Transcript Multi-cycling for SPS extraction septa 08.06.2005. B

Impact of multi-cycling on the
ZS electrostatic septa
B. Balhan, J. Borburgh
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
History (pre 2004)
18/7/2002: ZS5 damaged in LSS6, during / after scrubbing.
It was decided, that from then onwards, to reduce the risk for
the ZS’s when using high intensity beams, systematically to:
– continue to retract the girders of the SPS ZS septa
– switch off the High Voltage
– keep switched on the Ion traps at Unom and strap the
vacuum interlocks (Itraps and fast valves) to avoid trips
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
2004
During the 2004 run for the first time a P1 cycle could be
followed by a P2 cycle with a high intensity CNGS beam.
Hence SFTPRO and CNGS cycles could occur in the same
super cycle. No problems were encountered with low
intensity beams, but as the intensity was increased and the
beam ramped to nominal energy, the ZS1 in LSS2 started
to spark each cycle...
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Spark timing during SFTPRO/CNGS high
intensity run of 22/09/04
6000
SPARK TIMING [ms]
14000
5600
12000
5200
10000
8000
4800
6000
4000
2000
4400
SPARK TIMING
INTENSITY@ 400 GeV [10^10 ppp]
16000
INTENSITY
0
19:40:48
20:24:00
21:07:12
21:50:24
22:33:36
4000
23:16:48
TIME
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Parasitic tests (MDs?) since Sept. ’04
Tests took place during the nights to reproduce the
phenomenon.
Initially the same beam characteristics with girder IN
and nominal voltage on ZS were used, later on
even higher beam intensities were tried, but the
high spark rate was not observed again anymore.
Is this a result of scrubbing????
More tests are needed to answer and
understand the problem.
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Multi-cycling consequences
For combined SFTPRO and CNGS and LHC
operation:
• Impossible to move (ZS) girder when mode
of operation changes: girder displacement
not designed for multi-cycling (low
reproducibility of the position, very slow, risk
of fatigue)
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Potential solution
• Ramp down the HV on the ZS before a
high intensity beam passes by (less risks
of fatigue on HV cable and HV feedthrough or damage to cathode).
• Presently the system is not designed to do
so rapidly, and the actual response time to
go to an estimated “safe value” naturally
(for example 220 kV-> 180 kV) is 8 sec.
• The ramp up time is limited by the current
of the generator.
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
HV circuit diagram
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
HV circuit response time
To study the ramp down of the ZS HV, the
equivalent RC circuit consists of:
- a capacitor of ~70 nF (mainly the long coaxial
cables)
- a resistor of 600 Mohms (internal resistance of
HV generator)
This RC circuit has a time constant of ~40 s, and
to decrease from 220 kV to 180 kV ~8 s are
needed.
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Ion trap circuit diagram
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
LSS2 septa HV circuit discharge
220.00
200.00
180.00
Voltage (kv)
160.00
140.00
120.00
100.00
80.00
60.00
40.00
20.00
0.00
0.00
20.00
40.00
60.00
80.00
100.00 120.00 140.00 160.00 180.00 200.00
Time (sec)
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Potential solution to ramp down faster
To reduce this response time:
by connecting in // to the generator a passive
load (using HV divider or spare HV generator)
it’s possible to ramp down in half the time (for
example from 220 kV to 180 kV in ~4 s).
Side effects:
- a high DC current occurs permanently, the
monitoring of the ZS becomes more complex
- additional control and interlocks will be needed
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
HV circuit time response
t (sec)
Actual
//750M Ω
//600M Ω
0.50
217.40
215.33
214.82
1.00
214.82
210.77
209.76
1.50
212.28
206.29
204.83
2.00
209.76
201.92
200.00
2.50
207.28
197.64
195.30
3.00
204.83
193.44
190.70
3.50
202.40
189.34
186.21
4.00
200.00
185.32
181.82
4.50
197.64
181.39
177.54
5.00
195.30
177.54
173.36
5.50
192.98
173.78
169.28
6.00
190.70
170.09
165.30
6.50
188.44
166.48
161.41
7.00
186.21
162.95
157.61
7.50
184.00
159.50
153.90
8.00
181.82
156.11
150.27
220.00
215.00
Actual
Add haefly divider
210.00
Add generator
205.00
200.00
195.00
190.00
185.00
180.00
175.00
170.00
0.00
Multi-cycling for SPS extraction septa 08.06.2005.
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
B. Balhan, J. Borburgh
Other outstanding issues
Outstanding issues:
• Add interlock (BIC?) to prevent bumped beam if during
CNGS or LHC operation
• Interlock on girder position
• If ramping solution will be retained, additional interlocks
are needed to verify the absence/presence of Unom.
• ROCS may be a suitable tool to manage the
current/voltage settings
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh
Conclusions
• High intensity beam has damaged the ZS in 2002.
• Since then the Zs’s have been retracted with HV off during high
intensity operation.
• During a MD with high intensity beam in 2004, when the ZS were
not retracted and with the nominal HV on, a high spark rate was
observed. However, this could not be repeated in later sessions.
More data is required to understand the problem and eliminate the
risk for the ZS.
• Presently we cannot assume that the ZS as is, in position and with
nominal HV on, is ready for fast multi-cycling.
• A potential solution may be to ramp down the HV during high
intensity operation, but this is still a slow process that may not allow
full fast multi-cycling.
• More tests on the spark rate with high intensity beams are needed.
Multi-cycling for SPS extraction septa 08.06.2005.
B. Balhan, J. Borburgh