[hpsdr] Selectivity design/measurement

[Mark Amos]

Bob,

Great!  I'd like to see the results - part of the fun I have with this stuff is doing my own measurements and 
comparing them to those of the experts.  It's a great learning experience for me.

I know one of the challenges to achieving any chance of consistency between measurements is to insure that 
the configuration under test is controlled.  If the measurements you made include your test configuration and 
settings, that would be very useful.  

In some informal SDR-1000 tests I performed a few weeks ago, I know that there were some deficiencies in my 
setup that contributed to the numbers not being up to par. I'm working to eliminate these issues and I've 
seen some dramatic improvements already.

I'm excited to hook up Janus to see what he can do!

Thanks for the reply and your contributions.  Also, thanks to the HPSDR designers and developers for this 
interesting and historic work.

Mark

------- Original Message -------
>From    : Robert McGwier[mailto:rwmcgwier at gmail.com]
Sent    : 2/15/2007 5:56:09 PM
To      : kk7p at wavecable.com
Cc      : mark.amos at toast.net; hpsdr at hpsdr.org
Subject : RE: Re: [hpsdr] Selectivity design/measurement

 I did all of these measurements on the alpha Janus and Ozy boards.  The 
AKM5394A is  one of the the  best recommendations I have ever given to 
any project or collection of projects and the Phil's and Bill and Lyle 
have a done a great job.   The Janus board lowers the noise floor of my 
SDR-1000 and that system is not SDR-1000 noise limited rather than Delta 
44 noise limited and the "hump" is gone on my set up where I used to 
have a hump with my Delta 44.  I never had one with the Lynx L22 and it 
uses the AKM5394A with balanced inputs. 

I am in Boston all week at Mercury computing for Cell processor 
training  and will be at Virginia Tech next week at the wireless group 
but I invite any of the other developers to feel free to publish the 
numbers Hambly and I measured in the lab using the SDR-1000 and with 
with the fancy audio generators.

Bob
N4HY


Lyle Johnson wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
>   
>> ...My question should have been much more specific, as:  "Has anyone built a 
>> receiver out of HPSDR components and if so have they done any selectivity measurements?"
>>     
>
> Janus+Ozy have been used with PowerSDR and an SDR-1000.
>
> The QSD A/D converter system will likely impact the minimum discernible 
> signal (MDS) unless there is a low-noise RF preamp in use.  It will most 
> certainly have a great impact on the dynamic range and distortion products.
>
> The selectivity is done in DSP and is pretty independent of the soundcard.
>
> Mercury prototypes have been built and are documented on the wiki.  The 
> don;t use a "soundcard" but instead do direct digital conversion of RF.
>
>   
>> I think your point about the DSP defining the selectivity means that the hardware components do not 
>> contribute or detract meaningfully from selectivity. Is this correct... 
>>     
>
> That is correct.  In more general terms, a bandpass filter is needed at 
> RF to limit the signals the SDR is exposed to.  The usual selectivity - 
> e.g., a 500 Hz CW filter, or 2.8kHz SSB filter -- is done in software 
> and shoudl not be materially affected by the hardware.
>
> Lyle
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at hpsdr.org
> Subscription help:  http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org 
> HPSDR web page:  http://hpsdr.org 
> Archives:  http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/ 
>
>   


-- 
AMSAT Director and VP Engineering. Member: ARRL, AMSAT-DL,
TAPR, Packrats, NJQRP, QRP ARCI, QCWA, FRC. ARRL SDR WG Chair
"Taking fun as simply fun and earnestness in earnest shows
how thoroughly thou none of the two discernest." - Piet Hine





 1171642510.0