EMRFD Message Archive 11149

Message Date From Subject
11149 2015-06-08 19:43:10 jwolczanski 2009 presentation on high-level mixers
Presentation uploaded by KB1GMX

Spoiler Alert!
The 4 JFET DBM has become so good that it doesn't pay to improve it until we improve phase noise at its output.

OK, it's now 2015.  Is this still a valid statement?

At FDIM, Elecraft talked about their new K3 with improved phase noise, if I understood that correctly.


11154 2015-06-09 08:44:07 jmlcs2000 Re: 2009 presentation on high-level mixers
Hi Jerry....I'm at a loss and/or missing something...I've not able to find the presentation you are referring to in your email....Perhaps my old age is creeping in.......please lead me in the right direction so I can take a look at it....many thanks, John K5IRK
11156 2015-06-09 11:45:12 jwolczanski Re: 2009 presentation on high-level mixers
It's in Allison's (KB1GMX) Folder at:


It's the first one "K8KI..."

11157 2015-06-09 12:52:31 John Lawson Re: 2009 presentation on high-level mixers
Jerry thanks, I overlooked it....guess I need a new pair of glasses, :).  John

Sent from my iPad

11158 2015-06-09 16:26:54 kb1gmx Re: 2009 presentation on high-level mixers
Yes, it still applies.

Generally the LOs used for most hobby designs are not sufficient to exceed that level.
Especially in the minimum art radios.

The logic that may have been missed is if you improve the mixer everything else needs to reach that new level or the improvements will be smaller if at all.

Its very important to review designs one wants to improve as the seemingly obvious
can yield anything from disappointing results to a negative result.  The converse is if you improve a key circuit what other improvements are required to take advantage of it.

An example: if BCI  is a problem a new mixer may help, or it may create a whole new class of distortion products because of lack of enough LO or maybe excess harmonics.  Or succeeding there the next stage has to handle the result and can't.  So the process is one of understanding the original design to see where and what needs to be done.  Maybe a 7th order high pass works better to eliminate the offending signal rather than trying to process it.