Errata in ECAC Doc 29
These are the 5 entries of the errata registry on ECAC Doc 29, 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.
ECAC Doc 29, 5th ed., Volume 2, Appendix B, Eq. (B-41) (descent deceleration)
Section titled “ECAC Doc 29, 5th ed., Volume 2, Appendix B, Eq. (B-41) (descent deceleration)”- Location: Appendix B, section B7.1.1, the deceleration defined under Eq. (B-41), on the page that carries Eq. (B-40) and Eq. (B-41).
- The print: , that is both ground speeds divided by over twice the slant length of the segment.
- The problem: the descent slope is counted twice. The mean deceleration along the flight path is the change in the square of the along-path speed over twice the path length; the printed expression converts the speeds to along-path values and uses a path length that is already the slant one, so it overstates by . The 4th edition’s Eq. (B-21) is self-consistent, and the denominator is not what changed: it reads with “the ground distance covered”, which is the same slant length the 5th edition writes as . What changed is the numerator. The 4th edition’s Eq. (B-22) defines the speeds it divides as groundspeeds, , that is the true airspeed resolved into the horizontal plane, so dividing each by correctly restores an along-path speed. The 5th edition feeds Eq. (B-41) the profile points’ own , which is along-path already, and kept the division. Doc 29’s own reference results decide it: of the twelve points of Volume 3 Part 2 case 2D, flown entirely at that step type, nine are reached by the deceleration. The plain ground speeds reproduce the tabulated thrust at every one of the twelve, worst deviation 0.047 lb, while the printed divided speeds fall short at all nine, by 6.05, 5.47, 4.02, 3.81, 4.56, 4.45, 0.29, 6.35 and 6.41 lb in profile order: always low, and never within the workbook’s own 0.05 lb of printed precision. The drag term beside it does keep the the same equation prints, which the same points confirm to the same 0.05 lb.
- Evidence: reproduction of Volume 3 Part 2 sheet
D1-(Arrival_Results)case 2D under each reading. Verified on PDF page 104 (printed p. B-31) of ECAC.CEAC Doc 29, 5th ed., Volume 2: Technical guide, which carries Eq. (B-40), Eq. (B-41) and the deceleration under it, and on PDF page 90 (printed p. B-15) of ECAC.CEAC Doc 29, 4th edition, Volume 2, which carries Eq. (B-21) and, immediately under it, the Eq. (B-22) that defines its and as groundspeeds and so decides the reading. - Library behaviour:
flight_performancecomputes the deceleration from the plain ground speeds over the slant length, and the helper’s docstring carries the departure and the numbers above.test_arrival_case_reproduces_every_profile_pointpins all 124 arrival points, and the conformance row ECAC Doc 29 Appendix B approach thrust pins the descent thrust of case 2A. - Status: unreported.
ECAC Doc 29, 5th ed., Volume 2, Appendix B, Eq. (B-18) (runway gradient)
Section titled “ECAC Doc 29, 5th ed., Volume 2, Appendix B, Eq. (B-18) (runway gradient)”- Location: Appendix B, section B6.1.1, the average acceleration defined under Eq. (B-18).
- The print: ” is the average acceleration (ft/s) along the runway, equal to: ”, with ” is the Calibrated Airspeed (kt) at Point2” and in feet on the same page.
- The problem: the expression is declared in ft/s and evaluates in kt/ft. The missing factor is , the square of the knots-to-feet-per-second constant Doc 29 fixes in B2.2 and carries explicitly in Eq. (B-24) and Eq. (B-41), which build accelerations out of the same kind of expression. It is not cosmetic: enters only through , so understating it by 2.85 overstates the gradient correction, and at a 1 % upslope with kt and ft the dimensionally correct 7.69 ft/s gives a factor of 1.0437 against the literal reading’s 1.1353: 8.8 % of take-off distance. The 4th edition carries the same omission, so it is inherited rather than introduced, and prints where the 5th prints ; only the 5th edition’s placement is a speed the aeroplane has, since is the true airspeed of Eq. (B-7).
- Evidence: dimensional analysis against the same document’s Eq. (B-24)
and Eq. (B-41). Verified on PDF page 90 (printed p. B-17) of ECAC.CEAC
Doc 29, 5th ed., Volume 2: Technical guide, and, for the inherited half, on
PDF page 86 (printed p. B-11) of ECAC.CEAC Doc 29, 4th edition, Volume 2,
where the same definition sits under Eq. (B-11) and reads
, ft/s.
This one cannot be
arbitrated against the reference results: Volume 3 Part 2’s departure case
sheet
C8-(Departure_Cases)has no runway-gradient column, so all 17 reference cases are flown at , where Eq. (B-18) is the identity. - Library behaviour:
flight_performancerestores and takes the 5th edition’s placement; the helper’s docstring states both departures and that no reference case can detect either. - Status: unreported.
ECAC Doc 29, 5th ed., Volume 2, Appendix B, Eq. (B-21) (mid-step airspeed)
Section titled “ECAC Doc 29, 5th ed., Volume 2, Appendix B, Eq. (B-21) (mid-step airspeed)”- Location: Appendix B, section B6.1.2, the mid-step corrected net thrust defined under Eq. (B-21), in the branch that computes it from Eq. (B-12), that is for every aeroplane the propeller coefficient table carries. B4.1 and B4.2 split the turboprops between them without stating a rule, so this is not the same set as “the turboprops”: of the 20 in ANP v2.3, 11 sit in the propeller table and reach Eq. (B-12), the other 9 sit in the jet table and reach Eq. (B-9), and the 8 piston aeroplanes are all in the propeller table.
- The print: “is the Corrected Net Thrust of the aircraft when being located at mid-step, i.e. at the altitude ”, and then, under “In the case of Eq. B-12,”, , the root mean square of the two endpoint true airspeeds.
- The problem: the speed contradicts the altitude named one line above it. A Climb step is flown at a held calibrated airspeed, so the true airspeed at the mid-step altitude is fixed and is evaluated at the mid-step (Eq. B-7); the root mean square of the two endpoint values is a different number. The two branches of the same list therefore describe two different aeroplanes at the one point they both call mid-step: the jet branch prints , which is the step’s own held calibrated airspeed and so places the aeroplane at halfway up, while the Eq. (B-12) branch places it at the root mean square of the ends. The mean also reads as transplanted. Section B6.1.3, for the Accelerate step, is built on exactly this root mean square and is self-consistent about it, giving both branches the same and converting it for the jet form with the mid-step ; B6.1.2 keeps the Eq. (B-12) line but replaces the jet line with a Point 2 quantity, and only one of the two survives the substitution. Of the candidates the printed one is the largest: at constant the true airspeed rises convexly with altitude, so the root mean square exceeds the arithmetic mean, which exceeds the mid-altitude value. Eq. (B-12) makes thrust inversely proportional to , so the printed speed understates the mid-step thrust, understates and lays the climb down long.
- Evidence: reproduction of Volume 3 Part 2 sheet
D2-(Departure_Results)under each reading. The four turboprop departure cases are the only reference data that reach this branch, and they are unanimous. On case 56 the final profile point is printed at 400814.3 ft: the mid-step altitude reading lands it 0.001 ft away, the arithmetic mean 323.944 ft long and the printed root mean square 544.944 ft long, against the 0.15 ft the departure distances are otherwise matched to. Cases 8, 28 and 68 put the same final point 172.333, 223.003 and 544.944 ft long under the printed reading and 102.535, 132.673 and 323.944 ft long under the arithmetic mean, always long and never near the printed precision; the mid-step altitude reading is within 0.049 ft of every point of all four cases, worst case 28. The departure grows with the height of the step, as a convexity error must: on case 8 the printed reading puts 0.0225 kt above the mid-step value on the 1500 ft climb and 0.1066 kt above it on the 2500 ft one. Verified on PDF page 92 (printed p. B-19) of ECAC.CEAC Doc 29, 5th ed., Volume 2: Technical guide, which carries the mid-step altitude sentence, the jet branch’s and the Eq. (B-12) branch’s on the one page, and on PDF page 95 (printed p. B-22) of the same document for the B6.1.3 pair the Eq. (B-21) line appears to be drawn from. - Library behaviour:
flight_performanceevaluates the propeller form at the true airspeed the aeroplane has at the mid-step altitude, and the Climb-step helper’s comment quotes the printed expression, says the model departs from it and points here.test_departure_case_reproduces_every_profile_pointpins all 190 departure points, four cases of which are flown on Eq. (B-12). - Status: unreported. Of the three Appendix B departures recorded here this is the one the reference results decide most sharply, and the only one that changes a shipped profile.
ECAC Doc 29, 5th ed., Volume 2, Appendix D, Table D-6c (the last row is the last row of Table D-6b)
Section titled “ECAC Doc 29, 5th ed., Volume 2, Appendix D, Table D-6c (the last row is the last row of Table D-6b)”- Location: Appendix D, Table D-6c “Revised JETW NPD data using SAE-ARP-866A”, last row (JETW, SEL, operation D, power setting 22500).
- The print: 109.8, 106.0, 103.2, 100.2, 95.2, 89.8, 85.8, 81.4, 76.8 and 72.3 dB from 200 ft to 25000 ft: the ten values of the last row of Table D-6b, the same aeroplane revised with SAE-ARP-5534, printed a few lines above it on the same page.
- The problem: the text under Table D-5 builds Table D-6c as the Table D-6a levels plus the SAE-ARP-866A increments of Table D-5, and the other six rows of the table are exactly that, to the last digit. The last row of Table D-6a (109.7, 105.7, 102.8, 99.5, 94.0, 88.0, 83.7, 79.0, 73.9 and 68.7 dB) plus the DEP_103 increments of Table D-5 (+0.2, +0.3, +0.4, +0.6, +0.9, +1.3, +1.5, +1.8, +2.2 and +2.8 dB) gives 109.9, 106.0, 103.2, 100.1, 94.9, 89.3, 85.2, 80.8, 76.1 and 71.5 dB, and the unrounded increments give the same ten values. Eight of the ten printed cells differ from these, by up to 0.8 dB at 25000 ft; the two that agree, at 400 ft and 630 ft, are the two where the increments of the two routes round to the same sum. The row has been copied from Table D-6b rather than computed.
- Evidence: recomputation from Tables D-6a and D-5 of the same example. Verified on PDF page 133 (printed p. D-9) of ECAC.CEAC Doc 29, 5th ed., Volume 2: Technical guide, which carries Tables D-6a, D-6b and D-6c on the one page, and on PDF page 132 (printed p. D-8) of the same document for the increments of Table D-5.
- Library behaviour:
revise_npd_curvesadds the increment to Table D-6a and gives the recomputed row.test_revised_npd_data_is_table_d6c_but_its_misprinted_rowpins it, and the conformance row ECAC Doc 29 Appendix D Table D-6c holds the other 60 cells of the table. - Status: unreported.
ECAC Doc 29, 5th ed., Volume 2, Appendix D, text under Table D-2 (Tables D-3b and D-3c paired with the wrong methods)
Section titled “ECAC Doc 29, 5th ed., Volume 2, Appendix D, text under Table D-2 (Tables D-3b and D-3c paired with the wrong methods)”- Location: Appendix D, the paragraph under Table D-2 that introduces the three attenuation tables of the worked example.
- The print: “together with the absorption coefficients in Table D-1 for the AIR-1845 atmosphere and using attenuation values calculated using SAE-ARP-5534 and SAE-ARP-866A for the specified atmosphere […]. All three sets of attenuation values are listed in Tables D-3a, D-3b and D-3c respectively.”
- The problem: “respectively” gives SAE-ARP-5534 to Table D-3b and SAE-ARP-866A to Table D-3c, and the tables are titled the other way round: Table D-3b “calculated using SAE-ARP-866A”, Table D-3c “calculated using SAE-ARP-5534”. The values follow the titles. SAE ARP 5534 reproduces all 240 cells of Table D-3c to the printed digit, and SAE ARP 866A 208 of the 240 of Table D-3b, the other 32 (the longest high-frequency paths) within half a unit of the printed digit plus 7 parts per million, while the pairing the sentence states is off by as much as 177 dB, at 10 kHz and 25000 ft (921.851 dB against 744.810 dB). The text after the tables, which gives Table D-4 to SAE-ARP-5534 and Table D-5 to SAE-ARP-866A, follows the titles too.
- Evidence: comparison of the sentence with the two table titles, and recomputation of both tables by both methods. Verified on PDF page 128 (printed p. D-4) of ECAC.CEAC Doc 29, 5th ed., Volume 2: Technical guide for the sentence, and on PDF pages 130 and 131 (printed pp. D-6 and D-7) of the same document for the titles and values of Tables D-3b and D-3c.
- Library behaviour: none needed; the library pairs each table with the method its title names, which is the pairing its values confirm. The conformance rows ECAC Doc 29 Appendix D Table D-3b and Table D-3c hold the two tables.
- Status: unreported.
Created and maintained by José Manuel Requena PlensORCIDAbout the maintainerRelease notesChangelogAll projects