[hpsdr] GIBRALTAR LO generation question

[Philip Covington]

On 7/6/06, Lyle Johnson <kk7p at wavecable.com> wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
> Hello Alex!
>
> > Umm, so where does the LO come from for CASMIR and HORTON?
> > I thought we were going to do the centralized uWave LO in order to support N4HY's phasing request?
>
> One of many ideas being bandied about.
>
> Gibraltar is a frequency and timing reference as I understand it,
> nothing more.
>
> I suspect Horton may need its own, localized LO.  Or perhaps the LO is a
> separate module, or ???
>
> I am not sure what is required for CASMIR.  I haven't had the time to
> dig into the Analog Devices data sheets on the chips you are proposing
> to see what their requirements are.
>
> 73,
>
> Lyle KK7P
>

I expect we will have a PLL based solution where the VCO is divided
down from the 2000+ MHz range and a DDS based solution like the
AD9958/9 chip.  The reference for both could come from Gibraltar or an
on board source.

Speaking of the AD995x... I see where Giuliano I0CG
<http://it.geocities.com/giulianoi0cg/ad9951.html> used a 500 MHz
source with his AD9951 that was derived from a 100 MHz oscillator with
the 5th harmonic filtered by a Toko helical resonator filter.  The
helical filters are expensive.  I wonder if you could get the same
results with a 80 MHz oscillator's 5th harmonic filtered by a ECS SAW
filter at 400 MHz?  The SAW filter has a bw of 80 kHz centered on 400
MHz and cost about $4.  This would drive the AD9958 at 400 MHz would
still should be good up to 160 MHz or so.

73 de Phil N8VB