Global Issues: Commissioning,
Operations and Availability
WBS 2.2.1
Tom Himel
Co-conspirators
U.S.
International
Tom Himel
Janice Nelson
Marc Ross
Alberto Fasso
Eckhard Elsen
Nobuhirio Terunuma
Sebastian Schaetzel
Syuichi Ban
Toshiya Sanami
G. Xia
F. Poirier
We are part of an international collaboration.
Will emphasize U.S. work.
Tom Himel
2
Contents
What we do and interface to other
groups
Availability simulation
improvements, new results
Radiation rules and calculations
Machine Protection System
Budgets and Summary
Tom Himel
3
Our Responsibilities and Interfaces
Our main job is to make sure the ILC can be commissioned and run
efficiently
Himel is the contact person for our group to all the other groups.
For our actual work, each subject is headed by the person in bold mainly
with resources from their region.
Availability: Himel, Elsen – Specify subsystem and device availability (mostly
in BCD already). Advise on design issues that might effect availability.
Commissioning: Elsen – Work with CFS on construction and commissioning
schedule and specify temporary dumps, shield walls, bypass beam lines
needed for commissioning
MPS and fault recovery: Elsen, Himel – Do high level MPS design including
fault analysis and the effects of the faults. Detailed design and costing will be
done by the controls group for the electronics and the area groups for the
kickers and dumps.
PPS: Teranuma, Himel – Shielding design and radiation calculations.
Electronics costs will be done by the controls group. Shielding and beam
stoppers will be costed by the area groups.
Tuning: Elsen – Check that each area has enough tuning and diagnostics built
in and check how they all work together.
Transportation, people and supply depots: TBD
Number and location of dumps: Himel – Dumps are used for running, tuning,
MPS, and PPS. Coordinate decisions on their location and power handling.
Summary: Our job is to make other groups spend money to make ILC operable
Availsim Introduction (1 of 3)
The ILC will be an order of magnitude
more complex than any accelerator ever
built.
If it is built like present HEP
accelerators, it will be down an order of
magnitude more.
That is, it will always be down.
The integrated luminosity will be zero.
Not good.
Tom Himel
5
Availsim Introduction (2 of 3)
Availsim is a Monte Carlo simulation
under development for 2 years.
Given a component list and MTBFs and
MTTRs and degradations it simulates
the running and repairing of an
accelerator.
It can be used as a tool to compare
designs and set requirements on
redundancies and MTBFs.
Tom Himel
6
Availsim Introduction (3 of 3)
It includes
Component lists down to the level of magnets, power
supplies, power supply controllers, and AC breakers
Tracking of energy overheads and DR kicker overhead (20
of 21 kickers)
Repairs need access or not or can be done hot
Cool-down and start-up time for accesses
PPS regions: beam in one, people in next
Downtime planning: fix things with most bang for the buck
first. Fix more than just the item which caused the
downtime.
Recovery time is proportional to the time without beam.
Machine development (opportunistic and scheduled)
Summary outputs which tell what regions and components
caused the downtime.
Tom Himel
7
FY06 Improvements (1 of 2)
All regions now have detailed component lists,
not just DR and linac. Only cryo-plant and site
power are lumped systems
Program features added to handle more complex
decks
Sped up factor of 10 (had slowed down due to extra
components)
Allow comments in decks
Allow sub-decks which get variables set and then
copied to main component list
Add concept of subregions
Make component properties object oriented
Make it easy to change tunnel configurations
Specify minor variants all in the one excel
component file.
Simulate e+ keep-alive source
Tom Himel
8
FY06 Improvements (2 of 2)
DR in separate tunnel from linacs (but still dogbone
magnet count)
Bunch Compressors now in DR region, not linac
Keep-alive source is on e+ side
Broken global system (site power, global controls) 
keep-alive broken
E+ transport line is in both linacs and both BDS’s
Numbers of components not updated yet. Will wait until
number of magnets and power supplies settle down (2
months?)
Randomized recovery times
Tom Himel
9
Recovery times for medium downtimes
Use exponential distribution to simulate
Average Recovery Time vs
Average Downtime
Keep assumption of recovery proportional to time
Tom Himel
without beam
11
Is e+ keep-alive source needed?
Run
Number
ILC1
ILC2
ILC3
ILC4
ILC5
ILC6
ILC7
LC description
2 tunnels with min in accel tunnel;
conventional e+; Nominal MTBFs
ILC1 but table A MTBF's
ILC2 but with undulator e+ and no
keep alive e+ source
ILC2 but with undulator e+ and keep
alive e+ source 1
ILC2 but with undulator e+ and keep
alive e+ source 2
ILC2 but with undulator e+ and keep
alive e+ source 3
ILC2 but with undulator e+ and keep
alive e+ source 4
Simulated
% time
Simulated fully up
% time
integrating
down incl lum or
forced MD sched MD
Simulated
Simulated Simulated % time
% time
% time
actual
integrating scheduled opportunis
lum
MD
tic MD
Simulated
% time
useless
down
Simulated
number of
accesses
per month
30.1
14.9
69.9
85.1
67.5
80.0
2.4
5.1
4.6
1.9
25.5
13.0
7.7
2.9
20.5
79.5
68.6
10.9
1.6
18.9
3.3
16.5
83.5
78.0
5.5
1.7
14.8
3.4
17.0
83.0
78.3
4.8
2.8
14.2
3.4
16.8
83.2
78.5
4.8
2.6
14.2
3.4
20.4
79.6
69.1
10.5
1.6
18.8
3.3
Any e+ keep-alive source with bunch intensity high
enough for diagnostics to work well is OK
Tom Himel
12
Are 2 tunnels needed?
Simulated
% time
down incl
forced MD
Simulated
% time
fully up
integrating
lum or
sched MD
Simulated
% time
integrating
lum
Simulated
% time
scheduled
MD
Simulated
% time
actual
opportunis
tic MD
Simulated
% time
useless
down
Simulated
number of
accesses
per month
Run
Number
LC description
ILC8
everything in 1 tunnel; no robots ;
undulator e+ w/ keep alive 2; Tuned
MTBFs in table A
30.5
69.5
64.2
5.3
2.2
28.3
18.1
ILC9
1 tunnel w/ mods in support buildings; no
robots; undulator e+ w/ keep alive 2;
Tuned MTBFs in table A
26.5
73.5
68.1
5.5
2.0
24.4
11.1
ILC10
everything in 1 tunnel; with robotic repair ;
undulator e+ w/ keep alive 2; Tuned
MTBFs in table A
22.0
78.0
73.0
5.1
2.4
19.5
5.9
ILC11
2 tunnels w/ min in accel tunnel; support
tunnel only accessible with RF off;
undulator e+ w/ keep alive 2
22.9
77.1
72.3
4.8
2.7
20.2
3.7
ILC12
2 tunnels with min in accel tunnel;
undulator e+ w/ keep alive 2; Tuned
MTBFs in table A
17.0
83.0
78.3
4.8
2.8
14.2
3.4
ILC13
2 tunnels w/ some stuff in accel tunnel;
undulator e+ w/ keep alive 2; Tuned
MTBFs in table A
21.3
78.7
73.8
4.8
2.7
18.7
9.7
ILC14
2 tunnels w/ some stuff in accel tunnel w/
robotic repair; undulator e+ w/ keep alive
2; Tuned MTBFs in table A
17.0
83.0
78.2
4.8
2.8
14.3
3.5
ILC15
ILC9 but table B MTBFs and 6% linac
energy overhead
14.7
85.3
79.4
6.0
1.5
13.1
5.6
ILC16
ILC15 but table C MTBFs and 3% linac
energy overhead
15.2
84.8
79.2
5.6
1.9 Tom13.3
Himel
6.5
13
Should keep-alive source be on e- side?
Pros
Might be able to share some of the main
e+ source accelerator
Cons
Availsim says Int Lum decreases 0.1% negligible.
e- source for e+ DR cannot share some of
the keep-alive source accelerator
Can’t be used for early commissioning
Conclusion: a wash.
Tom Himel
14
Should all 3 DR be in one tunnel?
Pros
Less tunneling cost
Rings would probably be near IPs and central site,
so transport time would be less when repairs are
needed
Cons
When access needed to one ring, no beam can be
in other. Availsim says Int Lum decreases 0.7%
3 rings in 1 tunnel could make maintenance
difficult if not very carefully engineered.
Prefer 2 separate tunnels, but all in 1 not a
killer.
Tom Himel
15
Should ability to have people in
linac and beam in DR be dropped?
Pros
Save money on stoppers and shield walls. Would
still need same dumps for tune-up.
Would save a lot of debate about how to make it
safe for people.
Cons
Availsim says Int Lum will drop by 1.3% if both
DR’s, linacs, and BDS’s are a single PPS zone
instead of separate.
Might restrict DR commissioning during linac
construction.
Conclusion: Keep the separate PPS zones.
Tom Himel
16
Needed MTBF Improvements
Improvement
factor
Device
magnets - water cooled
power supply controllers
flow switches
water instrumention near pump
power supplies
kicker pulser
coupler interlock sensors
collimators and beam stoppers
all electronics modules
AC breakers < 500 kW
vacuum valve controllers
regional MPS system
power supply - corrector
vacuum valves
water pumps
modulator
klystron - linac
coupler interlock electronics
linac energy overhead
A for 2
tunnel
conventional
e+ source
20
10
10
10
5
5
5
5
3
Improvement
Improvement
factor B for 1
factor C for 1
tunnel undulator tunnel undulator
e+ source, 6%
e+ source, 3% Nominal MTBF
energy overhead energy overhead
(hours)
20
20
1,000,000
50
50
100,000
10
10
250,000
10
30
30,000
5
5
200,000
5
5
100,000
5
5
1,000,000
5
5
100,000
10
10
100,000
10
10
360,000
5
5
190,000
5
5
5,000
3
3
400,000
3
3
1,000,000
3
3
120,000
3
50,000
5
40,000
5
1,000,000
3%
3%
Tom Himel
17
Radiation Safety Rules
Complex and different at different labs. Here list
amount a lab worker can be exposed to.
SLAC: Normal operation < .005 mSv/hr or 10
mSv/yr; misteering < 4 mSv/hr; worst failure (18 MW
loss) < 250 mSv/hr and < 0.1 mSv/incident (that is a
1.5 second loss at full power) (shield to < 0.014
mSv/hr/kW-loss)
DESY: Average operation < 1.5 mSv/yr. Assume
losses dominated by misteering causing 100 W/m
loss for 100 hours/yr (shield to < 0.03 mSv/hr/kWloss) (assuming 5 m of line loss is equiv to point loss)
KEK: Average operation < 2 mSv/yr (what loss to assume
not known)
Conclusion: Rules differ, but limits similar. Will use
tightest: shield to < 0.014 mSv/hr/kW-loss.
Tunnel Rad calc w/ FLUKA by Fasso
• Uses full tunnel geometry
• Loss is 1 m upstream of 35 cm diameter
penetration
• Support tunnel below 3.5 m has 0.003
mSv/hr/kW which is < 0.014 OK
• Conclusion: < 5 m between tunnels is
definitely not OK. 5 m is OK, if willing to
fence off area near penetrations.
19
MPS
Multi-level architecture top-level design done
and in the BCD
Made list of failure modes
Starting failure analysis + simulation in linac.
With various failure modes (phasing, magnet
shorts, magnet settings) what will beam hit
and will it destroy it. (Europe)
Will refine MPS design based on results
Have made list of dumps needed for MPS,
PPS, tuning, and other reasons.
Tom Himel
20
Budgets etc.
Budgeted 0.95 FTE and used about
that.
Now working on long term plans. Plan
to increase to 1.5 or 2 FTEs by FY09.
Depends on how we divide up the work
internationally.
On Budget
On Schedule
Tom Himel
21