-
Notifications
You must be signed in to change notification settings - Fork 67
Release Notes
Corrected errors in CData setting heat release rate, heat of combustion, target trigger values, scalar inputs for fires.
Added support for setting flow coefficients for natural flow vents (including leakage).
Improved initialization of external density for compartments above the lowest compartment.
Corrected an error in CEdit setting default locations for new targets and detectors.
Corrected errors in CData setting fire ignition criterion, setting fire location, and setting device location.
Increased the maximum number of output summary variables in CData from 400 to 1200 by user request.
Corrected an error in CEdit writing the &MFIR CData namelist input.
Corrected an error in CEdit setting options for CFAST execution. Limiting spreadsheet output is set by the -O: option, not the -S: option.
Fixed display display of target orientation for new targets that led to writing incorrect &DEVC lines for some targets. CEdit would flag the error although the input looked correct in the GUI.
Added the ability to vary &INIT namelist variables in CData.
Corrected an error in CData that was causing spurious error messages with correct input files. Fixed writing of CData fire inputs in CEdit.
Corrected an error in CEdit that prevented selecting only desired spreadsheet output files. Spreadsheet output is now specified using the -o option rather than the -s option.
Added a new program, CData, to the installation that supports Monte Carlo analysis with CFAST. CData generates multiple files using user-specified distributions, generates scripts to run the files with CFAST, aggregates specified data into a single csv file, and provides some simple tools to aid in analysis of the data. Volume 5 of the CFAST guide describes CData in detail. At this point, CData input is only supported by entering commands into an existing CFAST base input file with a text editor. Examples are included in the manual and example files folder with the installation.
Corrected error in CEdit reading and writing &DUMP commands. These are typically only used for internal V&V runs at this point so most users won't be effected by the error.
Corrected an error in CEdit that caused compartments to be written twice when surfaces were specified as adiabatic. Inputs were still read in correctly so there should not be any impact on calculations.
Corrected an error in CFAST where certain vent opening combinations for cases with multiple vents in a single compartment would not initialize correctly.
Corrected an error in the CEdit GUI where single layer calculations was not described correctly.
Added additional error messages on reading input file errors to make it easier to determine what is causing the error.
Added the ability to specify compartment surfaces with up to three layers of different materials.
Added the ability to specify material thickness for individual surfaces or targets to override default value in thermal properties inputs.
Added the ability to specify front surface temperature of targets for use in verification and validation tests. New input is "SURFACE_TEMPERATURE".
Added command line option to limit the spreadsheet output files as desired. Use -o:CDMVW for full output. C=compartments, D=devices, M=masses, V=vents, W=walls. Default is still all five output files.
Simplified the input of target normal vectors to allow the user to simply specify where the front face of the target is facing. For example, "CEILING" or "Fire 1". New input is "SURFACE_ORIENTATION". Original normal vector input is still available for special cases.
Changed the input for wall vents to be bottom of the vent and height of the vent rather than the older sill and soffit that many found confusing. Older namelist input files are converted automatically.
Changed the initialization of interior wall node locations to not depend on the simulation time. Thermal penetration depth is now taken to be 1/2 the wall slab thickness rather than a calculation that used the simulation time as an input.
CEdit now has the ability to capture and display any Windows system errors or program crashes when running CFAST and Smokeview from the GUI.
Added additional verification and validation tests to the CFAST Validation Guide.
Corrected an error in CEdit where unit conversions were not correct in vent opening specifications if units were not the default S.I. units.
Corrected an error in verification test VVentTests.in that changed the area in one scenario to be inconsistent with the rest of the input file.
Fixed an issue where the flame height was incorrect in the _compartments.csv spreadsheet file.
Fixed an error where target temperatures in the _devices.csv output spreadsheet would only update at the printed output time interval rather than the correct spreadsheet output interval.
Reorganized and renamed the spreadsheet outputs from CFAST to better group related outputs. Spreadsheets are now organized as
- _compartments: contains layer temperature and height, species concentrations, and fire-related outputs.
- _devices: contains outputs for targets and detectors (detailed target info with -V option)
- _masses: contains mass of species in compartment layers and total trace species masses (detailed unburned fuel info with -V flag)
- _vents: contains vent flows (additional flow details with -V option)
- _walls: contains compartment surface temperatures
Added the ability for CFAST to automatically add leakage to a compartment with user input of the leakage area for walls and/or floors per the ASHRAE Handbook of Smoke Control Engineering (no leakage by default).
Added ID= to all namelist inputs. These are used to provide the newly-reorganized spreadsheet outputs and must be unique throughout an input file. Opening and saving an input file will ensure all ID inputs are present and unique.
Added a FYI = 'string' input to all namelist inputs. This has no impact on the model calculations.
Added an OVERWRITE=.FALSE. input to the &MISC input. When present, it prevents CFAST from overwriting an existing calculation.
Corrected an issue in CFAST where wall vent opening and closing would be ignored if the user specified the outside as the first vent connecting a compartment to the outside. This is now flagged by CEdit and CFAST as a fatal error.
Corrected an issue where CEdit would incorrectly change the fire time curves when there was more than one fire in a compartment. This was only an issue when a saved file was opened and the selected fire was changed on the Fires tab.
Fixed issues in CEdit where the code failed when removing a compartment that had vertical heat transfer connections and to eliminate an incorrect error message when creating a target before creating an associated thermal property.
Corrected an issue in CFAST to prevent a divide by zero when calculating solid angles for radiation when the fire is located in the corner of a compartment.
Corrected an issue where CEdit would incorrectly point targets aimed at the base of the fire to the floor instead.
Corrected vent areas sent to Smokeview. Some values were initial values rather than current time values. Fixed an issue where Smokeview did not visualize closed vents for circular ceiling/floor vents.
Added additional verification and validation tests to the CFAST Validation Guide.
Added the ability for Smokeview to display the internal temperature of targets (on by default) and compartment surface temperature (off by default).
Added an option in the &DEVC input to specify internal target location as an actual depth from the surface rather than as a fraction of the thickness of the target. This is the default written out by CEdit. Older input files are converted to depth when the file is read.
Added the current vent opening fraction to the printed output for vents. Added an output for targets to indicate whether a target back surface is inside or outside the compartment for calculation of radiation and convection.
Corrected an issue in the calculation of target temperature for targets that are located on a compartment surface (i.e., target location in x, y, or z direction is zero or the width, depth, or height of the compartment respectively). Convection calculation for the back surface of the target was incorrectly using the layer temperature rather than the external ambient temperature. With this change, calculated target temperatures can be expected to be lower for highly conductive target materials, particularly near the back surface of the target.
Fixed an issue where CEdit did not write out the current version number when writing an input file.
Fixed an issue in CEdit where removing a compartment did not work if the compartment had ceiling or floor vents. Fixed an issue in CEdit where adding a target before adding the associated thermal property resulted in an error.
Fixed an issue where ceiling/floor vent opening data was not passed correctly to Smokeview for visualization.
Corrected an error that could cause CFAST to crash with certain combinations of fire ignitions in a single input file.
Corrected an issue where vents were not opening correctly by temperature or heat flux. Time-based openings were working correctly. Corrected the units for vent opening by temperature or heat flux for ceiling/floor vents and mechanical vents. In NAMELIST input files, input units of C for temperature and kW for heat flux were not being internally converted to K and W respectively.
Corrected cut, copy, and paste in spreadsheet inputs in CEdit.
Corrected units labels in the species spreadsheet where gas species were labeled as molar fraction and should have been molar percent. Data in spreadsheets are correct.
Added an error message in CEdit when wall vents are in the wrong order (outside --> inside or 2 --> 1, for example). First compartment for wall vents should always be the lower numbered compartment based on the input order of the compartments.
Corrected the default vent location for a newly created vent. This effected only visualization.
Added a column to the primary model output spreadsheet (_n.csv) for the expected HRR. This is the HRR as input by the user before it is modified for available oxygen or by sprinkler activation.
Moved rarely used details of unburned fuel from species (_s.csv) to mass (_m.csv) spreadsheet and make it only in validation output.
Fixed an issue in CEdit where vent opening by time was not working correctly.
Fixed an issue in CFAST where ISOsurface files were not being written correctly so Smokeview would crash visualizing ISOsurfaces.
Fixed several Namelist inputs for &DIAG in CEdit that should only effect NIST verification tests.
Corrected an inconsistency in the default maximum time step in CEdit that caused the parameter to be written to the CFAST input file even when it was left at the default value of 2 s.
Added the ability for HCN yield to be less than the notional yield indicated by the chemical formula of the fuel.
Corrected the units of temperature on vent opening/closing by temperature. Cedit assumed units were in K; correct is C.
Smokeview now visualizes wall vent flows as a parabolic velocity profile by default.
Added calculation of smoke obscuration at a target's location as an indicator of tenability.
Added the ability for output times and simulation times to be real rather than just integer values.
Added a namelist input to &MISC to allow setting of fuel-specific specific extinction coefficient for smoldering and flaming smoke.
Added a better default for fire area to CEdit. For a new test case, the fire area is calculated to be consistent with a Q* of 1. User-entered values are not changed.
Corrected an issue in underventilated burning where the fire could use more oxygen that was actually entrained into a plume. This may cause some scenarios to exhibit lower heat released in the fire room and remaining burning to take place in a vent fire from the fire room.
Corrected an error in the outside pressure offset initialization for mechanical vents connected to the outside. This only impacts scenarios where the outside ambient pressure is drastically different than the inside ambient pressure.
Corrected an issue where wall heat transfer connections would not work properly with the new namelist input file format.
Corrected input units for vents opening at a specified temperature or heat flux in namelist input. Input units should be °C and kW/m^2, respectively.
Corrected the default for mechanical ventilation orientation to be vertical consistent with the user's guide.
Corrected initialization of spreadsheet output for smoke obscuration triggered detectors.
Corrected an issue in CEdit where vent opening and closing by time would not display or edit properly if there was a vent in the simulation that did not have any open/close specification (i.e., the vent was open all the time).
Corrected an error in CEdit where old format spreadsheet inputs for thermal properties would not import correctly.
Corrected an error in CEdit where fires would not show up in the list of available calculated normal vectors for targets or other fires.
Fixed an error in the fire radiation calculation when a target was very close to the calculated location of the fire's point source for radiation.
Added new verification cases for trace species production.
Completely new input file format similar to that used for FDS. See Appendix A in the CFAST User's Guide for details. Some really old input files will need to be recreated from scratch to work with this version. Newer input files should convert to the new format when you save the file.
Added calculation of gas and convected heat tenability (based on ISO 13571) at all target locations. Verification cases added for both.
Added calculation of target temperature for adiabatic targets. Verification case added.
Added calculation of smoke concentration for both smoldering and flaming smoke generation. Verification case added.
CFAST now automatically determines whether to use a wall or corner fire plume algorithm based on the fire area and proximity to a wall or corner.
Updated calculation of door jet fires to correctly account for multiple fires. Previously, it was simply assumed that door jet fires were methane fuel.
Fixed an error where CFAST would crash with a very large number (>2500) of vertical vents in a single simulation.
Fixed an error in target radiation for compartments that are offset from the origin.
Fixed an error where CFAST would calculate a lower layer temperature hotter than the upper layer when there was a very small layer at near-ambient temperature.
Fixed an error in the specification of orientation for mechanical flow into compartments. Minor effect on layer mixing art the vent. Fixed an error in the printed output for mechanical vent cross-sectional area. Calculations and spreadsheet outputs were correct. Fixed an error in trace species filtering for mechanical vent flow into the lower layer of a compartment.
Added new validation cases for wall and corner fires from the NIST/NRC wall, corner, and enclosure fires tests. Added new validation cases for wall and ceiling/floor vents from the NIST vent study tests. Added additional verification cases for fire chemistry and radiation exchange to surfaces. Corrected fire location for UL/NIST Vents study tests.
Bug fix only in this release.
Corrected an error in CEdit that did not correctly write an input files when vent opening and closing ramps were used in more than one mechanical ventilation system. In this case, CFAST would not run the file. Only CEdit changed. CFAST.exe is still at version 7.2.3.
Bug fixes only in this release.
Corrected an error in calculation of target radiation view factor for compartments that are offset from the origin. This only makes a difference for multiple compartment cases where targets are included in compartments that are not located at (0,0,0). Impact is usually small, but can cause the results to be nonsensical.
Fixed adjustment to plume temperature calculation for fires along walls or in corners. Previously the plume entrainment had this adjustment but not the plume temperature calculation.
Fixed an out of bounds error that rarely occurred in calculation of layer absorbtivity.
Smokeview updated to version 6.5.4. Ceiling/floor vents now show as round or square as specified in an input file. Included option to show lower layer temperature in visualization.
Bug fixes only in this release.
Added input checks in CFAST to ensure vents specifications were correct.
Corrected reading of very old style vertical vents in CEdit.
Corrected display of units for thermal conductivity.
Fixed display of spreadsheet outputs to make them more consistent. Removed trailing comma in spreadsheet output files.
Corrected an error in CFAST that would cause test cases to fail if they had more than 100 vents in a single simulation.
Corrected ceiling jet heat transfer algorithm to ensure empirical correlation is used within stated limits.
Bug fix only in this release.
Corrected error that caused CFAST to stop with an error message with detectors in simulations with many compartments.
Added ability to include multiple ceiling/floor vents between compartment pairs. Previously, only one connection between compartment pairs was allowed.
Added ability to change the opening fraction for any vent as a function of time. Previously only a single change was allowed.
Added ability to change the vent opening fraction for any vent based on temperature or heat flux.
Added ability to place all vents at user-specified locations for visualization. Smokeview version 6.4.2 is included with this release.
Added spreadsheet output for species mass in layers.
Added more complete output describing vent geometry in printed output file.
Corrected variable dimensioning in Smokeview spreadsheet routine for large simulations.
Corrected reporting of fire ignition and detector activation when multiple activations occur during a single time step.
Corrected handling of discontinuities in vent flow to eliminate numerical failures in V&V cases with tiny layer volumes and when layer interfaces are near vent boundaries. All discontinuity smoothing now uses the same hyperbolic tangent function for consistency. This is a fix for Issue #491 and likely numerous program crashes over the history of CFAST.
Corrected handling of pressure difference for mechanical ventilation so pressure is correctly calculated at height of vent. Previously, pressure difference was calculated at compartment floor height. Entirely new mechanical ventilation calculation internally that's more consistent with other vent calculations.
Fixed an issue where CFAST would fail if the input file contained more than 10000 blank lines.
Bug fixes only in this release.
Corrected spreadsheet header information for smoke obscuration so that units are properly listed as 1/m.
Corrected incident flux used to judge fire ignition for secondary fires to be consistent with printed output values (emissivity was not included). Since emissivity is typically near unity, this should change calculated ignition times for secondary fires by only a few seconds at most. Added incident heat flux to printed and spreadsheet output.
Fixed visualization of mechanical ventilation in Smokeview for rooms that were not square. Length and depth position (only for visualization) were swapped. No change in calculated output.
Corrected species units in User’s Guide to be consistent with those now used in CFAST. All species are output as molar (volume) percent.
Fixed error in vertical flow routine that mixed up volume and mass flows. Since gas density is typically near unity, overall results for full V&V suite change by less than 0.5 %.
Bug fixes only in this release.
Fixed a bug in the GUI where the CFAST inputs for horizontal heat transfer (the HHEAT command) were not generated correctly by CEdit. This is a rarely-used option and only impacts scenarios with highly-conductive wall materials. It's important to note that to make horizontal heat transfer work properly, you need to have an input for each connected compartment. For example, if 25% of the wall surface of room 1 is connected to 25% of the wall surface of room 2, you need to specify a heat transfer connection for room 1-->2 AND one for room 2-->1. See Issue #425 for additional discussion.
Fixed an error in the GUI where a CFAST input file created outside of the GUI with negative inputs for the compartment position would be reset to zero without notifying the user. This does not impact input files created with the GUI CEdit.
Fixed an error in the GUI where incorrect values for vent offsets and fire position could be calculated by the GUI. This should not impact CFAST runs of impacted input files since CFAST would flag these previously incorrectly calculated values as incorrect.
Added the ability for detectors to activate based on the smoke obscuration in the layer where the detector is located (previously smoke detectors were simulated as a heat detector with a small temperature rise at activation; this is still included in the model). Added validation test results for these smoke-based detectors.
Added visualization for vertical and mechanical vents into the zone model visualization in Smokeview. Smokeview version 6.3.6 is included with this release.
Updated the column headers in a spreadsheet file to improve consistency in all spreadsheet output files. Added user-defined target names to printed and spreadsheet outputs. Added smoke obscuration at the detector location to output. Added total mass output as a default output so the old –T option is not longer required.
Fixed an error where CFAST would only report the internal temperature for plate targets at the center of target. Fixed an error where CFAST would fail with vertical flow vents where one layer was of minimal volume yet CFAST tried to take significant mass from that minimal layer. Fixed an error where CFAST would stop without an error message if one of the output files was open in another program. Fixed an error where sprinkler activation would cause a crash if there was only upper layer burning in the fire room. Fixed an error where CEdit would incorrectly flag an error for a mechanical vent near the ceiling of a compartment.
Corrected an inconsistency between the code and documentation on the maximum number of fires that can be included in a single scenario and the maximum number of lines allowed in an input file. Added discussion of the derivation of the model equations and their solution to the Technical Reference Guide. Revised the documentation to better describe target and surface heat flux calculations. Added labels to CFAST plots in guides that specify the repository version of the model used to create the data for each plot.
Added an analytical verification test of layer temperature and compartment pressure in a compartment with a fire.
Bug fixes only in this release.
Fixed bugs in the CEdit GUI where the vent offset was getting reset from user desired inputs (this only impacted visualization and did not change calculated results) and where the ignition criterion for secondary fires was reset to default values for certain scenarios and user mouse clicks.
Fixed an error in the calculation of target temperatures where gas temperatures for targets very near a just-ignited secondary fire would reach unrealistic values. These temperatures are now limited to the flame temperature. Fixed an error where target spreadsheet output for fire heat flux was not reported correctly if an input file had more than 151 user-defined targets.
Fixed several issues in the compilation and running of CFAST on Linux and OS X machines. All platforms, including Windows, now use a "-" rather than "/" to indicate a command line option (for example, cfast standard.in /V ---> cfast standard.in -V).
#Version 7.0.0 (October 31, 2015)
CFAST and Smokeview are now only available in 64-bit versions. Some older PCs may not be able to run the new versions.
New input file format that includes thermal properties and fire definition in the CFAST input file. Older input files may need to be re-created or at least have a careful review of the fire inputs since the fire inputs have changed significantly (see below) and species yields cannot be automatically update.
Completely updated combustion chemistry routine to a simple combustion of a hydrocarbon fuel that can include carbon, hydrogen, oxygen, chlorine, and nitrogen. The fuel is now a direct user input rather than implied by the ratios in earlier inputs. CO, soot, and trace species yields are now input relative to fuel burned. This will require older input files to be manually edited since the actual fuel composition is not known in an earlier CFAST input file. Also, since CFAST no longer creates automatic targets (see below), any fire that is intended to ignite by temperature or heat flux criteria will need to be updated to add a new target for the secondary fire. Secondary fires are now linked to a target and the conditions of that target (surface temperature or front face heat flux) is used to determine ignition.
Plume entrainment is now based on the Heskestad plume correlations to be consistent with other inputs for flame height and plume temperature. Validation results are slightly closer to experimental results with the Heskestad plume than with the older McCaffrey plume correlation used in earlier versions of CFAST.
Plume temperature and velocity are now calculated off the plume centerline.
Ceiling jet temperature and velocity are now calculated using the Heskestad correlations from the SFPE Handbook. This provides a much simpler calculation than the correlations used in older versions of the model. Validation results are slightly closer to experimental results compared to earlier version of CFAST. A new option for ceiling jet temperature and velocity in corridors is available.
Vent flow now accounts for the buoyancy of gases flowing through a vent. Hotter gases flow to the upper layer of the destination compartment; Cooler gases flow tot he lower layer. Little change in results is obsessed for nearly all test cases.
Eliminated thermally thin targets and implicit calculation of target temperature. All targets are now thermally thick with an explicit solution scheme. Eliminated automatically-created floor, ambient, and fire targets. All targets are now user-specified and calculated consistently. For fires, this means that secondary fires are ignited based on conditions (either surface temperature or front face heat flux) at a user-specified target.
Revised calculation of radiative heat flux to targets to better calculate the view factors for the front and back of plate targets.
Corridor flow delay is no longer included in the model. This could have a minor impact on the calculation of detector activation at large distances down a corridor.
Wind effects are no longer included in CFAST. The only effect included was an initialization of vent pressures for doors and windows. Impact on results is zero on inputs files that did not include a non-zero wind speed and small in ones that did.
Simplified the calculation of convective heat transfer to be consistent with the calculations by FDS. Small changes in target and wall temperatures can be expected.
Changed calculation of point source radiation to be consistent for fires, walls, and targets. All now assume point source is located at 1/3 the flame height, consistent with empirical correlations.
Added the ability to specify compartment surfaces as adiabatic (indicating that no heat is lost by conduction through the surfaces). While this is mainly of used in verifying the model, it does change the results for simulations where surfaces were specified as OFF since this also corrected a bug in the calculation of surface heat exchange for surfaces specified as OFF.
Spreadsheet output has been enhanced to include additional details for door jet fires and target calculations. All species output is now in volume percent/fraction rather than PPM for some species.
Changed default smoke detector characteristics in the GUI to a temperature rise of 10 C and RTI of 5, consistent with NRC NUREG 1805.
Smokeview was updated to version 6.3.1, consistent with current FDS release version. This includes new vent flow visualization, slice files that can show 3-D temperature and velocity within compartments, and isosurfaces of constant temperature within compartments. Only visualization of door and window vent flows are included at the moment. Visualization for floor/ceiling vents and mechanical vents will be included in a future release.
Fixed an error in the calculation of layer absorptivity where the calculation was using unburned hydrocarbon concentration rather than water concentration. Small changes in gas, wall, and target temperature can be expected.
Fixed an error in the initialization of relative humidity where the water content was not set for the outside.
Fixed an error in wall and corner entrainment where radiation would go to infinity at the wall resulting in a crash for some test cases.
Corrected an error loading certain files. The first listed detector RTI and activation temperature was set to default values for detectors. For some input files, thermal properties of the first compartment could be reset to OFF in the GUI on reading the file.
Corrected an error in the calculation of trace species filtering in mechanical ventilation systems where the systems had fans flowing in both directions between a pair of compartments (two fans flowing in opposite directions).
Updated the entire CFAST code to Fortran 2008.
Completely revised verification test cases and added numerous additional validation cases to the test suite. Automated the testing of model V&V consistent with the processes used for FDS.
Completely new documentation now in LaTeX to facilitate editing and to automate document creation.