[TenTec] OMNI VI Transmit IMD

[Tom Rauch]

> With regard to IMD performance, again from knowledge and experience,
> to achieve optimum, one must have very carefully matched beta devices.
> Thus the reason why some companies sell components in match pairs. 
> While ideal, and typically a bit costly, and I'd expect from a Tentec
> viewpoint the factor of production output (yield) vs. scale of
> economics and a few other factors, general devices will make the FCC
> guidelines all of the time.  (Tentec may wish to comment on this to
> the reflector or direct to me.)
> 
> Ideally, some standard production lot could be submitted for test, the
> results obtained and then some percentage of the high and low numbers
> discarded then publish the mean mathematical number as a
> specification.  It works, time consuming and expensive.  The question 
> to be answered, is it worth it for a hobby?

Geesh, it sounds like we are tying to get -80dB IM3 performance. That 
simply isn't the case!

A zero-bias class-C amplifier can do almost as well as some modern 
radios for IM3 performance. Is it really too much to ask that radios 
be 10 or 20 dB better than class C performance?

I think the important point is radios are going downhill, and are 
almost certainly doing so because designers either are not taking a 
few extra hours to "tweak" designs or don't have a clue how the radio 
should work or what is important.

For example?

1.) Radio manufacturers have released radios that click like hell, 
even though adjusting the value of a few components would virtually 
stop the problem.
   
2.) CW quality is getting poorer on some brands, even though a few 
changes in a few lines of software would correct most of the 
problems.

3.) SSB IM products are varying from acceptable performance to 
horrible performance barely better than a class C amplifier.

All of this has to be rooted in the fact they either don't care what 
they are doing, or don't know what to do. None of it is a cost issue, 
or a manufacturing QC issue.73, Tom W8JI
W8JI@contesting.com