Abstract
The main physics program of the CERN Super Proton Synchrotron (SPS) is dedicated to the fixed target physics experiments hosted in the North experimental Area (NA). Protons are delivered to the NA via third-integer resonant slow extraction over an almost 5 s flattop. In order to maximize the usable intensity delivered to the experiments, the flux of extracted particles should be kept as constant as possible. This is a very general requirement for fixed target experiments served by synchrotrons. Power supply ripples are a well-known issue in resonant slow extraction, affecting the quality of the spill. A long-standing effort is ongoing at CERN to characterize the SPS slow extraction frequency response to its main power supplies. In this paper, beam dynamics simulations are employed to understand and characterize the process, combined with dedicated beam based measurements.
Highlights
The CERN Super Proton Synchrotron (SPS) is the last accelerator in the Large Hadron Collider (LHC) injector chain
The bottom-left plot shows that only the first harmonic of the main frequencies is generated, as opposed to the case of Fig. 3. This comparison demonstrates how the low pass filter effect of the slow extraction process can be beneficial for the quality of the extracted spill: if properly exploited, it can be a very powerful tool to suppress the power supply ripples. These results show that the transfer function is amplitude independent only for small enough ripples satisfying the monotonicity of the tune ramp
While the small amplitude transfer functions are brought nearer to each other with respect to the case of Fig. 2, a nonlinear behavior is still evident, being the obtained transfer functions still amplitude dependent. These results show that the nonlinearity is coming from Eq (4), but either it is present in the low-pass filter effect of the slow extraction, or the two effects cannot be disentangled in such a way
Summary
The CERN Super Proton Synchrotron (SPS) is the last accelerator in the Large Hadron Collider (LHC) injector chain. Both iterative feed-forward methods described here apply the corrections estimated from the measured spill signal at the subsequent slow extraction machine cycle, until the optimum solution is reached This process can take several machine cycles to converge (usually about ten) and the stability of the reached solution relies upon the reproducibility of the magnets’ response and power supply ripples in the SPS. This approach could be considered as a real-time feed-forward system, advantageous when the ripples are not stable from one machine cycle to another Another possible way to improve the spill quality is to increase the smoothing effect of the time intervals distribution of the extracted particles, by acting on the machine parameters (from a frequency point of view, it corresponds to increasing the attenuation of the low pass filter of the extraction process).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
