Prsentation PowerPoint

1
Introduction

• Cavity Lpol
• Present Status
• Idea of optimisation

S. Baudrand Polarisation Optimisation
2
Cavity LPOL
3
Cavity fabry-perot principle
e beam
L
Linearpolar
Circularpolar.
When nLaser ? c/2L ? resonance
But Dn/nLaser 10-11 ? feedback laser/cavity
By tuning the laser frequency.

4
(No Transcript)
5
ZEUS
HERMES
6
New polarimeter using a fabry perot cavity
about 60 m
96 ns
Cavity
diffused g
Calorimeter
Laser
electron bunch
driver
electron beam

• Infrared laser (l1064mm)
• Fabry-Perot cavity (2m)
• Compton scatering (e-g angle3.3 )
• Slow control light polarisation, temp
• Fast control Laser frequency for the cavity

7
Present status
8
Effect of radations at the calorimeter location
2003/2004
Hermes magnet if Off
Hermes magnet if On
9
Effect of radations at the calorimeter location
2003/2004
Hermes magnet if Off
Hermes magnet if On
We are waiting for a new calorimeter. Absorbers
in place should reduce the noise.
Cavity remain locked
10
Ideas of optimisation

• Fabian and Christian did some very rough
  estimation on simple model 3 years ago

11
Polarisation measurment

• 3 possibles experiments
• Laser on Left or Right (2s each). Time to change
  polarity is only few ms.
• Laser off (will not be implemented)

On line implemetation possible
12
Estimators

• The Polarimeter will be running continuously and
  the two (left and right) histograms will be
  issued every 4 seconds.
• dP of about 1 per mill (stat.) for summing all
  histograms for colliding bunches.
• dP of about 3 per mill (stat.) summing all
  histograms for non colliding bunches.
• dP of about 3 per mill (stat.) for individuals
  bunches but summing during one min.

?2 Estimation
Histograms are fitted on T in order to obtain a
precise value of P? and ?P? . Change can be
checked by redoing the same procedure again
Looking at the ?2 probability or change in P?.
13
Time in minits needed to get Q1
Direct comparison
T depends on the scenario chosen. The largest it
is, the more efficient is the optimisation on the
plateau. But the smallest is T the earlier the
polarisation changing can be seen
14
Conclusion

• A Fabry-Perot cavity will be used for a new LPOL
• 2003-2004 Radiation damage. 2005 shielding.
• NOW, very stable running condition (with
  temperature control in the cavity house)
• Ready to perform the alignment with the e- beam
• A likelihood program is used to extract the
  polarisation from a first principle fit (few
  photon mode) to the photon energy spectra
• The expected precision is ? 1/min/bunch (stat.)
• 10/00 min (stat.) for all bunch to optimize the
  polarisation
• Effect of synch. rads. are under studies but
  dont seems to have a dramatic effect on pol.
  measurement.
• New calorimeter needed and should be installed
  soon