I’ve had my KX2 twelve months now and thinks it’s the best portable/SOTA HF CW rig I’ve used. From the Elecraft spec and all the glowing reviews I had expected its receive sensitivity (-136 dBm for SSB/CW, or 0.0355uV) to outclass my Yaesu FT-857D (spec’d as 0.20uV, or -101 dBm) and my first-generation FT-817 (0.25uV, or -99 dBm) or even my Youkits HB-1B.
But, after months of A vs B testing using the same antenna on weak stations and beacons from 80m to 17m, I hear no noticeable difference, no superior KX2 performance.
In fact, the 857 S-meter consistently shows signals about two S points higher, although to be fair it also shows a higher reading on background noise for the same filter bandwidth - maybe my KX2 S-meter is pessimistic.
The KX2 is used with internal ATU and pre-amp ON; the 857 or 817 with external LDG Z-11 Pro ATU – both set-ups getting the same or very similar VSWR.
Some months ago, I sent the KX2 back to my UK dealer who installed the latest firmware and ran tests, which apparently it passed. No change on A-B testing on return.
Has anyone done sensitivity comparisons of the KX2 with these or similar rigs?
You’re not going to find any noticeable difference in weak signal readability (i.e., sensitivity) on 80-17M between well designed radios in actual use when connected to an antenna. They will all be noise-limited. The important difference will be freedom from distortion/blocking from nearby strong signals. That’s where the KX2/KX3 will outclass the other radios.
73, Barry N1EU
Barry, thanks for the explanation.
I did those tests in my urban home surrounding by neighbouring houses. When I started SOTA activations I was pleasantly surprised to find the background noise was often S0. In fact, the very first time I thought the dipole might be broken or disconnected until the first signal came through.
Does this suggest, had I conducted those comparison tests at the summit where the noise floor is much lower, the KX2 could have resolved a very weak signal with a better S/N than the ‘less sensitive’ rigs?
I believe you still would have been atmospheric noise limited and would have seen little difference between the radios. Rig differences (well designed modern radios) are simply NOT about sensitivity. In terms of pulling out weak signals, it’s all about the antenna, not the radio.
And how good your lugs (Scottish term for ears) are - Barry
I’m in a hurry so I’ll make it quick:
- the math on uV / dbm conversion is wrong. Also, when comparing, make sure you are using the same mode, bandwidth and S/N ratio.
- most S-meters are extremely bad (including FT-817/857, not sure about Elecraft). Unless it’s Flex or some other decent SDR radio, just ignore the reading for measurement purposes.
- receiver sensitivity doesn’t matter on HF, even a receiver built with a few transistors has enough. The band noise level is much higher than receiver internal noise.
The flaw in the receiver sensitivity comparisons outside a lab is that the receiver cannot improve the signal to noise ratio of an incoming signal with noise.
Better quality filters can make a difference. The ceramic filters in a conventinal superhet receiver often have passband ripple that affects readability. The kx2 and kx3 effectively convert to audio and dont use such ceramic filters as they dont have an IF. In that respect they and other SDR type receivers will produce a better quality signal compared with a superhet with ceramic filters. But it isnt due to having more gain or being able to detect a lower signal level.
My 703 uses mechanical filters on both rx and tx, and a mech filter for cw. These improve readability on rx and tx due to having lower passband ripple. Every little bit helps.