New 10 GHz LNB

[g0mjw]

Probably the phase noise. What is good for SSB might not be good for wider bandwidths. The far out phase noise can often plateau and this is integrated with the wideband signal. To a first approximation at least, so if your oscillator phase noise drops off quickly away from the 25MHz frequency, but then reaches a constant level, by the time it is multiplied by 400 or so to 10 GHz it can be quite high and the MER becomes dominated signal to phase noise rather than signal to noise. There is very little to beat a well implemented crystal oscillator phase noise performance, but there may be other issues, e.g. the level into the PLL in the LNB.

Mike