[Halld-cal] Newer plots

[Elton Smith]

Hi Tegan,

I was looking over David's shoulder when he was looking at the code to 
see where/how MC energy was converted to hits. It appears that the geant 
generated energies are passed onto mcsmear where they are converted to 
hits simply by applying an attenuation factor. (The only other 
parameter/constant involved is the ceff, which does no affect energies). 
By using the variation=mc for both mcsmear and hd_root, the programs 
should be getting the same value for generation and reconstruction from 
ccdb.

There is one other possibility for loosing energy (besides genuine 
leakage) and that is a threshold that is applied in mcsmear. If I 
remember correctly, hits are only passed onto to the output if the 
energy in a cell is greater than 3.5 MeV. On average this corresponds to 
3.5 MeV/0.03 MeV/integral count = 116 integral counts / 15 integral 
counts/peak counts = 7.8 counts. This is in the ball park, but for our 
daq threshold of 5 counts, we should have 75 integral counts * 0.03 
MeV/int count = 2.25 MeV.

I think we need to verify that the simulated hits add up to the entire 
energy deposited by geant into the active volume of the detector. My 
suggestion would be to
1) For testing, set the threshold value to 0.
2) Generate photons with a fixed energy (pick 1 GeV) and fixed angle 
(start at 90 deg, i.e. perpendicular to the BCAL) and shoot these into 
the BCAL.

- Check that all the energy generated into the active volume make their 
way though mcsmear into the output file.
- Turn the threshold on (set to 2.25 MeV) and compare.
- Consider whether (non-linear) corrections should be applied to the 
deposited energy or it is sufficient to scale outputs to match the input 
energy.
- Check the width of the energy distribution to see if it is consistent 
with our expectation, i.e. fluctuations from sampling, photostatistics, 
pedestal width, etc. If not, adjust parameters in mcsmear to match 
expectations.
- Repeat exercise with photons at other angles.
- Take a look at the pi0 peak again, but it should match up. (If not one 
needs to look into position reconstruction).

I know... I have just outlined a project, not just a task.

Cheers, Elton.





Elton Smith
Jefferson Lab MS 12H3
12000 Jefferson Ave STE 4
Newport News, VA 23606
(757) 269-7625
(757) 269-6331 fax

On 6/4/15 5:24 PM, beattite wrote:
> Yes, both the mcsmear and hd_root commands were done while the 
> variation=mc variable was set.
>
>
> On 04.06.2015 15:17, David Lawrence wrote:
>> Hi Tegan,
>>
>>   Just double checking that you used variation=mc when you ran
>> mcsmear as well as
>> when you ran the reconstruction.
>>
>> Regards,
>> -David
>>
>> On Jun 4, 2015, at 3:44 PM, beattite <beattite at uregina.ca> wrote:
>>
>>> Hello.
>>>
>>> Here are the mcsmear comparison plots using the variation=mc 
>>> variable.  There are two files.  One is the mass spectra using 
>>> cluster energies, the other is using shower 
>>> energies.<BCAL_inv_massEcluster.pdf><BCAL_inv_massEshower.pdf>_______________________________________________
>>> Halld-cal mailing list
>>> Halld-cal at jlab.org
>>> https://mailman.jlab.org/mailman/listinfo/halld-cal
>
> _______________________________________________
> Halld-cal mailing list
> Halld-cal at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-cal