[esa-t474] BPM frequencies

[Bino Maiheu]

	Dear All,

As a first step in the analysis of the July 07 data, I've started by
having a real close look to the frequencies of the waveforms as they are
needed for the calibration of the BPMs and I want to understand any
systematics as good as possible.

I've fitted every BPM waveform with a decaying exponential, using a
constrained fit with the frequency and width of the fourier transform
peak as initial input. This fit, although somewhat slow, seems to be
working fine. 

The FFT spectra, frequencies etc are available from

http://www.hep.ucl.ac.uk/lc/T474/EsaData/fits.0/

where the fit parameters of each fit are stored in the root files as
well. I'll upload macros into cvs later on to plot the fit functions on
top of the waveform and FT. 

So far, I've plotted these frequencies versus pulse number and updated
the data quality web-database with the plots. So far runs 2725->2850 are
there, with the others on the way (takes about 50-60 hours for a single
average run). I've made a cut on the chi2/ndf for plotting the
frequencies, showing only 'decent' quality fitted frequencies. 
As Yury and Zen figured out, we were running on internal clock, so I
have taken this into account, the IF frequency values shown are in MHz
( note the fact that we were undersampling for all but BPM7 ).

http://www.hep.ucl.ac.uk/lc/T474/EsaRunInfo/show_dq.php

( currently 2735 -> 2850 )

I'll make a plot of the average frequency per run with the error, but i
will need to look a bit in detail since some of the distributions are
not nicely Gaussian, so that I pick the correct frequency for accurate
calibration of the DDC algorithm. 

One thing i noticed, see e.g.:

http://www.hep.ucl.ac.uk/lc/T474/EsaRunInfo/show_dq.php?run=2725&plot=7

is that the frequency for BPM 12/24 waveforms seems quite unstable, I
suspect this to be due to changes in the LO for these channels. Also,
once can clearly see a double frequency structure for BPM 7. 

Anyway, as I said, I'll determing the averages to use as input for the
DDC from this as Mark learned that this gives the best resolution. I'll
add the plots for the decay constants and t0 from the fits as well. 

I'll keep you informed of my progress.

Bino

-- 
                                                  !!!!!
                                                 ( o o )
          @ -- Bino Maiheu, Dr. ;) ---------oOO----(_)----OOo------- @
         /  :: WORK                .::  /  :: HOME             .::  /
        /  University College London   /                           /
       /  Physics and Astronomy       /  Flat 23, Harwood Point   /
      /  Gower Street                /  307 Rotherhithe Street   /
     /  London WC1E 6BT             /  London SE16 5HD          /
    /  [tel] +44 20 7679 3454      /  United Kingdom           /
   /  [fax] +44 20 7679 7145      /  [mob] +44 7747 75 58 78  /
  @ -------------------------------------------------------- @