# Errata in Harris (1995)

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

Canonical: https://docs.jmrp.io/phonometry/reference/errata/harris/
Language: en
Alternate: https://docs.jmrp.io/phonometry/es/reference/errata/harris/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 Harris (1995): 2 defects in the printed text, with the evidence and what phonometry does about each.

---

These are the 2 entries of the [errata registry](https://docs.jmrp.io/phonometry/reference/errata/) on Harris (1995), 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.

## Harris 3e (1995), Tables 32.1 to 32.8 (nineteen double-unit pairs whose two halves are not each other)

- **Location:** Tables 32.1 to 32.8, the impact insulation of floor-ceiling
  constructions, on printed folios 32.8 to 32.15 (PDF pages 750 to 757), in
  chapter 32, "Aislamiento del sonido transmitido por estructuras", of the
  Spanish edition. Non-normative source: a handbook.
- **The print:** every dimension and every mass of these eight tables is
  printed twice, SI first and the US customary value in parentheses after it,
  inside the running description of each construction: "Losa de 10 cm (4 in)",
  "cada 40,6 cm (16 in)", "alfombra de 1,5 kg/m2 (44 oz/yd2)". Three hundred
  and twenty-one such pairs are printed over the eight tables.
- **The problem:** nineteen of the three hundred and twenty-one pairs are not each
  other. The inch is 2.54 cm exactly, the pound is 0.45359237 kg exactly and
  the yard 0.9144 m exactly, so one pound per cubic foot is 16.0185 kg/m3, one
  pound per square yard 0.54249 kg/m2 and one ounce per square yard 33.906
  g/m2, and each pair is decided by arithmetic alone. On these pages the US
  customary half is the measurement and the SI half its translation, so a pair
  whose imperial half is a whole number or a fraction is read in that
  direction only, and one whose imperial half is itself printed as a rounded
  decimal is allowed either. A pair counts as sound when the SI half is the
  conversion rounded **or truncated** to the precision it is printed to, which
  forgives the whole of this chapter's loose rounding: "60 cm (24 in)" and
  "2,5 cm (1 in)" are truncations of 60.96 and 2.54 and no more than that, and
  row 17 prints "36,9 cm (14,5 in)", where 36.9 cm is 14.53 in and the page
  would print that as the 14.5 in beside it. The nineteen below survive that
  test. Each one names the row the page numbers, the pair as it is set, and
  the conversion that fails:
  - **Row 9** (PDF page 750, printed folio 32.8): "30,8 cm (16 in)" for the
    batten spacing. $16 \times 2.54 = 40.64$ cm, which these eight tables
    print as 40.6 cm eighteen times over.
  - **Row 11** (PDF page 751, printed folio 32.9): "53,2 cm (21 in)" for the
    rib spacing. $21 \times 2.54 = 53.34$ cm, which row 38 prints as
    53.3 cm.
  - **Row 14** (PDF page 752, printed folio 32.10): "15,6 cm (6 in)" for the
    slab. $6 \times 2.54 = 15.24$ cm, which these eight tables print as
    15.2 cm eleven times over.
  - **Row 18** (PDF page 752, printed folio 32.10): "36,7 cm (14,5 in)" for
    the beam spacing. $14.5 \times 2.54 = 36.83$ cm, which rows 13 and
    15 print as 36.8 cm.
  - **Row 22** (PDF page 753, printed folio 32.11): "60,1 cm (24 in)" for the
    joist spacing. $24 \times 2.54 = 60.96$ cm, which rows 26, 27, 33 and
    38 print as 61 cm.
  - **Row 23** (PDF page 753, printed folio 32.11): "60,1 cm (24 in)" for the
    joist spacing, the same pair again.
  - **Row 26** (PDF page 754, printed folio 32.12): "32,3 cm (11,75 in)" for
    the total thickness. $11.75 \times 2.54 = 29.85$ cm, and 32.3 cm is
    12.72 in, so neither half is the other.
  - **Row 27** (PDF page 754, printed folio 32.12): "1,89 cm (0,78 in)" for
    the oak strip floor. $0.78 \times 2.54 = 1.98$ cm, which row 26
    prints as 1.98 cm for the same floor.
  - **Row 28** (PDF page 754, printed folio 32.12): "1,89 cm (0,78 in)" for
    the same oak strip floor.
  - **Row 28** (PDF page 754, printed folio 32.12): "31,6 cm (12,5 in)" for
    the total thickness. $12.5 \times 2.54 = 31.75$ cm, which rows 16
    and 18 print as 31.8 cm and row 25 as 31.7 cm.
  - **Row 29** (PDF page 754, printed folio 32.12): "60,8 cm (24 in)" for the
    resilient channels. $24 \times 2.54 = 60.96$ cm.
  - **Row 31** (PDF page 755, printed folio 32.13): "10,1 cm (2 in)" for the
    batten section. $2 \times 2.54 = 5.08$ cm, which these eight tables
    print as 5.1 cm twenty-two times over, the same row included.
  - **Row 34A** (PDF page 755, printed folio 32.13): "7,5 cm (3 in)" for the
    furring strips. $3 \times 2.54 = 7.62$ cm, which these eight tables
    print as 7.6 cm ten times over.
  - **Row 35A** (PDF page 756, printed folio 32.14): "410 kg/m3 (26,1 lb/ft3)"
    for the compressed paper-pulp floor board. $26.1 \times 16.0185 =
    418.1$ kg/m3, and 410 kg/m3 is 25.6 lb/ft3, so neither half is the other.
    The page does not say which of the two carries the defect, and no other
    cell of the eight tables settles it: row 5 prints 35.2 kg/m3 (2.2 lb/ft3)
    and row 37A 2370 kg/m3 (148 lb/ft3), and both of those convert correctly,
    which makes this a property of this cell and not of the chapter.
  - **Row 35B** (PDF page 756, printed folio 32.14): "60,1 cm (24 in)" for the
    steel joist spacing.
  - **Row 36B** (PDF page 756, printed folio 32.14): "60,1 cm (24 in)" for the
    steel joist spacing.
  - **Row 37A** (PDF page 756, printed folio 32.14): "1,5 kg/m2 (3,4 lb/yd2)"
    for the diamond mesh and metal lath. $3.4 \times 0.54249 = 1.84$
    kg/m2, and row 38 converts 4.14 lb/yd2 to 2.25 kg/m2 with that same
    factor.
  - **Row 38** (PDF page 756, printed folio 32.14): "1,81 kg/m2 (40 oz/yd2)"
    for the hair underlay. $40 \times 33.906 = 1356$ g/m2, which row 28
    prints as 1.4 kg/m2 for the same cloth.
  - **Row 38** (PDF page 756, printed folio 32.14): "1,99 kg/m2 (44 oz/yd2)"
    for the wool pile carpet. $44 \times 33.906 = 1492$ g/m2, which rows 28
    and 30 print as 1.5 kg/m2 for the same cloth. The two carpet cells of this
    row are both 1.334 of their own conversion, so one wrong factor was
    applied to the pair rather than two digits slipping.
- **Evidence:** Verified on PDF pages 750 to 757 (printed pp. 32.8-32.15) of
  Harris (ed.), *Manual de medidas acústicas y control del ruido* 3e (1995),
  the Spanish edition of *Handbook of Acoustical Measurements and Noise
  Control*. Two readers transcribed the eight tables independently and agreed
  on every one of these nineteen pairs; each was then read again on its own
  page, enlarged, before being listed here. The three hundred and twenty-one pairs
  were converted and compared one by one rather than by sampling, which is
  what makes the list closed rather than a sample of what a reader happened to
  notice. The near misses it leaves off are the ones the rule above forgives,
  and three of them are named there.
- **What the library does:** publishes every one of these descriptions exactly
  as the page sets them. The seventeen rows that carry one of the nineteen
  pairs are marked in
  [`PUBLISHED_IMPACT_INSULATION`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/building/impact_catalogue.py):
  the cell is recorded in `misprinted`, quoting the printed pair and the
  conversion that fails, so a reader reproducing the book sees what the book
  says and a reader using the catalogue is told not to. Eighteen of the
  nineteen sit inside the running description, where this catalogue serves no
  quantity from them and nothing else follows. The nineteenth is row 35A's
  density, which is the one cell of the nineteen this catalogue would lift
  into a field of its own, so `layer_density_kg_m3` is left empty there and
  `why_missing` hands back both printed halves rather than choosing one. Rows
  5 and 37A, whose densities convert, are served normally.
- **Status:** unreported.

## Harris 3e (1995), Table 30.2 (five pile weights whose two halves are not each other)

- **Location:** Table 30.2, "Absorción del sonido de alfombras sobre hormigón
  desnudo", column "Peso del pelo kg/m$^2$ (oz/yd$^2$)", on printed folio
  30.22 (PDF page 704), in chapter 30 of the Spanish edition. Non-normative
  source: a handbook.
- **The print:** every pile weight is printed twice, SI first and the US
  customary value in parentheses: "1,2 (35)", "1,5 (43)" and so on, eleven
  pairs in the table.
- **The problem:** one ounce per square yard is $0.033906$ kg/m$^2$ by
  definition, and five of the eleven pairs are not each other under the
  criterion the Chapter 32 entry uses (the SI half is the conversion rounded or
  truncated to the precision it is printed to):
  - "2,3 (66)", "3,1 (88)" and "2,1 (60)": $66 \times 0.0339 = 2.24$,
    $88 \times 0.0339 = 2.98$ and $60 \times 0.0339 = 2.03$. All three
    are what $0.035$ kg/m$^2$ per oz/yd$^2$ gives, a factor that also
    reproduces every pair of Tables 30.2 and 30.3 that does hold, which is how
    these three come out a tenth high: the metric column follows a factor 3 per
    cent above the definition.
  - "1,3 (32)": $32 \times 0.0339 = 1.08$, and $0.035$ gives $1.12$;
    neither reaches $1.3$.
  - "1,1 (3,2)": $3.2$ oz/yd$^2$ is $0.11$ kg/m$^2$. The same carpet,
    knotted, cut nylon with a pile of 14 mm, is printed "1,1 (32)" in Table
    30.3, so the imperial half has a stray decimal comma.
- **Evidence:** Verified on PDF page 704 (printed p. 30.22) of Harris (ed.),
  *Manual de medidas acústicas y control del ruido* 3.ª ed. (1995); both
  independent readings of the page and a crop of the column print all five
  pairs as quoted. All eight pairs of Table 30.3, on the same page and the
  next, hold.
- **What the library does:** in
  [`PUBLISHED_CARPETS`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/materials/absorbers/carpets.py) the
  four rows whose kilograms are in doubt serve no pile weight and mark the
  cell `misprinted`, quoting the pair; the fifth serves its 1.1 kg/m$^2$ and
  says in its note that the ounces lost a digit.
- **Status:** unreported.
