EXCALIBUR [75] is the most widely used generator. It contains not only the Standard Model matrix elements, but also the option to switch on a number of the anomalous couplings seen in the general Lagrangian, equation 3.1. The couplings that can be varied are; , , , , , and . They may be set at any value. It is not possible to implement the CP-violating couplings.
EXCALIBUR generates four-fermion final states through all possible electroweak four-fermion processes [74]. However, the QED two-photon diagram is omitted from the OPAL version because this process is not well understood and is better modelled by dedicated two-photon generators. This means that the interference terms of this process with the other NC48 diagrams are neglected. This is thought to be a small effect, especially compared to the uncertainty associated with the two-photon process itself.
All fermions are assumed to be massless when they are generated. The generator also includes the width of the and bosons. QED initial state corrections are implemented using a structure function formalism [76]. Interfering QCD backgrounds [77] are also possible. An example of these diagrams are as in figure 5.5. These interfering backgrounds are only relevant for the channel and in the OPAL version of EXCALIBUR they are not implemented. This is to avoid double-counting because they are also contained in the standard PYTHIA quark pair production Monte Carlo that is used as the background generator for the channel. EXCALIBUR has a Coulomb correction [78] for the CC03 WW production. This correction accounts for the fact that if the two W bosons are travelling slowly with respect to each other, the pure Coulomb attraction between them is not negligible, and it changes the W boson momentum distribution. This effect is particularly important at the W-pair production threshold of 161 GeV.A naive QCD correction is also implemented in EXCALIBUR. This is because, in addition to the four-fermion production that EXCALIBUR models, there is also four-fermion plus one-gluon production which enhances the WW production cross-section. The correction is applied naively to all final-states with W q decays and it boosts the cross-section by for each W q decay. It should be noted that the W q g topology is simulated in the events by the parton-shower part of the event generation, so multiplying the cross-section by this correction is a reasonable thing to do.
The naive QCD correction in principle should be applied to all diagrams with a vector boson V q decay. However to achieve this (Q) has to be evaluated at the correct (vector boson mass) scale Q. This is extremely involved, so the naive QCD correction is only applied to the WW diagrams, where Q is well defined ( ).