EMRFD Message Archive 7779

Message Date From Subject
7779 2012-09-14 05:39:16 Ashhar Farhan UR3IQO and the Chris' KISS mixer
I ran into Skydan's design site on at http://t03dsp.skydan.in.ua/
If you click on his Blocks > RF link on the left bar, you can see the
front-end topology that he has used. It is almost the same mixer as Chris's
KISS mixer (someone correct me if I am wrong). However, the interesting bit
is that he doesn't use any elaborate diplexer at all. Rather, the mixer is
terminated with zero termination at all frequencies except the IF with just
a tank circuit! Does anybody here have experience with this configuration?

His other project NEON at http://neon.skydan.in.ua/ is pretty interesting
too.

- farhan VU2ESE


[Non-text portions of this message have been removed]
7780 2012-09-14 06:27:06 Chris Trask Re: UR3IQO and the Chris' KISS mixer
>
>I ran into Skydan's design site on at http://t03dsp.skydan.in.ua/
>If you click on his Blocks > RF link on the left bar, you can see the
>front-end topology that he has used. It is almost the same mixer as Chris's
>KISS mixer (someone correct me if I am wrong).
>

Yes, that's a variation. I've come across numerous examples of the basic SPDT switch mixer since I published that paper.

>
>However, the interesting bit
>is that he doesn't use any elaborate diplexer at all. Rather, the mixer is
>terminated with zero termination at all frequencies except the IF with just
>a tank circuit! Does anybody here have experience with this configuration?
>

It should work fine with little degredation in performance. The problem with diode mixers being improperly terminated is that the reflected signals go back to the diodes and interact with everything else, degrading IMD performance mostly.

Chris
7781 2012-09-14 07:45:52 Ashhar Farhan Re: UR3IQO and the Chris' KISS mixer
chris,
that simplifies the 'no early amplifiers' approach quite a bit. one could
then directly drive the crystal filter with a 'nominal' diplexer instead of
the hybrid coupler followed by twin roofing filters. of course, it wouldn't
give the best possible performance. at the cost making martein and the
triad ill at the thought, one could settle with a simplification of the
front-end design at the cost of slight degradation of the receiver factor.
- farhan

7782 2012-09-14 08:21:06 AD7ZU Re: UR3IQO and the Chris' KISS mixer
There is an excellent article in the June 2010 QEX covering this design when expanded to cover the unterminated transformer issue. the QEX design utilizes diplexers.  
 
I found it interesting that Martein's TLT measurements (sans transformer terminations) are very very good.  As I remember he used off the shelf mini-circuits transformers and one spdt switch. 
 
There was a long EMRFD discussion a few months back
7783 2012-09-14 09:32:55 Chris Trask Re: UR3IQO and the Chris' KISS mixer
>
> >
> > >
> > >However, the interesting bit
> > >is that he doesn't use any elaborate diplexer at all. Rather, the mixer
is
> > >terminated with zero termination at all frequencies except the IF with
> > >just a tank circuit! Does anybody here have experience with this
> > >configuration?
> > >
> >
> > It should work fine with little degredation in performance. The problem
> > with diode mixers being improperly terminated is that the reflected
signals
> > go back to the diodes and interact with everything else, degrading IMD
> > performance mostly.
> >
>
> that simplifies the 'no early amplifiers' approach quite a bit. one could
> then directly drive the crystal filter with a 'nominal' diplexer instead
of
> the hybrid coupler followed by twin roofing filters. of course, it
wouldn't
> give the best possible performance. at the cost making martein and the
> triad ill at the thought, one could settle with a simplification of the
> front-end design at the cost of slight degradation of the receiver factor.
>

You might want to consider the less demanding approach I use for
narrowband mixer terminations:

http://www.home.earthlink.net/~christrask/Paper022.html

Chris