Optimising the Linear Collider Luminosity: Feedback on Nanosecond Timescales

linear collider currentlyunder consideration. ∆t is the time separation between the N bunches in the train.The nanometre-level vertical beam overlap is a particularly challenging goal for all these designs, most notablyCLIC. Any source of beam motion which results in relative vertical offsets of the two beams at the interactionpoint (IP) at the nm level will clearly reduce the luminosity from the nominal value. In all of the collider designsstabilisation below the 1σ level is required to keep the luminosity loss below 10%.The many kinds of potential beam motion may be characterised in two classes: (i) slow drifts resulting fromeg. thermal excursions or component settling, with characteristic timescales varying from seconds to months;(ii) jitter on a timescale comparable with the machine repetition time. Both kinds of motion were experiencedin the decade-long experience at SLC, and were dealt with by employing slow- and fast-feedback systems,respectively.We are addressing the design of an intra-bunch-train fast-feedback (FB) system for the next-generation linearcollider (LC). The system comprises a fast beam position monitor (BPM) to detect the relative misalignmentof the leading electron and positron bunches at the IP, a feedback loop, and a fast kicker for kicking the trailingbunches back into collision.The system time-response requirements for J/NLC and CLIC are clearly very different from those for TESLA.We have therefore chosen the more challenging case of J/NLC and CLIC as the primary focus of our efforts todevelop a working prototype hardware system. However, from the timing point-of-view, a system which workson the 10ns scale could clearly be applied to the less demanding TESLA timescale of 300ns. With currenttechnology it is clear that a system for J/NLC and CLIC must be based on fast analogue signal processing,rather than digital processing [1]. However, given the rapid advances in signal processing speeds over the pastdecade it is quite possible that digital technology will be fast enough by the time the real LC system is deployedc. 2010; if so, the FB electronics could look almost identical for any of J/NLC, CLIC or TESLA. We aim tomonitor this situation and to take advantage of technological developments as they arise.