[en] Along the general run of the vacuum chamber synchrotron radiation strikes the wall at a glancing angle of about 5.6⁰. The heat source is well-approximated by a ribbon of uniform power density having a small vertical height and an infinite azimuthal length. The heat transfer problem reduces to one in two-dimensions and it has been considered in a previous note. At the corner of a beam port the angle of incidence becomes 90⁰, so the temperature rises much higher than elsewhere. Since the power density at the corner is not uniform in its azimuthal dependence, but is strongly peaked at the point of normal incidence, two-dimensional heat flow is not a good approximation. The rectangular 3d problem is considered. This is easily solved and yields a good first estimate of the temperature rise at the corner