Klaus Lieutenant

1
Tests of Fermi Chopper Components in McStas

• Bei dieser Präsentation wird sicher eine
  Diskussion mit dem Publikum entstehen, die zu
  Aktionsschritten führt. Verwenden Sie PowerPoint,
  um diese Aktionsschritte während Ihrer
  Präsentation festzuhalten.
• Klicken Sie in der Bild-schirmpräsentation auf
  die rechte Maustaste.
• Wählen Sie 'Besprechungs-notizen'.
• Wählen Sie die Registerkarte 'Aktionsschritte'.
• Geben Sie die zur Sprache kommenden
  Aktionsschritte ein.
• Klicken Sie auf OK, um dieses Feld zu schließen.
• Hierdurch wird automatisch eine Folie mit
  Aktions-schritten am Ende Ihrer Präsentation
  erstellt, auf der Ihre Notizen erscheinen.

McStas

• Klaus Lieutenant

2
Test by R Peacock and others

• Tests presented on MsStas user meeting in Feb
  2005
• Chopper_Fermi (A. Garrett)
• triangular shape instead of top-hat shape for 1
  slit
• FermiChopper (M. Pöhlmann)
• TOF problems
• (transmission 90 )
• Fermi_Chopper_Straight (B. Roure)
• wrong intensity
• wrong shape

3
New tests

• Test method
• Black box test
• Compared to analytical functions
• Marseguerra and Pauli
• Peters
• own calculations
• Components and treatment before tests
• Chopper_Fermi (A. Garrett)
• change function that creates triangular shape
  removed
• FermiChopper (M. Pöhlmann)
• set zero time only as an option
• efficiency introduced as a parameter that can
  be set
• Vitess_ChopperFermi (G. Zsigmond)
• converted from VITESS
• Fermi_Chopper_Straight (B. Roure) not treated

4
Config. 1 Point source, no diverg.

• Configuration 1 (nearly as tested by R. Peacock
  et al.)
• Source
• 0.01 x 0.01 mm²
• Energy 1, 10, 100, 1000 meV (?0.286, 0.9045,
  2.86, 9.045 Å)
• no divergence
• starting time 0 10 ms ( 2 T)
• Chopper
• straight, r200 Hz
• 1 slit 30 mm wide, 100 mm long, 80 mm high
• min. diameter 104.40 mm
• initial phase 0 and adjusted to reach 0
  position, when first neutrons arrive at centre

5
Check 1a time structure behind FC
Pulses at mon. 2 for E 1 meV and phase 0

• Check 1a (TOF, pulse length, intensity)
• Result for all components
• TOF correct
• pulse length correct
• pulse repetition correct
• intensity correct

Pulses at mon. 2 phase adjusted for first
neutrons
Analytical calculations
6
Check 1b blade in the centre

• Idea
• If a blade is in the centre of the beam, it
  should block the beam (of the point source)
  completely
• Simulation
• 3 blades of 2 mm (and 6 channels of 6 mm) use
• Result
• Zsigmond transmission 0, perfect

7
Check 1c Zero-time option

• Idea
• Chopper creates one pulse by setting the time
  close to zeroThis is a (virtual) source for a
  TOF instrument
• Simulation
• 10 meV (2.860 Å)

• Expected
• 1 pulse of 0.3884 ms length and rel. intensity 4
  (-0.1508 to 0.2376 ms)
• First results
• Poehlmann 2 pulses 0.232 ms and 0.156 ms of rel.
  intensity 4 - failed
• Zsigmond 2 pulses 0.156 ms and 0.232 ms of
  rel. intensity 4 - failed
• After corrections
• 1 pulse of 0.388 ms length for both components -
  correct

8
Config. 2 area source, no diverg.

• Configuration 2
• Source
• 20 x 60 mm² (W x H)
• Wavelengths 0.5, 4, 7.5, 11 21.5 Å (E327.2,
  5.113, 1.454, 0.6761, 0.3891, 0.2525, 0.177 meV)
• no divergence
• starting time 0 1.6667 ms ( 0.5 T)
• Chopper
• straight, r300 Hz
• 20 slits 0.9 mm wide, 19.1 mm long, 60 mm high
• walls 0.1 mm wide
• min. diameter 27.73 mm
• initial phase 30
• Realization
• Garrett 18 mm width of Fermi Chopper
• Poehlmann 22 channels of 0.9 mm, efficiency 0.9
• Zsigmond direct (total width 20.1 mm)

9
Check 2a pulse shape, TOF
Pulse shape for a fixed wavelength (4 Å)
Pulse shape for a fixed wavelength (4 Å)rotation
in negative direction (starting at -30)

• Results
• Garrett failed - taken out of
  consideration
• Poehlmann and Zsigmond (for both directions of
  rotation)
• time-of-flight correct
• transmission correct
• shape of pulse correct

Analytical function calculated according to
Marseguerra and Pauli using Excel
10
Check 2a different wavelengths
Pulse shape for a fixed wavelength of 0.5 Å
Pulse shape for a fixed wavelength of 11 Å

• Short wavelengths
• shape well represented
• Poehlmann Walls partly transparent - bug
• Zsigmond Still perfect absorption assumed

• Long wavelengths
• Poehlmann width too large
• Zsigmond practically perfect

11
VITESS Fermi-Ch transm. vs gate positions
No difference
V2.6
V2.5.1 overestimation
12
Check 2b large shadowing cylinder

• Configuration 2b
• same as configuration 2a except
• diameter 100 mm
• Source 4 Å, no divergence
• Results
• Pöhlmann slightly too high transmission for t lt
  tcentre.
• Zsigmond no change to config. 2a. Influence of
  cylinder is not treated

Analytical function calculated as product of
transmission of inner and outer channel(s)
13
Check 2c curved Fermi Chopper

• Configuration 2c
• same as configuration 2a except
• blades curved for 4 Å (r262.3 mm)
• Source 1.5, 4, 7.5 Å, no divergence
• Result
• Zsigmond practically perfect

Analytical function calculated according to
Marseguerra and Pauli using Excel
14
Check 2d prototype neutron (Pöhlmann)
Comparison simulations using different speeds of
prototype neutron

• Configuration 2d
• same as configuration 2a except
• prototype neutron with a velocity of 989 m/s (4
  Å) in a simulation of 0.5 Å neutrons
• Result
• Pöhlmann noise is suppressed, rest seems to be
  identical

15
Config. 3 area source, divergent

• Configuration 3
• Source
• 90 x 60 mm² (W x H)
• wavelengths 1, 4, 10 Å (E81.80, 5.113, 0.8180
  meV)
• hor. divergence 1 or 2, no vertical divergence
• starting time 0 4 ms ( 0.667 T) or 0 ms
  (pulse)
• Chopper
• straight, r166.67 Hz
• 40 slits 0.5 mm wide, 20 mm long, 60 mm high
• no walls
• min. diameter 28.28 mm
• initial phase 0 or adapted that pulse passes
  centre in position fully open (t0)

16
Check 3 divergent case
Comparison of analytical calculations and
simulation
Divergence distribution behind chopper for pulse
of incoming divergence 2.0 arriving at t0
Comparison pulse width and height in divergent
(2) and non-divergent case

• Pulse shape
• principal change correct
• maximal pulse width correct
• ratio of integral over pulse correct for 1 and 4
  Å
• Escape angle (divergence behind chopper)
• correct for Zsigmond component
• overestimated for 10 Å in Pöhlmann component

17
Summary

• Chopper_Fermi (Garrett) do not use it
• FermiChopper (Pöhlmann)
• for straight FC with absorbing and reflecting
  (not tested here) walls
• good, if missing blades width can be treated
  otherwise
• Vitess_ChopperFermi
• for straight and curved Fermi choppers with
  absorbing walls
• good, if there is sufficient computer power and
  the shadowing cylinder is not important

18
Validation

• Validation
• Chopper_Fermi (Garrett) removed from McStas
  package
• FermiChopper (Pöhlmann) validated (though not
  perfect for long wavelengths)
• Vitess_ChopperFermi validated

Thank you for your attention !
Literature R. Peacock et al., internal report at
ILL M. Marseguerra, G. Pauli, Nucl. Instr. Meth.
4 (1959) 140-150. J. Peters, Nucl. Instr. Meth.
Phys. Res. (2005)

• This research project has been supported by the
  European Commission under the 6th Framework
  Programme through the Key Action Strengthening
  the European Research Area, Research
  Infrastructures. Contract n RII3-CT-2003-505925