TenTec
[Top] [All Lists]

Re: [TenTec] Jupiter Extended Report - Up Conversion

To: tentec@contesting.com
Subject: Re: [TenTec] Jupiter Extended Report - Up Conversion
From: Duane Grotophorst <n9dg@yahoo.com>
Reply-to: tentec@contesting.com
Date: Wed, 24 Dec 2003 14:35:52 -0800 (PST)
List-post: <mailto:tentec@contesting.com>
It's not so much up conversion in and of itself but
the side efects of the reason(s) RX designers choose
to up convert in the first place.

One of the main the reasons to up convert is to
provide continuous coverage from the low kHz range to
30 mHz without gaps and to have very good first IF
image rejection. The technique does work well for
those specific goals, for example compare the IF
rejection figure for the Orion's main RX vs sub RX,
the sub RX has considerably better IF rejection than
its main RX. All current model JA radios also up
convert first, that way they can provide both ham band
RX/TX and general coverage RX, just like the
Pegasus/Jupiter/RX320/RX350/516 also do. A design
trade off decision nothing more nothing less.

The other thing that typically goes with general
coverage RX is "all mode" capability, so that means a
relatively wide filter for FM ~15-20kHz. Also as it
turns out creating sharp filters in the 45-70 mHz
region is both a bit challenging and expensive. So
they all use a wide first IF roofing filter that is
typically ~15-20kHz wide and then depend on providing
sharper selectivity at the 2nd, 3rd, or 4th IF, -
either by using traditional crystal filters or DSP.
The problem with this design approach is that whatever
unwanted signals and broadband noise that has already
made it through that first wide roofing filter is
still available to cause grief further downstream in
the IF chain.

By looking at the Orion's main vs. sub RX swept BDR
plot you will see the direct effects of both of these
two design aproaches (VHF vs. 9 mHz 1st IF) in the
same radio. The Orion uses more or less the exact same
DSP code for both the main and sub RX, the majority of
the performance differences between them is due to the
analog component choices, 1st IF frequency choice
(therefore roofing filter BW's), and 1st LO frequency
generation method.

One of the cornerstone design decisions of the Orion
is to limit the IF bandwidth at the first IF (the main
RX only), so far it is the only DSP IF based ham
transceiver to do so.

The other down side of up converting is that the first
LO is in the low VHF region, it is therefore
inherently noisier from a phase noise perspective.
That is because it gets progresively harder to achieve
a clean LO the higher up you go in frequency. Again
looking at the Orion's main RX design the first LO
frequency is never any higher than ~39 mHz. All of the
up conversion designs start with a 1st IF in the
45-70mHz region, and have LO's that tune up 30 mHz
above that. The Orion's sub RX IF frequency scheme is
the same as the Pegasus/Jupiter et al. so it does have
its first IF LO starting at ~45mhz (tunes from ~45-75
mHz). The Orion sub RX is also a more straight forward
PLL design vs. the main RX's DDS/PLL/divider design.
The effects of those differences can be seen in the
main RX's absolutely phenominal phase noise spec being
< -136dBc/Hz (better than many analog only crystal
mixer schemes) vs. the sub RX. The dynamic range specs
for the sub RX are not nearly as good, and
specifically in the context of the LO source's simpler
PLL design are actually "noise limited" in some cases.
The Orion main RX is not phase noise limited in any
way.

Wide roofing filter = poorer close in BDR and IMD
performance. 
Radios like: Pegasus/Jupiter/RX320/RX350/516/Orion sub
RX, all JA "general coverage" radios to date.

   
Narrow roofing filter = great close in BDR and IMD
performance.
Radio's like: Orion (Main RX), Icom 7800 (but to only
6kHz spacing in all liklihood), K2 (by virtue of ham
band only design, a single high level mixer, and no
wide BW modes).

So in a nutshell that is the 'rest of the story'.

Duane
N9DG

--- Steve Baron - KB3MM <SteveBaron@StarLinX.com>
wrote:
> Guess I don't understand what up conversion
> has to do with the problem.
> 
> ----- Original Message ----- 
> From: "Duane Grotophorst" <n9dg@yahoo.com>
> To: <tentec@contesting.com>
> Sent: Wednesday, December 24, 2003 16:45
> Subject: Re: [TenTec] Jupiter Extended Report
> 
> 
> > 
> > --- Ben K8DIT <benk8dit@serv.net> wrote:
> > > Not there. We are lead to believe that the swept
> BDR
> > > and IMD are half that
> > > of the Main Rx on the Orion! There must be a
> 'rest
> > > of the story' there.
> > 
> > Yes there is, the
> > Pegasus/Jupiter/RX320/RX350/516/Orion sub RX all
> > suffer the same malady as all other radio designs
> do
> > that use up conversion and wide roofing filters
> do.
> > That is they are prone to less that stellar BDR
> specs
> > within the bandwidth of the first IF roofing
> filter.


__________________________________
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing.
http://photos.yahoo.com/
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec

<Prev in Thread] Current Thread [Next in Thread>