Bose 2.2 User Manual Page 135

  • Download
  • Add to my manuals
  • Print
  • Page
    / 155
  • Table of contents
  • BOOKMARKS
  • Rated. / 5. Based on customer reviews
Page view 134
strong oscillations of the Feshbach current which are damping out after around 2
ms
.
These oscillations are reduced when using a linear ramp (with a ramping time of 1
ms
),
and almost vanish when using an s-shaped ramp of the form
B
prog
(
e
t) = B
f
(B
f
B
i
) ·
h
(k + 1) · (1
e
t )
k
k · (1
e
t )
k+1
i
, (A.33)
where
e
t
=
t/T
ramp
is the time normalized by the ramping time
T
ramp
and
k
2 is the
ramping parameter that characterizes the “steepness” of the s-shaped curve. The particular
form of the ramp given by Eq.
(A.33)
ensures a smooth behaviour of the programmed
field, as both the derivatives of
B
prog
at the beginning (
e
t
= 0 ) and at the end (
e
t
= 1 ) of
the ramp vanish. For the measurements presented here, we have used the parameters
T
ramp
= 1 ms and k = 3.
Eddy currents
As mentioned before, eddy currents in the metallic vacuum chamber lead to a delay of
the real magnetic field
B
(
t
) at the position of the atoms with respect to the programmed
field B
prog
(t). The field B(t) is well described by the differential equation [36]
τ
˙
B(t) + B(t) = B
prog
(t) (A.34)
and we can measure its value at different times of the s-shaped ramp by performing
Zeeman spectroscopy
118
. As expected for the s-shaped ramp, we see no oscillations in the
measured magnetic field and the behaviour of
B
(
t
) is well described by the numerical
solution of Eq.
(A.34)
with the parameter
τ
= 0
.
25
ms
. At the time
t
= 1
ms
, i.e. at the
end of the programmed ramp, the real magnetic field has accomplished around 90 % of
the ramp. We can therefore experimentally realize well controlled ramps of the Feshbach
magnetic field on time scales around 1 ms.
118
Each value of the magnetic field shown in Fig. A.2 is the result of a spectroscopy measurement: We
vary the detuning of the probe laser in the absorption imaging and extract the resonance frequency
from a Lorentzian fit to the recorded atom numbers. From the known Zeeman shift of the magnetic
sublevels, we can eventually calculate the magnetic field at the position of the atoms.
135
Page view 134
1 2 ... 130 131 132 133 134 135 136 137 138 139 140 ... 154 155

Comments to this Manuals

No comments