# Errata in VDI 2081

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

Canonical: https://docs.jmrp.io/phonometry/reference/errata/vdi-2081/
Language: en
Alternate: https://docs.jmrp.io/phonometry/es/reference/errata/vdi-2081/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 VDI 2081: 6 defects in the printed text, with the evidence and what phonometry does about each.

---

These are the 6 entries of the [errata registry](https://docs.jmrp.io/phonometry/reference/errata/) on VDI 2081 Blatt 1:2001, VDI 2081 Blatt 1:2001-07 and VDI 2081 Blatt 2:2005-05, 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.

## VDI 2081 Blatt 1:2001, Section 6.7.3 (the symbol list of Equation (36) sends A back to Equation (36))

- **Location:** Section 6.7.3, the symbol list under Equation (36), the entry
  for the equivalent absorption area ``A``.
- **The print:** "A  äquivalente Absorptionsfläche; in $\text{m}^2$, Gleichung (36)" /
  "A  is the equivalent absorption area; in $\text{m}^2$, Equation (36)".
- **The problem:** Equation (36) is the level equation the list belongs to,
  $L_P = L_W + 10\lg[Q/(4\pi r^2) + 4/A]$, in which ``A`` is an input. It
  does not define ``A``. The guideline defines it twice further down the same
  section: Equation (37), $A = 0.163\,V/T$, and Equation (39),
  $A = \sum \alpha_i S_n + \sum A_n$. The reference is a self-reference, and
  it stands in both language columns, so it is a typesetting slip in the
  original rather than a translation one.
- **Evidence:** verified on PDF page 43 (printed p. 43) of VDI 2081
  Blatt 1:2001-07, with Equations (37) and (39) on PDF pages 44 and 44
  (printed pp. 44 and 44) of the same print.
- **Library behaviour:** unaffected. The slip is in a cross-reference, not in a
  number;
  [`room_effect`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/noise_control/hvac.py) takes ``A`` as an
  argument and
  [`sabine_absorption_area`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/room/steady_field.py)
  implements Equation (37).
- **Status:** unreported.

## VDI 2081 Blatt 1:2001, Section 6.7.3 (the English column calls a hemispherical propagation spherical)

- **Location:** Section 6.7.3, the sentence stating where the reverberation
  field begins, immediately after Equation (36b).
- **The print:** German, "Der Nachhallbereich beginnt bei **halbkugelförmiger**
  Schallausbreitung in einer Entfernung, die größer ist als
  $r_H = 0{,}2\sqrt{A}$"; English, "The reverberation area begins as a
  **spherical** sound propagation at a distance which is greater than
  $r_H = 0.2\sqrt{A}$".
- **The problem:** *halbkugelförmig* is hemispherical, not spherical, and the
  printed constant sides with the German. The reverberation radius is
  $r_H = \sqrt{Q A / 16\pi}$, which is $0.199\sqrt{A}$ at the $Q = 2$ of a
  half space and $0.141\sqrt{A}$ at the $Q = 1$ of a full one. Only the
  first rounds to the printed $0.2$. A reader following the English column
  would take $0.2\sqrt{A}$ for the spherical radius and place the
  reverberation field 41 % too far out.
- **Evidence:** verified on PDF page 44 (printed p. 44) of VDI 2081
  Blatt 1:2001-07, both columns of the same sentence read side by side.
- **Library behaviour:** unaffected.
  [`critical_distance`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/room/steady_field.py) takes ``Q`` as
  an argument and states the hemispherical reading in its own text.
- **Status:** unreported.

## VDI 2081 Blatt 1:2001-07, Section 6.4 (the English column says the opposite of the German)

- **Location:** printed folio 40 (PDF page 40), Section 6.4 "Verzweigungen" /
  "Junctions", the sentence directly under Equation (35).
- **The print:** the German column reads "Diese in Bild 27 dargestellte Senkung
  des Schallleistungspegels ist **frequenzunabhängig**." The English column of
  the same page, translating the same sentence, reads "This sound power level
  reduction shown in Figure 27 **depends on the frequency**."
- **The problem:** the two say opposite things, and the German is the
  authoritative one: the cover of every VDI guideline states that the German
  version shall be taken as authoritative and that no guarantee is given for
  the English translation. The German is also the one the rest of the document
  agrees with. Figure 27 on the same page plots $\Delta L_W$ against the
  cross-section ratio $S_1 / \sum S_{1,2,3}$ alone and carries no frequency
  axis; Equation (35) itself, $\Delta L_W = |10 \lg (S_1 / \sum_i S_i)|$,
  contains no frequency; and the worked example of VDI 2081 Blatt 2:2005-05
  prints a junction's level reduction as a single number rather than as an
  octave spectrum, in each of its three junctions (Table 1, elements 3, 7 and
  16, printed folios 13 and 15: $5.6$, $4.8$ and $3.0$ dB).
- **The likely mechanism:** the negating prefix of "frequenzunabhängig" is
  absent from the translation, which turns "independent of the frequency" into
  its opposite. Nothing else in the sentence differs.
- **Consequence:** a reader working from the English column alone would look
  for a frequency dependence that neither the equation nor the figure has, and
  might conclude that the guideline is incomplete rather than that the sentence
  is mistranslated.
- **Evidence:** the two columns of the same printed page read against each
  other; Figure 27 on that page; Equation (35) above it; and the three junction
  rows of the worked example in Blatt 2. Verified on PDF page 40 (printed p. 40)
  of VDI 2081 Blatt 1:2001-07 and PDF pages 13 and 15 (printed pp. 13 and 15) of
  VDI 2081 Blatt 2:2005-05.
- **Library behaviour:**
  [`split_loss`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/noise_control/hvac.py) with
  `model="vdi2081"` returns one value for the junction, the German reading, and
  reproduces all three printed junctions of the worked example.
- **Status:** unreported. Both prints are superseded (Blatt 1:2022-04 and
  Blatt 2:2022-10) and neither successor is held, so whether the translation was
  corrected is not known here.

## VDI 2081 Blatt 2:2005-05, Table 1, element 2 (the hydraulic diameter it prints is not the one it computes with)

- **Location:** Table 1, printed folio 12 (PDF page 12), element 2, the splitter
  silencer: the rows "Hydr. Durchmesser $d_\mathrm{h}$ (m)" and "Strouhalzahl
  $St$".
- **The print:** $d_\mathrm{h} = 0{,}171$ m, and the eight Strouhal numbers
  $0{,}9$, $1{,}7$, $3{,}4$, $6{,}8$, $13{,}5$, $27{,}0$, $54{,}0$ and
  $108{,}0$ over the octaves 63 Hz to 8 kHz, for a clear gap $s = 0{,}100$ m,
  a splitter height $H = 0{,}600$ m and a gap speed $v = 14{,}81$ m/s.
- **The problem:** the two rows disagree. VDI 2081 Blatt 1 Section 7.2.4.2
  defines $St = f_\mathrm{m} d_\mathrm{h} / v_\mathrm{i}$, so the printed
  $d_\mathrm{h}$ and the printed $St$ determine each other. With the printed
  $0.171$ m the eight numbers would be $0.73$, $1.45$, $2.89$,
  $5.79$, $11.57$, $23.15$, $46.29$ and $92.59$: not one of them
  rounds onto the printed row. With $d_\mathrm{h} = 2s = 0.200$ m they come
  out as $0.851$, $1.688$, $3.376$, $6.752$, $13.504$, $27.009$,
  $54.018$ and $108.035$, which round onto all eight.

  Both values are defensible as a hydraulic diameter, which is why this is an
  internal inconsistency rather than a wrong number: $4A/P$ for a $0.100$ m
  by $0.600$ m gap is $0.171$ m, while the parallel-plate limit that a long
  narrow gap tends to is $2s = 0.200$ m. The table prints the first and
  computes with the second.
- **Consequence:** following the printed $d_\mathrm{h}$ reproduces neither the
  Strouhal row nor the flow-noise spectrum beneath it. With $2s$ the whole
  element falls out to the last printed decimal: $L_\mathrm{WA} = 52$ dB from
  Equation (49) and the eight octave levels $62.7$ down to $35.6$ dB from
  Equations (46), (50) and (51), the worst of them 0.046 dB from its printed
  cell.
- **Evidence:** the two rows of the same printed element read against Section
  7.2.4.2 of Blatt 1 (printed folio 53); both candidate diameters evaluated
  over the eight octaves; and the flow-noise spectrum recomputed from each.
  Verified on PDF page 12 (printed p. 12) of VDI 2081 Blatt 2:2005-05 and PDF
  page 53 (printed p. 53) of VDI 2081 Blatt 1:2001-07.
- **Library behaviour:**
  [`silencer_self_noise`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/noise_control/hvac.py) with
  `model="vdi2081"` takes the clear gap and uses $2s$, so it reproduces the
  worked example. The docstring says which of the two it takes.
- **Status:** unreported. Both prints are superseded and neither successor is
  held.

## VDI 2081 Blatt 2:2005-05, Table 1, element 14 (the element's own row contradicts the one above it)

- **Location:** Table 1, printed folio 15 (PDF page 15), element 14, the round
  bend: the rows "$\Sigma L_\mathrm{W}$" and "$\Sigma L_\mathrm{W}$ (log)",
  8 kHz cell.
- **The print:** the two rows read, over the octaves 63 Hz to 8 kHz, $68{,}5$
  $61{,}7$ $55{,}7$ $43{,}4$ $30{,}1$ $28{,}5$ $34{,}0$ $34$ and $68{,}5$
  $61{,}7$ $55{,}7$ $43{,}4$ $30{,}1$ $28{,}5$ $34{,}0$ $33{,}6$. Seven cells
  agree and the eighth does not.
- **The problem:** the second row is the first with the element's own flow noise
  added, so it can never be lower. Element 13 hands 8 kHz over at $37.0$ dB,
  the bend attenuates $3$ dB, and the first row prints the $34.0$ dB that
  leaves. The bend's own noise in that band is $-14.4$ dB, which the same
  element prints two rows higher, and adding it moves the level by less than
  $0.0001$ dB. The second row should therefore print $34.0$ dB and prints
  $33.6$.

  Read the other way the cell is equally unreachable: $33.6$ dB would need
  $36.6$ dB to arrive from element 13, and element 13 prints $37.0$ dB in
  its own $\Sigma L_\mathrm{W}$ row on the same page.
- **Consequence:** $0.4$ dB at 8 kHz, carried into element 15 and everything
  after it. The A-weighted total of the element is printed as $50.9$ dB in
  both rows, which is what hides it: at 8 kHz the A-weighting is $-1.1$ dB and
  the band is $16$ dB under the 4 kHz one, so $0.4$ dB there does not reach
  the first decimal of the total.
- **Evidence:** the two rows of element 14 and the $\Sigma L_\mathrm{W}$ row of
  element 13, with the sum recomputed at full precision from the printed
  hand-over, attenuation and flow noise. Verified on PDF page 15 (printed
  p. 15) of VDI 2081 Blatt 2:2005-05.
- **Library behaviour:** the conformance rows for element 14 compare the flow
  noise the bend makes, $26.9$ down to $-14.4$ dB, against the row that
  prints it, and the chain row carries the $34.0$ dB the arithmetic gives
  rather than the printed $33.6$.
- **Status:** unreported. The print is superseded and the successor is not held.

## VDI 2081 Blatt 2:2005-05, Table 1, element 2 (a cross-reference to the wrong clause)

- **Location:** Table 1, printed folio 12 (PDF page 12), element 2, the box
  reading "Tabelle aus VDI 2081 Blatt 1/7.3.2" beside the coefficients
  $a_1$, $a_2$, $b_1$ and $b_2$.
- **The print:** the coefficients $0{,}255$, $0{,}015$, $-2{,}82$ and $-2{,}91$
  are credited to Section 7.3.2 of Blatt 1.
- **The problem:** Section 7.3 of VDI 2081 Blatt 1:2001-07 is "Luftschalldämmung
  eines Bauteils", the airborne sound insulation of a building component, and
  has no such table. The coefficients are printed in Section **7.2.3.2**,
  "Kulissenschalldämpfer", on printed folio 52, whose table gives exactly those
  four values in its 200 mm row, which is the splitter thickness the element
  uses.
- **Consequence:** a reader following the reference lands in the wrong chapter.
  The values themselves are right.
- **Evidence:** the cited clause and the actual one, both read from the printed
  pages. Verified on PDF page 12 (printed p. 12) of VDI 2081 Blatt 2:2005-05
  and PDF page 52 (printed p. 52) of VDI 2081 Blatt 1:2001-07.
- **Library behaviour:** none; the library cites Section 7.2.3.2.
- **Status:** unreported.
