rahmans1
commented
2 years ago @michael-pitt @ajentsch I have put in a initial template for the B0 ECal ( xml file and constructor. It's basically a copy of the homogeneous calorimeter under epic repository. I have chosen nominal placeholder values for the parametrization and will need some help to tune them to expected values. Here's how the system in the PR looks currently.
As you can see, there are still some issues with the parametrization. I had some ideas but haven't had the time to pursue them. I am going to focus on the B0 tracker tomorrow. So, if anyone wants to take a shot at fixing the B0 ECal parametrization tomorrow, they are more than welcome to.
I tried to divide the pacman into inner and outer radial sectors with the inner radius going full azimuth while the outer radius covering some angular span. Each ring is populated by the cuboid modules but as you can see it's not populating the inner radial sector properly and leaving gaps. We can pursue the following remedies:
1) Tweak the disk constructor. The relevant functions are in this case add_disk and fill_rectangle in the constructor.
2) Tweak the xml (play around with radial boundaries) and see if we can use the existing disk constructor to get a close enough layout.
3) Add individual modules in the xml if the position of the modules are known previously. In this case, under placements in the xml file, I believe we can add individual modules instead of a disc constructor using the following recipe. The key would be to figure out the relevant rotations and positions with respect to the global reference frame beforehand for individual modules so that it takes the pacman shape.
<individuals
x="<user_defined_x>"
y="<user_defined_y>"
z="<user_defined_z>"
rotx="<user_defined_rotx>"
roty="<user_defined_roty>"
rotz="<user_defined_rotz>"
envelope="true"
sector="1">
<module
sizex="B0ECal_CrystalModule_width"
sizey="B0ECal_CrystalModule_width"
sizez="B0ECal_CrystalModule_length"
vis="GreenVis"
material="PbWO4"/>
<wrapper
thickness="B0ECal_CrystalModule_wrap"
material="Epoxy"
vis="GrayVis"/>
</individuals>I think 3 is the least ideal solution. So, I would attempt 1 or 2 before I try 3.
wdconinc
Is your feature request related to a problem? Please describe. For the B0, either a PbWO4 EMCAL OR the preshower will be used. We will need to look at the performance of both. So when one runs a simulation with the full Far-Forward geometry, one can comment out the one not in use in the “far_forward.xml” file. Right now there is no implementation of the EMCAL in the far forward region.
Describe the solution you'd like Comment from @ajentsch For the B0 EMCAL, I would recommend looking at both the current ZDC and B0_tracker XML files to see how the geometry is setup, and reference the tutorials as needed. As a starting point, you can take the B0 tracker geometry, only use one layer, change the active material to PbWO4, change the length to the correct crystal length (like 10cm or something), adjust the segmentation to match the transverse size (2.5cm x 2.5cm, I think), and then ensure the readout/detector type is set to be a calorimeter (right now it’s set to be a tracker). Comment from @rahmans1 The B0 tracker is implemented as an assembly of trapezoids. The ZDC implementation has an envelope volume with modules placed inside. We can follow either template but depending on the number of modules, the first approach may be slow down simulation performance according to @wdconinc.
Additional context Excerpt from ECCE proposal In addition, a PbWO4 calorimeter will be positioned behind the fourth tracking layer at 683 cm from IP. The calorimeter constructed from 10 cm long 2x2 cm2 PbWO4 crystals and positioned to leave 7 cm for the readout system.