>
> |
| META TOPICPARENT |
name="NotesInPreparation" |
Changes made after the Physics Meeting
| Author |
Title |
Files |
Version |
| Ferenc Siklér |
Reconstruction of low pT charged particles with the pixel detector |
 |
25 Sept |
| Ferenc Siklér |
Reconstruction of V0s and photon conversions with the pixel detector |
|
25 Sept |
General: Magnetic field
The plot shows the Bz component as function of z coordinate at the beam-line from OSCAR_3_9_8.
The field value use in ORCA is 4 T, which leads to a ratio of 4.068/4 = 1.017. This means that the reconstructed pT was smaller and was expected to be around 0.983 of the simulated value, explaining the pT bias and the problems with the invariant mass spectra. The corresponding plots are changed.
CMS Note AN-2006/100: Reconstruction of low pT charged ...
Section 2: Standard algorithm
Original text: "The standard pixel hit triplet algorithm \cite{standard} first finds hit
pairs, then for each pair it predicts the range of possible coordinates
($\phi$ and $z$ or $r$) of the third hit. The method uses a straight line
through the inner hit and outer hit of the pair in order to get the $z/r$
prediction. The $\phi$ prediction is simply the corresponding value of the
outer hit (Fig.~\ref{fig:comparison}). Tolerances are set such that they take
into account the effects of multiple scattering."
Section 3: Cluster shapes
New text: "If the maximum
and the minimum move in opposite directions or if there are empty pixels in
the column, the cluster is not compatible with a single particle hit. (These
type of clusters might be skipped, but it was not done in this analysis.) ...
(The
cluster might also be skipped if it is at the boundary of a silicon unit, but
it was not done in this analysis.)"
Section 4.5: pT bias
New text: "The reconstructed $\pt$ tends to the simulated one for higher
$\pt$, their ratio converges to 1. There is an increasing bias with
decreasing $\pt$ and for heavier particles. It is due to the energy loss of
the particle which increases with decreasing momentum and scales with $p/m$
of the particle."
CMS Note AN-2006/101: Reconstruction of V0s and photon ...
Section 2.1: Definition of truncated mean dE/dx
New text: "The energy loss
of the particles, in this case the easily calculable truncated mean $\dedx$
is obtained as follows. The ADC values of the each pixel clusters are summed.
The sum is divided by the path of the track inside the silicon. In the end
there are three values corresponding to the three hits. They are ordered in
increasing order $\DeDx_1 < \DeDx_2 < \DeDx_3$ and truncated by taking the
average of the lower "half" of the numbers (50\% truncation), that is $\dedx
= (\DeDx_1 + 0.5\DeDx_2)/1.5$. The resulted $\dedx$ value as function of
total momentum is shown in Fig.~\ref{fig:vzeroDedx}."
Section 3.2: Invariant mass distributions
New text: "The $\PKzS$ is reconstructed with the
resolution of 16~MeV/$c^2$ and at an average mass of 0.496~GeV/$c^2$. This
agrees with the nominal mass value. At the same time $\PgL$ and $\PagL$ have
a resolution of 6~MeV/$c^2$ and they are located at 1.114~GeV/$c^2$."
Section 3.3: Mass spectra at various conditions
-- FerencSikler - 25 Sep 2006
| META FILEATTACHMENT |
attachment="oscarMagneticField.png" attr="h" comment="" date="1159177466" name="oscarMagneticField.png" path="oscarMagneticField.png" size="6343" stream="oscarMagneticField.png" user="Main.FerencSikler" version="1" |
|
