://orcid.org/0000-0003-1250-6212)
Published: 2026-10-28
Updated: 2026-11-16T21:36:14Z
License: MIT (https://github.com/jmrplens/phonometry/blob/main/LICENSE)
Summary: Errata in ISO 5136: 8 defects in the printed text, with the evidence and what phonometry does about each.

---

These are the 8 entries of the [errata registry](https://docs.jmrp.io/phonometry/reference/errata/) on ISO 5136:2003, in the order the registry lists them. The registry's introduction says how each claim is checked against the printed page and what the status of an entry means.

## ISO 5136:2003, Table A.5, 5 000 Hz row (the leading digit of $a_3$ is missing)

- **Location:** Annex A, Table A.5, "Values of coefficients $a_i$ for the
  determination of the combined mean flow velocity and modal correction
  $C_{3,4}$ of the sampling tube for duct diameters 0,8 m $\le d <$ 1,25 m",
  row 5 000 Hz, column $a_3$.
- **The print:** $- ,24 \times 10^{-05}$: a minus sign, a space, a decimal
  comma and two digits, with no digit before the comma. Every other cell of
  the twelve coefficient tables of Annexes A, H and I prints one digit before
  the comma.
- **The problem:** the coefficient cannot be read from the document, and the
  row is inside the normative range of the standard (5 000 Hz,
  $|U| \le 40$ m/s). The $a_3$ of the same band in the two neighbouring
  tables is $-1.17 \times 10^{-5}$ (Table A.4, 0.5 m to 0.8 m) and
  $-1.27 \times 10^{-5}$ (Table A.6, 1.25 m to 2 m), which brackets
  $-1.24 \times 10^{-5}$; a leading digit of 2 or more would move
  $C_{3,4}$ at 40 m/s by 0.64 dB per unit of the digit
  ($a_3 U^3$ with $U^3 = 6.4 \times 10^4$), which no neighbouring band or
  table supports.
- **Evidence:** the cell as printed. PDF page 39 (printed p. 29) of
  ISO 5136:2003, against the same cell of Table A.4 on PDF page 38 (printed
  p. 28) and of Table A.6 on PDF page 40 (printed p. 30).
- **Library behaviour:** reads $-1.24 \times 10^{-5}$, the value the
  neighbours bracket, in `_TABLE_A5` of
  [`_sound_power_in_duct.py`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/emission/_sound_power_in_duct.py).
  The table's comment and
  `test_table_a5_5000_hz_reads_the_missing_digit_as_one` in
  [`tests/emission/test_sound_power_in_duct.py`](https://github.com/jmrplens/phonometry/blob/main/tests/emission/test_sound_power_in_duct.py)
  say that it is a reading and not the print; a copy of the standard in
  which the digit survived would settle it.
- **Status:** unreported.

## ISO 5136:2003, Annex D, Annex H and Annex I ($C_{3,4}$ "according to Equation (3)")

- **Location:** the first sentence of Annex D, and the sentence of Annex H
  and of Annex I that introduces their coefficient tables.
- **The print:** "For $d$ = 0,5 m, the values of the coefficients $a_i$ for
  the calculation of $C_{3,4}$ according to Equation (3) are given in Table
  A.4" (Annex D); "Values for the coefficients $a_i$ necessary to compute the
  mean flow velocity-modal corrections $C_{3,4}$ according to Equation (3)
  are given in Tables H.1 to H.3" (Annex H) and "... in Tables I.1 to I.3"
  (Annex I).
- **The problem:** Equation (3) is the cut-on frequency of the first cross
  mode, $f_{1,0} = 0.586\,(c/D)\sqrt{1 - (U/c)^2}$, in the definition of
  3.10. The polynomial in $U$ whose coefficients the tables hold is Equation
  (7) of clause 5.3.3.4. The same wrong number is printed three times.
- **Evidence:** PDF pages 45, 64 and 68 (printed pp. 35, 54 and 58) of
  ISO 5136:2003, against Equation (3) on PDF page 16 (printed p. 6) and
  Equation (7) on PDF page 28 (printed p. 18).
- **Library behaviour:** evaluates Equation (7);
  [`flow_modal_correction`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/emission/_sound_power_in_duct.py)
  cites it. No number changes.
- **Status:** unreported (cross-reference defect, no numerical consequence).

## ISO 5136:2003, Annex B, B.2 step 4 ($\Delta L_{\max}$ "given in Table C.1")

- **Location:** Annex B, clause B.2, "Comparative procedure using a
  microphone fitted with a nose cone and a microphone fitted with a sampling
  tube", Step 4.
- **The print:** "Check whether the difference between the circumferentially
  averaged sound pressure levels obtained with the nose cone and the sampling
  tube ($\overline{L_{p\mathrm{NC}}} - \overline{L_{p\mathrm{ST}}}$) is
  smaller than or equal to the maximum allowable difference $\Delta L_{\max}$
  given in Table C.1."
- **The problem:** Table C.1 is the A-weighting $C_j$ of Annex C and holds no
  $\Delta L_{\max}$. The table of the maximum allowable difference against the
  turbulence noise suppression $\Delta L_\mathrm{t}$ of the sampling tube is
  Table B.1, on the page after the step, and the paragraph two above the
  steps already sends the reader to it ("see Table B.1").
- **Evidence:** PDF page 41 (printed p. 31) of ISO 5136:2003, with Table B.1
  on PDF page 42 (printed p. 32) and Table C.1 on PDF page 44 (printed
  p. 34).
- **Library behaviour:** the signal-to-noise procedure of Annex B is a
  qualification of the measurement, not a term of $L_W$, and is not
  implemented. No change was needed.
- **Status:** unreported (cross-reference defect, no numerical consequence).

## ISO 5136:2003, Annex B, B.1 ("the determination of the combined mean flow velocity")

- **Location:** Annex B, clause B.1, "General", the first sentence.
- **The print:** "Two procedures for the determination of the combined mean
  flow velocity are given in B.2 and B.3."
- **The problem:** the annex is titled "Determination of the signal-to-noise
  ratio of sound vs. turbulent pressure fluctuation in the test duct", and
  B.2 and B.3 determine that ratio; nothing in the annex determines a "combined
  mean flow velocity", a phrase that is a fragment of the "combined mean flow
  velocity and modal correction" of clause 5.3.3.4. The sentence also counts
  two procedures where the annex, by the coherence method it closes with,
  gives three.
- **Evidence:** PDF page 41 (printed p. 31) of ISO 5136:2003, the annex
  title and the sentence on the same page, and the coherence procedure on
  PDF page 43 (printed p. 33).
- **Library behaviour:** Annex B is not implemented; nothing to change.
- **Status:** unreported (wording defect).

## ISO 5136:2003, clause 7.4 NOTE (the "hydraulic diameter" $D_\mathrm{h} = \sqrt{S_{\mathrm{f}2}/\pi}$)

- **Location:** clause 7.4, the NOTE that follows the outlet-duct rule for
  large fans in installation category D.
- **The print:** "The hydraulic diameter of the fan outlet area, $S_{\mathrm{f}2}$,
  is given by $D_\mathrm{h} = \sqrt{S_{\mathrm{f}2}/\pi}$".
- **The problem:** $\sqrt{S/\pi}$ is the radius of the circle of area $S$;
  its diameter is $\sqrt{4S/\pi} = 2\sqrt{S/\pi}$. Followed as printed, the
  "2 $D_\mathrm{h}$" the clause asks the outlet duct to be is one equivalent
  diameter long, not two, and whether the rule intended is two diameters or
  two radii cannot be settled from the document.
- **Evidence:** PDF page 33 (printed p. 23) of ISO 5136:2003.
- **Library behaviour:** the duct lengths of clauses 5.2 and 7.4 are facility
  geometry and are not computed; nothing to change.
- **Status:** unreported.

## ISO 5136:2003, Table A.2, coefficient header (the $a_9$ column heads $a9_0$)

- **Location:** Annex A, Table A.2, "Values of coefficients $a_i$ for the
  determination of the combined mean flow velocity and modal correction
  $C_{3,4}$ of the sampling tube for duct diameters 0,2 m $\le d <$ 0,3 m",
  the header row of the coefficient columns, tenth column.
- **The print:** an italic $a$, an italic 9 on the baseline and a subscript
  0, between an $a_8$ and an $a_{10}$ of the same row that both carry their
  index as a subscript.
- **The problem:** a stray subscript zero on a column that is $a_9$. The same
  column is headed $a_9$ in Tables A.1 and A.3 to A.6, the NOTE under every
  one of them sums $a_i U^i$ from $i = 0$ to $i = 10$ over the eleven columns
  the row has, and the single cell this one holds, the
  $4.09 \times 10^{-14}$ of the 20 000 Hz row, is the coefficient of
  $U^9$: an $a_{90}$ would have no place in that sum at all.
- **Evidence:** PDF page 36 (printed p. 26) of ISO 5136:2003, against the
  header row of Table A.1 on PDF page 35 (printed p. 25).
- **Library behaviour:** the column is read as $a_9$. `_TABLE_A2` in
  [`_sound_power_in_duct.py`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/emission/_sound_power_in_duct.py)
  carries the 20 000 Hz row as the ten coefficients $a_0$ to $a_9$, and
  `test_table_a2_20_khz_row_reads_the_last_column_as_a9` in
  [`tests/emission/test_sound_power_in_duct.py`](https://github.com/jmrplens/phonometry/blob/main/tests/emission/test_sound_power_in_duct.py)
  multiplies the row out. No coefficient value changes.
- **Status:** unreported (typographic, no numerical consequence).

## ISO 5136:2003, Table A.6, 16 000 Hz row ($a_1$ printed with a doubled multiplication sign)

- **Location:** Annex A, Table A.6, "... for duct diameters 1,25 m $\le d \le$
  2 m", row 16 000 Hz, column $a_1$.
- **The print:** $4{,}52 \times\!\times 10^{-01}$, two multiplication signs
  where every other cell prints one.
- **The problem:** typographic only; the mantissa and the exponent are
  legible and the value is $4.52 \times 10^{-1}$, in line with the
  $4.51 \times 10^{-1}$ of Table A.5 and the $4.52 \times 10^{-1}$ of
  Table I.1 at the same band. The row is in the informative range above
  10 kHz.
- **Evidence:** PDF page 40 (printed p. 30) of ISO 5136:2003.
- **Library behaviour:** $4.52 \times 10^{-1}$ in `_TABLE_A6` of
  [`_sound_power_in_duct.py`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/emission/_sound_power_in_duct.py).
- **Status:** unreported (typographic, no numerical consequence).

## ISO 5136:2003, Table I.2 (continued), 20 000 Hz row (the exponents of $a_8$ and $a_9$)

- **Location:** Annex I, Table I.2, "... for duct diameters 3,55 m $\le d
  \le$ 5 m", the continuation page, row 20 000 Hz, columns $a_8$ and $a_9$.
- **The print:** $a_8 = -5{,}88 \times 10^{-10}$ and
  $a_9 = 2{,}25 \times 10^{-10}$.
- **The problem:** at $U$ = 40 m/s the printed $a_9$ alone contributes
  $2.25 \times 10^{-10} \times 40^9 \approx 5.9 \times 10^4$ dB to
  $C_{3,4}$, which no correction can be. The same row of the neighbouring
  tables prints $a_8 = -5.90 \times 10^{-12}$ and
  $a_9 = 2.25 \times 10^{-13}$ (Table I.1) and
  $a_9 = 2.25 \times 10^{-13}$ (Table I.3), so the exponents are
  $-12$ and $-13$ and the print is short by two and three decades. Annex I
  is informative and the row is in the informative range above 10 kHz.
- **Evidence:** PDF page 72 (printed p. 62) of ISO 5136:2003, against the
  same row of Table I.1 on PDF page 70 (printed p. 60) and of Table I.3 on
  PDF page 74 (printed p. 64).
- **Library behaviour:** the informative Annexes H and I are outside the
  scope the standard states for itself (0.15 m to 2 m) and are not
  implemented; a duct above 2 m is refused. Recorded so that an
  implementation of Annex I does not carry the exponents as printed.
- **Status:** unreported.
