[Hps-analysis] SVT ECal relative alignment

[Graf, Norman A.]

Hello Nathan & Holly,
I appreciate your looking at these plots. I'm glad to hear that the qualitative 
behavior is as expected and that the quantitative agreement is that good.
If you can send me a correction function which takes a Cluster and returns the
corrected position I'd be more than happy to incorporate it.
Thanks, 
Norman

-----Original Message-----
From: Nathan Baltzell [mailto:baltzell at jlab.org] 
Sent: Wednesday, July 29, 2015 1:31 PM
To: Graf, Norman A.
Cc: Stepan Stepanyan; hps-software; hps-analysis at jlab.org
Subject: Re: [Hps-analysis] SVT ECal relative alignment

Hi Norman,

We took a look at this and I think this is due to lack of depth/position correction (since the one in the reconstruction code is for data with b-field).

There is an analytic correction that requires only geometry (track and crystal angles) and shower depth as inputs.  And we have a measure of the shower depth for these same crystals measured from a previous experiment (albeit with photons, average energy around 1.5 GeV).
If we estimate the average geometry from the 2-d plot you showed and remember that target is 4m upstream from ecal, this gives average expected y-errors of -2.5 mm for top and +2.5 mm for bottom, and -3 mm for x.  This happens to be a good match with the residuals you measure.

Need to parameterize the crystal angles as function of x/y and then send a correction for you to try.

There also is a small x-rotation for top half of ecal from the survey which is not in lcsim.
That could explain the difference in dx for top/bottom.

-Nathan & Holly


On Jul 29, 2015, at 14:40, Graf, Norman A. <ngraf at slac.stanford.edu> wrote:

> Hello Stepan,
> Yes, the residuals are cluster - track.
> Norman
> 
> -----Original Message-----
> From: Stepan Stepanyan [mailto:stepanya at jlab.org]
> Sent: Wednesday, July 29, 2015 11:38 AM
> To: Graf, Norman A.; hps-software; hps-analysis at jlab.org
> Subject: Re: SVT ECal relative alignment
> 
> Norman,
> 
> On X-axis, is it really cluster_position-track_postion?
> 
> Stepan
> 
> On 7/29/15 1:13 PM, Graf, Norman A. wrote:
>> Hello All,
>>  I've taken the straight tracks from the field-off run 5784 and 
>> projected them to the Ecal. The residuals are plotted in the 
>> attachment using the z position of the cluster (which was essentially 
>> constant at 1393.7). Although the accuracy is quite good (~2.5 to ~3
>> mm) there are clear systematic shifts in both x and y for both top 
>> and bottom. Do these patterns make sense to anyone? I've also 
>> attached the cluster x-y positions, showing the fiducial cuts imposed on the clusters.
>> 
>> I should note that these events were reconstructed using a 
>> non-production steering file. The calorimeter positions were taken 
>> from the uncorrected cluster collections pointed to by the 
>> ReconstructedParticle objects. The reconstruction did use the
>> 
>> HPS-EngRun2015-Nominal-v2
>> 
>> detector which has the SVT survey incorporated. Have the ECal survey 
>> numbers been incorporated into this detector? If not, are these 
>> shifts commensurate with any of the measured offsets?
>> Thanks,
>> Norman
>> 
>> #####################################################################
>> #
>> ##
>> Use REPLY-ALL to reply to list
>> 
>> To unsubscribe from the HPS-SOFTWARE list, click the following link:
>> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
> 
> 
> _______________________________________________
> Hps-analysis mailing list
> Hps-analysis at jlab.org
> https://mailman.jlab.org/mailman/listinfo/hps-analysis