sPHENIX HCal discussion

[Eric Mannel <mannel AT bnl.gov>]

Stefan-

At some level the channel numbering is fixed and a channel map will be 
needed to map a hardware channel into an analysis channel (phi-eta 
space?).  The data coming  up will have three identifiers: 1) a channel 
number (0-63) 2) a slot number (0-20) and a packet ID. The first 2 are 
fixed in the digitizer hardware and the last is fixed by which DCM-II 
channel that reads out the XMIT board associated with the digitizer 
board group and assigned in the configuration of the DCM-II group. Which 
tower maps into which digitizer channel/module is defined by the 
internal wiring and the way we connect up the external signal cables.

The other issue that will drive the cabling of iHCal/oHCal is that each 
digitizer board needs to read out an 8x8 array of towers to generate the 
2x2 tower sums at the digitizer level to provide the trigger primatives 
for the LL1 system.  Since there are only 48 channels in an HCal module, 
this means that each digitizer board needs to read out 1/3 of 4 
consecutive modules in phi. Assuming 2 digitizer boards per XMIT group, 
an XMIT group will cover will cover an region of 16x8 towers in eta-phi. 
The digitzers crates are located on the east and west sides, so each 
crate will readout the 16 modules on the corresponding side.

As you pointed out, phi=0 is in the center of an HCal module, so 
channel=0 for that module will not be phi=0.

Eric

On 1/4/2022 9:10 PM, Stefan Bathe via sPHENIX-HCal-l wrote:
> Dear All, (specifically Xiaochun, Eric, Chris)
>
> Xiaochun asked in today’s HCal Meeting about the HCal channel mapping for his 
> Geant simulation.  As far as I can tell, this was last discussed in the 
> 8/27/19 HCal Meeting:
>
> https://indico.bnl.gov/event/6688/
>
> where both Eric and I gave presentations on our ideas of how to map hardware 
> channels to digitizer channels.  I think we basically agreed.  In the end, 
> though, I think, it might be up to the software group (Chris?) to decide 
> which numbers to assign to the channels.  I guess the two options would be to 
> either adopt the digitizer channel numbering (no additional map needed, but 
> geometrically more complex) or to have numbers simply increment in z and phi 
> (geometrically simple, but additional map needed).
> The other open question is where to start counting.  Naturally one would like 
> to start counting at phi = 0 (and at the South end, so along the z axis).  
> But the sector at phi = 0 (and at phi = 180 degrees) have their outer radius 
> combs vertical.  Therefore phi = 0 is in the center of a sector, not at the 
> edge.  This would introduce an offset with respect to the digitizer channel 
> number, which is ugly.  The answer to this question might also depend on 
> which sectors get plugged into which digitizer boards/crates, which Eric and 
> Steve would know.  (I’m saying this because it might make sense to start 
> counting at channel 0 of a digitizer board.)
>
> Regards,
> Stefan
>   
> -
> ---------------------------------------------------------------------------------
> Stefan Bathe
> Professor of Physics
> Baruch College, CUNY
> and RIKEN Visiting Scientist
>
> Baruch:                                     BNL:
> 17 Lexington Ave                      Bldg. 510
> office 940                                  office 2-229
> phone 646-660-6272                phone 631-344-8490
> ----------------------------------------------------------------------------------
>
>
>
> _______________________________________________
> sPHENIX-HCal-l mailing list
> sPHENIX-HCal-l AT lists.bnl.gov
> https://lists.bnl.gov/mailman/listinfo/sphenix-hcal-l
--
Eric J. Mannel, Ph.D.
Research Engineer
Dept of Physics/PHENIX
Brookhaven National Laboratory
631/344-7626 (Office)
914/659-3235 (cell)