Re: Attenuation length for GEANT

["George Lolos" <George.Lolos@uregina.ca>]

Hi Matt and Elke:

I also agree with Matt that we have not seen results from simulations on energy and timing resolution from cosmic rays.  However, the results from the tests we have on the number of photoelectrons can be incorporated but we don't have yet any results from SiPM's for the well known reasons of gain non-uniformity by the arrays.  So we will have to use the results from PMT's.  We are now in the process of repeating the cosmic ray measurements with the calibrated PMT's we will get from Carl when he's done and we will glue the light guide-WC systems on Module 2 to remove any varying effects due to contact variations with cookies or optical grease.  If the SiPM array we just received from SensL shows adequate gain uniformity, then we'll also couple it and extract number of photoelectrons. 

Cheers,

George

>>> Matthew Shepherd <mashephe@indiana.edu> 01/29/08 9:40 PM >>>

Hi Elke,

See answers below:

On Jan 29, 2008, at 6:51 PM, Elke-Caroline Aschenauer wrote:

> But I have a question which I have asked several times already if we  
> run
> muons through the detector and compare the results we get for the  
> geant
> based simulation for energy and timing resolution, do we come close to
> what was measured in the bcal test beam or for the current cosmic  
> tests
> in regina.

Zisis, Alex, or George would probably be best to answer this.  Note  
that in the mainstream simulation we haven't yet incorporated light  
yield and threshold information from the current, most accurate,  
cosmic tests done with SiPMs into the simulation mainly because this  
info is quite fresh.  After the review, we plan to go back and do all  
of this, but it seems like the changes aren't quite so drastic.

> The attenuation length of the FCal has a big impact on the light  
> output
> and such what the lowest energy photons we can see. Can we really  
> neglect
> all this issues if we have a final design review and want to show our
> detector has the performance we need.

Changing attenuation length from 166 cm to 90 cm as suggested by  
Richard's notes according to my estimate corresponds to a degradation  
in light yield of about 10% for showers that peak in intensity in the  
middle of the block.  Seeing as how Mihajlo recently changed the per  
block threshold in the FCAL from 100 MeV to what we think might be a  
better estimate of 30 MeV (although admittedly still under  
investigation) and noticed only small effects in reconstruction  
efficiency for physics channels such as eta pi0.  I think these  
changes are at the level of other uncertainties in our current  
understanding of the FCAL simulation.  For instance, we are proposing  
a light coupling that increases the light yield by a factor of 2-3  
from RadPhi, but again, this is based on study and simulation with  
some notable systematic error.  I'm not dismissing the need for a  
better understanding here, but I don't necessarily think that the  
attenuation length itself is the most pressing issue.  In other words,  
yes, I think it is safe to lump this issue in with others that need to  
be addressed regarding FCAL performance.  A full scale prototype with  
the actual proposed light coupling and readout and/or a beam test  
would be an excellent way to try to address these issues.  This is why  
we are enthusiastic about pursuing prototype and beam test studies.

Cheers,

Matt