Transcript Energy Upgrade to 9 T - CARE-HHH

Machine Protection
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
Can the upgraded LHC be protected?
Scenarios
Intensity upgrade (from 0.58 A to 1 or 1.7 A)
Energy upgrade I (from 8.33 T to 9 T)
Energy upgrade II (from 8.33 T to 15 T)
1
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
Intensity Upgrade
On Monday 25/10/04, during an MD period to test LHC collimator material
samples placed in TT40, an extraction fault occurred. Probably due to
EMC interference the MSE extraction septa switched off the power
converter. As a result, a nominal LHC extraction pulse of 3.4e13p of
450 GeV pierced a hole in the vacuum chamber inside the QTR 4002
and damaged its coils.
 This corresponds to 0.061A or 11% of the nominal 0.584 A (respectively 4%
of the intended maximum 1.7 A)
 The stored energy of a QTR is incomparably small to a typical
superconducting magnet. The damage likewise.
 How would the beam pipe look after such an incident in LHC?
 (Don’t tell me, it can not happen!)
2
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
3004, winding short, anticryostat
Picture taken from A. Siemko, MTM
3
Intensity upgrade
 Naively speaking, intensity upgrade is an issue of collimation.
 The steady intensity, seen by the machine as losses, may not
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
increase beyond the presently set limits.
 If the current increases by up to a factor 3 or 4 the collimators
have to be effective accordingly.
 Note that there is no proof that we will be able to achieve the
promised 0.58 A! Hence, any extrapolation includes too many
unknowns.
 Caveat: The losses do have a time structure, which makes
calculations, based on averages, doubtful.
– The bunch filling scheme is relevant only for the
injection/dump elements (next talk)
– Collimator movements
– Unknown sources
4
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
Time structure of beam losses @HERA
Pictures from NIM A 351 (1994), pg 284
5
Intensity upgrade, summary
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
 The intensity upgrade may be possible, if
– everything runs smooth,
– the collimators catch all additional losses.
 However:
– accidental damage is increased,
– radiation damage in the warm parts is increased and
the machine lifetime reduced (personal dose is
correspondingly increased).
6
Energy Upgrade to 9 T
 Moderate energy increase to 7.5 TeV
 Assumption: beam current unchanged.
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
 Key issue is now the temperature margin of the magnets.
Picture from Stephan
Russenschuck
7
Energy Upgrade to 9 T
 Moderate energy increase to 7.5 TeV
 Assumption: beam current unchanged.
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
 Key issue is now the temperature margin of the magnets.
Picture from LHC Design
Report, pg 160
8
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
Temperature Margin
4 * more sensitive to
beam loss
Delta T
0.2 K
Joints
0.27 /0.32 K
Beam loss 1.12 / 0.28
9
Moderate Energy Increase
 The energy increase to “ultimate” decreases the
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
temperature margin to beam losses (all kind) by a
factor 4
 In summary:
– The machine will be able to run safely (after some
training)
• If the losses are further reduced and not
modulated
• If the losses are spread evenly and mainly in
the warm parts of the machine
• No energy extraction or quench protection
issue
10
The big step to 15 T
 No magnet design yet.
– Some rough scaling can be done.
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
 1. Assumption: The Nb3Sn magnets can absorb there own
stored energy (i.e. no energy extraction needed for each
individual magnet).
 2. Assumption: No major civil engineering possible (i.e.
confined space).
 3. Assumption: Lateral dimensions of the magnets comparable
to those of the existing magnets. (At most moderately bigger).
– Compared to 8.33 T the stored energy/length has increased by a
factor 3.25 @ 15T, assuming the same volume.
– Hence the energy density has increased by 3.25, raising the hot
spot temperature considerably.
– Larger volume seems unavoidable, which increases the stored
energy further.
11
The big step to 15 T
 No magnet design yet.
– Some rough scaling can be done.
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
 1. Assumption: The Nb3Sn magnets can absorb there own
stored energy (i.e. no energy extraction needed for each
individual magnet).
 2. Assumption: No major civil engineering possible (i.e.
confined space).
 3. Assumption: Lateral dimensions of the magnets comparable
to those of the existing magnets. (At most moderately bigger).
 First naïve attempt to protect the magnets:
– Quench detection and heaters improved, but essentially as
is.
– Energy bypass with diodes
– Energy extraction with switches and resistors.
12
Things to be considered
 There are no cold diodes above 13 kA. Unlikely that industry is
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
interested to develop radiation tolerant diodes with much
higher current, which work as expected at 1.9 K. Note that the
only manufacturer of our diodes has discontinued production.
It would be difficult to get 13 kA diodes!
 The insulation of Nb3Sn coils seems more tricky than that of
NbTi coils. The voltage is presumably limited to 1 kV.
 Keeping the max current at 13 kA, the inductance has to
increase by 3.24 to get to 15 T (same volume). Let’s assume an
increase by 4 (very demanding).
 The dump resistor is R=Umax/Imax=77mOhm.
 Using the same layout as presently the decay time  =L/R
increases by 4 (as the inductance).
 Hence the energy to be absorbed by the diodes increases by 4
and the blocks will have a mass of 4*60 kg =240 kg! Impossible
to handle in the tunnel.
13
Next attempt
 Second attempt to protect the magnets:
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
– Quench detection and heaters improved, but essentially as
is.
– Energy bypass with diodes
– Energy extraction with switches and resistors but more
often.
 Subdivide the sector not in 2 (forth and back) but 8 subdivisions
14
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
Normal operation
PC
Energy
extraction
15
Issues arising from the subdivision
 4 times more 13 kA leads, 4 times more switches, 4 times more
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
resistors
 Serious space problem in the tunnel
 What about recooling? Water pipes needed!
 Consider superconducting (HTS) switching current leads?
 In conclusion, the 15 T are quite unhandy -with 64 dump
resistors for the dipoles- but feasible, if the current is kept
around 13 kA.
 We need to invest in solutions to replace the diodes (by HTS
switches ?) if higher currents are needed or energy has to be
extracted from the magnets.
 We need to look into solutions to replace the many bulky
switches.
16
Summary
 The beam intensity creates dangerous problems already with
K H Meß AT-MEL CARE HHH-2004 Session2 Machine Protection
the present design. A further increase poses even stronger
demands on the collimator system. At higher energies things
may be become even worse, if new physics pops up.
 A moderate energy increase will be of little use, because the
lee-way (temperature margin) decreases dramatically.
 A big energy increase is possible (as far as machine protection
is concerned), if some commonly known rules are obeyed.
However it will be very difficult to fit into the present tunnel. We
may need to develop alternatives for the cold diodes (back to
the Tevatron solution with HTS switches) and for the very bulky
switches (semi- or superconducting).
–
And finally also the correctors will now need active
protection……
17