Updating README.md

examples/03.IAXO/README.md

@@ -1,9 +1,3 @@
-
-## Available configuration files
-
-- `BabyIAXO.rml`: It will serve to reproduce the sensitivityof BabyIAXO data taking phase. We include vacuum phase and a combination of vacuum and 73 density settings that extend the axion search up to 0.25 eV.
-- `IAXO.rml`:
-
 This example is used to reproduce the different scenarios considered inside the "E. Armengaud et al JCAP06(2019)047, Physics Potential of the International Axion Observatory". Those scenarios are summarized in the following table, where some values, concerning detector efficiency, $\epsilon_d$, have been replaced by just window detector efficiency, $\epsilon_w$, since the detailed detector response is considered and constructed using [restG4](https://github.com/rest-for-physics/restG4/tree/master), see example [14.DetectorResponse](https://github.com/rest-for-physics/restG4/tree/master/examples/14.DetectorResponse).
 
 Parameter | Units   | BabyIAXO | IAXO baseline | IAXO upgraded |
@@ -17,8 +11,10 @@ $\epsilon_w$ |         |   0.85      |      0.95     |     0.95      |
 $\epsilon_o$ |         |   0.35      |      0.7      |     0.7       |
 a            | $cm^2$  |  2 x 0.3    |    8 x 0.15   |   8 x 0.15    |
 $\epsilon_t$ |         |   0.5       |      0.5      |     0.5       |
-t            | $year$    |   3+3       |      6+6      |    10+10      |
+t            | $year$    |   3+3       |      5+5      |    10+10      |
+ File       |     |     BabyIAXO.rml  |  IAXO.rml  |  IAXOPlus.rml  |
 
+We consider a tracking of 12 hours which is why we define $\epsilon_t$ efficiency, and we consider a data taking efficiency of 300days out of 365 natural days available.
 
 ## Vacuum sensitivity curve generation
 
@@ -38,9 +34,17 @@ restRoot
 [1] GenerateSignalComponents( "BabyIAXO.rml", "GasSignal" );
 ```
 
-Then, inside `TRestSensitivity` we define an experiment list, with a
-column which defines the signal to be used, with common background
-and common exposure time.
+The exposure times at each density setting are adjusted to follow the KSVZ line trend. It is `GenerateSignalComponents.C` the responsible to perform this calculation and to create a `settings` file that will define the exposure times, and the corresponding TRestComponent signal, that will be also generated by this macro for each density setting, and stored in a ROOT file.
+
+The `GenerateSignalComponents` macro defines a default total exposure time that will be distributed between the different density settings. For example, in order to generate the signals and settings file for IAXO we would need to specify that the gas phase takes in total 5 years of 300days and 12 hours, and express it in seconds.
+
+```
+restRoot
+[0] .L GenerateSignalComponents.C
+[1] GenerateSignalComponents( "BabyIAXO.rml", "GasSignal", 5*300*12*3600 );
+```
+
+The first density setting will be skipped since it exceeds the total time required to reach the KSVZ line. We may also change the number of skipped settings. Then, inside `TRestSensitivity` we define an experiment list, the `settings` file with a column which the calculated exposure time, and the signal to be used, with a common background defined inside <TRestSensitibvity>.
 
 ```
 [0] TRestSensitivity sens("BabyIAXO.rml", "CombinedPhase");
@@ -51,3 +55,4 @@ and common exposure time.
 **Hints**
 - A detailed x-ray detector response is included in the signal calculation. The pre-calculated response matrix will be convoluted with the axion energy spectrum resulting in a calculation that is 150 times (number of reponse matrix bins) more expensive. If removed, calculation should achieve better computational timing.
 - The number of parameter nodes, mass values at which the signal is calculated is of the order of 500 points, which will lead to a HD curve. Reducing the number of points by increasing the parameter `stepParameterValue` will also reduce the computational cost.
+- The parameter `useAverage` has been enabled for all examples, meaning that the tracking mock MonteCarlo data will be generated using exactly the average number of background counts, thus representing an average experiment but not a realistic data taking program. Still good enough to present prospects. A more complete exclusion could be achieved by disabling the `useAverage` parameter and generating several sensitivity curves, using `TRestSensitivity::GenerateCurves(N)` and then drawing them using `TRestSensitivity::DrawLevelCurves`.