Known Problems with C07.02
This page lists the known problems with this version of Cloudy. A known problem means that we know about the problem but the fix is too difficult to implement in the current version.
Format for known problems: We should include a simple description of the problem itself, avoiding computer or physics jargon, an estimate of when and how the problem will be fixed, an acknowledgment to the person who pointed out the problem if appropriate, and the date when the problem was entered on this page.
Energy is not conserved in vastly optically thick environments.
The total radiated power is less than the input luminosity when Av is 100 or larger. This will be fixed in the next major release. Thanks to Nick Abel and Norm Murray for pointing this out. 2007 Nov 26
The set temperature floor and constant temperature commands cause insanity when combined
The set temperature floor and constant temperature commands will cause constant-temperature simulations to become thermal equilibrium on second and later iterations if they both appear in the same input stream. The fix is to not include both commands. They would not normally occur in the same input deck since the set temperature floor command is only needed for thermal-equilibrium simulations. This has been fixed on the trunk and the fix will appear in the C08 release. 2007 Nov 16
Punch output may be incomplete if the code aborts during a set of grid calculations
The grid command allows a large number of calculations to be done in which one or more parameters are varied. If a particular model aborts while doing a grid the code should do whatever punch output is requested then move on to the next model. The most frequent abort in a grid will be for extreme parameters where the kinetic temperature is above or below the 3 K - 109 K temperature limits of the code. In version C07.02 the abort process may not be correctly handled in some cases, and the punch output for an aborted simulation may not occur.
Be careful to examine the output to discover any aborts that occurred during a grid run. Examine the punch output to make sure that all data are in the file. Aborted models may not have punch output.
The next release, C08.xx, will improve some details of the output produced during a grid to better aborted models. The punch grid command will indicate error conditions. Other punch output will return zero for models that did not terminate. 2007 Sept 17
The stop line command needs two emission lines
The stop line command can be used to stop a matter-bounded cloud (see AGN3 Chapter 2) when a certain line ratio is reached. The test suite has included an simulation that stops when a [C II] 157.6 micron / [O III] 5007 Angstrom intensity ratio is reached. At one time the code accepted the option to only specify one line. The Balmer beta line would be used as the second line. The test suite only tested the two-line option since this was the more difficult to implement. Use it or lose it. Dick Henry discovered that the single-line option no longer works.
The code that implements the stop line' command is based on Fortan 77 constraints and is not worth fixing. The entire command has been recoded using C/C++ features. This is much simpler with about 1/3 the code. The command will work correctly in the next major release but these changes cannot be implemented in a service release. The test suite in the next major release will include both one- and two-line tests so this will not happen again.
For now you need to specify both lines, the numerator and the denominator, of the ratio. You can specify the Balmer beta line of hydrogen by writing the second line as "H 1" 4861, but you do have to specify it.
Thanks to Dick Henry for reporting this problem. 2007 July 24.
Using the Rauch hydrogen-only model atmosphere grid
This applies to C07.02.01 as well as C07.02.00. Thomas Rauch has recently updated his hydrogen-only model atmosphere grid. The test case in the auto test suite is still based on the older version of this grid. When used with the new grid, this test case will yield botched asserts. This is not a problem with the code, it simply reflects the fact that we were testing an older version of the grid. If you want to fix the botched asserts, follow the directions on the HotFixes page.
If you want to update to the new grid, remove the old rauch_hydr.ascii and rauch_hydr.mod files, download the new hydrogen-only grid (there is no need to download anything else) and recompile the ascii and binary atmosphere files. For links and more detailed instructions see the StellarAtmospheres page.
Compiling Cloudy using the Makefile
If you compile Cloudy using the Makefile in the source directory (see MakefileDescription) and have not set the CLOUDY_DATA_PATH environment variable prior to compiling, the resulting executable will not work. You will get this cryptic error message instead:
Could not open phfit.dat for reading Is the path set correctly? [Stop in t_ADfA]
To solve this you can do two things:
1) You can keep the executable and make sure that the CLOUDY_DATA_PATH environment variable is set to the correct path every time you run it. The easiest way to do this is to include it in the startup script for your shell.
2) You can set the CLOUDY_DATA_PATH environment variable to the correct path, remove the file path.o and recompile Cloudy. After that you will not need to set the CLOUDY_DATA_PATH environment variable again.
More details (including instructions on how to set environment variables) are on the MakefileDescription page. This issue will be fixed in the c07.02.01 release.
Illuminate command and the grid / optimize options
The illuminate command can accept illumination angles in degrees or radians. When this is used with the grid or optimize options the angle must be given in radians. The next edition of Hazy will make this clear.
Molecular opacities in dust-free environments
Most molecular environments are also dusty. This assumption is behind nearly all ISM chemistry codes (see the comparison given in Roellig et al. 2007, available here). Cloudy includes a complete treatment of x-ray opacities for all elements in grains and molecules, and a detailed treatment of opacity for grains, H-, H2, and CO. But the code follows the UMIST approach for minor molecules. This includes writing many molecular photo rates as a function of the incident radiation field and the total dust extinction. The opacity contributed by minor molecules is ignored.
The innermost regions of dense circumstellar disks may be molecular and, if hot enough, dust free. Species such as TiO may become important opacity sources. Their opacities are not now included. The code is not now applicable to this environment.
This is on our to-do list but the technical problems are vast. The chemistry of TiO in the ISM has not been worked out in detail since Ti is very highly depleted. Stellar atmosphere-based codes do TiO chemistry assuming thermodynamic relationships, which do not apply to low-density environments. For now, the code is not applicable to dust-free molecular clouds when minor molecules are important opacity sources.
Many thanks to Andrea Isella for pointing out this problem.
2007 June 15
Return to KnownProblems page
Return to StepByStep instructions
Return to main wiki page
Return to nublado.org