Operating the LHC Initially at a Lower Energy?

1
Operating the LHC Initially at a Lower Energy?

• from Chamonix 2004
• Rüdiger Schmidt AB-CO
• with M.Calvi, V.Kain and A.Siemko

Reminder Chamonix 2003 Discovery potential for
Higgs Transient losses and quench
level Reliability of machine / magnet operation
Proposals and Conclusions
presentation to LEADE 26 March 2004
2
Operating LHC at lower energies - main
conclusions from André Verdier

• Heat load for cryogenic system only marginally
  reduced when operating at lower energy
• In case of energy deposition due to continuous
  beam losses of given value, the discovery
  potential increases with the machine energy
• In case of energy deposition due to transient
  losses, the discovery potential may decrease with
  the machine energy

3
Operating LHC at lower energies

• During initial operation, say, the first year,
  the discovery potential for Higgs production is
  considered
• For a light Higgs (120 GeV), the relative
  potential R is decreased by 10 for 6 TeV
  compared to 7 TeV assuming the same luminosity
  (from particle physics, A.Verdier 2003)
• R7 TeV 1, R6 TeV 0.9
• Discovery potential and luminosity with the same
  beam parameters the luminosity at 7 TeV would be
  16 higher compared to 6 TeV. Assuming the same
  efficiency and the same luminosity lifetime, the
  Higgs discovery potential at 7 TeV is larger by
  about 20.

• The LHC is a frightening complex accelerator
  both from the hardware point of view and from
  operation
• Initially, the efficiency of the operation will
  to a large extent determine the integrated
  luminosity

challenging
4
Higgs discovery potential for case A

• assuming
• not full beam intensity (collimation and machine
  protection)
• not limited by beam-beam effects (due to limited
  intensity)
• not limited by aperture (no full squeezing to 0.5
  m, too delicate, and non-linearities in the
  triplet)
• given emittance of the beams at injection
• given number of protons per bunch at injection
• the luminosity scales with energy
• assuming same efficiency of operation

?
?
5
Transient beam losses when, where and why?

• at injection and during the ramp, due to beam
  parameters that are not constant (orbit, tune,
  chromaticity)
• at 7 TeV when collimators are adjusted
• collimators will be moved in small steps, and
  during such steps protons will be touched and
  lost
• during initial operation the collimation
  efficiency will not have been fully optimised
• transient losses at collimators particle showers
  in downstream dipole and quadrupole magnets
• at 7 TeV during squeezing and adjustments for
  luminosity preparation (shifting from dipoles to
  low-beta triplets)
• see also experience from HERA fluctuation of
  beam losses
• During all this time, transient losses could well
  cause magnet quenches

6
Beam losses and machine operation

• Quenches induced by beam losses will inevitably
  occur
• at 7 TeV, a fast loss of a tiny fraction of the
  nominal beam (10-8 - 10-7 ) would induce
  a quench in a superconducting magnet
• at 450 GeV, a fast loss of some 10-4 of the
  nominal injected beam would quench a magnet

• Quenching due to magnets for LHC start. should
  never happen
• being too close to their margin
• retraining
• etc.
• It must be easy to understand if quenches are
  beam induced or from other causes
• in particular during initial operation, quenching
  of magnets should only happen due to beam losses
  and due to hardware failures that cannot be
  avoided (quench protection, cryogenics, etc.)

See later
7
Operational margin of a superconducting magnet
Applied Magnetic Field T
Bc critical field
Bc
Normal state

Applied field T
Superconducting
state
Tc critical temperature
Tc
9 K
Temperature K
8
Increased quench margin at lower energy for
transient losses

• More margin at 6 TeV with respect to transient
  losses, determined by several factors
• Part of the beam hits a magnet less energy
  density beam energy is lower and beam size is
  larger
• more temperature margin
• Calculation for proton shower assumes (Verena
  Kain)
• 6 TeV versus 7 TeV, emittance decrease with
  energy
• Calculations for quench limit assumes (Marco
  Calvi)
• heating in inner layer
• magnet field for 7 TeV corresponding to 8.34 Tesla

9
Energy deposition by 6TeV and 7TeV proton shower
820
620
V.Kain
10
Field map of LHC dipole magnet superconducting
cable for 12 kA 15 mm / 2 mm Temperature 1.9 K
cooled with Helium
56 mm
11
Temperature margin versus beam energyinner layer
M.Calvi and A.Siemko
12
Energy margin versus beam energyinner layer
(
M.Calvi and A.Siemko
13
Energy margin versus beam energy no heliuminner
layer
(
M.Calvi and A.Siemko
14
Margin for fast beam loss between 6TeV and 7TeV

• both together, the increased margin for fast
  losses and the decreased energy deposition from
  protons showers need to be combined
• the margin could be up by a factor of 2 twice
  more protons are tolerated to be lost in magnet
  for fast losses at 6 TeV
• the real situation is more complex when the
  margin becomes smaller at higher energy (closer
  to the critical surface)
• the length of the minimum propagation zone
  decreases
• the effect of helium cooling might be reduced
• there could be less current sharing
• excellent Project Report 44 Quench levels and
  transient beam losses in the LHC magnets
  J.B.Jeanneret, D.Leroy, L.Oberli and T.Trenkler -
  1996

15
Luminosity for case B LHC operation limited due
to transient losses

• assuming
• not full beam intensity (collimation and machine
  protection)
• not limited by beam-beam effects (due to limited
  intensity)
• not limited by aperture (no full squeezing to 0.5
  m, too delicate, and non-linearities in the
  triplet)
• given emittance of the beams at injection
• number of protons per bunch could be increased,
  possibly by up to a factor of two if operating at
  6 TeV
• the luminosity scales with the energy and N2

16
. this demonstrates that operation of the LHC
initially at 6 TeV could be of interest

• depends on
• assumption for Higgs mass
• LHC limitation during initial operation (is it
  really fast losses?)
• Today coupling with MAD including a (partial)
  aperture model of the LHC and FLUKA for loss
  calculations (V.Kain)
• Tomorrow (?) coupling of the output with a
  program for quench studies (for example SPQR
  F.Sonnemann and M.Calvi)
• With such an approach it might be possible to
  calculate the quench limits with more realistic
  assumptions for beam losses

17
Other arguments for / against 6 TeV operation

• Quench propagation from quenching magnet to
  adjacent magnet is reduced less magnets will
  quench
• Faster recovery of the cryogenic system
• Faster ramping up and down (initially, due to
  quenches and hardware failures the average length
  of a run might not be very long)
• At 6 TeV, less synchrotron radiation and less
  heating of the cryogenics (marginal effect)
• At 6 TeV, less energy loss by synchrotron
  radiation
• abort gap cleaning less efficient
• if operation is with reduced current, this should
  be acceptable

18
Magnets and machine operation

• Operation with beam should start at an energy for
  which the magnet system is very robust
  regardless of the beam
• At an energy of say, 6 TeV, magnet quenches
  without beam are very unlikely
• At 7 TeV, magnet quenches are more likely
• Operation without beam should be performed with a
  current that is 200-300 A above the level
  envisaged for luminosity operation
• if hardware commissioning is done at 7 TeV, start
  at 6.8 TeV
• for example, if we intend to operate for
  luminosity at 7 TeV, we should operate the
  hardware at 7.2 TeV before
• This ensures that magnets will NOT quench if
  there is no beam loss and no hardware failure
• post mortem recording to understand origin of
  quenches !

19
Quench antenna signals

• the quench antenna (pick up coils inside the
  magnet) measures mechanical activity in the coils
• this mechanical activity does not directly
  correlate with a quench, but when there is not
  activitiy, in general no quench is observed
• the data shows the signals from the quench
  antenna for two ramps performed with the same
  magnet
• during the first ramp, constant activity and
  quench at 11.633 kA
• during the second ramp there is no activity up to
  10.8 kA, the activity starts at 11.4 kA, full
  activity at 11.8 kA
• at constant current, no activity is observed

M.Calvi and A.Siemko
20
M.Calvi and A.Siemko
21
M.Calvi and A.Siemko
22
Statistical analysis of magnet quenches

• many magnets have been cold tested
• simple statistical look at magnet quenches
• this has been done by MTM all data are provided
  from the MTM database by M.Calvi
• to get an idea about the ability of a magnet to
  operate at 7 TeV (or higher), it is of interest
  to have a look at the training curves
• it is absolutly normal for a magnet to require
  some training quenches to reach nominal field (or
  ultimate field) most magnets show an excellent
  training (few quenches to ultimate)
• for some magnets training is more difficult

23
M.Calvi and A.Siemko
24
M.Calvi and A.Siemko
25
Training of one dipole magnet
M.Calvi and A.Siemko
26
Training before / after thermal cycle of another
dipole magnet
M.Calvi and A.Siemko
27
Conclusions

• LHC experiments tell us do not change the energy
  frequently
• the LHC should start operation with an energy
  where the magnets are rock-solid
• not to decide now, wait for magnet tests and
  hardware commissioning (the magnets will tell
  us)
• beam loss studies
• before 7 TeV with beam, operate magnet system at
  7.2 TeV
• should be possibly in the shadow of physics
  operation due to sectorisation, e.g. when part of
  the machine is down
• limited retraining of few magnets would be
  acceptable
• for few weaker magnets conditioning / training
  with the LHC when other sectors are down
  (quenched) ramp to higher current
• the objective is still to go to 7 TeV after some
  time - maybe one year (and later possibly higher)

28
is it politically correct to start operation at 6
TeV ?

• . if initial operation is at 6 TeV no change
  of the name
• fortunatly we named it LHC and not the Sevatron
  neither Sixatron