Summary of panel discussion on beam-beam codes and simulations

This is a summary of a panel discussion following a presentation on beam-beam simulation codes [1]. It reflects the main issues raised and considered important during the discussion by the panel members and the audience. Due to the limited time available for the formal discussion, some relevant deliberations during informal discussions mainly between and with panel members have been included as well. BEAM-BEAM SIMULATIONS Beam-beam simulations are an essential tool to design and run a colliding beams facility. They are used in all phases of the design and during the operation, basically until the machines are switched off. Applications of beam-beam simulations In the very early phase they are usually used to define the basic performance parameters. In this stage some of the studies are of more ”academic” nature and possible problems should also be unraveled and possible solutions proposed. Unfortunately, and this is true for most other simulations, it is rather difficult to predict the future or the unexpected. Surprising effects and the lack of proper simulation techniques and resources have led to the reputation that beam-beam simulations can only predict the past, as one of the panel members has formulated a provokative statement. Recent progress in understanding and advanced techniques has largely improved the predictive power and the limitations are well understood. In the second stage of the design the use of beam-beam simulations is vital for the optimization of the parameters and the definition of the operational conditions. During the (usually many) years of operation of a collider the beam-beam simulations become one of the most important tools to understand and improve the performance. This was demonstrated in many machines, e.g. SPS proton-antiproton collider, LEP and the Tevatron. Many observations were only understood with the help of such simulations and helped to improve the simulation programs themselves [1]. Including the much improved computing resources we have now a number of reliable simulations codes and tools available or under development [1]. Types of beam-beam simulations The beam-beam effects act on single particles as well as on an ensemble and we distinguish two basic types of codes. Weak-strong simulations which are used to study single particle behaviour under the influence of a static beam-beam force and strong-strong simulations where both beams are allowed to change their parameters under the influence of the opposing beam. In the latter case we look for self-consistent behaviour to study collective motion and emittance growth. Both types of simulations are used to evaluate the performance in all phases of a collider. WEAK-STRONG SIMULATIONS Typically single particles are subjected to beam-beam forces in this types of simulations. Single particle simulations A straightforward and immediate application is the evaluation of the amplitude dependent detuning caused by the opposing beam. The results are usually presented in the form of ”tune footprints” and are used to estimated the required tune space in the working diagrams. They also serve as a benchmark for the overall strength of the incoherent beam-beam effects. The stability of particles under the influence of the non-linear fields is usually estimated by tracking particles through the machine elements which can include the beambeam interaction as a source of very non-linear fields. The evaluation of the particle loss, dynamic and diffusive aperture and chaotic behaviour allows to estimate the expected stability region. The definition of tunes, i.e. the working point, of the machine must be done taking the beam-beam effects into account. Since such simulations are often done for machines with imperfections (linear and non-linear, self-consistent optical functions) their correction and the tracking can be integrated into general purpose optics programs. Fluctuations bunch by bunch In colliders with many bunches fluctuations must be expected from bunch to bunch and properly taken into account in simulations. Different collision pattern of the bunches can lead to significantly different dynamic behaviour, e.g. due to unsymmetric filling schemes. The properties of ALL bunches must be studied in these cases Bunch by bunch effects and self-consistent treatment Partially as a consequence of bunch to bunch fluctuations but also due to beam-beam effects in colliders with many bunches with different collision schemes, one has to expect different parameters of the bunches. In a weakstrong configuration these parameters can be evaluated using perturbative methods. However, in a strong-strong situation a perturbative treatment can lead to completely wrong results. In the case of significantly different interaction pattern of many bunches in a collider, the parameters might have to be computed in a self-consistent form when the beam-beam interaction is very strong. Such a selfconsistent treatment was essential to understand the LEP performance and the predictions were experimentally verified. It has been demonstrated that a self-consistent treatment is possible, even for a machine with as many bunches as in the LHC. The computation of parameters such as tunes, orbits and chromaticities for all bunches is reliable even in the presence of intensity or beam size fluctuations and allows to set tolerances on these fluctuations.