01.02.2015 Views

Beam extraction - 50th anniversary of the CERN Proton Synchrotron

Beam extraction - 50th anniversary of the CERN Proton Synchrotron

Beam extraction - 50th anniversary of the CERN Proton Synchrotron

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

7 Implementation <strong>of</strong> MTE at <strong>the</strong> <strong>CERN</strong> PS<br />

7.1 Design principle<br />

The experimental campaign was completed at <strong>the</strong> end <strong>of</strong> 2004. The analysis <strong>of</strong> <strong>the</strong> required modifications<br />

to implement <strong>the</strong> proposed multi-turn <strong>extraction</strong> took place during <strong>the</strong> 2005/2006 long shutdown <strong>of</strong> <strong>the</strong><br />

PS machine. The conceptual design <strong>of</strong> <strong>the</strong> proposed multi-turn <strong>extraction</strong> can be sketched as follows (see<br />

Ref. [44] for more details and Ref. [45] for a complete account <strong>of</strong> <strong>the</strong> hardware commissioning):<br />

– <strong>Beam</strong> splitting: two pairs <strong>of</strong> two sextupoles and one octupole each will separate <strong>the</strong> initial<br />

single beam into <strong>the</strong> five beamlets prior to <strong>extraction</strong>. Contrary to <strong>the</strong> experimental set-up,<br />

where only one set <strong>of</strong> two sextupoles and one octupole was used, <strong>the</strong> choice <strong>of</strong> two pairs is<br />

mainly dictated by <strong>the</strong> need to control and adjust <strong>the</strong> phase <strong>of</strong> <strong>the</strong> islands at <strong>the</strong> <strong>extraction</strong> point.<br />

– Extraction: <strong>the</strong> <strong>extraction</strong> point is in SS16, where <strong>the</strong> magnetic septum for <strong>the</strong> beam <strong>extraction</strong><br />

towards <strong>the</strong> SPS is located. In <strong>the</strong> proposed scheme, <strong>the</strong> electrostatic septum, currently used to<br />

slice <strong>the</strong> beam in <strong>the</strong> context <strong>of</strong> <strong>the</strong> CT <strong>extraction</strong> in SS31, is not required, thus simplifying <strong>the</strong><br />

overall scheme. Two bumps will be used to displace <strong>the</strong> beam towards <strong>the</strong> magnetic septum<br />

blade (slow bump) and to extract <strong>the</strong> beamlets over five turns (fast bump).<br />

– Slow bump: a set <strong>of</strong> dipole magnets (bumpers) will be used to generate <strong>the</strong> slow bump around<br />

<strong>the</strong> magnetic septum. Currently, four bumps powered with a series/parallel circuit are used to<br />

extract <strong>the</strong> beams towards <strong>the</strong> SPS. In <strong>the</strong> proposed scheme, six independently powered<br />

magnets are foreseen. The large number <strong>of</strong> bumpers is imposed by <strong>the</strong> aperture constraints, as it<br />

will allow a careful shaping <strong>of</strong> <strong>the</strong> bump to overcome <strong>the</strong> potential aperture bottlenecks.<br />

– Fast bump: three new kickers will generate <strong>the</strong> fast bump used to displace <strong>the</strong> beam beyond <strong>the</strong><br />

blade <strong>of</strong> <strong>the</strong> magnetic septum. The pulse length should correspond to five PS turns. Since <strong>the</strong><br />

centre core <strong>of</strong> <strong>the</strong> beam needs to be ejected, an additional kick will have to be imparted at <strong>the</strong><br />

fifth turn. For this <strong>the</strong> fast <strong>extraction</strong> kicker will be used.<br />

– Trajectory correction in <strong>the</strong> transfer line towards <strong>the</strong> SPS: even though <strong>the</strong> <strong>extraction</strong> conditions<br />

for <strong>the</strong> novel multi-turn <strong>extraction</strong> do not change from turn to turn, as one single island is used<br />

to extract <strong>the</strong> beam, <strong>the</strong> feed-down effects <strong>of</strong> <strong>the</strong> machine non-linearities (particularly from <strong>the</strong><br />

pole face windings in <strong>the</strong> main magnets) due to <strong>the</strong> <strong>extraction</strong> bumps could generate turn-byturn<br />

variation <strong>of</strong> <strong>the</strong> beamlet position at PS <strong>extraction</strong>. Such an effect could have a negative<br />

impact on <strong>the</strong> emittance after filamentation in <strong>the</strong> SPS. Hence two kickers capable <strong>of</strong> generating<br />

deflection changing from turn to turn will be used in <strong>the</strong> TT2 transfer line to correct for <strong>the</strong><br />

variation in <strong>the</strong> <strong>extraction</strong> conditions (position and angle). These two devices are already being<br />

used for <strong>the</strong> CT <strong>extraction</strong> mode.<br />

7.2 Implementation<br />

The implementation <strong>of</strong> MTE required different interventions, such as design, production, and installation<br />

<strong>of</strong> new octupoles; construction <strong>of</strong> new power converters for <strong>the</strong> new <strong>extraction</strong> bump; design,<br />

construction, and installation <strong>of</strong> new vacuum chambers for increasing <strong>the</strong> beam aperture in particular in<br />

<strong>the</strong> <strong>extraction</strong> region; construction <strong>of</strong> new <strong>extraction</strong> kickers and <strong>the</strong>ir Pulse Forming Network (PFN). A<br />

large fraction <strong>of</strong> <strong>the</strong> PS circumference was affected by those interventions, summarized in Fig. 29, and<br />

24

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!