SPS Cycles for LHC Injection
The LHC baseline filling scheme has been changed
for Protons.
The following is the new scheme based on the so-called
72 bunch scheme in the PS.
These are schematic views of the SPS cycles for LHC filling and don't
pretend to be completely accurate.
Both cycles are conceived 'dedicated', in that during the time
they are run, the SPS will only serve the LHC.
Protons
The proton cycle in the SPS will consist of either 3 or 4 PS injections
at 3.6s intervals. Each PS batch will contain 72 bunches, spaced
by 25ns. The momentum at injection will be 26 GeV/c. For the
nominal LHC beam the intensity per bunch will be ~1.1x10+11.
The injection plateau in the SPS will therefore last (up to) 10.8s.
The acceleration phase is about 8.3 s and brings the beam up
to 450 GeV/c. The speed is determined primarily by the RF power requirements.
A 1 s flat-top is presently assumed. This will be used to prepare
the extraction equipment (bumpers etc.) and perform any RF re-phasing necessary
to put the beam in the correct location in the LHC.
Fast extraction is used (all beam extracted within the
same SPS turn). It will either send the beam to TI 2 (via the West Extraction
channel), or to TI 8 (via the new East Extraction channel).
For a 3-batch cycle the total intensity in the SPS will
be 2.38x10+13 protons with 26% of the SPS circumference containing
beam.
For a 4-batch cycle the total intensity in the SPS will
rise to 3.17x10+13 protons and will occupy around 35% of the
SPS circumference.
The above cycle is repeated 12 times for each LHC ring. 3-batch
and 4-batch cycles will be interleaved in the form...
334 334 334 333
... in order to fill each ring with a total of 2808 bunches.
Hence the LHC filling time will be 4.3 minutes per ring.
The bunch pattern for the LHC, SPS and PS therefore looks like
this ...
Neglecting the beam dump kicker hole, the above pattern has 4-fold symmetry.
This is important for questions of beam-beam. In this scheme, with
the exception of the beam dump hole, each of the 4 LHC experiments will
see full holes in the collision schedule. There are no crossings
where a bunch passes through an experiment during a hole in the other beam.
This, however, will occur during the beam dump hole.
Heavy
Ions (Pb82+)
In each cycle of the PS, 4 bunches of fully stripped Pb82+ ions
will be produced. Each bunch will contain 1.6x10+8 ions.
The injection energy will be 5.11 GeV/c/u. Thirteen (13) such injections
spaced by 3.6 s will be made into the SPS, giving a total of 52 bunches.
At injection the bunch spacing is 125.31 ns. As the beam is accelerated
to 177 GeV/c/u, the bunch spacing will shrink to 124.75 ns.
Once again a 1 s flat top is foreseen and fast extraction will be used
to send the beam to either TI 2, or TI 8.
The bunch pattern for Pb82+ is a little more complex! It looks
like this ...
Description
The PS will transfer 4 bunches of ions during each 3.6 s cycle into the
SPS. These will be stacked in the SPS to form a continuous train
of bunches, 52 bunches long. This batch is accelerated and passed
into the LHC.
In the LHC 3 of these batches are injected with a gap of 7 bunches
between each one (to allow for the LHC injection kicker rise time). This
ensemble forms an LHC bunch train.
There will be 4 of these bunch trains in the LHC with 8 bunches missing
between each one PLUS 25 ns. The last bunch train is shortened to
allow for the LHC beam dump kicker rise time. In this case
the final batch delivered to the LHC will consist of only 9 injections
from the PS, instead of the usual 13. In total 608 bunches per ring
will be injected. The filling time of the LHC with this scheme will be
9.5 minutes per ring.
The bunch harmonic number with 125 ns spacing is not an integer at 712.8.
However the arrangement of the bunches in the above scheme is once again
designed to generate 4-fold symmetry (neglecting again the beam dump hole).
This is acieved by arranging the bunches into 4 trains. The spacing
between the trains is not an integer number of bunches. It corresponds
to 8.2 bunches.
For comments and changes send e-mail to PaulCollier@cern.ch
Copyright
CERN
modified 11/09/2000
