Dynamic range of frequency mixer

[oac]

Hi,

I'm building a receiver for FM radar:
Received signal is about -60dBm strong. First stage is an LNA with noise figure 0.7dB and gain 14dB. Second stage would be another amplifier with a gain that i don't yet have decided. Third element in the chain would be the downconverting mixer (conversion loss 6.9dB, so noise figure of about 7.4dB according to minicircuits). I'm worried about the RF dynamic range of this mixer. The second stage has to have a gain that suits this dynamic range.
Minicircuits (https://www.minicircuits.com/app/AN00-009.pdf) says:

Dynamic range is the signal power range over which a mixer provides useful operation.
The conversion compression point signifies the upper limit of the dynamic range. The
noise figure of the mixer signifies the lower limit of the dynamic range.

1) The upper limits makes sense to me, but i don't understand the lower limit. To me, noise figure have to do with signal to noise ratio at output and input, and not with abolute input powers... What noise power are they referring to?

2) If I treat the mixer as a block with gain -6.9dB and noise figure 7.4dB in the Friis formula for noise figure, is it still necessary to check this dynamic range?

Thank you very much,
Oac

 

[chuckey]

"Dynamic range is the signal power range over which a mixer provides useful operation." When the IF output is below the noise level of the mixer, you could say that the mixer is not working anymore, so mixer noise level does set the lowest practical signal level.
Frank

 

[oac]

I agree. Do you agree with the following statement: if we assume the mixer has a noise figure of say 8dB, this means SNR_in/SNR_out=8dB. So you rather need a minimum input signal-to-noise-ratio than a minimum input power. If we want the useful IF output signal power to be stronger than the noise floor, we need an SNR_in of at least 8dB.

Thank you for your answer, oac

 

[chuckey]

Noise figure is defined as the ratio of the noise power compared to the thermal noise generated by a resistor of the same system impedance. i.e. a noise figure of 3 dB would mean that the amplifier generates twice as much noise as a 50 ohm resistor. So for a given bandwidth an equivalent noise voltage can be calculated, this in turn would lead to the signal to noise ratio for a given signal strength.
So in a word I disagree with your statement, the attenuation of the mixer also affects the noise from the front end and the mixers own noise is then added in.
Frank