NJOY 99.396 16 August 2013 ----------------------------------------------------------------------- This file describes updates to NJOY99. See Readme0 for more general information on NJOY99, installation, test problems, and so on. *ident up1 */ leapr -- 04apr00 */ this fixes errors that destroy the calculation of s(alpha,beta) */ for cold para-hydrogen and for both cold deuteriums. the first */ was introduced in 97.53 while making the constants symbolic, */ and the second was introduced in the cleanup process for njoy99. We found these problems this March and April. LEAPR runs for the cold hydrogen and deuterium cases made with versions earlier than 97.53 should be OK. *ident up2 */ groupr -- 04apr00 */ there is an error in the indexing for the xmas 172-group structure */ that throws all the group bounds off by one. this structure is */ used in europe for thermal reactor calculations. This one is only important for people using this group structure. */ fix a problem introduced with the activation patch of 97.102. the */ nk parameter is only used when doing mf values for activation */ products, and it is not appropriate for fission nubar values. *d groupr.3824 */ don't strip off the upscatter groups for the neutron spectra */ coming from delayed neutron emission (mt=455). */ add a missing save statement and fix an unset variable in anased. */ these problems affect delayed neutron spectra. These changes are important for anybody doing multigroup calculations for fissionable materials. *ident up3 */ acer -- 04apr00 */ the acer consistency checks include an option to readjust */ eprime values that are greater than e, when appropriate. */ there are some problems with the logic, especially for */ cases that use histogram interpolation for the distribution. This change improves the appearance of the outputs, but it doesn't actually have any effect on calculations. MCNP only uses the differential probabilities for linear-linear interpolation, and it doesn't make any difference if they are incorrect at the upper limit for histogram cases. */ when using the old format (mcnp4b and earlier), some angle-energy */ distributions from file 6 are converted into the law 67 format, */ because these earlier versions of mcnp couldn't use all the */ file 6 representations. when converting from the cm to the lab, */ the methods used in subroutine fix6 are a little crude. they get */ confused when cm energies are so small that lab cosines of -1 */ are not reached. this patch tries to fix that in a rough way, */ but evaluations that use the cm frame in file 6 will work best */ if most of the cm energies are greater than e/(awr+1)**2 for */ each incident energy e. for mcnp4c and later, the code can */ sample directly from tabulated cm representations, and the */ approximations of the fix6 routine are avoided. this patch */ is needed for one evaluation from JEFF-3. */ increase the available storage to handle the very large */ mf6/mt16 tabulation in JEFF-3 Be-9. */ increase the space available for discontinuities in convr */ to allow for JENDL-3.2 si-nat These three sets of changes are only needed for very specific cases. We already know that there will be a few more like this coming up! *ident up4 */ reconr -- 05apr00 */ be sure to count subsections of file 12 before allocating */ storage for the elements of the new directory. otherwise, */ some materials with many sections of file 12 will overflow. */ this is a longstanding problem that we never noticed before. This is another problem that only showed up for one evaluation. It has been this way in NJOY for many years. *ident up5 */ purr -- 7may00 */ fix a problem introduced while installing the heating part */ of the probability tables. it shows up when doing elements */ that have unresolved data. also, increase the number of */ resonance sections allowed to handle the very large cd-nat */ evaluation from JENDL. PURR marks regions where the unresolved range for one isotope of an element overlaps with the resolved or smooth range from other isotopes by putting a negative sign on the energy. We have to be sure to recover the energy correctly by taking the absolute value! Note that this problem only shows up for elemental evaluations. There are only two of these in ENDF, but many more in JENDL. We also had to allow for more sections in the resonance range to handle JENDL's Cd-nat evaluation. *ident up6 */ acer -- 30may00 */ fix a typo in up3 (reported by bunde, anl) */ acer -- 30may00 */ fix problems with converting cm distributions to law=7 */ and problems reading law=7 into the ace file. these problems */ show up when running newfor=0 with njoy2000, especially on */ some materials from jef-2.2. This one is pretty important. NJOY99 was pretty thoroughly tested for producing MCNP4C libraries, but we neglected to do as much checking for MCNP4B cases (newfor=0). *ident up7 */ viewr -- 30may00 */ increase the allowed size of 3d plots. */ pushed by pb-nat from jef-2.2. In most cases, NJOY just truncates 3-D plots that are too big for viewr, but in the pb-nat case, it actually overwrote some memory. The storage available for 3-D plots has now been increased, and some plots that were truncated before without causing the code to fail will plot better now. *ident up8 */ acer -- 3jun00 */ subroutine ptleg2 does not need the dynamic array xat. */ this problem was first noted by Waclaw Gudowski for ENDF/B-VI */ si-nat. it shows up as "id xat not defined". This is a fairly isolated problem. It is a case of evolution leaving behind an obsolete organ. *ident up9 */ acer -- 07jun00 */ add the capability for processing anisotropic charged particle */ emission using tabulated legendre coefficients into the */ mcnp4c law61 format. this is needed for jeff-3 cr-52. */ allow for multiple interpolation ranges in file 6. this */ also occurs for jeff-3 cr-52. currently, the neutron */ energy-angle distribution only allows for combinations of */ histogram and linear linear interpolation, but the */ charged-particle sections allow for general combinations of */ all allowed interpolation laws. There are only one or two such cases in current evaluations, but more will probably appear in the future. The limitation to interpolation laws 1 and 2 for neutrons comes from the occasional need to handle multiple subsections in File 6 for neutron emission. The increased generality for charged-particle subsections comes about because multiple subsections are not handled. Someone will probably do that someday.... Although we did work fairly hard to handle the JEFF Cr-52 evaluation, the actual representation is fairly strange, and the treatment of the particle yields near the effective emission threshold should really should be changed. *ident up10 */ leapr -- 13jun00 */ fix two incorrect constants in leapr. one affects cases with */ diffusive effects, and it has been incorrect since njoy97.0 */ (oct 97). the other affects cold hydrogen calculations, and it */ has been incorrect since njoy97.53 (dec98). NJOY99 was exhaustively checked using ENDF/B evaluations and the newfor=1 option (to support MCNP4C). Unfortunately, we didn't test it with newfor=0 (for MCNP4B and earlier versions)! Several users have reported problems (especially Waclow Gudowski in Sweden. Some changes that affect running with newfor=0 were made above. The following changes (up11) are especially noticable with ENDF/B-VI Release 6. *ident up11 */ acer -- 26jun00 */ fix an error in determinining which reactions have to be */ converted into law=7 format when using newfor=0. because of */ overzealous code cleanup, acer is trying to convert sections */ with the kalbach representation in addition to sections with */ tabulated angular distributions. */ acer -- 26jun00 */ during the cleanup of the topfil routine, the logic to process */ sections of file 6 using law=2 (two-body distributions) into */ equally probable bins for newfor=0 was omitted. this shows up */ for evaluations that use mf6/mt51, etc., to represent inelastic */ levels. */ acer -- 26jun00 */ fix an error in the reaction naming. it affects mt=44 (n,n2p) and */ mt=45 (n,npa). this problem was introduced in may of 1995. */ examples of cases that use these reactions are later releases of */ endf/b-vi al-27. */ acer -- 26jun00 */ add missing external statement. reported by bokyun seo (kaeri) */ acer -- 26jun00 */ add missing line for the sequential (n,2n) reactions for be-9. */ this line was accidentally removed in njoy 94.19 (jan96). the */ error was continued through njoy 97 and 99. discovered by */ bob little (lanl). */ acer -- 26jun00 */ fix anisotropic photon production (e.g., endf c,n,o) *ident up12 */ njoy -- 27jul00 */ fix two typographical errors in the 64-bit version of the */ slatec math library. reported by piet de leege (delft). *ident up14 */ acer -- 20jul00 */ acer fails if you run it on a pendf tape that only has the */ single reaction mt=2 (elastic). this can happen for he-4 if you */ don't run heatr, thermr, or gaspr first. found by gudowski. */ acer -- 20jul00 */ acer fails for mf=6, law=2, lang>0 (angular distribution with */ tabulated cosines). the only known example is mt=51 for pb-208 from */ release 6 of endf/b-vi. found by waclaw gudowski. */ acer -- 21jul00 */ an error was included in up9, which added a capability to handle */ anisotropic charged-particle emission represented using legendre */ polynomials. the update disabled the case of isotropic */ charged-particle emission, which occurs in a number of important */ materials from release 6 of endf/b-vi. the symptom is a serious */ clobbering of the ace file, such that it cannot even be read into */ mcnp or even printed using acer. also reported by gudowski. *ident up15 */ heatr -- 31jul00 */ incorrect initial value found by m.mattes (u.stuttgart). */ increase the allowed number of legendre terms in h6ddx */ to handle the new jeff-3t fe-56 evaluation. */ watch for ill-defined vertical segments in distributions. these */ have been seen in zr90 from cendl3 and fe56 from jeff3. actually, */ the evaluations should be fixed to avoid such features, because */ we don't really know what y value to select in the vertical */ segment. we choose to just move the second energy of the double */ point up a little. we only print the diagnostic once to keep the */ output cleaner, but there could be more than one vertical segment. *ident up16 */ groupr -- 31jul00 */ watch for ill-defined vertical segments in distributions. these */ have been seen in zr90 from cendl3 and fe56 from jeff3. actually, */ the evaluations should be fixed to avoid such features, because */ we don't really know what y value to select in the vertical */ segment. we choose to just move the second energy of the double */ point up a little. we only print the diagnostic once to keep the */ output cleaner, but there could be more than one vertical segment. *ident up17 */ acer -- 31jul00 */ fix an incorrect index in the law=61 section for the primary */ particle. the effect of this error is to give an incorrect */ angular distribution for the energy points with scattering */ probability zero (which should be isotropic). this change */ is strictly cosmetic and shouldn't affect any results. */ acer -- 31jul00 */ we want to use the compact law=4 for isotropic charged particle */ distributions, and the more general law=61 for anisotropic cp */ distributions. unfortunately, we can't tell which is which */ without reading past the first few energies for some evaluations. */ the need for this patch was first noticed by jeff3 fe56. *ident up18 */ acer -- 03aug00 */ some jef, eff, and jeff evaluations contain a redundant reaction */ mt=10 that gives the continuum neutron production. it is */ necessary to exclude this reaction from the reconstructed total */ cross section and to omit the associated energy-angle distribution. */ otherwise, the continuum neutron production will be counted twice. */ acer -- 03aug00 */ this change can fix an infinite loop during acer plotting */ acer -- 03aug00 */ for file 6 sections with only one subsection, the mt number is */ set to zero for the messages about energy-dependent yields. */ this is a trivial cosmetic patch and doesn't affect results. */ acer -- 03aug00 */ this change is needed to handle nubar for jendl-3.2 u-235. it */ was originally made at los alamos in april, but somehow didn't */ make it to the official update file. *ident up19 */ heatr -- 03aug00 */ as noted above, some jef, eff, and jeff evaluations use the */ redundant mt=10. this value needs to be excluded from the */ heating and damage calculations. *ident up20 */ acer -- 16aug00 */ there is an error in processing angular distributions using ltt=3 */ when newfor=0 (mcnp4b compatibility). the extra tosend causes the */ code to skip over the first reaction after the elastic mf=4. */ this leads to a bad tyr=0 value in the ace file, which causes */ mcnp to issue a confusing error message about "sabcol," even */ when s(alpha,beta) is not being used. this problem occurs only */ when processing the 150-mev evalutions from endf/b-vi.6. it */ is probably best to use release 5 with mcnp4b anyway. the */ release 5 and 6 data are identical below 20 mev. It should be reiterated that Release 6 of ENDF/B-VI differs from Release 5 only in the addition of data from 20 to 150 MeV. People who only need the conventional 20-MeV data, or people processing for MCNP4B, may want to run from Release 5 and avoid the additional complexity and bulk in the files that are produced from Release 6. */ acer -- 16aug00 */ the code is finding the wrong value for the lct parameter (lab */ or cm frame) when processing file 4 angular distributions if */ the section is fully isotropic. this can result in an incorrect */ value for the ace tyr parameter, which can result in an apparent */ error from sabcol, even with no s(alpha,beta) data in the problem. */ this problem was introduced while the njoy97 coding was being */ converted to block structuring. */ acer -- 16aug00 */ the consistency check for incorrect reference frame should take */ place for isotropic distributions also. sometimes, this check is */ not a real error. users should check the evaluation to see if the */ reference frame is really as intended by the evaluator. as fixed, */ this check would have found the two problems above! */ acer-- 16aug00 */ we are not currently handling law=5 for energy distributions. */ this occurs for u-233 fission from jef-2.2. the evaluation can */ be patched by converting the lf=5 part of the distribution to */ lf=1, which is sampled much better by mcnp using cummulative */ probability distributions anyway. */ acer -- 16aug00 */ there are some additional places where skip6 should be skip6a. */ see up6 above. this shows up for endf be-9 with newfor=1. */ acer -- 16aug00 */ there is an error in the law=7 part of up11. this shows up */ when processing endf be-9 using newfor=0 */ acer -- 16aug00 */ there is an error in the new skip6a routine introduced by up6 */ that shows up when processing sections with law=7 with newfor=0. */ acer -- 16aug00 */ missing initialization in ptlegc (this could affect incident */ charged particles on some systems). NJOY99.20 has been tested rather extensively with MCNP4B and MCNP4C. It has successfully processed all the materials in ENDF/B-VI Release 5 (which has Emax=20 MeV), many important materials from JENDL-3.2, and a few problem materials from JEF and JEFF libraries into ACE formats for both 4B and 4C. All of these ACE files have been staged into MCNP and run for a few cycles to make sure that the format is reasonably correct. It has also been used recently for high-energy and charged-particle processing, but without testing in MCNP or MCNPX as yet. A few problems are known, and several materials had problems with the evaluation (e.g., carbon). *ident up21 */ acer -- 28sep00 */ the self-shielded cross sections are not being printed out */ correctly for the reactions, but total is ok. the gendf */ file is ok, so libraries made with njoy99 are ok. *ident up22 */ reconr -- 28sep00 */ add capability to handle the new extension to the reich-moore */ resonance format that uses the sign of aj to designate which */ channel spin to use for a particular resonance. based on */ coding provided by nancy larson, ornl. This is a new ENDF format approved at the 1999 CSEWG meeting. There are no official evaluations that use it as yet, but it is felt to be important for Al-27 and Si-29. We are adding it to NJOY now so as to be ready when the new evaluations appear from ORNL. *ident up23 */ gaminr -- 28sep00 */ allow for up to 400 groups (added by request) *ident up24 */ dtfr -- 28sep00 */ allow for up to 400 groups (added by request) *ident up25 */ groupr -- 11oct00 */ fix the section that reduces the number of sig figs in getdis. */ it was only acting on the in-group probabilities. this helps */ to make the results for elastic and discrete inelastic matrices */ the same on different machines. the basic idea is that these */ numbers are obtained by subtraction of numbers on the order of */ unity, so any results less than about 1e-7 are just random */ numbers and can be removed. */ groupr -- 12oct00 */ change the size of common groupg to agree with the changes */ made in gaminr above. *ident up26 */ acer -- 12oct00 */ the current coding sometimes gets the threshold for charged */ particle production off by one point. *ident up27 */ dtfr -- 27oct00 */ fix problem with finding right material and temperature */ on the pendf tape. the goto loop was not translated correctly! */ dtfr -- 27oct00 */ fix error made in up24 *ident up28 */ acer -- 05nov00 */ the pointer into the a array is not being correctly incremented */ for the "call moreio" line. this only affects the new JEFF */ evaluation for beryllium, which has exceptionally detailed */ angulur tabulations. found by fischer (fzk). */ fix an indexing error in adjusting the normalization and */ precision for the pdf of angular distributions for law67 charged */ particle production that causes the pdf to be the same as the */ cdf. this problem shows up for beryllium (n,2n) alpha production */ in endf/b-vi, for example. identified by konno (jaeri). *ident up29 */ ccccr -- 05nov00 */ the pointer in the e array for moreio is wrong. the result of */ this is that larger group structures cannot be handled correctly */ for delayed neutrons. found by broeders (karlsruhe). *ident up30 */ heatr -- 05nov00 */ the insert of the data value for nlmax was incorrectly done into */ the "sw" conditional block instead of after the conditional */ block was complete. thus, it was only available to 32-bit */ versions of the code. this was discovered by deleege (delft) */ when running in 64-bit mode on a vax/alpha. *ident up31 */ groupr -- 05nov00 */ fix two problems with the ltt3 option for 150 mev evaluations. */ the incorrect index for the c array leads to findex problems */ caused by clobbering the index for the dynamic storage system. */ you also have to make sure that the "over" option that allows */ getfle to extrapolate to energies slightly higher than the */ upper limit of the table doesn't act at the break between */ the two energy ranges with ltt3. this problem was reported */ by wienke (sck-cen). *ident up32 */ reconr -- 05nov00 */ some fission products from the jendl-3.2 library include */ an unresolved resonance range with no corresponding resolved */ range. trkov (iaea) proposed the following fix. *ident up33 */ gaminr -- 18jan01 */ the photoatomic group cross sections are not printed out */ correctly for a p-order greater than 5. *ident up34 */ groupr -- 29jan01 */ need more storage in groupr to handle mt=91 for am243 from */ endf/b-vi release 5, which goes to 30 mev. the symptom was */ "storage exceeded" from cm2lab. *ident up35 */ groupr -- 08feb01 */ when we increased the common block for photon group structures */ to allow as many as 400 groups (see up25), we forgot to update */ the parameter ngmax. this causes a "too many groups" error if */ you run with more than 150 gamma groups. *ident up36 */ acer -- 08feb01 */ in up17, we checked for isotropic distributions in order to use */ a more compact presentation. the logic misses one special case, */ namely, pb208 from endf/b-vi release 6. *ident up37 */ reconr -- 09feb01 */ all through njoy, we have been using 1e10 ev as our idea of */ an infinite energy. progress happens, and red cullen at llnl */ is putting out an endf version of the evaluated photon data */ library (epdl97), which contains data to 100 gev. the following */ change prevents reconr from going into an infinite loop in the */ emerge routine with 100 gev data. This and the following related changes will only become important with Release 8 of ENDF/B-VI, which will include the new 100 GeV data. *ident up38 */ njoy -- 09feb01 */ keep on increasing infinity for the 100 gev data. the routines */ gety1, gety2, and terpa return an "infinite" energy at the end */ of the table, and we now increase that to 1e12 ev. this doesn't */ seem to cause any problems in njoy modules (such as groupr) that */ still check for return values of 1e10 or more; all the standard */ test problems still work fine. *ident up39 */ gaminr -- 09feb01 */ keep on increasing infinity for the 100 gev data. *ident up40 */ acer -- 23mar01 */ due to a bad if clause, the contribution to heating from charged */ particles is not being included for mf=6, law 3 or 4. this was */ noticed in the run for endf/b-vi be-9 by lanl/x-5. the errors in */ this particular case are quite small because of the small cross */ sections for charged-particle emission. this error will only */ effect mcnpx calculations for coupled neutron-proton transport. The MCNPX error here is extremely small because of the very small cross section for proton emission. *ident up41 */ acer -- 27mar01 */ the value "nr = 0", implying linear interpolation over all points, */ is not printed on the acer output listing for two cases, as reported */ by lanl/x-5. these errors do not affect mcnp results, but the */ repair makes the printout for photon yields and energy distributions */ match those for other types of data. *ident up42 */ purr -- 27mar01 */ remove the timers that are given as each ladder is processed */ in order to reduce the number of diffs that show up when */ succesive runs are checked for qa purposes using the same */ sequence of random numbers. for lanl/x-5. Remember that runs using PURR will always show lots of diffs between different machines because the sequence of random numbers will by different. We intend to fix this in NJOY2001. *ident up43 */ heatr -- 27mar01 */ the roundup applied to the first energy grid point should */ be smaller now that we are routinely working with 7-digit */ energies. the effect if this in current files is that the */ first energy in any of the heating and damage reactions is */ a little larger than the normal 1e-5. this shows up as a */ zero heating or damage value for the first point in the mcnp */ ace files, which is strange looking, but of little significant */ impact on real calculations. reported by lanl/x-5. *ident up44 */ acer -- 29mar01 */ lanl/x-5 has requested that the main container array be increased */ in size to allow bigger ace files to be generated. it is also */ necessary to increase the i7 length field on the xsdir cards to i8 */ to accomodate the larger ace files. *ident up45 */ acer -- 08apr01 */ as discovered by jean christophe sublet, sun forte6 f95 is */ finiky about opening a scratch file that is already open, */ although all other compilers used for njoy thus far were more */ accepting. we just have to be careful to close a unit used */ as a scratch file before reusing the unit for another purpose. This one shows up as an error trying to open tape13 in ACER. So far, it has only be observed with Sun Forte6 f95. *ident up46 */ gaspr -- 09apr01 */ close another scratch unit. This one shows up as an error trying to open tape10 in ACER if GASPR has been run. *ident up47 */ acer -- 09apr01 */ the length published for thermal data files is too long by one */ for cases including incoherent elastic scattering. for endf, */ this is poly, h(zrh), and cold solid methane. discovered by */ roberto orsi (enea-bologna). This error only affects thermal ACE tables for hydrogenous solids where the true data-block length is divisible by 4. *ident up48 */ acer -- 09apr01 */ the landh parameter should be zero (not -1) for isotropic */ subsections of mf=6 described using law=3. this occurs for */ the reactions (n,p0) through (n,a0) in be-9 from endf/b-vi. */ Noted by bob little (lanl/x-5). This error only affects runs with MCNPX. The symptom is a cosine > 1.0. *ident up49 */ acer -- 09apr01 */ the representation for ace law3/33 should use -q instead of */ abs(q) in order to handle two-body reactions for isomeric */ targets. this change in the ace specifications was recommended */ by bob little (lanl/x-5) after a query by waclaw gudowski. it */ only affects a small number of evaluations. *ident up50 */ acer -- 09apr01 */ most of the jendl photonuclear evaluations currently available */ from http://iaeand.iaea.or.at/photonuclear/ crash with an i/o */ error because they use a non-conforming format where mf=6, */ mt=201-27 are used to represent particle production. we are */ providing a clearer error message for the user's convenience. */ these evaluations cannot be used in njoy or mcnpx in their */ current form. *ident up51 */ acer -- 12apr01 */ add a capability to handle a two-body recoil subsection of mf=6 */ for photonuclear files. this may be useful for representing the */ photodisintegration of the deuteron with full distributions for */ both neutron and proton. we tested the patch using a modified */ version of the g+2H evaluation from JENDL. This patch is also needed for the preliminary LANL evaluation for the photodisintegration of the deuteron based on EDA R-matrix calculations. *ident up52 */ acer -- 13apr01 */ add a capability to handle tabulated sections of File 5 (lf=1) */ for photonuclear files. Such sections are used in the Russian */ evaluations for three isotopes of plutonium included in the */ iaea photonuclear compilation. this also fixes a bug in the */ storage of fission nubar. the first point for energy distributions */ often has a nonrealistic sharp triangle given for the spectrum. */ this can cause problems with the vertical scale for plots */ because the emission probabilities get very large for small */ ranges of secondary energy. therefore, we ignore the first */ incident energy in determining the vertical scale for the plot. The original Russian evaluations of photonuclear reactions on the actinides had a few formatting problems. These were fixed, and the modified files were referred back to the evaluators for comments. With those formatting changes and this patch to handle the fission spectra, all of the BOFOD photonuclear evaluations can be processed. *ident up53 */ groupr -- 11jun01 */ if the file6 distribution is fully isotropic (law=3), the getfle */ routine doesn't realize that when doing a discrete recoil (law=4). */ we create a special flag of law=-4 to pass the fact of isotropy */ into getfle. this problem only affects runs that compute a */ transfer matrix for the recoil particle when the first particle */ emitted is given as totally isotropic (for example, mt=701 for */ endf be-9). the error message is "desired energy above highest */ given." found by dieter leichtle (fzk). *ident up54 */ reconr -- 12jun01 */ allow for the series of mt numbers 875-891 that can be used */ to represent different levels of the (n,2n) reaction in the */ same way that 600-649 are used to represent different levels */ of the (n,p) reaction. the code expects that mf=3/mt=16 */ contains the sum of mt=875 through 891 in the same way that */ mt=103 contains the sum of 600-649. this representation is */ used for be-9 for eff-3.1 and jeff-3.0. A complex new evaluation for Be-9 is being prepared for the JEFF library, and it uses a new representation for the parts of the (n,2n) reaction. A large number of changes to NJOY are required to handle this new evaluation, and additional patches will be found below. *ident up55 */ heatr -- 12jun01 */ if mt=875-891 appears in the file, mt=16 is redundant. this */ is analogous to the way mt=107 is redundant if mt=800-850 */ is present. */ the integration over secondary energy for law 7 in getsix */ must allow for histogram interpolation as used in be-9 */ from eff-3.1. the effect of this is to get especially */ bad particle energies for the discrete neutron in mt=876. *ident up56 */ acer 12jun01 */ the angle-energy law in file 6 is always causing trouble. */ it is especially difficult when more than one subsection */ is used to describe the emission for a particle, because */ an overall angular distribution for the reaction must be */ contructed. with the new formats, it is easy to eliminate */ law=7 sections by converting them to law=1 with tabulated */ angular distributions. This particular representation is currently only used for Be-9 evaluations. For each incident energy, it gives secondary energy spectra at various emission angles in the lab system. With this patch, we convert these distributions into ones that give angular distributions at various values of the secondary energy, which can be represented by the new ACE LAW=61. This makes it easy to handle one current case where multiple distributions for neutron emission are given in File 6. However, it could be argued that this change violates the intent of the evaluator. We may explore other approaches to handling these representations in the future. */ add support for law1 with tabulated angular distributions */ to the charged-particle section. The capabilities for handling incident charged particles are gradually evolving as new evaluations get tested. *ident up57 */ acer -- 12jun01 */ changes to acer needed to support the eff-3.1/jeff-3.0 */ representation for be-9. we have to allow for the series */ of mt numbers 875-891. if present, mt=16 is redundant and */ must appear after the distributions in the reaction list. */ the section for mt=876 has two subsections for neutron */ emission. this problem is handled by the previous update. */ add the new reaction names to the mtname routine */ need more storage for eff-3.1 be-9 This new evaluation is very large! *ident up58 */ broadr -- 10jul01 */ increase the storage area in broadr to reduce paging and make */ comparisons between njoy99 and njoy2001 easier. there will */ normally be a small difference in the grids produced by broadr */ each time paging takes place, and this makes it hard to compare */ files using diff. In BROADR, the large sections of cross-section versus energy are paged through the Doppler broadening process. In NJOY2001, these pages breaks occur at different energies than in NJOY99. This leads to small differences in multigroup cross sections near these breakpoints that show up in diff runs. More importantly, it changes the energy list in ACE listings so that diff gets lost after the breakpoint. This makes it hard to compare results from the two codes for QA purposes. Now that everyone has lots of memory, it is practical to increase the amount of storage allocated to BROADR to reduce the number of materials that show these breaks. Some materials with very large numbers of energy points will still not compare properly using diff. *ident up59 */ acer -- 20jul01 */ in charged-particle emission, the first point for energy */ distributions often has a nonrealistic sharp triangle given for */ the spectrum. this can cause problems with the vertical scale */ incident energy in determining the vertical scale for the plot. We use this same trick other places. Without it, plots will sometimes just show the first big peak and the more interesting data at higher energies will be pushed below the floor of the plot. *ident up60 */ reconr -- 24sep01 */ occasionally, reactions are given with a nonzero cross section */ at threshold, even though this violates endf procedures. reconr */ had some logic for handling this that was being overwritten by */ another change. we fix it here by inserting an extra energy */ point just above the threshold and zeroing the cross section at */ the threshold. a diagnostic message is provided. one example */ of a place were this occurs is gd158 from endf/b-vi. reported */ by frankle (lanl). */ reconr -- 24sep01 */ reconr contains some logic that tries to avoid doing work on */ very small charged-particle cross sections by defining a */ "pseudo-threshold" when the cross section rises to more than */ 1e-15 barns. however, this scheme isn't carried out completely, */ and it only results in the omission of the threshold energy for */ reactions that have less than this cross section just above */ the threshold. this effect shows up for the (n,n't) reaction */ mt=33 for cd-110 from endf/b-vi.4. at the request of bob */ little (lanl), we are changing the constant "ssmall" that */ triggers this effect to a smaller number. in the long term, */ we should reconsider this logic. *ident up61 */ acer -- 25sep01 */ kisako kazuaki (sumimoto) has observed that the common variable */ coeff in eval is not set. actually, eval is not really used in */ tabize anymore. it is just leftover as an intialization for a(iy). */ the only other place it is used is in islin2, and islin2 is not */ called anymore! This update removes these leftover remnants. */ kazuaki also noticed that the photoatomic heating value was */ being stored in ev instead of mev and that the atomic number */ aw0 was not being set. Note that the photoatomic processing in ACER is still incomplete. When this was written, we had hoped that there might be an upgrade to MCNP to handle fluorescence in a more complete way. It is also important to make sure that the fluorescence energies produced are consistent with the photoelectric edge energies. This is because these photons are properly produced with energies just below the edges where cross sections are low; thus they can travel further in the material. As a result, it is impractical to use some of the older sets of fluorescence data with newer cross sections. New data for atomic relaxations are now becoming available in ENDF format, and we hope to fix this problem in the future. *ident up62 */ purr -- 28sep01 */ lanl group x-5 has noted that the conditional heating cross */ section in the mcnp probability tables is not quite what they */ expected. we change the calculation here to get results that */ are consistently given as eV/reaction for lssf=0 and fluctuation */ factors for lssf=1. People interested in self-shielded values for heating in the unresolved range should pay attential to this patch. *ident up63 */ acer -- 15oct01 */ add a capability to include delayed neutron data in the ace */ file as allowed for mcnp4c. With the release of MCNP4c, the code now can generate the spectrum of neutrons from fission accurately. Previous versions used the prompt spectrum for all of the fission neutrons. Now, the prompt spectrum is used with prompt nubar, and the delayed spectrum is used with the delayed nubar. We have tested the effect of this for some fast critical assemblies. It is very minor except for BIGTEN, where it causes a decrease in k-eff of 0.15 to 0.2 percent. In addition, the new code can do time dependent calculations involving delayed neutrons. This patch provides the new data blocks in the ACE format to support these capabilities. It would be important for people who need to move beyond the data provided in the ENDF61 library included with MCNP4c. *ident up64 */ acer -- 07dec01 */ there was a mistake introduced with up57 that only affects */ 64-bit compiles (when "*set sw" is not used). In addition, there */ were three mistakes related to the delayed neutron patch of up63. */ it's bad practice to use different names in common blocks in */ different places! *ident up65 */ groupr -- 10dec01 */ fix groupr to handle radionuclide production using the new */ endf file 8 and data from eaf2001. *ident up66 */ groupr -- 12feb02 */ if you mix automatic reactions with manual reactions where */ the mtname string is not given, the mtname on the manual */ reaction will have whatever string was left from the previous */ case. we fix that here. This only affects output listings for some special cases. */ groupr -- 13feb02 */ there is an error in the calculation of the kalbach a factor */ for the photonuclear case. it is necessary to convert e to */ mev for this formula. the symptom is results for a that are */ so large that sinh(a) overflows with a floating point error. This one is important for people using multigroup methods with photonuclear evaluations. *ident up67 */ matxsr -- 12feb02 */ add photonuclear capability With this photonuclear capability in MATXS files, it is possible to use the TRANSX code to generate full coupled sets for n-gamma transport. *ident up68 */ reconr -- 18apr02 */ the name "pi" is being used for both the imaginary part of */ p and the pi constant. the former is changed to "pim" as */ used in the version of this subroutine in purr. the symptom */ is a floating point error when sqrt(pi) is calculated with */ pi changed to a negative value. */ allow for more digits in the temperature printout to */ handle the usage of reconr being made for eaf-2001. the */ temperature field is not being used for resonance */ reconstruction but only passed to the output file for */ later use. *ident up69 */ acer -- 07dec01 */ add a capability to generate fluorescence data for mcnp using */ the existing cashwell-everett format with new numbers obtained */ from the endf versions of eadl and epdl. the data produced with */ this method should give reasonable results for transport and */ heating for energies above 1 kev. the new evaluations allow */ for lower incident photon energies and for more detail in photon */ and electron distributions from photoabsorption, and future */ versions of njoy and mcnp can take advantage of this. *ident up70 */ acer -- 20feb02 */ add consistency checks for delayed neutrons */ add plots for nubar and delayed neutron spectra *ident up71 */ reconr -- 25aug02 */ if the cross section or yield in a section is zero at all */ energies, the union grid is spoiled. we fix lunion so the */ so called pseudo-threshold search cannot go past one less */ than the number of points in the section. this problem was */ seen in n-15 of jendl-3.3 for 12/104 and 12/105. *ident up72 */ acer -- 3sep02 */ allow for up to 8 groups of delayed neutrons as used in some */ of the materials in jeff-3. */ there were some small problems with the use of endf law=1 in */ addition to the more common law=5. we also have to be sure to */ renormalize distributions that don't have the proper normalization. */ watch for a special case in plotting *ident up73 */ groupr -- 3sep02 */ allow for up to 8 groups of delayed neutrons as used in some */ materials of jeff-3. Up72 and up73 are only a beginning. It will also be necessary to modify CCCCR to produce a more complicated DLAYXS file. There is also current interest in providing energy-dependent time constants for delayed neutrons, but it is not clear how to pass this information on to the applications. *ident up74 */ groupr -- 03sep02 */ fix editing error in the lwr epri weight function. */ reported by skip kahler (bechtel bettis). */ fix an incorrect boolean statement in the removal of upscatter */ in subroutine getsed. reported by kazuaki (sae, japan). *ident up75 */ acer -- 09oct02 */ fix some typographical errors in recent updates *ident up76 */ plotr -- 09oct02 */ allow for more energy resolutions in plotr output */ for looking at the details of resonance reconstruction *ident up77 */ plotr -- 11oct02 */ add a capability to plot percent difference or ratios This capability has been available in the online Nuclear Data Viewer for some time, but having it available for normal NJOY runs increases flexibility. It should be noted that the plotter does thinning on the data it plots, so it is not so good for looking at envelopes of differences (peak values) over wide energy ranges with lots of resonances. The ratio capability is handy for plotting fission ratios. *ident up78 */ broadr -- 10oct02 */ tighten up the tolerances for integral thinning a bit */ broadr -- 10oct02 */ fix some problems related to keeping computed cross sections */ on printable 7- or 9-digit grids. this was discovered when */ doing testing at 0.01% for u235 above 1 kev, and it doesn't */ usually make itself evident for easier cases. This problem tended to produce small artifacts with widths of a few digits in the ninth place and heights on the order of .05% while making high precision code comparisons. The integrated effect of these artifacts was very small, so this problem shouldn't require major reprocessing of older libraries. These changes together with some other changes in Red Cullen's prepro codes have enable us to get very accurate agreement at the .01% level with nice plots of the differences showing only the effects of different choices of the energy grid. Integral thinning was supressed for these comparisons. *ident up79 */ reconr -- 10oct02 */ tighten up the default tolerances for integral thinning a bit The default values of the reconstruction tolerances cause the code to thin out some of the regions with small contributions to the resonance integrals in order to reduce the total size of the reconstructed file. Thus, the percent error in the cross sections can be larger than expected for very small cross sections, such as those in interference minima or for very small cross sections between resonances. This thinning normally has very small effects on observables, but users should do their own tests to be sure that the setting are OK for their applications. */ reconr -- 10oct02 */ make some improvements in grid generation to make sure that */ all cross sections are computed on printable 7- or 9-digit */ energies and that the original nodes for resonance reconstruction */ are reasonable. these changes were made to remove some small */ artifacts on the order of 0.05% above 1 kev for some materials. */ the printout for errors caused by significant figure reduction */ are obsolete now that we can go to 9 digits when necessary. */ we remove them and add a column for the error introduced into */ the fission integral by integral thinning. *ident up80 */ acer -- 10dec02 */ fix the photoatomic energy grid to include the discontinuities */ at the photo edges. reported by morgan white (lanl x-5). */ fix the values of the production cross section and heating */ at the particle production thresholds. */ modify the acer listing to allow 6-digit energy indexes with */ the first column always blank. this allows for materials with */ more than 99999 energy points. */ the global value of ntr is being overwritten when charged particle */ production is present. allow both fission flags. this only */ affects the plots. it showed up when processing a high-energy */ u-238 case with proton production by causing a bad plot of capture */ and fission resonance contours. *ident up81 */ moder -- 2dec02 */ fix up the processing of the covariance sections to match */ the endf-6 format specifications. This patch is needed to use MODER on JENDL-3.3 files, which have the more advanced version of the format for resonance-parameter covariances. However, NJOY cannot now process these covariances. *ident up82 */ acer -- 28jan03 */ the heating value in the photoatomic ace file is supposed to be */ used in an "f6" tally with the computed photon flux to compute */ the energy deposition from electrons, positrons, and atomic */ recoils. the original heating value was designed to be used */ with an explicit calculation of heating during a coupled */ electron-photon calculation, and it only included the atomic */ recoil. this update adds in the other parts of the heating. */ it was also necessary to modify the quadrature in iheat to */ handle the very high incident energies now given in endf. Most of this work was provided by Morgan White (LANL X-5), and the results are included in the photoatomic libraries provided for MCNP5. *ident up83 */ plotr -- 12feb03 */ fixes an error in plotr that prevents it from printing out */ negative values of percent differences. zero percent */ difference confuses the thinning logic, so we add a small */ amount to all the percent differences. Plots of percent difference are very useful in comparing two evaluations, and this patch makes such plots work properly for more cases. *ident up84 */ purr -- 30sep02 */ change the sampling scheme to use a different set of total cross */ section bins for each temperature. change the binning logic to */ use approximately equally probable bins over most of the range */ with two bins of lower probabilities on the low and high ends. When the methods implemented in PURR were originally developed, we were worried about representing the effects of temperature correlation. That is, if a neutron is traveling through one region and the unresolved table assigns it a low cross section, it should also see a low cross section when it enters another region at a different temperature. However, the method we used is faulty, leading to different results for the tables depending on the starting temperature. This showed up when developing ENDF66, which used a starting temperature of 77K for some materials. This patch fixes that problem, but the arguments about temperature correlation are not as strong as we would like. We also changed the binning strategy to try to improve the sampling of small cross sections in the interference minima of the elastic reaction, but it is very difficult to get enough samples into these minima. This can show up at low sigma-zero values in the Bondarendo integrals. In addition, the change in the binning increases the run time. Users may want to decrease the number of ladders calculated in order to bring the run time back to the previous level. *ident up85 */ heatr -- 25feb02 */ add mt=442 to hold the total photon production ev-barns This option is provided to support a planned modification to MCNP that will allow the heating values to be corrected for the local deposition of photon energy when users don't want to do a complete coupled neutron-photon-heating calculation. Of course, current materials that don't have photon production in the evaluation already have the photon energy deposited locally in ACE files. *ident up86 */ leapr -- 10dec02 */ add the fcc lattices for aluminum and lead to the */ coherent scattering option. This work was done in support of cold-moderator work at the Lujan neutron scattering center. We prepared evaluations for cold aluminum and iron that seemed to perform fairly well. Most NJOY users won't have to worry about these changes. *ident up87 */ acer -- 2jul03 */ the mt numbers used for determining charged particle production */ have some errors. this affects some light-isotope runs for */ incident charged particles. */ acer -- 02jul03 */ for the particle production sections, the acer logic fails for */ some cases where two identical particles are produced by */ file 6, such as p+t->alpha+alpha. the acer job seems to run */ ok, but it fails when the file is read back in for checking */ and plotting. this patch fixes the problem. this problem */ only shows up for a few light-isotope cases with incident */ charged particles. */ acer -- 02jul03 */ patch the phase-space option for the primary particle. this */ problem showed up for the proton distribution for 3He(p,2p). These problems showed up when processing new light-isotope charged-particle evaluations from Gerry Hale (LANL) that have been proposed for EMDF/B-VII. The changes will be of interest to some users in the future when MCNP and MCNPX gain the capability to use these files. *ident up88 */ matxsr -- 28jul03 */ the update that introduced a photonuclear capability for */ matxsr on 12feb02 accidently removed the assignment of the */ scratch file used for computing the self-shielding delta */ cross section values. any matxs files that include temperature */ and sigma-zero data and that were generated with versions of */ njoy99 from 99.67 up are incorrect. */ the read statement that brings in the group structures has a */ scale factor in it by mistake. this distorts group energies */ that are written on the gendf tape in f format. the result is */ an incorrect group structure on the matxs file. this error */ has existed for many years without causing problems. The first of these changes is very important for MATXS libraries, and any users that generated such libraries recently should check on this problem. The self-shielding effect will be incorrect. The second change doesn't seem to cause problems with TRANSX, because it doesn't use the group bounds in its calculations. Other users of the MATXS format should check on this one. *ident up89 */ groupr -- 29jul03 */ increase the main container array to allow for up to p5 for */ the lanl u233 evaluation proposed for endf/b-vii. */ increase the number of gammas allowed to handle w-182 and */ w-186 from endf/b-vi release 8. As the complexity of evaluations increases, we are always having to make changes like these! *ident up90 */ purr -- 6aug03 */ the purr module is zeroing out the lssf and intunr flags in */ the section mt=152 that contains the bondarenko self-shielding */ information. this has the dramatic effect giving different */ infinitely dilute cross sections when nsigz is one or greater */ than one, and incorrectly interpolated values. note that purr */ is substituting its bondarenko values for the ones from unresr */ if both unresr and purr are run in that order. these errors */ can have important effects on multigroup results from groupr. Older versions of NJOY used the output from RECONR when computing multigroup self-shielded cross sections. More recently, GROUPR has been using the results from PURR for this. Unfortunately, PURR was not passing on two flags that are important for this processing. People who have used recent versions of NJOY to prepare self-shielded multigroup cross sections should check for problems in the unresolved energy range. *ident up91 */ purr -- 06dec03 */ alain hebert noticed that the normalization of the sampled */ bondarenko tables to match the infinitely dilute cross sections */ was being done incorrectly and provided this fix. *d purr.2390 do i=nsig0,1,-1 *ident up92 */ reconr -- 10dec03 */ reuven perel noticed that zero=0 needed to be added to subroutines */ rdf2aa and rdf2hy in reconr. *ident up93 */ thermr -- 01mar04 */ allow for higher incident energies in calcem. this also requires */ some tightening up of the calculation near e'=e. also, add */ some energies near the zrh einstein oscillator. It is important to match the major features of the thermal scattering data with energy points, and these new grids do that in addition to making incident energy interpolation easier. With current methods, there is normally a break from scattering treated with S(alpha,beta,T) to a free scattering law at some point. This leads to discontinuities in the scattering cross section and the emission spectra. Current MCNP libraries break at 4.5 eV. The size and impact of these discontinuities can be reduced a bit by going to higher breakpoint energies. This set allows breakpoints as high as 10 eV. The size of the discontinuity goes as 1/E for the coherent elastic component, and similar dependences are seen for thermal inelastic scattering, so one would have to go very high energies really elimate the effect! *ident up94 */ acer -- 01mar04 */ allow for up to 1024 secondary energies for thermal scattering. This was done to allow for convergence studies on the MCNP discrete sampling method, but grids this large are probably not really useful in practice. Using future improved sampling schemes would be better. */ acer -- 15mar04 */ increase storage available Evaluations get larger and larger! *ident up95 */ viewr -- 20sep04 */ make symbol size scale with character size in legend. */ fix viewr to handle colors of filled symbols correctly. *ident up96 */ acer -- 30mar05 */ correct treatment of delayed neutrons to handle jeff-3.1t This patch is important for people working with the new JEFF-3.1 evaluations. They use 8 time bins rather than the traditional six, and they also use some different tabulation laws and interpolation methods. *ident up97 */ thermr -- 11apr05 */ changes recommended by m.mattes, ike-stuttgart */ increase array size for ike evaluations */ fix reading long tab1 and list records */ fix print of max energy transfer at higher temperatures There are new evaluations for H in water, D in heavy water, and ZrH that were developed by IKE under IAEA auspices. They use large grids that solve some integration problems that occur with the existing ENDF thermal evaluation for H(H2O), thereby stressing the old coding a bit. *ident up98 */ leapr -- 11apr05 */ changes recommended by m.mattes, ike-stuttgart */ allow for up to 400 beta values as needed for ike h2o */ correct directory entry These changes to LEAPR support the preparation of the new IKE/IAEA evaluations for thermal scattering S(alpha,beta,T). *ident up99 */ reconr -- 11apr05 */ add forgotten initializations of zero (a.hogenbirk, nrg) *ident up100 */ heatr -- 11pr05 */ increase number of legendre terms to handle new mo95 evaluation *ident up101 */ gaspr -- 11apr05 */ need more space for for ni-58 from jeff-3.1t */ provided by d.l.aldama, nds/iaea consultant New evaluations continue to get more complex. *ident up102 */ matxsr -- 11apr05 */ changes recommended by kazuaki kosako (shimizu corporation) More storage was requested for larger group structures. */ changes for processing be-9 from jeff-3.0 from iaea. The JEFF-3.1 evaluation for Be-9 has an extended representation for the (n,2n) reaction that uses partial reactions in the series MT=875, 876, .... A number of changes in several modules are required to implement these new reaction types. See below also. */ remove a problem in case of multi-temperature matxs files processing. */ scratch tapes iref and nscr should be simultaneously forwarded. */ the problem seems to affect only multi-temperature runs, if the mf=6 */ of gendf tape was not ordered by mt numbers. */ recommended by d.l.aldama, nds/iaea consultant, july 2005 *ident up103 */ groupr -- 11apr05 */ changes to groupr recommended by d.l.aldama, nds/iaea consultant */ process be-9 from jeff-3.0 *ident up104 */ purr -- 11apr05 */ increase scratch space to handle pu-239 from jef-2.2 (trkov,iaea) */ allow pt for multi-isotope materials when */ not all the isotopes have unresolved resonance data */ from d.l.aldama, nds/iaea consultant *ident up105 */ groupr -- 11apr05 */ patch the initialization of getsig, getflx, and getff */ provided by shimuzu corp. Setting the energy requested to zero is used often in GROUPR to request initialization for a given ENDF section. */ changes recommended by a.trkov, nds/iaea */ increase the maximum number of legendre coefficients allowed */ provide more space for the flux calculator */ fix inconsistent usage of output weighting flux unit number */ fix logic when searching for the right flux point */ fix a test that has misbehaved for lahey pc compilers */ accept old proposal from c.dean This changes the results from the flux calculator slightly, and it has been vetted by both winfrith and the iaea. *ident up106 */ acer -- 18aug05 */ fix a problem with the patching of distributions with */ e'>e. the distributions with too high e' values are */ moved down to e' values below e. the previous version */ generated non-sequential e' values for jeff-3.1 be-9. Values of e'>e are not uncommon in evaluations. They often arise from the binning used in model codes. In most cases, correcting them to protect subsequent codes like MCNP doesn't change things very much because only small portion of the secondary neutrons violate kinematics like this. But the correction has to be done right! *ident up107 */ groupr -- 18aug05 */ add the ecco33, ecco1968, tripoli315, xmas172 and vitamin-j */ group structures with 7 significant decimal digits. those */ group structures are used in europe for fast breeder and */ thermal reactor neutronics calculations. for compatibility */ with calendf and apollo */ increase the size of egn from 641 to 15000 i.e dice 13193 apollo 11276 */ increase the size of storage array to handle certain groupr figures These changes are provided to support our European users. *ident up108 */ acer -- 18aug05 */ provide more space for angular distributions in ptleg2. */ recommended by aldama (iaea nds). */ format extension for negative energies (overlap) */ recommended by aldama (iaea nds). */ declare "error" external to avoid conflict with intrinsic function */ provided by trkov (iaea). */ if pointwise representation in cm system, csn should be used */ from trkov, iaea (pointed out by harry wienke). */ fix6 may run out of space without warning */ provided by trkov (iaea). */ provide for larger output files to handle jeff-3.1 u-238. */ recommended by aldama (iaea nds). It's a good thing that computers are getting larger faster than evaluations are growing! *ident up109 */ wimsr -- 18aug05 */ Prepare (n,2n) from mt=875-891 to correct absorption */ d.l.aldama, nds/iaea consultant, 2005 More changes to support the JEFF Be-9 evaluation. *ident up110 */ errorr -- 18aug05 */ several patches prepared by a. trkov and i. kodeli, iaea, feb-2005 */ - Fix fatal error converting code to f90 */ - increase storage arrray (namax:30000->300000,nxmax:150->450) */ - Allow suppression of spikes resulting from lb8 contribution */ - Allow extension for mt 850-891 */ - Activate mt=261 processing */ fatal error - misplaced statement incrementing reserved array index This update provides several changes for covariance processing, including lumped covariances in mt=850-891 and spectra in mt261. *ident up111 */ covr -- 18aug05 */ several patches prepared by a. trkov (iaea), feb 2005. */ - Increase storage (iamax 50000->300000, ipat 1000>14000)) */ - Fix energy-truncation of covariance matrix plots for small functions */ (spectra from irdf-2002 dosimetry library) */ needed to process matrices on input energy grid with many groups */ increase to 2000 is sufficient if no subdivisions are requested */ alternative subdivision of levels for colour scheme */ cannot explain this one, but one of the irdf-2002 spectra crashed Corresponding changes in covr for the iaea. colors in the correlation contour plots were changed to emphasize the lower correlations values. *ident up112 */ acer -- 20oct05 */ increase dimension for nubar tabulations in pn data */ to handle the new lanl am-241 evaluation for endf/b-vii. *ident up113 */ acer -- 15feb06 */ add continuous sampling method for thermal distributions */ energy distributions from thermr are converted to pdf/cdf */ form with some panels that have small cdf contributions */ eliminated. angular distributions are left as discrete */ for later smoothing in mcnp. this method is experimental */ and not standard in mcnp5 as of this date. this patch */ also provides extended plotting capabilities. The existing method for sampling thermal scattering distributions was to use a set of discrete emission energies and cosines chosen to preserve average energy and angle changes. This method works well for criticality calculations, but it gives unreasonable looking spectra in the thermal range containing sharp spikes. This problem has been significant for people doing MCNP analyses of accelerator experiments, such as cold neutron sources. A new method is provided here that uses a continuous secondary-energy distribution that can be sampled in MCNP or MCNPX using a pdf/cdf method. Angular distributions are still discrete, but they can be smoothed during MCNP/MCNPX sampling. A patch to MCNP to handle this format is in the process of being adapted to the standard distributions of MCNP and MCNPX. The new plotting capability adds plots of the outgoing distributions. *ident up114 */ leapr -- 01mar06 */ put out a few more discrete lines at low alpha */ weight is wrong when adding continuum parts for discrete lines */ allow for smaller sab numbers in endf file */ add skold method for handling intermolecular correlations * as used for d(d2o) by keinert and mattes. These changes to leapr support some of the new evaluations provided for ENDF/B-VII and JEFF3. *ident up115 */ heatr -- 3apr06 */ skip over heating calculation when there is no distribution */ allow for fission distributions in file 6 as in 7beta1 th-232. */ the yld becomes nubar. *ident up116 */ groupr -- 05apr06 */ fix reading tab1 record in conver (trkov, iaea) *ident up117 */ moder -- 05apr06 */ process compact format representation in mf32 (trkov, iaea) */ see new utility routine intgio in up118 below. The original formats for resonance covariances have been expanded to allow for more detail correlations between the various resonance parameters. The data are so complex that a compact representation has been devised. Changes in this update set are designed to read through the new format in File 32, but not to process them. The alternate ERRORJ modules will take care of that in the future. *ident up118 */ njoy -- 05apr06 */ new utility routine to process compact format representation */ for mf32 (trkov, iaea). see also up117 which uses this subroutine. *ident up119 */ acer -- 05apr06 */ fix bug that switched off ddx plots (trkov, iaea). *ident up120 */ errorr -- 05apr06 */ temporary patch to skip processing unrecognized options */ for mf32 (trkov, iaea). *ident up121 */ heatr -- 13apr06 */ add tosend in nheat when the last subsection of file 6 */ section 5 does not include particle distributions. *ident up122 */ acer -- 13apr06 */ fix bad message for renormalizing the delayed spectrum */ the undefined index lxx can cause crashes *ident up123 */ acer -- 19apr06 */ increase storage limit for preVII thermal evaluations *ident up124 */ thermr -- 19apr06 */ increase storage space for big preVII thermal evaluations */ allow for up to 32 angle bins as used in tripoli *ident up125 */ acer -- 19apr06 */ patch acer to handle the discrete anisotropic mf6 capture */ photon in endf/b-vii h-1. temporarily, the relativistic */ effect will be ignored and it will be treated as a */ simple primary photon. The new ENDF/B-VII evaluation for neutrons on H-1 includes an anisotropic photon given in File 6. This photon requires a relativistic treatment of the outgoing photon energy and angle for completeness. In the interim, we have provided an approximate treatment. The complete treeatment will require changes in MCNP. *ident up126 */ broadr -- 22apr06 */ watch out for tt(0) A computer compatibility fix. *ident up127 */ heatr -- 22apr06 */ the routine getsix doesn't have any coding to handle law 6 */ (phase space distributions). temporarily, we set ebar */ and dame equal to zero. this affects endf h-2 (n,2n). This is a temporary workaround. The heating and damage for deuterium will be wrong above the n2n threshold. */ missing save */ fix initialization of df for tabulated distributions Computer compatibility problems that show up with compiler checking on. *ident up128 */ thermr -- 24apr06 */ fix ending point for coherent scattering, which sometimes */ goes into an infinite loop. *ident up129 */ njoy -- 26apr06 */ restructure do loop so tht bounds checking doesn't fail */ need large printed fields for storage diagnostics Problems showed up when compiler bound checking was used. *ident up130 */ acer -- 27apr06 */ increase Legendre array to 65 and add test to avoid array bound overflow */ increase argument value from 100 to 300 to match */ nethr definition in subroutine first. */ restructure if test since e1 and e2 are only defined */ when ithopt=1. */ move egamma statement to assure it is initialized */ prior to subsequent usage. */ provide a dummy law value so subsequent if statements */ have a defined value to test. */ a do nothing change to keep the compiler happy. */ existing code logic (ltt3=3) is ok. */ make sure pp1 is initialized to zero. This changes also correct problems found when detailed compiler checking was enabled with the Intel compiler. *ident up131 */ groupr -- 9may06 */ watch out for sed(_,0) */ save some additional variables. */ this has not been a problem in the past, but its easy to */ include an additional else branch test to abort if an */ invalid quadrature order is ever set. */ add "save ipd,ird" in getyld so that initial call to */ terpa has all needed info. */ delete un-needed contio call. This record was read back */ in getmf6 and c(_) has been passed to f6psp. Actually */ as currently coded, nin in the contio argument list is */ undefined here. Most compilers will have set it to zero */ and the contio call does nothing, hence previous processing */ of 2h (the only evaluation using law=6 at this time) is */ probably ok. More changes from detailed compiler checking. *ident up132 */ moder --12may06 */ matd is undefined unless loop>0, therefore break if test */ into two pieces. *ident up133 */ reconr -- 12may06 */ make sure there is a ym to go with this xm. *ident up134 */ broadr -- 12may06 */ make sure ctev is initialized before subsequent if test */ dealing with thermal g-factor edit. *ident up135 */ heatr -- 12may06 */ make sure izap is saved for subsequent use in h6dis and bacha. */ fix typo (fn should have been f) in this update. */ make sure tt=0 unless subsequent if clause changes it. */ make sure ihi and ilo are initialized prior to subsequent */ if test for output. */ fix the check plots Without this patch, the photon plots from the heatr energy-balance checks didn't show the results. These energy-balance tests have been useful during ENDF/B-VII testing. *ident up136 */ heatr -- 30aug06 */ change energy dependent fission pseudo-q calculation. the */ coefficient used with the incident neutron energy in previous */ versons of njoy does not agree with the format manual energy */ dependent equations. see the endf7 paper in nuclear data sheets */ for a discussion. The average energy of the fission fragments tends to be energy dependent, and formulas estimating the dependence have been included in the ENDF format manual for some time. Recently Madland revisted some of this work. While studying his work, we realized that we had not implemented the older work quite correctly, and this patch fixes that. The difference is modest in the MeV range, but a little more significant at 14 MeV. */ same as heatr.799, but for single precision code. */ rewrite the energy release equation with a plus sign. this lets */ the sign of the fq1 and fq2 coefficients given in the data */ statement control whether its truly addition or subtraction. */ make sure nwmax is large enough to handle 239pu prompt */ make sure mf1 is defined for the hgtyld call and pass mtd (not mt1) */ for the file type. */ make sure the tab1 record fits into the allocated space. */ initialize these variables in double precision, as done for others. */ eliminate redundant if tests. */ restructure these "if" tests since x is only defined when the */ gety2 call has been made. Needed to keep the compiler happy */ when processing endf/b-viib2 15n. *ident up138 */ groupr -- 31aug06 */ tweaks in groupr's bach so that it tests the same list of iza */ possibilities as are tested in heatr's bacha and acer's bachaa. Helps with consistency, but this capability is not so important now that ENDF/B-VII and other systems have moved to an almost exclusive use of isotopic evaluations. *ident up139 */ acer - 31aug06 */ a tweak in acer's bachaa to make it consistent with heatr's bacha. As above. *ident up140 */ groupr -- 31aug06 */ make sure all variables that were changed while processing the */ previous reaction are restored to their defaults before starting */ on the next reaction. */ de is only defined with (lf.ne.12), therefore only calculate */ xc under the same condition. If not, endf/b-viib2 241,243am will */ abort due to undefined de. *ident up141 */ njoy -- 31aug06 */ initialize xold prior to first use. a value of zero means the */ "do while" loop is executed at least once. *ident up142 */ moder -- 31aug06 */ add coding to read the new beta-delayed photon data that may */ be found in files 1, 12 and 14, section 460 (beginning with */ endf/b-vii). This coding matches the specifications given */ by brown et al at the nov2004 & 2005 csewg meetings. */ note that no changes are needed for file 14 processing. These changes just allow the code to flow when MT460 sections are present, but they don't do any processing of them. *ident up143 */ acer -- 31aug06 */ (these mods in response to c. broeders, karlsruhe) */ up113 uses "nsyso" in subroutine thrlod, but it was never defined. */ add error to the external subprogram list or some compilers will */ complain that our use doesn't conform to the intrinsic error function. *ident up144 */ reconr -- 01sep06 */ already initialized "zero" in up92, delete the redundant up99 code. *ident up145 */ broadr -- 01sep06 */ (from "patch_japan_njoy99_112a.txt") */ need to put this if test inside the do loop. *ident up146 */ heatr -- 01sep06 */ (from "patch_japan_njoy99_112a.txt") */ current code overloads "nz" and can never update loc. *ident up147 */ acer -- 01sep06 */ (several corrections, from "patch_japan_njoy99_112a.txt") */ maximum legendre order was changed from 20 to 64 years ago. */ change comment and hardwired do loop terminating constant to "ni". */ need reaction q value when call acelf5 or acelf6. Therefore */ shift location of current code. */ 2*nint(spi) is wrong; need nint(2.*spi). Here and line 7299. */ need a bigger integer field for the error message. */ keep making various arrays larger. Use parameter statement so it */ will be easier the next time. */ 2*nint(spi) is wrong; need nint(2.*spi). Here and line 5463. *ident up148 */ njoy -- 05sep06 */ change asend, afend amend and atend utility routines to set the */ last section record counter to 99999 or the last file, material */ or tape section record counter to zero. Reset the counter to */ one for the next initial section record. This convention has been used for a long time in the standard ENDF files--we are just getting up to date. *ident up149 */ broadr -- 19sep06 */ use unique variable to allocate nubar space. *ident up150 */ reconr -- 19sep06 */ update 142 fixed moder to handle the new mt460 beta-delayed */ photon data, but we also need patches so that we don't look for */ a corresponding mf=3, mt=460 when executing other modules. */ In particular, need patches in reconr, heatr, groupr and acer. */ Take care of reconr in this update, then heatr, groupr and */ acer in the next three updates. */ tested with endf/b-viib3 239pu. reconr output files are */ identical whether mf1, mf12 & mf14, mt460 are present or */ not in the endf input file (21sep06). */ eliminate mf12,mt460 from lunion's output tape, which propagates */ through to recout. Put this test before existing if(mfh.eq.12) */ test to make sure it gets executed. As stated above, we are just passing over MT460 and not making any use of the new data. *ident up151 */ heatr -- 20sep06 */ mods needed to skip over mf12, mt460 so that heatr doesn't */ try to find the non-existent mf3, mt460. */ tested with endf/b-viib3 239pu. heatr output files are */ identical whether mf1, mf12 & mf14, mt460 are present or */ not in the endf input file (21sep06). */ expand test, only true when mtd.ne.460 now. */ omit mt460 as a reaction to include in the sum. */ expand test, only true when mth.ne.460 now. *ident up152 */ groupr -- 21sep06 */ mods needed to skip over mf12, mt460. */ tested with endf/b-viib3 239pu. groupr output files are */ identical whether mf1, mf12 & mf14, mt460 are present or */ not in the endf input file (21sep06). *ident up153 */ acer -- 21sep06 */ mods needed to skip over mf12, mt460. */ tested with endf/b-viib3 239pu. acer output files are */ identical whether mf1, mf12 & mf14, mt460 are present or */ not in the endf input file (21sep06). *ident up154 */ acer -- 22sep06 */ set nu-bar for neutron multiplicity on MF6 fission (trkov, iaea). */ force lab coordinate system. */ (Th-232 from ENDF/B-VII with anisotropic fission neutron distributions) This caused problems with MCNP criticality results for Th assemblies. */ exclude redundant lumped cross sections mt 851-870, if present. */ (not allowed by ENDF rules but needed by ERRORR). */ initialize unset nxs and jxs elements. */ otherwise there can be problems if multiple acer runs */ are done in one njoy deck. */ allow thermal names as long as six characters (trkov) */ set default values for rkal and akal (kosako) *ident up155 */ reconr -- 22sep06 */ be more precise in skipping redundant lumped reactions (Ivo Kodeli) *ident up156 */ leapr -- 22sep06 */ fix problems in recent updates reported by Lahey compiler (A. Trkov) *ident up157 */ groupr -- 22sep06 */ from a.trkov, ijs */ groupr A. Trkov, IJS, June 2006 */ allow processing of lumped reactions defined for covariances */ (problem pointed out by a.bidaud */ processing th-232 from endf/b-vii) *ident up158 */ errorr -- 22sep06 */ further updates to process redundant lumped reactions (ivo kodeli) */ fix for si-28 endf/b-vi evaluation - more than 40 mt. (ivo kodeli) */ tidy-up for consistency (a.trkov) */ make sure l has the right value with multiple ni subsections *ident up159 */ viewr -- 22sep06 */ a.trkov, IAEA, April 2006 */ fix colour shades in covariance plots *ident up160 */ covr -- 22sep06 */ from a.trkov, iaea, April 2006 */ - fix labels for unassigned mt numbers */ - improve trapping of large correaltion coefficients */ move trapping of large correlations from level to matshd */ fix for si-28 endf/b-vi evaluation - more than 40 mt. (ivo kodeli) */ delete redundant common/fig/ *ident up161 */ groupr -- 22aug06 */ add a capability to handle int=22 */ for evaluations taken from jendl. Some of the newer evaluations from the JENDL system use the 2-D interpolation scheme called "unit base" for distributions. Many of these evaluations also appear in JEFF3, and some in ENDF/B-VII. This patch handles this int=22 method in File 5. File 6 processing already handles this correctly. njoy99 up204 readme notes: This update is the next official release after NJOY99.161. A major correction from up115 for heating calculations is made in up185. Another patch affecting charged-particle partial heating values was made in up201. Users of NJOY99.115 or later that work with heatr or acer output should re-generate those files. Other important changes include: - ident172 extends the up62 purr modifications; - ident173 (extended in ident183) eliminates redundant zeroes at the beginning of file 5 TAB1 functions (this eliminates a possible distortion of the low energy function when unit base interpolation is applied); - ident175 increases the maximum number of reactions that can be broadened from 10 to 40; - idents 180 to 182 extend use of the "f" format in pendf output to energies between 0.1 eV and 1.0 eV. This allows energy differences in the 8th significant digit to rendered; - ident200 adds single precision equivalents for the GROUPR group structures that were added with double precision mode in up107. Many of the remaining modifications relate to making sure the single precision and double precision versions of the code are consistent. We found a number of instances where prior idents had placed generic code within the *set sw construct, which meant that it only appeared in the double precision version. These code sections have been moved and compiler errors that were encountered when attempting to create the single precision version have now been eliminated. Also, the new group structures in up107 were only given in double precision; single precision equivalents have now been added. Finally, use of "d" or "e" exponents in selected data or assignment statements that was not always consistent with the required single or double precision specification has been corrected. Users are cautioned that several data or assignment statements use values that are too small to represent in single precision on 32-bit machines. Our experience is that the Intel compiler will successfully produce an executable NJOY file (with on screen warning messages stating that these variables have been set to zero). At other times, single precision data or assignment statememts may use numerical data with ten or more significant digits. Single precision accuracy to this level cannot be rendered on 32-bit platforms and code logic that depends upon these least significant digits may not (and more likely will not) execute correctly. We note that either the single precision or the double precision version of NJOY may be used on 64-bit platforms, but THE SINGLE PRECISION VERSION OF NJOY SHOULD NOT BE USED ON 32-BIT PLATFORMS. In the NJOY module, several inconsistancies between single and double precision versions of selected slatec modules were eliminated. The double precision version of these routines was correct, and they remain unaltered. In GROUPR, differences in the 9th and 10th digits of some of the the "qp8" array variables in "getdis" were eliminated. The change is technically insignificant, but these variables should have been (and now are) identical. This update also includes portions of upnea12 (from http://www.nea.fr/html/dbprog/Njoy/upnea12), as follows: - upijs65: implemented in up165. - upijs66: implemented in up166. - upnea001: not implemented. We'll leave this for local users to incorporate. It is very compiler specific and a side effect is that all scratch files with a status of "unknown" remain on the system after job execution. - upnea002: not implemented. Neither this coding nor the original (upfzk2.dat) coding seems necessary. The call to CONTIO that follows call findf in lines matxsr.1880 and matxsr.188x will have no impact upon the values of mat, mf and mt. These changes do not seem to be necessary. Received email from Andre Trkov (5/21/2007) and Cornelis Broeders (5/23/2007) agreeing with this position. - upnea003: implemented in up170. - upnea004: not implemented. Changes to existing group structures can have unexpected consequences to the total njoy user community. If specific users need such a change, they can implement it locally. - upnea005: not implemented. The save statement at groupr.5334 has been corrected in up167, which should eliminate the infinite loop error. The remaining test on "inow" was arbitrary and since the size of array a, c or cnow (depending upon subroutine or function) has been increased, it is no longer necessary. Received email from Andre Trkov (5/21/2007) and Cornelis Broeders (5/23/2007) agreeing with this position. - upnea006: implemented in up171, except that the maxw variable has not been changed. The current 5000 size is expected by TRANSX. - upnea007: partially implement in up167. We have chosen to keep all information messages. - upnea008: implemented in up169. - upnea009: partially implemented in up167. We do not believe that all of the changes made in "009" are appropriate and that the coding as now implemented is correct. - upnea010: implemented in up189. - upnea011: implemented in up190. - upnea012: implemented as part of up175 where the number of reactions that can be broadened is increased to 40. The heatr module has a capability to do detailed energy-balance tests. One usually needs to ask for the partial kerma mt303 to get complete results and all the plots. We provided up204 to protect the code from crashing when mt303 is not asked for. The photon energy-balance plots will be omitted in that case. Additional changes beyond 99.204 as issued on 27 June 2007: up205 modifies a single-precision quadrature point array in groupr to exactly match its double-precision counterpoint. up206 explicitly initializes an acer variable to keep the intel compiler happy. up207 stops a loop iteration at with the proper index to avoid an array bounds overflow in the groupr flux calculator. up208 adds an approximate treatment for the relativistic discrete gamma in the endf/b-vii evaluation for n + h-1. we are using a simple isotropic discrete primary photon for now. see also up212. up209 fixes a double counting of the charged-particle heating when the charged-particle is represented as discrete recoil in file 6 with law 4. an example is li6(n,t)alpha from endf/b-vii. this problem was reported by little and trellue (lanl). up210 fixes a problem observed when the weight function energies provided by a user use 7 or more digits. the symptom is an infinite loop (reported by Broeders). up211 fixes the heating and damage calculation associated with the phase-space representation of the (n,2n)p reaction for neutrons on h-2 from endf/b-vii. up212 implements an approximate treatment of the relativistic discrete gamma ray in the endf/b-vii evaluation of neutrons on h-1. for now, it is treated as a simple isotropic primary photon. see also up208. the proper relativistic formulas for treating the anisotropic photon distribution will be added at a future date. currently, mcnp can't handle sampling from the photon distribution, and we will want to coordinate the njoy and mcnp changes for this advanced representation. up213 makes some changes in acer plotting. the important thing is a new plot of the "recoil heating," that is, the normal heating minus the charged particle heating given with the particle production sections (when present). in order to do a coupled calculation including charged particles (such as a coupled neutron, photon, proton calculation), you have to correct the normal heating value by subtracting the heating from the charged particles (or particles) that you want to transport. the heating for that charged particle will then be deposited by the slowing down of that particle rather than as local heating. because of various energy-balance problems that show up in evaluations, it is possible that this "recoil heating" will show problems, such as negative values. this new plot can help the user to recognize these problems. of course, negative values for the heating are often compensated for when the excess energy of the emitted particle is deposited, unless the particle reacts far from the site of the reaction. the normal acer heating value for neutron-photon transport can also have unreasonable values (including negative values), and the linear version of the heating plot has been modified to show the negative region. more information on energy-balance problems is available from heatr checking runs. up214 patches a format statement in powr (which is rarely used these days). up215 modifies reconr to be able to process photonuclear files in a graceful way. this is needed for the russian actinide evaluations. see also up216. up216 modifies the photonuclear option in acer to be able to use both an endf input and a pendf input from reconr. this allows the russian actinide evaluations to be processed correctly without the awkward work-arounds that people have had to use. this problem was diagnosed by morgan white (lanl). u217 fixes the sizes of some arrays in heatr associated by legendre polynomials. up218 increases the size of an integer format in acer to handle cases with more than 100k points in the energy grid. up219 is an extension to acer that is turned off by default. it can be activated by changing the value of ismooth. many endf evaluations use histograms to represent outgoing neutron spectra as produced by model codes. however, the expected shape of the low-energy region for these distributions is sqrt(e). the edges introduced by the histogram steps in computed flux spectra can be noticeable. the the lowest bin can lead to strange looking overestimates of the flux at the lowest energies. this patch can be used to smooth out these bumps. it modifies the first few bins by adding additional bins with log spacing to better approximate sqrt(e). the effect on criticality of this smoothing seems to be minimal, but it does produce more physical looking graphs of the flux or reaction rates. similarly, the endf delayed-neutron spectra are represented using histograms. the histogram break at 10 kev can sometimes be seen in plots of the flux or reaction rates, and more delayed neutrons appear at very low energies than one would expect. this patch divides the 0-10 keV bin into smaller bins with log spacing to approximate a sqrt(e) shape. finally, the energy spacing for some tabulated fission spectra increases from .1 meV spacing just below 10 mev to 1 mev spacing above. with its linear interpolation, this larger spacing doesn't approximate the exponential shape of the function very well. this patch interpolates additional points on a 200 kev grid using the exponential shape to smooth out the interpolation cusps in the spectrum. up220 for acer fixes the plotting of fission nubar when it is described as a polynomial. up221 expands upon up219. fission spectrum processing for energies greater than 10 MeV was inadvertantly turned on in up219; this update turns it off as stated in the comments above. The algorithm has also been revised to only execute when fission spectrum energy values in the original endf-formatted file change by at least 200 keV (as they do for endf/b-vii 235U). up222 is a minor revision to viewr that allows exponents of -10 or less to be properly printed on the y-axis. Previous coding specified an "i2" format which was insufficient in these cases. up223 is an update to up215. up215 modified the total xs mt value written to reconr's pendf tape from 1 to 3, as necessary when processing photonuclear files, but was coded such that the dictionary was incorrectly modified when processing neutron or photo-atomic files. This update corrects that error. up224 is an update to up77. If the compiler has not defaulted unitialized variables to zero, then nin2 can be undefined. We explicitly initialize it to zero. up225 is an update to up187. up187 assumed we could get the maximum energy from file1/451 if this was an endfb6 formatted file. That is not true for early endfb6 releases and so this assignment statement has been modified to default to no lower than 20 MeV. We also delete a no longer needed data statement. up226 makes the corresponding change for acer. up227 makes the fixes to the energy range for discrete levels described by transition probabilites discussed above for up174, up187, up225, and up226. NJOY99.245 is the next public release after 99.227. The following patches are included: up228 explicitly initializes a viewr variable (as with many of the explicit initializations that have been made recently and that we continue to make, this patch has no impact if the code is compiled with uninitialized variables set to zero). up229 expands the field for a thermr print statement (Broeders). up230 fixes several groupr issues, including a save in f6ddx that might impact evaluations that use the Kalbach-Mann formalism, although our test jobs on the 393 endf/b-vii.0 files show no difference after this patch is implemented. up231 adds a missing save in gaminr. up232 to up234 include explicit variable initializations. up235 provides better checking of real variables as used in up180. The previous coding could erroneously print a number like 100 keV as 10.000000+5 rather than 1.000000+5 on pendf tapes. up236 adds logic to ccccr so that gendf files with either six or eight delayed neutron groups are properly processed (Broeders). up237 eliminates several inadvertant implicit real*8 declarations that appear in heatr when the user does not include *set sw in the upn deck. We also make sure that epn is defined prior to calling h6cm. up238 makes similar changes in groupr as up237 made in heatr. up239 makes similar changes in acer as up237 & up238 made in heatr and groupr. up240 extends the logic from up176. Before up176 we could reconstruct the total cross section by summing over all broadened cross sections. With the up176 modifications we now may broaden partial charged particle cross sections (i.e., mt103 plus any mts from 600 to 649 in a version 6 formatted file), but we need to omit these partial cross sections when reconstructing the total xs. Conversely, if mt103 is not present but mt600 and above are, then we need to make sure they are included when reconstructing the total cross section. up241 provides a correction to thermr to properly account for kernels where both lasym and lat flags are set (such as liguid hydrogen and liquid deuterium). This is essentially upnea015 by M.Mattes. up242 modifies reconr to properly reconstruct the total cross section when one or more of mt103 through mt107 are not present but the partial cross sections are present. In endf/b-vii.0 there are seven such evaluations (7Be, 74,75As, 90Y, 232Th, 231,233Pa). up243 extends the up242 modifications to acer. Without them, acer's total and absorption cross sections do not include the partial charged particle cross section contributions. up244 extends the up242 & up243 modifications to gaspr. up245 modifies purr to allow for better sampling when the evaluation defines narrow and widely spaced (keV D-bar values) resonances. NJOY99.259 is the next public release after 99.245. We continue to correct errors and add new capabilities. The following patches are included: up246 corrects a typographical error in up240 where we used "m4max" rather than "m4min" in an if test. broadr pendf tapes for nuclides with both mt107 and discrete threshold reactions in the mt800 to mt848 series (such as 10B) are affected. up247 expands the logic from up244 to properly handle input pendf tapes with multiple temperatures. up244 (gaspr) works fine with a single temperature pendf tape but goes into an infinite loop for a multiple temperature pendf tape. up248 adjusts some reconr format statements to allow more digits. up249 adjusts several acer format statements (from "i2" to "i3") to properly print all possible mt values. up250 expands a viewr array so that 9 possible colors are available when making selected plots. This allows one color for the plot boundary and up to 8 colors, such as is needed when plotting jeff-3.1 delayed neutron curves. up251 expands an if-than-else block in reconr to include file 10 when determining how much space to allocate for the pendf dictionary. Previous coding lead to bad mt values in the pendf dictionary if 6 or more file 10 sections were present in the original endf-formatted file (e.g. jeff-3.1 99Tc, 103Rh and 127,129I). up252 adds a test in heatr's gambar routine to adjust the elo and ehi variables being passed to terp1 when the cross section energy is exactly equal to a discontinuity energy in file 12 and histogram interpolation was not specified (e.g. mf12, mt102 in jeff-3.1 151Eu). up253 corrects the maximum energy definition used in subroutines disbar and hgtfle so that it does not exceed the maximum energy in the evaluation. Past coding could exceed the maximum evaluation energy when calculating mt=301 and mt=444. Users will see small (less than a fraction of a per cent) differences in test01 mt=301 and mt=444 output and in selected evaluations, typically at energies within one or two MeV of the maximum evaluation energy. up254 removes old groupr coding that affected processing of analytic fission spectra. up255 and up256 affect the leapr and thermr modules. Past documen- tation (LA-12639-MS (ENDF-356)) has not been clear that thermr will not correctly process pendf tapes produced by leapr when the isabt=1 option has been selected. Worse, thermr would read such a file, run to comple- tion and produce an output file that is properly formatted for process- ing by other njoy modules. Thermr will now recognize the presence of such an input tape and abort. We also recently realized that the flag denoting ln(S(a,b)) usage was not included in the leapr pendf tape, nor did thermr make any attempt to read this flag or properly process ln(S(a,b)) data. This omission has been corrected. We also added one more instance of "moreio" to handle large tab1 files, as initially done in up98 (Altiparmakov, aecl). up257 includes several modifications to the groupr module. First we revise the conver routine to use file 3 cross section data for photon production calculations, similar to the coding introduced in up186 for acer. We override the linear-linear interpolation specification for mf6/law1 so that unit base interpolation (int=22) is used. MCNP already does this and therefore this change will lead to better consistency between multigroup and continuous energy monte carlo. This is really a deficiency in the original evaluation. As these deficiencies are gradually eliminated these modifications will eventually become moot. We also make a correction in hnab affecting high energy (> ~15 MeV) calculations (of energy-dependent Watt spectra such as used with 233U). Finally, similar to what was done in up219, we add coding that is currently disabled (ismooth=0 now; change to ismooth=1 to enable) that will force linear(E)-log(probability) interpolation for 235U and 239Pu fission spectra above 10 MeV where the energy mesh may be too sparse for the currently specified linear-linear interpolation. This same variable also controls the option to increase the mesh density at low energy so that the histogram segments more closely approximate a sqrt(e) shape. up258 (formally called ident errorj in the upn file for purposes of identifying individual lines of source code) replaces njoy's errorr module with Go Chiba's errorj code. The comments provided at the beginning of this code section describe the modifications made to integrate errorj into njoy. Users should note that 1) the formal module name, as specified in NJOY input decks, remains "errorr"; 2) the input requirements for this module are different than those for the original errorr module. See the "Userinp" link for the latest input specifications; 3) Links to revised (and original) input files for test jobs 4 and 5 are available. The output from these jobs is virtually identical to that obtained previously; 4) Longer running test jobs have been provided by Go Chiba. NJOY i/o files have been added to the t2.lanl.gov/codes/njoy99/index.html web site. These jobs require JENDL-3.3 235,238U and 239Pu input tapes and when run for the first time must include the reconr and broadr modules. Users may want to save the broadr pendf tapes from these jobs since they account for much of the cpu time used when executing these problems. Future cputime requirements for these test problems will be significantly reduced if they can start with groupr processing an already existing pendf tape. up259 adds a getyld call in groupr when processing mf=6 fission spectra so that we properly account for the prompt fission neutron multiplicity. This is not an issue for other file 6 neutron producing reactions since the multiplicity is properly determined from the initial TAB1 yield function. This impacts endf/b-vii.0 232Th and 231,233Pa (Sinitsa). NJOY99.296 is the next public release after 99.259. The following patches are included: up260 incorporates several updates to errorj to consistently define selected variables in grpav4, add external error declarations to avoid any conflict with the fortran intrinsic error function, define selected variables consistently between reconr and errorj and to eliminate the currently unused subroutine ggrmatorg. The need for many of these upgrades was identified by J-Ch Sublet and/or C. Broeders and some, but not all, were adopted from upfzk14. up261 replaces a data statement introduced in up257 with an assignment statement. This eliminates a annoying Lahey compiler warning. up262, 263 and 264 implement changes in reconr, unresr and purr, respectively, to handle energy-dependent scattering radius data in the unresolved resonance region, including channel and scattering radii calculations per the naps and nro flags read from the endf input file. up265 replaces groupr's abort condition (introduced in up167) when reading negative URR cross sections with a warning, thereby permitting the calculation to continue. Not necessarily the smartest thing to do in my opinion, but this change is in response to strong user feedback. up266 revises the legal minimum mtref number from 201 to 221 in thermr. As a practical matter, the 221 lower limit was imposed long ago, but never formally codified and we do so now so inexperienced user's do not request an illegal value in thermr's input. up267 eliminates extraneous "thermal" related text from the standard output when running a "fast" acer job. We also add a check that amin .ne. amax in acer's standard plots (which could lead to a crash in viewr (for jendl-3.3 16o, for example)). This revision makes a non-technical change in the standard output for test jobs 7, 8, 10, 13 and 14. up268 upgrades moder to more generally handle mf7, mt4 thermal files. The current coding properly handles existing files but if a future file included the full functionality of this format it would not be handled properly. up269 performs a more accurate average bin energy calculation in acer/acesix to avoid calculating a decreasing average bin energy when dealing with large group structures and low energy data. We also correct an indexing error that impacted the Bragg edge energy standard output (Mattes, ike-stuttgart). up270 includes minor updates to assure all variables are properly initialized in acer/photonuclear processing and a revise to coding logic that eliminates an array overflow error when creating plots with more than 16 curves on a frame. up271 eliminates a potential double counting of particle production that can be inferred from mt102 through mt107 and also explicitly given via mf6 data by acer. At present we only know of deuteron production from photon capture in endf/b-vii.0 1h being affected by this error. Reported by LANL (Trellue, Little). This correction was available and used by LANL when generating the endfb7 cross section library that accompanies mcnp5.1.50 and later releases. up272 includes a number of changes in errorj that were accumulated throughout 2008. These are summarized in the comments following ident up272 in the up296 file. up273 includes a number of changes in covr. Arrays are increased to allow processing of larger group structures, increase plot titles and labels to 80 characters, revise plot logic so that the linear 0% to 60% uncertainty scale shifts to a 0.1% to 100% log scale if the min-to-max uncertainty data differ by 10x or more, add a warning when the uncertainty data do not fit within these limits, plot the few group xs rather than repeating the uncertainty plot when mt=mt1, produce plots for MF34 (mt251 only) and MF35 files, increase the color shade array so that virtually all non-zero elements are plotted; previously values <0.1 appeared as white. In this upgrade, mf35 few group plots are the spectrum value by group and the incident neutron energy included in the plot title is always 0.0253 eV. Up283 changes this to include the true incident energy and to plot spectrum/eV so that discontinuities due to variable group widths are not introduced. Many of these visualization improvements were developed following discussions and recommendations from Andre Trkov and colleagues at ANL. up274 expands viewr to handle plot and axis titles up to 80 characters in length (the previous limit was 60). This change was made in parallel with a corresponding change in covr (up273). up275 implements 3d plotting of secondary emission spectra (mf6) from groupr input in plotr. This capability was only partially implemented previously and the resulting output tape could not be processed by viewr. Spectra are plotted at the lower group boundary energy, but commented coding is provided that will produce the same histogram at the upper group boundary if desired by users. See the eight commented lines beginning with einc=0.9999*ey3 ... . up276 is a technically insignificant change in groupr to define the variable enext for iwt=2 in the same manner as for other iwt values. We also make sure that ehigh is always defined in panel. up277 upgrades moder to handle compact covariance formatted data containing a variable, 2 to 6, number of digits. This upgrade incorporates coding previously distributed to 99.259 users via upnea034 and upnea037 (Trkov/Hebert). up278 upgrades njoy's intgio routine to handle 2-to-6 digit compact covariance formatted data. This patch was implemented in parallel with up277 and was previously made available to 99.259 users via upnea033 (Trkov). up279 incorporates several patches to errorj that were developed in upnea036 (Trkov). This includes (i) revised output to the user terminal; (ii) revisions in Resprx_RRR_Lcomp2 to properly read mf32 data and (iii) revised explicit read statements with intgio calls to obtain compact covariance data. up280 upgrades the plotr module to allow creation of ratio plots while automatically looping over all available mt values. The comment section has also been expanded to further explain this option. up281 is a minor upgrade to viewr that allows an additional digit to appear on the y-axis plus inclusion of a test to eliminate a possible degeneracy in the axis limits (similar to that incorated in up267 for acer). up282 upgrades errorj to properly handle mf35 spectrum data per the formal endf6 definition. We have also upgraded processing of User input for "ifissp" and "efmean" to properly perform the appropriate calculations. Much of this upgrade was made available to 99.259 users through upnea041, 043 and 045. USERS WHO PROCESS JENDL-3.3 MF35 DATA ARE CAUTIONED THAT THESE FILES DO NOT CONFORM TO THE ENDF6 FORMAT AND NEED TO BE CORRECTED USING ANDRE TRKOV'S "CHMF35" CODE, AVAILABLE AT http://www.nea.fr/html/dbprog/njoy-links.html, PRIOR TO PROCESSING WITH THIS AND LATER VERSIONS OF NJOY. We also apply the "zero-sum" rule test to the mf35 covariance matrix and apply the endf6 defined correction to these matrix elements when that test fails. Finally, we include additional information (za, awr and efmean) on the errorj output tape for subsequent use by other njoy modules. up283 upgrades covr mf35 plotting to display the few-group spectrum/eV and to include the proper incident neutron energy in the plot title. Plotting of mf35 data was initially implemented in up273, but we did not have the incident neutron energy (and defaulted to saying it was 0.0253 eV in the plot title) and we plotted the multigroup spectrum. As the group energy width varied this caused artificial discontinuities to appear in the plot. By plotting spectrum/eV this is eliminated. We also change the plotted curve to be blue and thicker so that it is not hidden by the plot frame. Finally, when plotting threshold reactions, we restrict the energy axis to the higher threshold value to minimize white space in the correlation matrix plot. up284 includes additional comments at the beginning of thermr to guide the user in defining "icoh". up285 includes several minor updates throughout acer to assure that njoy99 coding and the soon to be released f90/f95 njoy are consistent. up286 initializes viewr/hatch's xs variable. This is not an issue for users who compile with uninitialized variables set to unity. up287 upgrades errorj to eliminate the restiction of mpar=3 when lcomp=2. We also make sure that the l-dependent scattering radius is properly passed to all routines needing this parameter and revise an explicit error message to use njoy's error routine. This update incorporates coding that implements the intent of upnea047 (Trkov). up288 upgrades groupr to include the original tape title in groupr's output tape (per Doug Muir request) and we save an additional variable, nqp, in panel (this change done for consistency with the soon to be released f90/f95 njoy version. up289 increases an array size in matxsr, an issue identified by Aldama/Trkov, to allow processing of larger group structures. This update was unofficially distributed in upnea031. up290 incorporates parts of unofficial updates upnea039 and 049 (Trkov). We clean up some coding related to the "iwt" variable which was inadvertantly left in place when initially copied from groupr, we incorporate explicit dimension statements for consistency among calling routines (even though the actual length is one due to errorj restrictions on nl and nz), we correct the group boundary calculation in resprx and provide an alternate step size in resprx_cal_pendf when dealing with negative energy resonances. up291 provides URR related updates in reconr to use the endf file specified interpolation (previously only used the recommended value, 2 or lin-lin) and we use a finer energy mesh to more accurately generate urr cross sections over the large urr energy interval that appears in many endf/b-vii.0 evaluations. up292 makes the corresponding changes in unresr as done by up291 in reconr. up293 makes the corresponding changes in purr as done by up291 in reconr. up294 updates acer to read the urr region interpolation flag from mt153 rather than always assuming int=2. up295 adds new urr region cross section plots (total, elastic, fission when present and capture) to the standard suite of plots created by acer. in the urr region we illustrate infinitely dilute, 100-barn and 1-barn self-shielded cross sections computed from the probability tables. these are log-log plots and we require that ymin be no less than 0.001*ymax. up296 revises purr to provide better bin sampling in regions of low probability. purr takes longer to run now (test problem 10 needs about 5-6x the runtime as before). another impact on users is that the input parameter nbin (fourth parameter on card 2) must be 15 or greater. up297 revises subroutine hout in heatr to properly define the dictionary portion of file 1. This corrects a long-standing error where the dictionary entry for the first data section following the last heating section has been omitted. Pendf tapes from tests 8 and 13 will differ from those created by earlier njoy99 versions. up298 revises an if test in groupr's getff to permit processing of activation files. up299 corrects a variable typographical error in thermr's calcem. we are unaware of any errors resulting from the old coding. up300 expands the mf35 processing introduced in up282 so that the the user may input either ascii or binary endf formatted input tapes. up301 revises up288 to prevent un-necessary, but harmless, text from appearing in groupr's output tape when ngout1 .ne. 0. up302 increases the size of selected arrays in errorj's covcal. we see no change in errorj calculations with this change. up303 revises purr so that energies outside of the urr range are not added to the list of energy nodes. up304 increases the size of selected arrays and includes more robust array bounds checking. Previous code could overflow the space defined with reserv, but did so in a harmless fashion since the remaining space in the large "a" array is unused at this time. up305 makes two corrections related to Reich-Moore processing in errorj so that these routines mimic what is done in reconr. Results for lrf=3 nuclides with nls>1 are affected. up306 refines the free gas beta mesh (from 3 points to 9 points) to be consistent with the coding that will appear in njoy2009. We also make a few minor code changes that yield more consistent results when optimized or unoptimize code is executed. up307 corrects a matn labelling error. Numerical calculations are unchanged but the standard output for test problems (4, 5 and 15 - 18) is slightly different. up308 adds the abundance fraction to reconr's pendf file. All test problems with reconr output are changed. up309 revised selected print statements so that njoy99 and njoy2009 output formats remain identical. up310 revises purr so that the minimum elastic scattering cross will be at least 1 microbarn. Previous coding allowed zero. The output from test problem 10 will exhibit small differences. up311 corrects an array overflow condition in acer's ptleg2 routine. Angular distribution processing for nord .le. 63 are unaffected but nord=64 results were wrong. up312 includes several revisions within heatr. We revise some sigfig calls so that optimized and unoptimized code yield more consistent answers; we correct a typo in the legendre expansion formula appearing in getsix and we make small revisions in h6cm, h6ddx and h6psp so that they and the corresponding routines in groupr (f6cm, f6ddx and f6psp) are consistent. up313 cleans up an array overflow issue in groupr when specifying the lanl 80-group structure. We also make minor code changes to remain consistent with the corresponding heatr coding. up314 increases the available scratch space for purr. We now test whether this space is sufficient early in the purr calculation. up315 is the first of three updates related to improved processing of IRDF files. In this update we allow reconr to recognize mt261 and also revise the processing to properly process an input file that contains file 10 but no file 3 data. This update performs the same function as upnea025 and upnea063. up316 permits groupr to recognize mt261 and revises the minimum allowed cross section to 1.e-9 (was 1.e-8). The reduced minimum means that output from test problems 2 and 11 are different. This update incorporates upnea026 and upnea029. up317 revises acer's dosimetry file processing to include mf10 cross sections, using revised mt values that incorporate the original mt number plus the lfs flag. We also include the temperature in this file (it was previously always set to zero). This update performs the same function as upnea027. up318 increases the array space available to acer for fast library creation. We also make sure the xss and jxs arrays are initialized so that mulitple acer output files may be created within a single acer call. This update incorporated upnea028 and upnea058. up319 revises acer to allow the iopp=0 option to run to completion. up320 eliminates potential use of an undefined variable in matxsr. up321 fixes the same array index error in errorj as we fixed earlier in groupr. up322 revises the photonuclear processing in acer to remain consistent with njoy2009 coding and extends the existing coding to process incident photon files such as those found in tendl-2009. up323 adds a moreio call in acensd and fixes an indexing error that affected emission spectra plots when multiple interpolation ranges are defined for those data. up324 expands errorj to process mf40 files. We include coding to read izap per the CSEWG 2009 meeting format revision, set a new flag in the output file to distinguish these data from mf33 data, and revise the gendf-like output file produced by errorj to more closely mimic that produced by groupr when processing mf10 data. This update mimics the coding found in upnea062. up325 patches an undeclared variable and corrects the mt values appearing in selected if tests for particle production when processing incident charged particle files. up326 patches broadr to avoid attempts to broaden a zero cross section. This is not a concern when processing incident neutron files but can be an issue when processing incident charged particle files. up327 revises the unionx routine in acer to read from both the endf and pendf files. When processing incident charged particle files we read the endf file for file 6 data but get the union mesh from the pendf file. Previously when processing an incident charged particle file unionx only read the pendf file which therefore had to be a copy of the endf file since file 6 data are not normally written to the pendf file by reconr/broadr. Test 14 which uses an incident proton + 14N input tape is unaffected by these changes but the combination of up325 + up237 now allows incident proton files such as those found in tendl-2009 to be processed. up328 extends covr to recognize errorj mf40 output files and to append either "mf40" or if available a numeric label (10*izap+lfs) to the covr produced plot file title. up329 expands errorj's processing to include the scattering radius uncertainty. Current coding accepts input files with isr=1, as approved at the Fall 2009 CSEWG meeting. We also extend the user input (card 7) to allow a single relative uncertainty to be defined. By default the user input is zero, but if present will override the data that would be extracted from the endf input file. This coding incorporates that in upnea048 plus the corrections found in upnea051. up330 increases the array space used by gaspr to permit processing of selected CENDL files. up331 increases selected array dimensions in unresr to allow processing of selected CENDL files. up332 increases selected array dimension in purr to allow processing of selected CENDL files. up333 corrects a long-standing typo in the 36/38 group photon structure. We also revise the "ismooth" option to remain consistent with its usage in njoy2010. up334 revises code logic to treat 255Fm (Z=100) as a special case of Z=99; we also revise thermal names to reflect endf/b-vii usage. up335 increases selected arrays in dtfr to allow processing of larger group structures. Previous limit was 400, curretn limit is 2500. up336 implements a recommended revision from Go Chiba in the errorj module to calculate point cross sections on a more dense energy mesh in the resprx_cal_pendf routine. up337 increases selected arrays in acer to support plotting of urr data. up338 upgrades reconr process of urr data for evaluations lacking rr parameters and specifying lssf=1. we have not seen this format combination before, but it is utilized in in some jendl-4 evaluations. up339 extends the range of allowed LFS values in GROUPR. previous coding created a packed integer with LFS as the units digit. recent TENDL files sometimes contain LFS values of 10 or more, requiring a revised algorithm to track this datum. up340 adds basic endf variables appearing in the first few records of file 1 to the reconr/pendf tape (Sublet request). up341 corrects a long-standing flaw in mixr, plus increases array space to support merging of more complex files. upnea65 is incorporated in this update. up342 makes several improvements to errorr, and patches an inadvertant change from up324 where the awr variable was deleted from the output tape. we increase selected arrays to allow for larger input files and upgrade coding that only read ascii tapes to now read either ascii or binary tapes. up343 updates moder to recognize the scattering radius uncertainty format revision accepted following the Fall 2009 CSEWG meeting, plus we finally update this code to handle either ascii or binary tapes when converting mf=34, 35 or 40 datap. up344 increases the working array in ERRORR to allow processing of more complex input files. up345 increases the working array in COVR to allow processing of more complex input files. up346 increases selected arrays in VIEWR to allow processing of more complex input files. up347 includes several GROUPR upgrades requested by the JAEA (Kosako) to support processing JENDL-4 files. up348 updates RECONR to recognize the many new mt values approved at the Fall 2010 CSEWG meeting. up349 updates GROUPR to recognize teh many new mt values approved at the Fall 2010 CSEWG meeting. There is more work needed here, but this update is a start. up350 adds another plotting option to PLOTR, this one for self- shielding factors from GENDF files. up351 upgrades HEATR to handle larger ZA values, switches from log to linear interpolation if negative coefficients are encountered and explicitly initializes a previous undefined variable to zero. up352 increases selected arrays in ERRORR to allow processing of more complex input files. up353 updates GASPR to recognize the many new mt values approved at the Fall 2010 CSEWG meeting. up354 updates ACER to recognize the many new mt values approved at the Fall 2010 CSEWG meeting. up355 updates ERRORR to recognize the many new mt values approved at the Fall 2010 CSEWG meeting. up356 updates BROADR to recognize the many new mt values approved at the Fall 2010 CSEWG meeting. up357 is the first of four ACER upgrades requested by the JAEA for processing selected JENDL-4 evaluations. this upgrade increases the sensitivity for renormalizing delayed neutron spectra up358 increases the working array length. we also increase the sensitivity of a threshold test and initialize a previously uninitialized variable. up359 increases the array space for mf and mt values, makes more consistent use of "awr" and recognizes the lnu=1 option when processing mt=455 data. up360 upgrades acer photon processing to handle a variable number of discrete photons for a given reaction. we also recognize the difference between primary and secondary photons from endf file 6. up361 implements upnea052 and upnea072 in COVR. the former corrects coding first introduced in up273 or up283 that was erroneously placed within the "*if sw" construct; the latter revises an existing error message. up362 implements upnea069 in GROUPR to allow smaller non-zero spectrum values before an arbitrary truncation to zero. subsequent covariance processing of these data will be more accurate. up363 inplements upnea051, 067, 070 and 071 into ERRORR. we also increase the size of selected arrays per the request of NRG/Petten to allow processing of more complex files. Note however that one variable, mxnpar, has only been increased to 7000. Petten recommended a value of 10000 for this variable, but such a large size cannot be supported by 32-bit systems. Users running on 64-bit platforms may wish to change the value the "...mxnpar=7000..." code to a larger value. up364 increases the working array in BROADR to support larger delta-T step sizes. we also initialize selected thermal related values to avoid a potential "nan" calculation. up365 - fixes several typos in gaspr associated with the many new mt values defined at the Fall, 2010 CSEWG meeting. up366 - adds a new input card to covr allowing the User to define the correlation matrix color intervals. This is an optional feature and the previous interval line-up, 0.001, 0.1, 0.2, 0.3, 0.6, 1.0, is a default. see the covr input comments for card 2'. up367 - revised coding to set the z-axis limits for 3D plots in the presence of the near delta function distribution encountered near reaction threshold. up368 - more scratch space in gaspr, plus correction of the final "izr" for reactions with endf "lr" flags. up369 - in acer, make an energy-dependent yield test more sensitive to avoid subsequent divide by zero error. up370 - in purr, move up332 coding that inadvertently put general coding within the update "set sw" block. up371 - in leapr, move up304 coding that inadvertently put general coding within the update "set sw" block. up372 - in groupr, move up257 & up339 coding that inadvertently put general coding within the update "set sw" block. up373 - several acer corrections, including moving general coding from up327 out of the update "set sw" block, correcting a typo in a WRITE statement, increasing the scratch space for discrete photon processing and a further expansion of the 3D axis coding revision in up367. Impacts z-axis appearing in test problem 13, figure 25. up374 - in heatr, implement NEA updates nea76, nea79 and nea94. we also revise the coding associated with printing selected data from nheat to eliminate duplicate energy results. This affects the output for test problems 8 and 13, but those pendf tapes are unchanged. a couple of least significant digit changes are seen the the test 19 pendf tape for some heatr generated mts. up375 - a revision to up360 to more accurately determine primary photon energies. without this users may see different results when running with ascii tape input versus binary tape input! up376 - more scratch space, again, in gaspr plus fix one more typo. up377 - the fission energy release format was revised by CSEWG several years ago to allow for polynomial functions to characterize the energy release. with endf/b-vii.1 this format is now being used and so we've added the appropriate coding to handle these data. NOTE: In ENDF/B-VII.1 files, quadratic coefficients, when present, are erroneously given in 1/MeV units rather than 1/eV. This update uses the mf1/mt451 nmod and lrel variables to identify endf/b-vii.1 files and applies the appropriate MeV to eV correction factor. All other files using this representation for fission energy release are assumed to provide the poly- nomial coefficients with the correct energy units. up378 - in acer change some sigfig calls to allow increased precision (was 7 digits, now 9) for some pdf and cdf distributions. also revise the coding to assure the initial pdf and cdf values are equal when dealing with a combined discrete plus continuous distribution. acer output changes will be seen in test problems 7, 13, 14 and 19. up379 - several updates in groupr, passed to LANL from the UK (Serco and Atomic Weapons Establishment). these include allowing a greater downscatter range for hydrogen, correction of energy intervals used in calculating mf6 gamma production, use of a more numerically robust erfc function (similar to a change made many years ago in broadr!), greater use of unit base interpolatin when computing secondary energy distributions even though the origianl endf file specified only linear-linear interpolation. These changes cause small differences in the groupr output of test problems 1, 2, 7, 11 and in subsequent modules using groupr's gendf tape (ccccr output in test 2; multigroup acer in test 7, wimsr in test11). up380 - when executing multiple heatr jobs we didn't update the dictionary properly, although all requested mf3 mt data were present. this update corrects the dictionary error. up381 - during tendl-2011 testing it was discovered that reconr didn't process urr data correctly when either NE or the energy mesh varied as a function of (l,j) state. we fix that error here. Note: examination of similar coding in unresr and purr indicates that this deficiency only occurred in reconr. up382 - in acer with iopt=7 the user has the option of changing the zaid suffix, but past coding only produced a correctly revised zaid value if the original suffix was less than .50; we modify the coding to correctly replace any old suffix with the newly specified value. up383 - purr has always been a long-running module. we add a print statement to the terminal allowing the User to see that the code is still running. up384 - initialize to zero a previously undefined variable appearing in the charged particle processing ptlegc subroutine. A common compiler option is to do this so it is unlikely that User's will see any change in output. Test problem 14 processes p on 14N and its output is unchanged. up385 - mf34 processing in errorr assumes that the endf input data are in the CM system (mf4,lct=2),are given a Legendre moments (mf4, ltt=1) and that the mf34 data are also given in the CM system. The code hasn't checked whether this is true before now. If its not true we abort and if the angular data are given in tabulated probability distribution format we further advise the User that NJOY2012 is needed. We also skip over any mf34 sections with mt.ne.2. up386 - in acer, up375 used the wrong line number when attempting to redefine eps ... undo that mistake here. up387 - in purr, the calculation of ctx in subroutine unrest includes a temperature ratio, tref/temp(_). we really should be taking the square root of this ratio, per the definition of cth in unresx. this impacts all probability tables that have not been calculated at 300K (ornl, wiarda). up388 - in acer, an endf evaluation with missing mt sections among mf12 through mf15 can get stuck in an infinite loop. add a test to avoid this. up389 - in heatr, upgrade nheat to process the new mt's (152 to 200) approved by csewg in 2010 ... mostly implemented in previous idents but we continue to find additional mods are necessary. also, upgrade testing for expected mf12 mt sections in hconvr. omitted sections such as found in jendl-3.3 93nb or 207pb, for example, only show up later with cryptic error messages that are hard to understand a companion update for acer's convr follows in ident up390. up390 - in acer, upgrade testing for expected mf12 mt sections in convr. omitted sections such as found in jendl-3.3 93nb or 207pb, for example, only show up later with cryptic error messages that are hard to understand. a companion update for groupr's conver follows in ident up391. up391 - in groupr, upgrade testing for expected mf12 mt sections in convr. omitted sections such as found in jendl-3.3 93nb or 207pb, for example, only show up later with cryptic error messages that are hard to understand. similar updates for heatr/hconr and acer/conver were made in up389 and up390, respectively. up392 - in heatr, detect cases where the capture recoil is given without an associated spectrum. use the photon recoil method to generate a capture recoil spectrum. without this fix, cases from TENDL2012 give zero for capture damage (reported by J.C.Sublet). up393 - in acer, we relaxed the value of eps in up375, but a side effect is that similar energy discrete photons might not be treated as being different (2.112380e6 versus 2.112370e6 in e71 50V mt16, for example). Make this check more sensitive. up394 - groupr. Update 379 included coding to replace a series expansion for the complementary error function with the function given in the SLATEC library (a change made many years ago in broadr but not here). The 379 change was not implemented correctly; we fix that error here. The output from test problems 2 and 11 will differ. up395 - acer. Need to increase the generic "a" array to allow processing of the latest natGd and 238U files in the IAEA's IRDFF. up396 - groupr. Continue with flux calculator corrections originally started in up105.