Rowan, Scott (2016) LHC Main Dipole Magnet Circuits: Sustaining Near-Nominal Beam Energies. Doctoral thesis, University of Glasgow.
Abstract

Crossing the Franco-Swiss border, the Large Hadron Collider (LHC), designed to collide
7 TeV proton beams, is the world’s largest and most powerful particle accelerator –
the operation of which was originally intended to commence in 2008. Unfortunately, due
to an interconnect discontinuity in one of the main dipole circuit’s 13 kA superconducting
busbars, a catastrophic quench event occurred during initial magnet training, causing
significant physical system damage. Furthermore, investigation into the cause found that
such discontinuities were not only present in the circuit in question, but throughout the entire
LHC. This prevented further magnet training and ultimately resulted in the maximum
sustainable beam energy being limited to approximately half that of the design nominal,
3.5-4 TeV, for the first three years of operation (Run 1, 2009-2012) and a major consolidation
campaign being scheduled for the first long shutdown (LS 1, 2012-2014).
Throughout Run 1, a series of studies attempted to predict the amount of post-installation
training quenches still required to qualify each circuit to nominal-energy current levels.
With predictions in excess of 80 quenches (each having a recovery time of 8-12+ hours)
just to achieve 6.5 TeV and close to 1000 quenches for 7 TeV, it was decided that for
Run 2, all systems be at least qualified for 6.5 TeV operation. However, even with all interconnect
discontinuities scheduled to be repaired during LS 1, numerous other concerns
regarding circuit stability arose. In particular, observations of an erratic behaviour of magnet
bypass diodes and the degradation of other potentially weak busbar sections, as well as
observations of seemingly random millisecond spikes in beam losses, known as unidentified
falling object (UFO) events, which, if persist at 6.5 TeV, may eventually deposit sufficient
energy to quench adjacent magnets.
In light of the above, the thesis hypothesis states that, even with the observed issues,
the LHC main dipole circuits can safely support and sustain near-nominal proton beam
energies of at least 6.5 TeV.
Research into minimising the risk of magnet training led to the development and implementation
of a new qualification method, capable of providing conclusive evidence that
all aspects of all circuits, other than the magnets and their internal joints, can safely
withstand a quench event at near-nominal current levels, allowing for magnet training to
be carried out both systematically and without risk. This method has become known as
the Copper Stabiliser Continuity Measurement (CSCM). Results were a success, with all
iii
circuits eventually being subject to a full current decay from 6.5 TeV equivalent current
levels, with no measurable damage occurring.
Research into UFO events led to the development of a numerical model capable of
simulating typical UFO events, reproducing entire Run 1 measured event data sets and
extrapolating to 6.5 TeV, predicting the likelihood of UFO-induced magnet quenches. Results
provided interesting insights into the involved phenomena as well as confirming the
possibility of UFO-induced magnet quenches. The model was also capable of predicting
that such events, if left unaccounted for, are likely to be commonplace or not, resulting in
significant long-term issues for 6.5+ TeV operation.
Addressing the thesis hypothesis, the following written works detail the development
and results of all CSCM qualification tests and subsequent magnet training as well as the
development and simulation results of both 4 TeV and 6.5 TeV UFO event modelling. The
thesis concludes, post-LS 1, with the LHC successfully sustaining 6.5 TeV proton beams,
but with UFO events, as predicted, resulting in otherwise uninitiated magnet quenches and
being at the forefront of system availability issues.

Information
Library
Documents
[img]
Preview
CERN-THESIS-2016-151.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (19MB) | Preview
Statistics

Downloads

Downloads per month over past year

Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email