Similar to Parameter variations this action visualizes the consequence of fluctuations of model parameters to optical spectra. It is activated by the main window command Actions|Parameter fluctuation.

Preparation

Before you use this SCOUT action you should have a working model that computes optical spectra. The list of fit parameters should contain only those parameters for which you want to study the influence on the spectra. For each parameter you have specified the range of fluctuation by setting appropriate values for its minimum and maximum.

An example could be the following: For a layer stack you have optimized the film thicknesses. Now you want to know what spectral variations you can expect when the thicknesses are realized in a production with some tolerances, i.e. random fluctuations around their design values. In that case, all relevant thicknesses should be selected as fit parameters, and the range of fluctuation must be entered in the list fo fit parameters.

In order to demonstrate the application of the 'Parameter fluctuation' action, start SCOUT and load the final configuration of SCOUT tutorial 1, example 3. This has been saved in the file tu1_ex3_step3.s98 (you need to select the SCOUT 98 file filter in the file dialog). After loading it with SCOUT 2 you should see the following main window:

The configuration computes a reflectance spectrum for a single Ag layer on glass. The Ag layer thickness is the only fit parameter in the list of fit parameters. Set its range to 0.015 ... 0.025 (thickness in microns):

Activation

Execute the command Actions|Parameter fluctuation in the main window of SCOUT. You are asked for the number of repetitions, i.e. the number of realizations of a random parameter value and subsequent spectrum computations:

At present, values between 1 and 250 are allowed here. Leave the suggested value of 20 and press OK. That's all you have to do!

What happens

For each spectrum in the list of spectra (in the current example there is only one, the reflectance spectrum) a new worksheet in the workbook is created. The worksheet gets the name of the spectrum with the extension 'fluctuation', in the example the name is 'Reflectance fluctuation'.

SCOUT will then draw 20 times a random number between 0.015 and 0.025, assign the value to the Ag layer thickness and re-compute the reflectance spectrum. The reflectance data are stored in workbook columns, and SCOUT will compute the minimum and maximum (and their difference) of the reflectance for each spectral point. The worksheet looks like this:

In addition, SCOUT creates a new view called 'Reflectance fluctuation' and displays the minimum, maximum and difference of the fluctuating reflectance data. In the case of several spectra in the list of spectra, SCOUT will create several views with corresponding names. Click the View button in the main window's speedbar to cycle through all views.

The data display in the view is done by an object of type 'Workbook display (columns)'.  In the example, the view is the following:

After the creation of the view object, you can modify (Click on the object while the Ctrl-key is pressed) its appearance (change the graphics parameters, line types, colors and names of the individual curves, legend settings) as described above. However, if you do a parameter fluctuation computation again the graphics parameters are copied from the original spectrum object and some of your changes may be lost.

Doing it again

If you have changed the model (e.g. the range of fluctuations of some of the parameters) you can re-do the fluctuation computation by Actions|Repeat parameter fluctuation. After this command, you are not asked for the number of repetitions but the value set previously is used again.

Exporting the fluctuation data

Since the spectral data of the computation are stored in the workbook you can use the workbook mechanisms for data export to save the data to files or copy&paste to other programs like Excel.