Caveats and tricks

  • CASSIS does not start smoothly : This is very very rare, if it happens, just quit CASSIS and relaunch.

  • What is CASSIS doing? In some cases, there is no indication of the activity of CASSIS, such as a running watch or an hourglass. This will be added soon.

  • You cannot find an answer to your questions, neither in the Help nor in the Documentation section. In this case, contact us : cassis-team at

  • The scripts variables may have changed. If your old scripts does not work anymore, please update them using the provided scripts in "CASSIS Data and scripts" (link Documentation and Data) as examples.

  • Computing your LTE model takes a very long time : Make sure that the bandwidth and the spectral resolution you chose are not too large / high. Keep also in mind that the larger the number of species used in your template is, the longer the computation will be. The same is true for the number of components used.

  • "Check the continuum file: frequency out of range" in LTE model: if you get this warning despite the fact that the minimum and maximum frequencies of your model are within the frequencies of the continuum file, it is due to the fact that, when calculating a model, CASSIS uses a slightly larger range than the one specified in the "Tuning" panel (see point 1 of this section), to avoid "cutting" a line profile in case it is close to one of the edges of the band. More specifically, the model will be calculated over [range_min-2*fwhmmax ; range_max+2*fwhmmax], where fwhmmax is the maximum value of the FWHM column. Therefore, you need to make sure that your continuum file extend sufficiently beyond the desired model range (or alternatively, reduce your model range).

  • A RADEX model can take a long time : The default number of maximum iterations for RADEX to compute is 10000. Sometimes RADEX needs many iterations to converge, depending on the used parameters, so just be patient...

  • The number of molecules for which RADEX can be used is smaller than the one of RADEX online : This is due to a different coding of the quantic numbers into the LAMDA and the CASSIS databases. This will be homogenized in the forthcoming updates of CASSIS.

  • Your favorite telescope is not in the list : You can easily create your own telescope file. This file should have the following format (values are "dummies"):
    // Diameter (m)
    // Frequency (MHz), Beam efficiency, IF (GHz), LSB gain (%), USB gain (%)
    7.0000e+05 0.48000 8.0000 1.0000 1.0000
    9.0000e+05 0.48000 8.0000 1.0000 1.0000
    1.1000e+06 0.29000 4.0000 1.0000 1.0000
    1.3000e+06 0.29000 4.0000 1.0000 1.0000
    with columns being separated by a tab.
    Alternatively, you can just duplicate one telescope file in YOUR_INSTALL_DIRECTORY/cassis/delivery/telescope and change the parameters for those of your favorite telescope. Take care however that you do not change the format of the file !
    Note that the file you create can be placed anywhere on your computer.

  • Fitting from the FullSpectrum window can take a very long time if the number of points of the spectrum is larger than 5000. Make sure to restrict the number of points by narrowing the x-axis range in the "Tuning" part of the FullSpectrum data-loading window.
    Warning: zooming *WILL NOT* reduce the number of points.

  • Other Species:
    • Warning, this can take a long time with the default parameters as the ISM template contains many species and a lot of transitions. It is advised to either restrict the search domain with the thresholds or to make your own smaller template.
    • If several spectra are present in the "InfoPanel" tab, the display of transitions from the image band ("Species" tab) is not allowed at this stage. This will be changed in a future version of CASSIS.

  • On Macintosh : To update CASSIS from the CASSIS help menu, you have to disable the java console view in the Java Preferences. This is the default, just consider this caveat if you have changed your default Java Preferences.