STAR QA Board

["Van Buren, Gene" <gene@bnl.gov>]

Hi, Geary

I'm not sure I understand what you mean by "not possible". The plots that you 
link to, which I assume are essentially showing the plot that I copied and 
pasted into an email earlier in my very first post in this thread (yesterday 
at 11:38pm eastern U.S. time), are made with a finitely-weighted BeamLine 
constraint (called "BeamLine3D") that pulls the vertex towards the z-axis 
(x=0,y=0) _and_ severely (but not infinitely) restricts its displacement from 
(0,0). That constraint prevents those vertices from being properly 
reconstructed ~2 mm away from (0,0). That is completely possible in software. 
But perhaps I've misunderstood what you are trying to say.

Wary of adding more possible confusion, it is however worth noting that there 
is an additional factor in looking at the distribution of reconstructed 
vertices: the number of tracks used. When I attached a plot of Vy vs. Vz to 
an email in this thread at 9:09pm eastern US time today, I was selecting 
high-quality vertices to use in determining a BeamLine, and to that end I 
deliberately cut on selecting high-multiplicity vertices. I expect that any 
plots you've seen from L4 have no such high-multiplicity cuts, and therefore 
have a distribution width that gets contributions not only from the size of 
the beam "spot", but also from the projection error of tracks to the primary 
vertex, which impacts the resolution as 1/sqrt(vertex daughter multiplicity).

-Gene

> On Dec 27, 2021, at 9:47 PM, Geary Eppley <eppley@rice.edu> wrote:
> 
> 
> 
> Hi Gene, et al.:
> 
> That x-y distribution is consistent with the beam position you sent out: 
> x=-0.5mm, y=-2mm. The x-y distribution has a radius of ~1 cm, similar to 
> L4. The plots shown at the QA meeting Friday
> 
> https://drupal.star.bnl.gov/STAR/system/files/1225_HF_QA_pp500_0.pdf
> https://drupal.star.bnl.gov/STAR/system/files/Bulkcorr_pp_500GeV_QA_week1_new.pdf
> 
> show a central x-y distribution with a radius of ~1mm. It is not possible 
> for that x-y distribution to have a displacement y=-2mm.
> 
> Is the scale of both those x-y plots off by a factor of 10 for some reason.
> 
> -Geary
> 
> 
>> Hi, Geary
>> 
>> We have FastOffline _with_ the BeamLine constraint. The result of that is 
>> the data that was shown earlier in this thread, though it essentially 
>> showed the two components of the two different BeamLine constraints.
>> 
>> Then we have data I processed _without_ any BeamLine constraint. That's 
>> what my (Vtx_z,tpc - Vtpx_z,vpd) plot was made from, and the attached Vy 
>> vs. Vx plot is made from.
>> 
>> Hope that helps,
>> -Gene
>> 
>> [cid:B7B612CD-CE31-4CF3-A0F8-43E3B32AD8BA@attlocal.net]
>> 
>> On Dec 27, 2021, at 5:58 PM, Geary Eppley 
>> <eppley@rice.edu<mailto:eppley@rice.edu>> wrote:
>> 
>> Hi Gene:
>> 
>> Thanks. The offset matches L4 which does VPD-TPC gets -2. The RMS is also 
>> about the same as L4 and about the same as Run 17.
>> 
>> If I understand the earlier emails, you use fast offline with modified 
>> parameters to get this data.
>> 
>> Do you have a Vx vs Vy plot for this data?
>> 
>> -Geary
>> 
>> Quoting "Van Buren, Gene" <gene@bnl.gov<mailto:gene@bnl.gov>>:
>> 
>> Hi, Geary
>> 
>> On Dec 27, 2021, at 1:02 PM, Geary Eppley 
>> <eppley@rice.edu<mailto:eppley@rice.edu><mailto:eppley@rice.edu>> wrote:
>> The central Vx vs Vy central distribution shown Friday has a radius of 
>> order 1mm. The L4 distribution has a radius of order 1 cm. Do you know 
>> what causes that difference?
>> 
>> I do not know exactly what L4 is doing, but my best guess is that they are 
>> using no BeamLine constraint at all.
>> 
>> 
>> We are interested in VPD-TPC Vz offset to see if any adjustments need to 
>> be made. L4 consistently shows and offset of -2 cm or less. The RMS is 
>> just slightly worse than Run 17, which is not bad considering the current 
>> problems with the active splitter.
>> 
>> What offset do you see in your analysis?
>> 
>> 
>> I'm attaching a histogram of (Vtx_z,tpc - Vtpx_z,vpd) from the data I used 
>> to determine the BeamLine last Tuesday. Hope that's useful.
>> 
>> -Gene
>> 
>> [cid:5FA1EB5A-4C18-43E5-B139-E2BF69938FDB@attlocal.net]
> 
> 
>