[G0HZU_JMR]

Yes, on some Mutek boards the aim was to minimise losses and get a low noise figure but this was combined with a diode ring mixer and a strong IF amplifier and a narrow IF filter.

On the mk1 290R the radio has to support the internal whip option and I think this must affect the front end losses a bit. However, I've never tried to measure these losses. So although the little preamp probably has a noise figure of <= 1dB I think the overall receiver noise figure will be degraded by the need to support the whip antenna. I tried to find a spec for the receiver NF when the Mutek board is fitted to a mk1 290R but couldn't find one. I suspect it will be somewhere between 2 or 3dB but this depends on those front end losses associated with the whip antenna.

When I measure the noise figure I do it with an Agilent noise source and I use the Y factor method. I look at the receiver AF output on an analyser and set a noise marker in the AF passband with the noise source off. Then I turn the noise source on and measure the noise level again. These two noise levels can be used to calculate the Y factor and if the noise source ENR is known (eg 5.73dB ENR) the noise figure can be calculated with fairly low uncertainty using a fairly simple and classic equation. The hot/cold VSWR of the noise source is really low so this minimises uncertainty. I also have a Noisecom noise source and the results with this noise source usually agree very closely with the Agilent noise source.

By contrast, I think a lot of technical reviewers just measure SSB 'noise floor' with a sig gen and they declare the noise floor to be the sig gen level that boosts the AF noise level 3dB. But this isn't a reliable method for various reasons.