|
Beam profile plot for tracker 1, where tracker 1 is 10x10 GEM detector. |
|
Beam profile plot for tracker 2, where tracker 2 is 10x10 GEM detector. |
|
Beam profile plot for tracker 3, where tracker 3 is 10x10 GEM detector. |
|
Hit position along x is shown for trackers 1, 2, 3 and GE1/1. |
|
Hit position along Y is shown for trackers 1, 2 and 3. |
Trk only: (png) (pdf)
|
Top left: showing two dimensional profile plot of tracker. Here, a hit is counted when a track is reconstructed, using a linear fit y=mx+b fit to the tracker hit positions, in the tracker with normalised 𝜒2 < 10. Top right: Showing hits if they are detected by all three tracker when a track is reconstructed and extrapolated to GE1/1 and if the residual of the extrapolated track on the GE1/1 surface and the actual hit found on the GE1/1 prototype is less than 5mm. Bottom: Showing the efficiency (It is the ratio of number of events detected by GE1/1 to number of events detected by all three tracker), which is nothing but the ratio of the two histogram bottom left divided by the top histogram.
|
Detection efficiency of the GE1/1 for two gas mixtures at various (iη,iΦ) sectors readout sectors as a function of Egain is presented. E_gain is the average electric field in the GEM foil and is given as E_gain = (I x R_avg^GEMgap)/D, where I is current supplied to high voltage divider, R_avg^GEMgap is average of gap resistance and D is the thickness of gem foil. Efficiency, ϵ is given as ϵ = N_{GE11+Trk}/N_{Trk}, where NTrk is the number of events where a track reconstructed, using a linear fit y=mx+b fit to the tracker hit positions, in the tracker with normalised 𝜒2 < 10, and N_{GE1/1+Trk} is the number of events where a GE1/1 hit within 5mm of the extrapolated track position is found. The filled circle series is taken with Ar/CO2 (70/30) gas mixture at (iη,iΦ) = (5,2) while the triangle, square, and open circle series are taken with Ar/CO2/CF4 (45/15/40) at (iη,iΦ) = {(1,2),(5,2),(8,2)}, respectively. At a fixed high voltage operating point the effective gain with the Ar/CO2 mixture is approximately one order of magnitude higher than with Ar/CO2/CF4 mixture. As a result the efficiency distribution shown in the closed circle series, obtained with Ar/CO2, is shifted to lower values of Egain.
|
|
|
we assumed that the time resolution of the detector behaves as a gaussian. Then we convoluted this gaussian with a piecewise function which models the 40 MHz clock. The resultant convolution gives an equation which is fitted to the experimental data. The parameters of this fit give, among other observables, the time resolution of the detector before convolution. To deconvolute the data entirely we need to know the mathematical expression of the filter that applies to the causes of convolution in the electronics. Since, We did not know the exact functional form so we used the method above, and it removes the impact of the clock. The time resolution with Ar/CO2 (70/30) reaches faster values at a lower Edrift; however a given point on the Ar/CO2 curve has a gain approximately one order of magnitude higher gain than the corresponding gain with Ar/CO2/CF4 (45/15/40) plot. We are able to reach faster timing at lower gains with the addition of the CF4 and this is important from the point of view of detector safety because at higher gains the discharge probability is higher. |