Skip to content

Errata in Ver & Beranek (2006)

These are the 4 entries of the errata registry on Ver & Beranek (2006), 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.

Ver & Beranek 2e (2006), TABLE 14.1 (three moduli whose e-notation is corrupted in the printing)

Section titled “Ver & Beranek 2e (2006), TABLE 14.1 (three moduli whose e-notation is corrupted in the printing)”
  • Location: TABLE 14.1, “Properties of Some Commercial Damping Materials”, on printed page 598 (PDF page 599), in chapter 14, “Structural Damping”, by Eric E. Ungar and Jeffrey A. Zapfe. Non-normative source: a handbook.
  • The print: the four moduli columns are set in a notation the table defines in its own footnote c: “The number following e represents the power of 10 by which the number preceding e is to be multiplied; e.g., 1.2e3 represents ”. Every cell of the block obeys it except three. Antiphon-13 prints 3.e3e under ; Soundcoat DYAD 606 prints 3G5 under ; and GE SMRD prints e35 under .
  • The problem: none of the three is a number in that notation. 3.e3e ends with an exponent marker that has no digit after it and has a decimal point with no fraction before the first marker; 3G5 puts a capital G where the marker belongs, and a capital G appears nowhere else in the table; e35 leads with the marker and has no mantissa at all. A reader cannot recover the intended value, because each one is consistent with more than one reading: 3.e3e could be with a stray marker or of some other setting, and e35 could be or .
  • Evidence: Verified on PDF page 599 (printed p. 598) of Ver & Beranek, Noise and Vibration Control Engineering 2e (2006). The three strings are legible and are what the page carries; the surrounding cells of the same block are equally legible and obey the notation, so the defect belongs to the printing. The neighbouring columns do not settle any of the three either. For Antiphon-13 the chapter’s own relation gives psi, which is consistent with a mantissa of 3 and an exponent of 4, but the printed string offers 3 and no legible exponent, and this library does not publish a value it had to finish itself.
  • What the library does: the three cells are held as misprinted in PUBLISHED_DAMPING, so the row keeps what the book prints and refuses to serve it as a number. Asking for one of them by printed() raises and names the glyphs. Every other cell of the seventeen rows is served normally.
  • Status: unreported.

Ver & Beranek 2e (2006), TABLE 14.1 (a transition modulus printed below the smallest modulus of its own row)

Section titled “Ver & Beranek 2e (2006), TABLE 14.1 (a transition modulus printed below the smallest modulus of its own row)”
  • Location: TABLE 14.1, “Properties of Some Commercial Damping Materials”, row “3M ISD-113”, on printed page 598 (PDF page 599), in chapter 14, “Structural Damping”, by Eric E. Ungar and Jeffrey A. Zapfe. Non-normative source: a handbook.
  • The print: the row reads , peak temperatures of , and , and the four moduli , , and psi.
  • The problem: the same page defines as “the smallest value of ”, in the paragraph under the table and again in footnote c, where is said to apply at low temperatures, at high ones and in the range of , which lies between them. Here psi is smaller than psi, so one of the two cells contradicts the definition the page gives for the other. The page does not say which. The printed is supported by its neighbour through the chapter’s own : psi, which is exactly what prints. Against that, every one of the other fourteen rows that prints all three moduli puts one to two orders of magnitude above , which is where would have to be for this row to behave like its neighbours. Neither reading can be had from the page alone.
  • Evidence: Verified on PDF page 599 (printed p. 598) of Ver & Beranek, Noise and Vibration Control Engineering 2e (2006). All four cells of the row are legible and unambiguous in the notation footnote c defines, and so is the sentence that defines ; the defect is a contradiction between two legible cells, not an illegible one. Taking as a shape test across the fifteen rows that print all three, the other fourteen fall between 0.976 and 1.054 and this row gives 0.031.
  • What the library does: both cells are served exactly as printed, because correcting either one would be this library choosing between two readings the page leaves open. The row carries a note saying so, which the published catalogue shows, and a test asserts the ordering holds on every other row so that a second occurrence cannot pass unnoticed.
  • Status: unreported.

Ver & Beranek 2e (2006), TABLE 8.5 (mass per unit area of the finest mesh, ten times too large in pounds)

Section titled “Ver & Beranek 2e (2006), TABLE 8.5 (mass per unit area of the finest mesh, ten times too large in pounds)”
  • Location: TABLE 8.5, “Mechanical Characteristics and Flow Resistance of Wire Mesh Cloths”, on printed page 262 (PDF page 266), in chapter 8, “Sound-Absorbing Materials and Sound Absorbers”. Non-normative source: a handbook.
  • The print: every quantity of the table is printed twice, but only three of the four pairs are one quantity in two systems of units: the wire count, the wire diameter and the mass per unit area. The fourth prints the flow resistance in N s/m3 and again as a multiple of . The mass per unit area column runs, in kg/m2 against lb/ft2: 1.6 / 0.32, then 1.2 / 0.25, then 0.63 / 0.13, then 0.48 / 0.1, and on the last row, the mesh of 80 wires per centimetre, 0.31 / 0.63.
  • The problem: one pound per square foot is kg/m2 by the definitions of the pound and the foot, so 0.31 kg/m2 is 0.063 lb/ft2 and not 0.63. The decimal point is one place too far right. Three things settle which of the two cells is the defective one. The four rows above convert to within the rounding of their own last digit, so the column is otherwise sound. The pound column as printed would make the finest mesh the heaviest cloth of the table, twice the mass of the coarsest, where every other column falls with the mesh. And the weave itself gives the mass: a square cloth of wires per metre of diameter carries per unit area, which for 8000 wires per metre of 57 m wire of density kg/m3 is 0.32 kg/m2, the printed 0.31 and not the printed 0.63. The page never says what the wire is made of, so that density is not read off it: 7800 kg/m3 is a stainless steel, and it is stated here because the argument is not reproducible without it. The same arithmetic at the same density reproduces the four rows above to within three per cent.
  • Evidence: Verified on PDF page 266 (printed p. 262) of Ver & Beranek, Noise and Vibration Control Engineering 2e (2006). Two readers transcribed the page independently and both read the cell as three glyphs, “0.63”, with no leading zero lost between them, and both read 0.31 in the cell beside it. The other four rows of the same column are equally legible and convert correctly, so the defect belongs to this cell and to the printing.
  • What the library does: this catalogue publishes the SI column of each pair, so the defective cell reaches no value served from PUBLISHED_RESISTIVE_SHEETS. The row holds the printed 0.31 kg/m2, which the column and the weave both support, and its note quotes the pound cell and points here, so a reader reproducing the book sees what the book says without it reaching a calculation. The note and not the misprinted hedge: that hedge says a number is not served, and it is read back only from the cell it empties, while the defective cell here is the customary restatement, for which this catalogue holds no column at all.
  • Status: unreported.

Ver & Beranek 2e (2006), TABLE 8.6 (the two surface-density columns disagree by one wrong factor on every row)

Section titled “Ver & Beranek 2e (2006), TABLE 8.6 (the two surface-density columns disagree by one wrong factor on every row)”
  • Location: TABLE 8.6, “Mechanical Characteristics and Flow Resistance of Glass Fiber Cloth”, on printed page 263 (PDF page 267), in chapter 8, “Sound-Absorbing Materials and Sound Absorbers”. Non-normative source: a handbook.
  • The print: the surface density is printed twice on each of the thirteen rows, in oz/yd2 and in g/m2: 3.16 / 96, 5.37 / 164, 6.70 / 204, 8.90 / 272, 19.2 / 585, 17.7 / 535, 12.3 / 375, 1.87 / 57, 1.94 / 59, 9.60 / 293, 14.5 / 442, 24.6 / 750 and 12.0 / 366.
  • The problem: one ounce per square yard is g/m2 by the definitions of the ounce and the yard, both exact. What is exact is the quotient; 33.906 is that quotient to three decimals, and this entry writes it that way whenever it is quoted short. The ratio the table prints is between 30.2 and 30.6 on all thirteen rows and never once 33.9, so the two columns cannot both be right, and the offset is the same eleven per cent throughout: one wrong conversion applied to the whole column rather than thirteen independent slips. Which of the two columns carries it the page does not say, and no other column of the table settles it: the weave and the flow resistance are printed once each, in one unit, and neither determines a surface density.
  • Evidence: Verified on PDF page 267 (printed p. 263) of Ver & Beranek, Noise and Vibration Control Engineering 2e (2006). Two readers transcribed the thirteen pairs independently and agreed on every digit. The other two tables of the chapter are not like this one: TABLE 8.7 converts its own mass column correctly on all eleven rows, and TABLE 8.5 on four of its five, the fifth being the single defective cell registered in the entry above. One cell of one row is a slip of the printing; thirteen rows off by one factor is a column, which is what makes this a property of this table rather than of the chapter.
  • What the library does: holds the gramme per square metre the page prints, as it prints it, in surface_density_g_m2. Every row of this table in PUBLISHED_RESISTIVE_SHEETS carries a note that quotes both printed values and the exact factor between their units and points here, so the contradiction reaches the reader with the number rather than instead of it. The library neither chooses between the two columns nor converts either of them: it publishes the one the page prints in SI and says what the other one says. The cell is not not_derivable, which is for a value this library declines to compute from cells the page did print and never for a quantity the page prints itself. The flow resistance of these cloths, which is printed once and in one unit, is served normally.
  • Status: unreported.