[G14_run] [EXTERNAL] Re: HD target lengths for Silver and Gold

[Alessandra Filippi]

Hi Tsuneo,
thank you for your message, I'll look at it into detail and come back to 
you soon. Just as a quick reply:

> Q1. At your analysis note of the chapter 4.1.6 at p. 44, you are writing the HD target lengths
> 
> of 4 cm (Gold) and 5 cm (Silver).  They may be right, but I wonder how do you define them ?

well, I don't define them. I took for good the description given in the 
wiki (I have never seen these targets myself!), and decided for a common 
fiducial volume that could make sense for both the targets excluding the 
effect of the entrance/exit windows (I assume that at least the second 
one could be "rounded", so that the volumes are not exactly cylindrical). I 
have also a presentation by Andy about the targets HD content/densities 
and the different contribution of the aluminum walls, and I tried to comply
to it (of course, now I cannot tell wher it retreived it from).


> Q2. When you use both of data, Silver and Gold, to extract asymmetries, I wonder if we have to consider the amount
> 
> of HD material lost for Silver target.   I did not read the whole anaysis note yet and you may have already considered
> 
> this.

yes, the different "length" (however you  define it) of the two targets is 
taken into account as an additional 4/5 ratio multiplying the ratio of 
luminosities. I cannot remember if this was already written somewhere,
or I still have to describe it in the (missing) paragraph on luminosities 
normalization. Men at work ;-)



> 
> 
>  I am attaching three PDF files.   I think I used the same cuts for (1) and (2), but don't remember which W-bins were
> 
>  used.
> 
> 
> (1) Silver_Zvertex.pdf
> 
> 
> (2) Gold_Zvertex.pdf
> 
> 
> (3) G14_tgt_cell_dimensions.pdf
> 
> 
>  For (1) z vertex distributions from MVRT for HD Silver target run (black), empty_b target run (blue)
> 
> and subtracted (red) corresponding to the pure HD.  Smaller entries between z = -11 and -8 cm than z = -8 cm
> 
> and -5 cm are coming from the fact that HD material was evapolated and lost during one of the target transfers.
> 
> 
>  For (2) z vertex distributions from MVRT for HD Gold target run (black), empty_a target run (blue)
> 
> and subtracted (red) corresponding to the pure HD..
> 
> 
>  We condensed the similar amounts of HD materials to the call 21a and 19b, but for 21a we lost some.
> 
> 
>  (3) shows the dimensions of the three target cells used for g14.   21a and 19b were used for Silver and Gold
> 
> runs, respectively.   According to the report, there are differences of location of the HD target materials 
> 
>  (Silver and Gold) due to the different dimensions of the target cells (21a and 19b).  
> 
> The data for empty-a and -b were taken with the cell, 21a; the locatios of entrance and exit windows for Silver 
> 
> and Empty-b are the same, while those for Gold and Empty-a are slightly different.    
> 
> From (1) and (2), it is hard to see clear differnces of the window position, because the differences are mm 
> 
> and may be too small to see on them.
> 
> You may consder these things for the empty subtractin of Gold target and vertex cut.
> 
>  
> 
>  I think for Gold target, the flat region of the distribution is between z = -10 and -6 or -5.5 cm and we may define
> 
> the length of 4 or 4.5 cm.
> 
> We condensed the HD material to the call to have 5 cm of length included the shrinkage of HD
> 
> material when frozen. For Silver target, it may be hard to define the length, because the evaporated HD may have
> 
> frozen in the upstream region like z = -10 cm.
> 
> 
>  My opinion is to use 4.5 cm for the both target lengths and to describe the HD material conditions in the analysis note.

thank you, I'll take a look to your material and add the description in the 
note (this part for sure is missing). The fiducial z for both targets was 
chosen as -9.5 < z < 5.8 cm, which means 3.7 cm, even less what you 
suggest, so this could be conservative with respect to 4.5 cm.
I have tested, to evaluate a possible systematic error, what happens
when this range is further narrowed (it is one of the largest sources of 
systematic error). It looked to me that widening this range could let
more background in through the windows (see for instance fig. 27), but I need
to check better now that you tell me.
In any case, this could be treated as a further source of systematics.
I'll investigate this further and let you know what I find out.
Thanks, cheers
     Alessandra