TT2 working group

1
TT2 working group

• S Baird, D Berlin, J Buttkus, D Cornuet, G
  Coudert, G Daems, G Metral

2
Beams considered

• 3.5 GeV protons to AD (direct and via TTL2) D3
  -1.2 sec cycles
• 14 GeV protons to SPS (CT) D3 - 1.2 sec cycles
• 20 GeV ions and MD beam to SPS D3 - 1.2 sec
  cycles
• 20 GeV protons to nTOF D3 in dedicated - 1.2
  sec cycles, and parasitic mode - 2.4 sec cycles
• 26 GeV protons for AD, LHC, SPS D3 - 2.4 sec
  cycles

3
Cycling

• Cycling is required if an element has more than
  two values (incl. zero)
• Method is-
• If new value is below current value - set New
  directly
• If new value above current value then cycle -
• Min - Max - New
• TT2 pulsed power supplies cannot pulse in less
  than 900 msecs.
• Therefore, for all operational beams, we assume
  a maximum of 900 msecs to cycle and set the
  elements of the TT2 line.

4
BHZ247
LEP/LHC
SPS
TT10 FTS
FT12
Dump D3
3.5 26 GeV Protons direct to AD
AD
nTOF
FTA
TT2A FTN
PS
TTL2 (ATP)
BTI247
TT2 (FT16)
5
One exception - BTI247
BTI247 has three operational values zero, 3.5
GeV direct to AD and 26 GeV direct to AD. AD
can never request 3.5 GeV and 26 GeV protons in
the same super-cycle, so BTI247 does not cycle.
BTI247 is equipped with a small auxiliary power
supply to compensate the remanent field when
setting the power supply to zero. This supply is
set to 18.5 amps (re-checked in 2001
start-up) BHZ327 BHZ377 have similar aux.
supplies, which are not used
6
Initial cycling times

• Initially the following power supplies did not
  meet our 900 msecs criteria-
• Supply Cycle time (msecs)
• BHZ117 980
• BHZ147 1200
• BHZ167 960
• BHZ377 2000
• BVT123 980
• BVT173 940
• QDE210S 960
• QFO205 960
• QFO215S 1000
• QFO375 960

7
This cycle time was found to include around 100
msec delay due to the sequential nature of the
power supply control. Once CO removed this delay
only two elements did not meet our 900 msec.
criteria

• BHZ147 1200 msecs
• BHZ377 2000 msecs

8
BHZ147
LHC
SPS
TT10 FTS
FT12
Dump D3
AD
Dump D2
nTOF
FTA
TT2A FTN
PS
TTL2 (ATP)
BTI147
TT2 (FT16)
9
BHZ147
D2 is never used, therefore the minimum value
needed for BHZ147 could be increased, reducing
the cycling time It was possible to reduce the
cycle time below 900msec. BHZ147 is no longer a
problem
10
BHZ377
LEP/LHC
SPS
TT10 FTS
FT12
Dump D3
All beams to SPS
AD
nTOF
FTA
TT2A FTN
BHZ377
PS
TTL2 (ATP)
TT2 (FT16)
11
BHZ377
The cycle time cannot be decreased significantly
This restricts the 1.2 second cycles that can
be placed before beams for the SPS. Current
solution Restrict super-cycle composition Ideal
solution A new, but identical power converter,
so that both BHZ 377 and BHZ 378 can be powered
separately. Approximate cost estimation 150 kCHf
(PO)
12
BHZ247
LEP/LHC
SPS
TT10 FTS
FT12
Dump D3
3.5 26 GeV Protons direct to AD
AD
nTOF
FTA
TT2A FTN
PS
TTL2 (ATP)
BTI247
TT2 (FT16)
13
BHZ247
BTI247 does not cycle but it has three
operational values zero, 3.5 GeV and 26 GeV to
AD. The problem is that, when ramping the main
supply down to zero, the secondary supply
oscillates for 350msec after reaching zero
field. Now the only 3.5 GeV beam is for AD but
AD can never request 3.5 GeV and 26 GeV protons
at the same time. Conclusion BHZ247 is not a
problem, but we need remote control status
information for the secondary supply
14
BHZ403
LEP/LHC
SPS
TT10 FTS
FT12
Dump D3
20 GeV Protons to nTOF
AD
nTOF
FTA
TT2A FTN
BHZ403
PS
TTL2 (ATP)
TT2 (FT16)
15
BHZ403
BHZ403 is not cycled. It has only two values.
There were some doubts about the setting time
being too long to switch between TOF and D3 for
1.2 second cycles. After work by PO the power
supply for BHZ403 can switch from 20 GeV to
zero and back as needed. However the remanent
field effect in the magnet still has to be
checked with beam under all conditions.
16
Conclusions
BHZ247 needs no further work unless we have
another 3.5 Gev user for TT2 (other than AD). We
still need remote control and status information
etc.
For BHZ377 the solution would be a new power
supply for around 150 kCHf
The remanent field at zero for BHZ403 needs to
be checked with beam.