# Errata in ISO 3744

> 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-3744/
Language: en
Alternate: https://docs.jmrp.io/phonometry/es/reference/errata/iso-3744/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 3744: 3 defects in the printed text, with the evidence and what phonometry does about each.

---

These are the 3 entries of the [errata registry](https://docs.jmrp.io/phonometry/reference/errata/) on ISO 3744:2010, 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 3744:2010, 8.3.4, Equation (21) (a time-integrated level compared with a time-averaged one)

- **Location:** clause 8.3.4, Equation (21) and the symbol list beneath it
  (PDF page 31, printed p. 25) of ISO 3744:2010, read against the definitions
  of clauses 3.3 and 3.4 (PDF page 9, printed p. 3). The same construction is
  printed as ISO 3741:2010 Equation (25) (PDF page 33, printed p. 24, with its
  symbol list on PDF page 34, printed p. 25) as ISO 3747:2010 Equation
  (14) (PDF pages 22 and 23, printed pp. 13 and 14), as ISO 3743-1:2010
  Equation (19) in clause 8.2.3 (PDF page 25, printed p. 16), and as ISO
  3746:2010 Equation (15) in clause 8.4.2 (PDF page 25, printed p. 16), which
  is the survey-grade route the library takes for `grade='survey'`.
- **The print:** $K_1 = -10 \lg\left(1 - 10^{-0{,}1\,\Delta L_E}\right)$ dB
  with $\Delta L_E = \overline{L'_{E(\mathrm{ST})}} - \overline{L_{p(\mathrm{B})}}$,
  where $\overline{L'_{E(\mathrm{ST})}}$ "is the mean frequency-band or
  A-weighted single event time-integrated sound pressure level" and
  $\overline{L_{p(\mathrm{B})}}$ "is the mean frequency-band or A-weighted
  time-averaged sound pressure level of the background noise", followed by
  "The integration time $T = t_2 - t_1$ and other measurement parameters shall
  be the same for the measurement of the single event time-integrated sound
  pressure level $L'_{Ei(\mathrm{ST})}$ and of the background noise level
  $L_{pi(\mathrm{B})}$."
- **The problem:** the two levels do not share a reference quantity, and the
  correction subtracts one energy from another. By clause 3.4, $L_E$ is
  $\int_{t_1}^{t_2} p^2\,\mathrm{d}t$ re $E_0 = (20\ \mu\text{Pa})^2\,\text{s}$;
  by clause 3.3, $L_{p,T}$ is $\frac{1}{T}\int_{t_1}^{t_2} p^2\,\mathrm{d}t$
  re $p_0^2$. Their difference is a ratio of energies only when $T = 1$ s. Over
  the common interval $T$ the background contributes the energy
  $L_{p(\mathrm{B})} + 10 \lg(T/T_0)$ (the identity of clause 3.4 NOTE 1), so
  the printed $\Delta L_E$ exceeds the signal-to-background energy ratio by
  $10 \lg(T/T_0)$ and $K_1$ is under-estimated for every $T > 1$ s: a burst
  whose energy is 6 dB above the background's in a 10 s interval reads as
  16 dB above it and earns no correction, where the criterion of 8.2.3 puts
  $K_1$ at its largest admissible value, 1.3 dB. The twin chain of 8.2, where
  both levels are time-averaged, has no such term, and 8.3.3 requires the
  single event levels to be averaged "in the same way as for the
  time-averaged sound pressure levels described in 8.2.2", so the intended
  reading is the one under which the two chains coincide for a source that is
  steady over $T$, $L_J = L_W + 10 \lg(T/T_0)$, and that is the reading under
  which the insistence on one integration time for both measurements does any
  work.
- **Evidence:** Verified on PDF page 31 (printed p. 25) of ISO 3744:2010 for
  the equation and its symbol list, and on PDF page 9 (printed p. 3) for the
  definitions of clauses 3.3 and 3.4 with NOTE 1; the same construction read
  on PDF pages 33 and 34 (printed pp. 24 and 25) of BS EN ISO 3741:2010, on
  PDF pages 22 and 23 (printed pp. 13 and 14) of BS EN ISO 3747:2010 and on
  PDF page 25 (printed p. 16) of BS EN ISO 3743-1:2010.
- **Library behaviour:** `sound_energy_pressure`, `sound_energy_reverberation`
  and `sound_energy_comparison` compare the background as its exposure over
  the same interval, $L_{p(\mathrm{B})} + 10 \lg(T/T_0)$, and require
  `integration_time` with the background of the source under test, which is
  the one compared against an event level. The reference source of
  `sound_energy_comparison` is steady, so `reference_background_levels` is corrected
  by the time-averaged rule of 9.1.2 instead and takes no window; the criteria and the
  clamp of 8.2.3 (and of 9.1.2 in ISO 3741) are then applied to that margin.
  `tests/emission/test_sound_energy.py` pins $K_1 = 1.2563$ dB for a 78 dB
  burst over a 62 dB background in a 10 s window, and
  $L_J = L_W + 10 \lg(T/T_0)$ field by field on both families; the
  conformance report carries the identity as "ISO 3744:2010 Eq. 23 / clause
  3.4 NOTE 1". `sound_energy_hard_walled` reads ISO 3743-1:2010 Equation (19)
  the same way and requires `integration_time` with the background;
  `test_steady_source_energy_is_power_plus_10_lg_t` in
  `tests/emission/test_sound_power_hard_walled.py` and the conformance check
  "ISO 3743-1:2010 Eq. 20 / clause 3.4 NOTE 1" pin the identity there.
- **Status:** unreported.

## ISO 3744:2010, 8.3.4 (the correction named K_1i in the text and K_1 in Equation (21))

- **Location:** clause 8.3.4, first sentence and Equation (21) (PDF page 31,
  printed p. 25).
- **The print:** "The background noise correction, $K_{1i}$, shall be
  calculated using Equation (21):" followed by
  $K_1 = -10 \lg\left(1 - 10^{-0{,}1\,\Delta L_E}\right)$ dB with $\Delta L_E$
  formed from the two means over the measurement surface,
  $\overline{L'_{E(\mathrm{ST})}}$ and $\overline{L_{p(\mathrm{B})}}$.
- **The problem:** the sentence names a per-position correction and the
  equation defines a single one from surface means. The twin clause 8.2.3
  names $K_1$ in both places and forms it from the same surface means
  (Equation (16)), and 8.3.5 subtracts the unsubscripted $K_1$ in Equation
  (22). The subscript is the per-microphone convention of ISO 3741:2010
  clauses 9.1.2 and 9.2.2 ($K_{1i}$, Equations (14) and (25)), where each
  position is corrected before the average, and does not belong to this
  clause.
- **Evidence:** Verified on PDF page 31 (printed p. 25) of ISO 3744:2010,
  against clause 8.2.3 on PDF page 29 (printed p. 23).
- **Library behaviour:** `sound_energy_pressure` forms one $K_1$ per band from
  the surface means, as Equation (21) prints it and as `sound_power_pressure`
  does for Equation (16); no per-position correction is applied in the
  ISO 3744 chain. No change was required.
- **Status:** unreported.

## ISO 3744:2010, H.4.2.7 (the altitude correction and the divisor under it)

- **Location:** Annex H (informative), H.4.2.7 "Meteorological and radiation
  impedance corrections", the paragraph that sizes $u_{C_1+C_2}$ from the
  Annex G correction.
- **The print:** "At 120 m altitude and 23 °C the correction is zero and at
  500 m altitude the correction is 0,6 dB. Assuming a triangular distribution
  for this uncertainty, the standard deviation is
  $s_\mathrm{met} = 0{,}6/\sqrt{6} = 0{,}3\ \mathrm{dB}$."
- **The problem:** two independent defects in one sentence pair.
  (a) Annex G, which is normative and which this paragraph points at, gives
  $C_1 + C_2 = 0.394$ dB at 500 m and 23.0 °C, not 0.6 dB. The reading is
  self-validating: the same two equations give $-4.6 \times 10^{-5}$ dB at
  120 m and 23.0 °C, which is the "zero" the same sentence prints, so the
  constants and the temperature terms are being read as the standard intends.
  0.6 dB is reached at about 697 m at 23.0 °C, or at 500 m only if the air is
  at 30.1 °C.
  (b) $0.6/\sqrt{6} = 0.245$, not 0.3. The quotient does not give the
  result printed beside it: 0.3 dB is exactly $0.6/2$, so either the divisor
  or the result is wrong. For a triangular distribution of half-width $a$ the
  standard deviation is $a/\sqrt{6}$, which is the divisor the sentence names.
- **Evidence:** H.4.2.7 read on PDF page 82 (printed p. 73), against Annex G
  Equations (G.1) and (G.2) with $a = 2.2560 \times 10^{-5}$ m$^{-1}$,
  $b = 5.2553$, $\theta_0 = 314$ K and $\theta_1 = 296$ K on PDF pages 73 and
  74 (printed pp. 64 and 65), all of BS EN ISO 3744:2010. Both values were
  recomputed from the printed equations alone.
- **Library behaviour:** the Annex H uncertainty budget is not modelled, so no
  published number depends on either figure. The Annex G correction itself is
  evaluated from Equations (G.1) and (G.2) by
  [`reference_atmosphere_correction`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/emission/sound_power.py),
  and the conformance check "ISO 3744:2010 Annex G / H.4.2.7" pins the half of
  the paragraph that is right: the correction vanishes at 120 m and 23 °C.
- **Status:** unreported.
