Appendix B
Quality Assurance and Quality Control

The draft report and associated material provided to the committee have not been adequately checked for correctness. This appendix documents many errors discovered by the committee, but the committee does not have the resources for a complete quality-assurance and quality-control check, so more errors can be expected.

The committee has included in this appendix all the errors that it found and that were apparently related to QA/QC. They are listed under four separate headings, in approximate order of decreasing significance within each heading, and with headings in approximate order of decreasing significance. The first heading, “Incorrect assignment of reactors”, could be considered an example under the second heading (inadequate verification of code), but this error alone invalidates all the calculations performed. Under the second heading, on verification of computer code, the first two bulleted items are major errors and resulted in computer code that can give substantial inaccuracies (off by a factor of more than 10). The following four bulleted items probably have small effects on the calculations; the last would have a trivial effect. The third heading describes inconsistencies between document and code whose effect cannot be determined by the committee. Finally, the fourth heading lists many transcription errors discovered by the committee; most of them probably resulted in trivial effects on the calculations, but a few could have a large effect (for example, if such an error coincides with a critical calibration measurement). Although some of the transcription errors occur in material that ultimately was not used in the calculations, the committee included them to warn against trusting the transcriptions in their current form in case more work is required by CDC.

1.Incorrect assignment of reactors.

The programs incorrectly assign reactor names, so most of the reactors are placed in the wrong locations. The correspondences are as follows:

Correct reactor name

B

C

D

DR

F

H

KE

KW

Read by programs as

B

C

KW

KE

D

DR

H

F

That happens in the interaction between the source file “reactor.dat” in the \reactor subdirectory and the “mksrc.PL” Perl script in the \rrates\scripts subdirectory, (“reactor.dat” is an ASCII file produced from the sheet “totals” of “reactor operations.xls” in the \reactor subdirectory.) The Perl script “mksrc.PL” expects “reactor.dat” to have release fractions (relative energy-production fractions for each month) in the order given as “Read by programs as” above, but “reactor.dat” contains them in the “Correct reactor name” order above. The script “mksrc.PL” produces the release-rate files for each individual nuclide in the various \rrates subdirectories. (There is also a “reactor.dat” file in the \rrates\scripts subdirectory, but it is a previous version that included the N reactor. The earlier file actually has the release fractions in the order required by the Perl script).

The effect of the mixup in ordering of reactors is to confuse the reactors in the way indicated in the correspondence above. For example, whenever reactor KW is required, values corresponding to reactor D are substituted, and the emission rates and times for reactor D are



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 26
Appendix B Quality Assurance and Quality Control The draft report and associated material provided to the committee have not been adequately checked for correctness. This appendix documents many errors discovered by the committee, but the committee does not have the resources for a complete quality-assurance and quality-control check, so more errors can be expected. The committee has included in this appendix all the errors that it found and that were apparently related to QA/QC. They are listed under four separate headings, in approximate order of decreasing significance within each heading, and with headings in approximate order of decreasing significance. The first heading, “Incorrect assignment of reactors”, could be considered an example under the second heading (inadequate verification of code), but this error alone invalidates all the calculations performed. Under the second heading, on verification of computer code, the first two bulleted items are major errors and resulted in computer code that can give substantial inaccuracies (off by a factor of more than 10). The following four bulleted items probably have small effects on the calculations; the last would have a trivial effect. The third heading describes inconsistencies between document and code whose effect cannot be determined by the committee. Finally, the fourth heading lists many transcription errors discovered by the committee; most of them probably resulted in trivial effects on the calculations, but a few could have a large effect (for example, if such an error coincides with a critical calibration measurement). Although some of the transcription errors occur in material that ultimately was not used in the calculations, the committee included them to warn against trusting the transcriptions in their current form in case more work is required by CDC. 1.Incorrect assignment of reactors. The programs incorrectly assign reactor names, so most of the reactors are placed in the wrong locations. The correspondences are as follows: Correct reactor name B C D DR F H KE KW Read by programs as B C KW KE D DR H F That happens in the interaction between the source file “reactor.dat” in the \reactor subdirectory and the “mksrc.PL” Perl script in the \rrates\scripts subdirectory, (“reactor.dat” is an ASCII file produced from the sheet “totals” of “reactor operations.xls” in the \reactor subdirectory.) The Perl script “mksrc.PL” expects “reactor.dat” to have release fractions (relative energy-production fractions for each month) in the order given as “Read by programs as” above, but “reactor.dat” contains them in the “Correct reactor name” order above. The script “mksrc.PL” produces the release-rate files for each individual nuclide in the various \rrates subdirectories. (There is also a “reactor.dat” file in the \rrates\scripts subdirectory, but it is a previous version that included the N reactor. The earlier file actually has the release fractions in the order required by the Perl script). The effect of the mixup in ordering of reactors is to confuse the reactors in the way indicated in the correspondence above. For example, whenever reactor KW is required, values corresponding to reactor D are substituted, and the emission rates and times for reactor D are

OCR for page 26
used when those for reactor KW are required. The effect is apparent if one examines the “*.rel” files in the \rrates subdirectories for individual nuclides. For example, examination of the “100KWmed.rel” files indicates that the reactor supposedly represented by files of these names started operation at day 335 in 1944. However, the KW reactor did not start operating until 1955; it was the D reactor that started at day 335. Data files for the RVRDSP program use the “*.rel” files as inputs specified in the “*.par” files controlling RVRDSP. The particular reactors required are selected by choice of the names of the “*.rel” files, and they are associated with locations in the “*.par” files. Those associations correspond to the names of the “*.rel” files, not to the reactors that are actually (erroneously) specified by the data in the “*.rel” files. In other words, there is no correction of the error introduced by the mismatch between “reactor.dat” and “mksrc.PL”. Assuming that the files provided to the committee correspond to those used (something the committee has not verified), this error alone invalidates the entire set of calculations. 2. Inadequate verification of the computer code. Examining the code of RVRDSP itself shows that it has not been adequately analyzed or debugged. For example, the committee located the following errors (there might be others): • The function “kernal” that is supposed to evaluate equation (2) of the draft report attempts to reduce computational effort by not evaluating the cosine series sum if (see line 1082 in the file “rvrdsp.for”) However, that test simply ensures that so the n=1 term of the sum has a magnitude that might be as large as 2×exp(-0.629)˜1.1, which can hardly be neglected when compared with unity, particularly if the sign of the term is negative (as occurs toward the far bank). It seems likely that what was intended was the test inasmuch as that would have ensured an error of less than 10-16. The reason for including the cosine sum in this series is to take account of an unmixed plume in the river (that is, to allow for the releases not being completely mixed across the river at calibration points and at points where dose has to be evaluated). Including the current test in function “kernal” removes any effect of the plume well before complete mixing has occurred. Ey was used as a calibration parameter (pages 32–34), so the entire calibration of the model would

OCR for page 26
probably have been incorrect (even in the absence of the other problems documented here) because of this error. • For small values of t, the code in function “kernal” will produce substantially incorrect results because of the limitation to 20 terms in the summation. This problem was recognized by Codell et al. (1982), from which much of the inspiration for the solution of equation (1) of the draft report appears to have been drawn. The committee does not understand why the solution given by Codell et al. (1982) was not used for this situation. The alternative forms of the solution given in Appendix A show how to handle all argument values rapidly with arbitrary precision. Alternatively, the draft report should prove that small values of t are never required for the application at hand or that any small values of t that do occur have negligible effect on the results. • In subroutine “simpson”, the following code segment occurs (at lines 842–847 in the file “rvrdsp.for’) just before a successful return:   S=(4.0*ST-OST)/3 C check for convergence   if (ABS(S-OS).le.eps*ABS(OS).and.j.gt.jstart) then   C=st   return   else The returned value (the value assigned to C) is the wrong one. It should be S that is returned, not st (that is, the assignment statement at line 845 should be C=S, not C=st). As a result, the integration is not by Simpson’s rule, but by a trapezoidal rule, and the error estimate is incorrect. If the function being integrated had smooth derivatives up to fourth order, the relative error of the result would be approximately the square root of eps, rather than eps. Here, eps is a symbolic name in the computer code. The value assigned to eps (10-6) was supposed to approximate the maximum possible relative error in the Simpson rule integration, but the error in the code means that the maximum relative error is of order the square root of eps (10-3). The functions being integrated in the code have finite jumps in their first derivatives,2 so the relative error cannot be easily predicted, but it cannot be expected to be much better than the square root of eps. For the example in Table 3 of the draft report, eps was set to equal 1×10-6, so the expected relative error of the result can be expected to be around 1×10-3. Even with this larger potential error, however, the results from the computation should be substantially more accurate than the 3% differences observed in the Draft report at Table 3 on page 32 (see the discussion above). • The integration ranges may be unnecessarily broad. There is no code in the subroutine “limits” that limits the integration range to the period of reactor releases. Whether that has any effect depends on the broadness of the ranges computed by the Codell et al. (1982) limits described in the draft report in equation (15) (page 31). If the integration range is set too broad 2   The function that should be integrated has finite jumps in its value—see below.

OCR for page 26
(with long periods of zero release rate at the beginning or end of the integration range), the numerical integration could (in principle) miss isolated months of high release. The effect of too broad an integration range just described occurred for the results shown in Table 3 of the draft report between RIVLAK and RVRDSP. RIVLAK performs the integration of the 0–11 s of the release by integrating over 0–11 s. RVRDSP, however, performs the integration over the range 0–973.9 s for the 5,000-s entry in Table 3 (and over the ranges 0– 2,973.9 s, 0–5,973.9 s, and 0–7,973.9 s for the 7,000-, 10,000-, and 12,000-s points, respectively). For most of the integration range, of course, the source strength was zero. For the 5,000-s entry in Table 3, therefore, about 98.87% [(973.9–11)/973.9] of the function evaluations in the integration routine returned the value zero. Even with the maximal number of function evaluations allowed by the input file (217˜130,000), only about 1,500 would have fallen within the 11-s source-release period. • The code linearly interpolates into the tables of release rates. However, those release-rate tables were generated month by month as the average release rates for individual months, so each represents the average release rate for a given month. The linear interpolation thus treats the release-rate estimates in a different way from the way they were estimated. Instead of a series of constant release rates in each month, the code interprets the tables to show a series of linear ramps in release rate. At the beginning of a month, the release rate is taken to be the average value for that month; then, throughout the month, the release rate is linearly changed until at the end of the month it is equal to the average release rate in the following month. The same linear interpolation has been performed for river flow rates (in subroutine “getflow”) to derive a “velocity”, although once again the input flow rates are supposed to be monthly averages. The river width and depth were derived from the flow rate external to the program and without any such linear interpolation, assuming that the flow rates were monthly averages. The program treatment of the flow rates is thus inconsistent with their derivation. The problem apparently was recognized at some point, inasmuch as the RVRDSP program contains a commented-out version of the interpolation routine that is supposed to perform the correct interpolation. The history documentation also includes the comment: “c 05/05/01: changed lint routine to use step function release instead on linear interpolation c   -commented out new lint routine 05/06/01”, but there is no indication why the correct approach was immediately rejected. • One effect of the linear-interpolation scheme used is the extension of the releases into the month preceding the start of releases (if the release did not start at time zero) and into the month following the end of releases (if the release did not end after the time at which the concentration is being computed). 3. Inconsistent parameter values used in the control files. Reactor and receptor locations are specified to the RVRDSP program in the “*.par” files. In many (perhaps all) of those files, some locations do not correspond to the locations

OCR for page 26
documented in the draft report. For example, page 33 and Table 5 (page 34) of the draft report indicate that the distance from the near shore for the Ringold sampling location is supposed to be 400 m. However, all the “*.par” files that the committee examined that contain Ringold as a sampling location have distance from the near shore set at 300 m (there are at least 116 such files, and the committee did not examine all of them). As a further example, page 14 of the draft report gives the location of the reactors on the river, but the “*.par” files indicate that the outfalls are apparently up to a mile away from the reactor locations, as indicated in the following table. Reactor River Mile (page 14 of draft report) m from RM 385 (based on draft report) *.par files Difference (m) B 384 1609.344 1609.344 0 C 383.6 2253.082 1609.344 -644 D 377.6 11909.146 12070.080 161 DR 377.6 11909.146 12070.080 161 F 369 25749.504 27358.850 1609 H 372.5 20116.800 20116.800 0 KE 381.4 5793.638 4828.032 -966 KW 381.8 5149.901 4828.032 -322 The number of significant digits given in the third column of this table is not warranted by the accuracy of the input information, but is included to match the number of digits used in the *.par files. The final column has been arbitrarily rounded to an integer number of meters—the only intent is to show discrepancies larger than expected from the information given. (Again, the committee has not examined all the “*.par” files.) If the outfalls were at such different distances down river from the reactors, that fact should be documented in the draft report. A similar inconsistency occurs in Table 4 (page 33) of the draft report, where the outfall location for the 100N reactor is specified as being 100 m from the near shore; but the 100N reactor did not discharge to the river (page 14 of the draft report, footnote c). 4. Incorrect transcriptions from HEDR documents. The data entered in the workbook “reactor operations.xls” contain numerous errors. The committee noted the following errors in the sheet “Reactor Data”, but the technique used by the committee to discover these errors does not guarantee the ansence of other errors. Moreover, the data entered in different sheets of this workbook are inconsistent; some errors are corrected between the “Reactor Data” and the “Power_data” sheets. Of these data, only the reactor energy-production data are used; but there are many errors among them. B Reactor power (MW) 8/51 Month is 8/51, not 9/51 8/51 Should be 398, not 399 2/55 Should be 841, not 941 12/55 Should be 895, not 95 (corrected in the “Power_data” sheet) 6/56 Should be 834, not 34 (corrected in the “Power data” sheet)

OCR for page 26
3/59 Should be 1468, not 1469 2/61 Should be 1783, not 1793 4/61 Should be 1787, not 1797 4/63 Should be 1805, not 1905 5/63 Should be 1785, not 1783 6/65 Should be 1818, not 1819 5/66 Should be 1938, not 1939 10/66 Should be 1805, not 1905 12/66 Should be 1982, not 1992 B Reactor energy production (MWd) 7/56 Should be 24080, not 24090 12/56 Should be 17800, not 17900 9/64 Month 9/64 should be 8/64 9/64 Should be 38920, not 33920 11/64 Should be 38210, not 39210 1/65 Should be 47580, not 47590 5/65 Should be 43280, not 43290 5/66 Should be 60080, not 60090 12/66 Should be 55050, not 55030 1/67 Should be 48380, not 48390 6/67 Should be 59820, not 59920 B Reactor water flow rate (Kgpm) 1/53 Should be 37.8, not 37.9 3/55 Should be 48.8, not 49.8 6/56 Should be 45.8, not 45.9 12/58 Should be 76.8, not 76.9 9/59 Should be 82.8, not 82.9 8/61 Should be 88.4, not 98.4 9/61 Should be 88.6, not 98.6 10/62 Should be 88.9, not 89.9 1/63 Should be 89.3, not 99.3 4/63 Should be 89.3, not 99.3 5/63 Should be 89.8, not 99.8 1/65 Should be 87.5, not 97.5 5/65 Should be 90.0, not 0.9 11/65 Should be 88.7, not 89.7 9/67 Should be 95.8, not 95.9 C Reactor power (MW) 6/46 Should be 0, not 1 1/47 Should be 0, not 1 5/53 Should be 876, not 976 11/59 Should be 1812, not 1912 10/60 Should be 1835, not 1935

OCR for page 26
2/66 Should be 2358, not 2359 5/66 Should be 2286, not 2296 8/66 Should be 2065, not 2063 C Reactor energy production (MWd) 4/56 Should be 36580, not 36590 1/58 Should be 41840, not 41940 6/58 Should be 41810, not 41910 8/59 Should be 44810, not 44910 8/64 Should be 52680, not 52690 C Reactor water flow rate (Kgpm) 5/54 Should be 88.7, not 98.7 2/55 Should be 82.8, not 2.8 3/55 Should be 81.8, not 91.8 12/55 Should be 83.8, not 93.8 1/56 Should be 84, not 94 7/56 Should be 85.8, not 85.9 1/57 Should be 81.2, not 91.2 2/57 Should be 81.6, not 1.6 3/57 Should be 82.9, not 2.9 11/58 Should be 93.5, not 95.5 6/60 Should be 95.8, not 95.9 9/60 Should be 95.8, not 95.9 5/61 Should be 98.5, not 99.5 2/63 Should be 97.8, not 97.9 2/64 Should be 97.8, not 97.9 5/65 Should be 100.8, not 100.9 5/66 Should be 107.8 not 107.9 D Reactor power (MW) 5/49 Should be 287, not 297 6/50 Should be 296, not 0.02 (corrected in “Power_data” sheet) 11/55 Should be 878, not 78 12/55 Should be 843, not 43 3/56 Should be 886, not 896 5/56 Should be 831, not 931 8/56 Should be 847, not 947 10/59 Should be 1548, not 1549 2/61 Should be 1864, not 1964 3/61 Should be 1871, not 1971 6/65 Should be 1889, not 1989 8/66 Should be 1804, not 1904 D Reactor energy production (MWd) 12/47 Should be 5810, not 5910

OCR for page 26
1/48 Should be 6880, not 6890 11/49 Should be 8300, not 9300 5/50 Should be 8480, not 9480 5/51 Should be 10880, not 10980 4/56 Should be 18890, not 19890 8/63 Should be 51840, not 51940 1/64 Should be 48270, not 49860 12/64 Should be 50390, not 30390 11/65 Should be 22280, not 22290 4/67 Should be 48980, not 48990 D Reactor water flow rate (Kgpm) 12/51 Should be 38.1, not 39.1 9/57 Should be 71, not 7.1 8/59 Should be 81, not 91 11/60 Should be 84.9, not 94.9 2/61 Should be 86, not 96 7/61 Should be 87.9, not 97.9 8/61 Should be 88.2, not 98.2 12/62 Should be 87.8, not 87.9 7/63 Should be 87.8, not 87.5 11/63 Should be 87.8, not 87.9 5/65 Should be 86.3, not 96.3 DR Reactor power (MW) On “Power_data” sheet the column labeled “DR Power” erroneously contains the DR reactor energy production in place of power. 2/52 Should be 380, not 390 5/52 Should be 548, not 549 12/55 Should be 862, not 962 2/56 Should be 828, not 929 4/61 Should be 1628, not 1629 5/61 Should be 1803, not 1903 2/62 Should be 1825, not 1925 4/62 Should be 1648, not 1649 4/63 Should be 1810, not 1510 DR Reactor energy production (MWd) 5/52 Should be 16480, not 16490 8/54 Should be 20480, not 20490 11/61 Should be 43780, not 43790 6/62 Should be 18600, not 19600 DR Reactor water flow rate (Kgpm) 8/51 Should be 40.8, not 40.9 4/52 Should be 40.8, not 40.9

OCR for page 26
12/53 Should be 42.8, not 42.9 6/54 Should be 42.8, not 42.9 10/54 Should be 42.7, not 42.0 6/55 Should be 49.8, not 49.9 8/55 Should be 50.8, not 50.9 6/57 Should be 70.5, not 703 2/60 Should be 80, not 50 7/60 Should be 83.9, not 93.9 12/60 Should be 84.2, not 94.2 5/61 Should be 87.8, not 87.9 9/61 Should be 88.7, not 98.7 11/61 Should be 87.8, not 97.8 12/61 Should be 87.7, not 97.7 F Reactor power (MW) 6/45 Should be 258, not 259 1/47 Should be 198, not 199 12/48 Should be 268, not 269 7/51 Should be 386, not 396 9/52 Should be 488, not 499 2/54 Should be 628, not 629 5/57 Should be 1058, not 1059 10/61 Should be 1738, not 1739 9/64 Should be 1681, not 1691 4/65 Should be 1877, not 1977 F Reactor energy production (MWd) 2/47 Should be 5250, not 525 12/49 Should be 7680, not 7690 5/50 Should be 8670, not 9670 6/50 Should be 8480, not 8490 3/51 Should be 11650, not 11630 6/52 Should be 10880, not 10550 3/53 Should be 14780, not 14790 8/53 Should be 16080, not 16090 10/53 Should be 18700, not 19700 1/54 Should be 16870, not 16970 8/54 Should be 20830, not 20930 2/57 Should be 23480, not 23490 3/59 Should be 28690, not 29690 12/59 Should be 40280, not 40290 4/61 Should be 48880, not 48890 8/61 Should be 38830, not 38930 11/64 Should be 44580, not 44590

OCR for page 26
F Reactor water flow rate (Kgpm) 2/47 Should be 31.04, not 11.04 7/47 Should be 31.78, not 31.79 9/48 Should be 32.48, not 32.49 4/51 Should be 31.85, not 31.95 9/52 Should be 38.1, not 39.1 3/53 Should be 38.7, not 39.7 8/53 Should be 38.5, not 39.5 6/56 Should be 43.8, not 43.9 11/56 Should be 48, not 49 2/60 Should be 83.5, not 93.5 1/61 Should be 86.7, not 96.7 4/62 Should be 89.5, not 99.5 6/62 Should be 89.3, not 99.3 7/62 Should be 89.4, not 99.4 4/63 Should be 89.3, not 99.3 8/63 Should be 87.5, not 97.5 10/63 Should be 87.6, not 97.6 11/63 Should be 86.3, not 96.3 2/64 Should be 88.8, not 98.8 4/64 Should be 89.6, not 99.6 5/64 Should be 89, not 99 7/64 Should be 89.3, not 99.3 8/64 Should be 89.9, not 99.9 11/64 Should be 89.8, not 99.8 H Reactor power (MW) 9/51 Should be 538, not 539 2/52 Should be 558, not 559 2/54 Should be 889, not 999 3/54 Should be 877, not 977 2/60 Should be 1708, not 1709 4/62 Should be 1698, not 1699 3/63 Should be 1861, not 1961 8/64 Should be 1832, not 1932 H Reactor energy production (MWd) 10/50 Should be 11430, not 0.1143 4/51 Should be 9450, not 94.5 6/52 Should be 14380, not 14390 9/52 Should be 16540, not 1650 2/53 Should be 17350, not 173.5 11/53 Should be 21540, not 2150 9/54 Should be 24180, not 24190 12/55 Should be 18410, not 19410 9/57 Should be 27820, not 27920

OCR for page 26
6/58 Should be 34350, not 343.5 4/59 Month 4/59 should be 4/58 7/59 Month 7/59 should be 7/58 7/59 Should be 49380, not 49390 9/59 Should be 38610, not 39610 7/61 Should be 47780, not 47790 2/63 Should be 48620, not 49620 9/63 Should be 27180, not 27190 12/63 Should be 46890, not 46990 7/64 Should be 43380, not 43390 H Reactor water flow rate (Kgpm) 6/50 Should be 39.8, not 39.9 3/51 Should be 39.48, not 39.49 1/59 Should be 81.4, not 91.4 6/59 Should be 85.2, not 95.2 7/59 Should be 85.4, not 95.4 9/59 Should be 85.8, not 85.9 11/59 Should be 83.4, not 93.4 2/60 Should be 84, not 94 10/60 Should be 87.2, not 97.2 1/61 Should be 86.2, not 96.2 11/62 Should be 85.5, not 95.5 2/63 Should be 84.8, not 94.8 5/63 Should be 84.6, not 94.6 8/64 Should be 92.8, not 92.5 KE Reactor power (MW) 9/55 Should be 1068, not 1069 9/56 Should be 1811, not 1511 2/58 Should be 2877, not 2977 1/61 Should be 3802, not 3902 3/61 Should be 3818, not 3918 9/61 Should be 3508, not 3509 12/61 Should be 3870, not 3970 5/62 Should be 3988, not 3998 12/63 Should be 4148, not 4149 9/66 Should be 4188, not 4198 2/67 Should be 4380, not 4390 7/69 Should be 3758, not 3759 KE Reactor energy production (MWd) 3/56 Should be 53850, not 53950 3/57 Should be 58660, not 59660 6/58 Should be 78910, not 79910 2/60 Should be 73930, not 73936

OCR for page 26
4/61 Should be 91830, not 91930 6/61 Should be 98830, not 98930 11/62 Should be 48140, not 49140 1/63 Should be 108380, not 109390 5/63 Should be 48680, not 49680 8/63 Should be 80320, not 90320 11/63 Should be 118790, not 119790 6/64 Should be 98020, not 99020 8/64 Should be 78380, not 79380 11/64 Should be 110850, not 110950 10/65 Should be 108510, not 109510 9/66 Should be 116850, not 116950 1/67 Should be 105820, not 105920 8/70 Should be 48920, not 49920 9/70 Should be 82980, not 92990 KE Reactor water flow rate (Kgpm) 6/57 Should be 161, not 161.1 2/60 Should be 180, not 190 9/60 Should be 180, not 190 1/61 Should be 181, not 191 7/61 Should be 183.5, not 193.5 1/62 Should be 188, not 155 2/65 Should be 208, not 209 2/69 Should be 188, not 15 KW Reactor power (KW) On “Power_data” sheet the column labeled “KW Power” erroneously contains the KW reactor energy production in place of power. 3/55 Should be 861, not 961 6/58 Should be 2814, not 2914 2/60 Should be 3598, not 3599 3/61 Should be 3810, not 3910 5/61 Should be 3821, not 3921 5/63 Should be 3804, not 3904 7/65 Should be 4148, not 4149 1/67 Should be 4387, not 4397 7/68 Should be 3486, not 3496 KE Reactor energy production (MWd) 10/44 Should be 0, not 1 (corrected in “Power_data” sheet) 9/55 Should be 18730, not 19730 8/57 Should be 68530, not 69530 2/59 Should be 78600, not 79600 10/59 Should be 81790, not 91790

OCR for page 26
2/60 Should be 83110, not 93110 8/60 Should be 87050, not 97050 12/62 Should be 108160, not 109160 4/63 Should be 98930, not 99930 11/64 Should be 109380, not 109390 2/65 Should be 98490, not 99490 7/65 Should be 80110, not 90110 4/67 Should be 108770, not 109770 8/67 Should be 102080, not 102090 3/69 Should be 84340, not 94340 4/69 Should be 118590, not 119590 KW Reactor water flow rate (Kgpm) 6/59 Should be 178, not 179 9/59 Should be 178, not 179 3/60 Should be 180, not 150 Month 5/60 should be 8/60 5/60 Should be 180, not 190 6/60 Should be 180, not 150 7/60 Should be 189, not 199 10/60 Should be 181.5, not 191.5 7/61 Should be 185, not 195 1/62 Should be 185, not 195 10/63 Should be 208, not 209 The committee examined the transcription of data from Tables B.1 and B.2 of PNWD 2223 HEDR to “\reactor\release rates.xls” in much less detail. The transcription accuracy appeared to be much better than that of the transcription of Tables A.1 and A.2 to “reactor operations.xls”, but there were obvious errors, such as those: 3/55 Med Cr-51 Should be 24100, not 2410 3/46 Min Np-239 Should be 14700, not 147000 In the reactor data sheet of the “reactor operations.xls” workbook, Month 1/49 should be 8/49 Month 9/51 should be 8/51 Month 9/61 should be 8/61 Month 9/66 should be 8/66