I have Kenwood TH-D72 2m/70cm HT. According to the manufacturer the sensitivity is "Less than 0.18 uV". I would like to check this, since the radio is not new and I'm not it's first owner.
To do this I decided to use the tracking generator of my Rigol DSA815-TG and a bunch of SMA attenuators (120 dB in total), because I don't have any other way to reliably generate a few uV signal. Naturally, zero span was used. I discovered that a -127 dBm signal opens the squelch on both 2m and 70cm and with squelch turned off I can hear the presence of -133..-134 dBm signal.
What surprises me is that if my calculations are correct -127 dBm is much less than 0.18 uV RMS:
>>> from math import pow, log10 >>> 10*log10(1000*pow((0.18/1000/1000),2)/50) -121.88424994129406 >>> 10*log10(1000*pow((0.05/1000/1000),2)/50) -133.0102999566398
I decided to check my HF QRP transceiver Xiegu X5105 as well and got similar results - the sensitivity is about 0.05 uV although the manufacturer promises only 0.25 uV.
Are my measurements wrong? Or maybe it's a common practice among radio manufacturers to specify worst-case sensitivity? According to the spectrum analyzer attenuators are fine. Also a -73 dBm signal gives S9 level on a S-meter as expected.