[1000mp] the battle of the front ends ...

[Tom Rauch]

I've seen some difference from rig to rig (with same models) and 
band to band, but not all that much that it is worth worrying about.

> Also, the K2 is ham band only.  While I know the front-end is
> designed to be truely robust (and it is), alignment over
> a narrow band with a actual measurement ( set-up to resolve
> IMD performance ) could push the envelope for a special case.

The K2 isn't so much a testimony how good the K2 is built, but 
rather it illustrates how POORLY the other rigs are designed.

The K2 ISN'T a particularly robust design, it really isn't anything 
special or modern at all. The technology that gives it reasonable IM 
and blocking DR is 1970's vintage. 

The K2 has single-ended 2N5109's in the post-mixer (I used them 
in the 70's in CATV projects and homebrew equipment) and pre-
mixer amplifiers, and uses a standard inexpensive packaged diode 
mixer (I have samples of the same mixers laying in boxes from the 
mid or late 70's). Even the filtering isn't all that special. 

Since the pre-amp is broadband, and since it uses wide front end 
selectivity, the fact it is better than our other radios really has 
nothing to do with it being ham-band only.

To see what a receiver could do with off-the-shelf components, I 
recently breadboarded a front-end and IF system using a high-level 
diode mixer and a push-pull 1.5dB noise figure post-mixer amplifier. 
This crude, cheap, mixer and IF system built in a few evenings 
easily makes over 100dB IMDR at 2kHz spacing!  It uses a 
broadband front end with very wide filters...like you would use in a 
general coverage receiver!

Any manufacturer could do that well or better without significantly 
raising costs, if they watched what they are doing.

I think the focus of radio manufactures is goofy. In many cases it 
seems they care less if the radios click or have excessive SSB 
IMD, as long as they have DSP and 50,000 stacked memories and 
dual-watch receivers one rarely uses. 

Now the bands are junked up with clicks and other problems, but 
by golly we have dual-watch rigs that operate from 1.8MHz to UHF 
with a dozen buttons we never even push! All the engineering went 
into push buttons instead of performance.

Most of this is our fault for letting marketing people tell us what we 
need and want, and because we stood by and allowed everyone to 
use nonsense tests for years and years to compare radios. 

Keyclicks had to get so bad it became almost impossible to 
operate in contests or near DX pileups before people started to 
complain, and I'll bet manufacturers STILL aren't seriously looking 
into the problem. Even SSB transmitter performance is much less 
than average, and we tolerate it like that's the best we can do.
73, Tom W8JI
W8JI@contesting.com