Break down the data file 1276014332.dat into parallel parts for easier side-by-side comparisons. The driver is runall.com which runs parallel.sh to create the files. These files do not have CCD calibration corrections.
The awk script xferlook.awk processes this to create oriented data for easy comparison.
Looking at Transfer Line 6, U/D components. The data is summarized in xfer_6.ud. This includes manipulation of the data to flip the direction for those DCOPS with opposite orientation. As usual, I look at the difference between the two profile positions as a function of distance along the line, in the expectation that differences in CCD calibration and DCOPS offset will drop out.
The residuals from a linear fit have rms 270 microns (190 without the flier at Z=-6816).
awk -f xferlook.awk xfer_2 > xfer_2.B
Look at the plots, 999999999.-out the ones that look bad.
awk '{getline;a=$1;ae=$2;getline;b=$1;be=$2;getline;if(ae<9999&&be<9999)print b-a;}' xfer_2.B
awk '{getline;a=$3;ae=$4;getline;b=$3;be=$4;getline;if(ae<9999&&be<9999)print b-a;}' xfer_2.B
Look at the differences nearest the laser to find the tilts. Include those tilts in getcenter.awk. Get the centers via:
awk -f getcenter.awk xfer_2.B > xfer_2.B.list
Find a linear fit to this via
awk '{if($2<999&&$3<999)print $1,$2-$3}' xfer_2.B.list |awk -f ~/public_html/cms/tools/fitzx.awk
Use the result in finding the residuals. I use
rootnt xfer_2.B.list z:x1:x2
and plot the obvious things.
For Transfer Line 1 we have 160microns rms xfer_1.ud
For Transfer Line 2 we have 130microns rms xfer_2.ud
For Transfer Line 3 we have 210microns rms xfer_3.ud
For Transfer Line 4 we have 280microns rms xfer_4.ud
For Transfer Line 5 we have 90microns rms xfer_5.ud
Modified 06-January-2011 at 12:40
http://hep.physics.wisc.edu/~jnb/cms/06Jan2011
| Previous notes | Next notes | Main slide directory |