The geometry of the SL75-20 accelerator shown in fig.5 can be found in references[14][13][12]. It has an exit window, primary collimator, scattering foils (there are three scattering foil positions in a carousel in the treatment head, the appropriate one is set according the beam energy), a beam monitor ion chamber, a steel shielding ring, a mirror, X and Y jaws, an accessory ring and electron applicators which are solid-walled flat tubes made of aluminium. The measured data are from Udale-Smith, reference [13].
Our results for the electron spectrum from the 5 MeV beam are similar to those of earlier study by Udale-Smith[14] although there are some unexplained differences[1].
Figures
19,
20,
21,
22 and
23
present the energy spectra and angular distributions
of electrons and contaminant photons and dose distributions
along with dose components contributed from each beam defining
component for
5, 10, 14, 17 and 20 MeV electron beams respectively.
Good agreement is obtained
between calculated and measured central-axis depth-dose
curves. Calculated dose profiles at the phantom surface and at 
(1.1 cm depth)
are shown in figure 24 for the 5 MeV beam.
It can be seen that even at , dose contributions from scattered
electrons are significant (33%).
The electron spectra from the SL75-20 have a wide spread partially because of its thicker scattering foils.
At higher beam energies there are two small peaks in the energy spectra. This is due to the geometry of the scattering foils which are stacked cylinders with increasing radii (see fig. 25). Electrons going through the scattering foils near the central-axis lose more energy. A simulation, which uses LATCH to separate components which have passed through various thickness of scattering foils, is shown in figure 25.
Again due to the design of the applicator of the SL75-20 there is a considerable amount of dose contribution from scattered electrons, particularly from the applicator (about 32% to 22% of total surface dose for 5 to 20 MeV beams respectively).