Also on behalf of V. Senaj & L. Ducimetière Effect of 25ns, 50ns & 75ns Beam Bunch Spacing upon MKDV Kickers: MD August 12th to 14th, 2008 M.J. BARNES, AB/BT Also on behalf of V. Senaj & L. Ducimetière Mike Barnes, AB/BT APC: October 3, 2008

Principle of Beam Dumping II. Principle of Beam Dumping When the beam cannot be extracted: dumping of the beam using the MKD beam dump system (MD, emergencies...) Function of the kicker magnets: Sweep the beam to distribute the beam energy over a large volume of the absorbed block. MKDV1: V=56mm, H=75mm, 5cells each ~51cm MKDV2: V=56mm, H=83mm , 5cells each ~51cm MKDH1: V=56mm, H=96mm MKDH2: V=56mm, H=83mm MKDH3: V=60mm, H=105mm H (vap) H (hap) V (vap) V (hap) Mike Barnes, AB/BT APC: October 3, 2008

MKDV Temperature & Pressure on 12, 13 & 14th August 2008 Normally: MKDV1 temperature>MKDV2 and pressure < 6e-8 mbar (as per 25ns beam); For 75ns beam MKDV2 temperature > MKDV1 and pressure < 6e-8 mbar; For 50ns beam MKDV temperature is not an issue, but pressure is an issue!. 8s resolution for pressure. See slides 6, 7 & 8 8s resolution Mike Barnes, AB/BT APC: October 3, 2008

MKDV 1 cell No measures taken to reduce beam coupling impedance to ferrite; No transition pieces, between magnet tank and magnet frame, installed in LSS1 (now installed in spare MKDV1 magnet)  ~16cm gap at each end. MKE magnet: Mike Barnes, AB/BT APC: October 3, 2008

MKDV: Longitudinal Impedance Courtesy of Giulia Papotti: All bunch spacing hit: ~440.6MHz and ~480.7MHz ZRL Amplitude is already low at 500MHz. 25ns bunch spacing. ~40.05MHz spacing of lines 50ns bunch spacing. ~20.03MHz spacing of lines Also hits: 420.6MHz and 460.6MHz ZRL 75ns bunch spacing. ~13.5MHz spacing of lines Also hits: 334.0MHz and 387.1MHz ZRL Longitudinal Impedance (ZRL) Theoretical Calculation: CERN-SL-2000-04 AP, by H. Tsutsui, pp7-10. MKDV1 theoretical ZRL fits baseline of measurement reasonably well. Cell length of 51cm  (λ/4) resonance, for open circuit, of 147MHz (assuming μr & εr of 1). “Spikes” may also be due to lack of transition pieces?? 11.5MHz Mike Barnes, AB/BT APC: October 3, 2008

MKDV Pressure on 14/08/2008, between 01:20hrs & 02:10hrs. 1s resolution Started 1 batch 50ns beam at 1:32hrs, continuously. Single shot of 4 batches of 50ns beam at 2:05hrs: interlock. One shot of 2 batches of 50ns beam at 1:42hrs. One shot of 3 batches of 50ns beam at 1:43hrs: interlock. Information from: Giulia Papotti: Summary: 50ns beam causes pressure rise; high pressure with 3 or 4 batches per shot. Mike Barnes, AB/BT APC: October 3, 2008

MKDV Pressure on 14/08/2008, between 05:00hrs & 07:00hrs. 1s resolution Started 1 batch 50ns beam at 5:13hrs. One shot of 3 batches of 50ns beam at 5:39hrs. Four shots of 4 batches of 50ns beam at 5:41hrs: interlock. Four shots of 4 batches of 50ns beam at 5:45hrs. Two shots of 4 batches of 50ns beam at 5:49hrs. Information from: Giulia Papotti: See next slide One batch per shot of 50ns beam from 6:05hrs to 7:00hrs. Summary: 50ns beam causes pressure rise; high pressure with 3 or 4 batches per shot. Mike Barnes, AB/BT APC: October 3, 2008

MKDV Pressure on 14/08/2008, between 05:49hrs & 05:50hrs. Limited (1s) resolution for pressure; May miss peaks on graph (interlock requires few 10’s of ms); Fast pressure rise and fall due to surface effect (e.g. electron cloud), or gauge misreading?? Two shots of 4 batches of 50ns beam at 5:49hrs?? Mike Barnes, AB/BT APC: October 3, 2008

Misc. Longitudinal Impedance (Note: impedance data in following plot is scaled according to length of tank rather than magnet length; therefore actual impedance per metre is larger than shown in plot) 47MHz For MKE: Is frequency of onset of spikes dependent upon aperture size? e.g. (MKE-L9 & MKE-S3) ; Serigraphy generally eliminates spikes up to and beyond 500MHz. Apertures (hap x vap): MKE-L: 147.7 x 35mm; MKE-S: 135 x 32mm; MKI: 54 x 54mm; MKDV1: 56 x 75mm. Cell Length: MKE: ~24cm; MKDV1: ~51cm. MKE: no stripes MKE (S6): stripes on 2 of 7 cells MKE (L10): stripes on all cells MKI: no screen MKDV1: no beam screen or transition pieces. Spikes due to cell length ??? MKI: 15 screen conductors Mike Barnes, AB/BT APC: October 3, 2008

SPS Kickers: Vacuum, August 14, 2008 1s resolution 75ns beam 3 batches, 50ns 4 batches, 50ns 4x4 batches, 50ns 50ns beam MKDV1: hap=56mm, vap=75mm MKDV2: hap=56mm, vap=83mm MKDH1: hap=96mm, vap=56mm MKDH2: hap=96mm, vap=56mm MKDH3: hap=105mm, vap=60mm MKE1: hap=147.7mm, vap=35mm MKE2: hap=147.7mm, vap=35mm MKE3: hap=135mm, vap=32mm MKQH: hap=135mm, vap=33.9mm MKQV: hap=56mm, vap=102mm Transition pieces  Ferrite  No transition pieces No transition pieces. 0.35mm laminated steel No temperature probes. MKQH less sensitive to 50ns & more sensitive to 75 bunch spacing…. Mike Barnes, AB/BT APC: October 3, 2008

SPS Kickers: Vacuum, August 14, 2008 1s resolution 75ns beam 3 batches, 50ns 4 batches, 50ns 4x4 batches, 50ns 50ns beam MKDV1: hap=56mm, vap=75mm MKDV2: hap=56mm, vap=83mm 12xMKP-S: hap=101mm, vap=61mm 4xMKP-L: hap=141mm, vap=54mm No transition pieces Ferrite  MKE1: hap=147.7mm, vap=35mm MKE2: hap=147.7mm, vap=35mm MKE3: hap=135mm, vap=32mm MKE4: hap=135mm, vap=32mm MKE5: hap=147.7mm, vap=35mm Transition pieces  MKE4 Mike Barnes, AB/BT APC: October 3, 2008

SPS Kickers: Temperature, August 12-14, 2008 MKDV: 25ns beam MKDV: 75ns beam to 01:30hrs, 50ns beam 05:13 to 07:00 75ns beam 25ns beam 50ns beam MKE6: 75ns beam to 01:30hrs, 50ns beam 05:13 to 07:00 MKE6 temp. not overly sensitive to 50ns or 75 bunch spacing. MKE6: 25ns beam 75ns beam 25ns beam 50ns beam MKE6 temperature rise limited by serigraphy. Mike Barnes, AB/BT APC: October 3, 2008

SPS Kickers: Temperature, August 12-14, 2008 MKE4: 25ns beam MKP: 25ns beam MKP: 75ns beam to 01:30hrs, 50ns beam 05:13 to 07:00 MKE4: 75ns beam to 01:30hrs, 50ns beam 05:13 to 07:00 25ns beam 75ns beam 50ns beam MKP temperature probes both on one magnet. MKP temp. not overly sensitive to 50ns or 75 bunch spacing. MKE4 temp. not overly sensitive to 50ns or 75 bunch spacing. Mike Barnes, AB/BT APC: October 3, 2008

Summary & Future Plans “Spikes” on measured MKDV1 longitudinal impedance, which are not present in analytical calculation, are probably attributable to magnet cell length, aperture dimensions and absence of transition pieces. An MKDV1 magnet is presently being HV conditioned (it has transition pieces). It is planned to make longitudinal impedance measurements on this magnet before installation (during the next shutdown); this will allow effect of transition pieces to be assessed. Impedance resonances of MKDV2 (not measured) probably cause heating of ferrite with 75ns bunch spacing. Shielding of MKE magnets eliminated high impedance resonances in longitudinal impedance and thus significantly reduces heating – possible long-term solution for MKDV1 ?? (MKDV2 no-spare exists, so difficult to modify). Could MKDV1 (H=75mm) be used in place of MKDV2 (H=83mm)?. But still see pressure rise with 50ns bunch spacing….. 50ns bunch spacing causes immediate pressure rise (surface effect-e.g. electron cloud) for both MKDV1 & MKDV2 – but MKDV1 is more significant. MKE6, MKDH, MKDH & MKQV kickers all show immediate pressure rise with 50ns bunch spacing ….. But MKQH sensitive to 75ns bunch spacing… Further MD with 50ns beam to confirm, with 3 or 4 batches per shot, as cause of high pressure rise?; electron cloud investigations ????. Log ion pump current? Mike Barnes, AB/BT APC: October 3, 2008

Questions & Comments Mike Barnes, AB/BT APC: October 3, 2008

MKDH3 Longitudinal Impedance Measurements MKDH3 no shielding or transition pieces. No temperature Probes on MKDH magnets; MKDH3 length=1.256m  λ/4 resonance of ~60MHz; Transverse impedance measurements planned for MKDH3. Mike Barnes, AB/BT APC: October 3, 2008

MKE: Beam Coupling Impedance Reduction Printed strips in MKE-L10 Interdigital comb structure 20mm spacing surface discharge Beam coupling impedance is reduced using conductive stripes (serigraphy), i.e. interleaved comb structure, directly printed onto the ferrite blocks and a reliable contact to the metallic HV plates at either side; Capacitive coupling between stripes (stripes carry beam image current). Mike Barnes, AB/BT APC: October 3, 2008