Re: test of Perseus on 160m and 80m
- --- In perseus_SDR@yahoogroups.com, Leif Asbrink <leif@...> wrote:
> Hi Alberto,
> It seems screen captures are not allowed on the list...
I see the capture perfectly reading the group on the Web site. You
don't see the capture if you read the group by email, and you have
HTML blocked in you email client...
>Yes, I did read that, but it is here where I have problems... the WAV
> Call Freq Real Freq Peak level
> unkn 51.44kHz 14028.44kHz 80 dB
> W5ZR 50.91kHz 14027.91kHz 84 dB
> OK1HB 48.74kHz 14025.74kHz 44 dB
> -------------------- snip ------------------------
> "Freq" means the frequency with the center frequency set to
> zero. "Real Freq" is the true frequency which means that the
> center frequency was 13.977 MHz when the recording was made
> but no information about that is contained in the standard
> .wav file.
file is an I/Q capture done at 96 kHz sampling rate. This means that,
with a central frequency of zero Hz, the span goes from -48 kHz to +48
kHz. How is then possible that a signal is placed at 51.44 kHz on that
scale ? That's why I posted my provious message. Thanks for any
73 Alberto I2PHD
- --- In perseus_SDR@yahoogroups.com, "adibene" <i2phd@...> wrote:
>Well, I have figured myself (and it was logical enough) that you had
> How is then possible that a signal is placed at 51.44 kHz on that
> scale ? That's why I posted my provious message. Thanks for any
just offset by 48 kHz the values, and so I have been able to find the
wanted OK1HB signal. This is the audio output from Winrad :
while this is the pointer found on your Web page to the audio output
processed by Linrad
to my highly untrained ears they sound of course different, having
been processed differently, but, to my ears, both unreadable... but as
said my CW skills are minimal. Maybe those with "Golden Ears" will
have more luck than me.... :-)
73 Alberto I2PHD
- Listening on both transmissions I find that the latter file processed by LinRad does allow me to copy the call twice, but to be honest I simply cannot get the call when listening on the WinRad processed file.
/Lasse CW op for 35 years :)
On Fri, 01 Feb 2008 13:27:55 -0000
> --- In perseus_SDR@yahoogroups.com, "adibene"
>> How is then possible that a signal is placed at 51.44
>>kHz on that
>> scale ? That's why I posted my provious message. Thanks
> Well, I have figured myself (and it was logical enough)
>that you had
> just offset by 48 kHz the values, and so I have been
>able to find the
> wanted OK1HB signal. This is the audio output from
> while this is the pointer found on your Web page to the
> processed by Linrad
> to my highly untrained ears they sound of course
> been processed differently, but, to my ears, both
>unreadable... but as
> said my CW skills are minimal. Maybe those with "Golden
> have more luck than me.... :-)
> 73 Alberto I2PHD
> Yahoo! Groups Links
> <*> To visit your group on the web, go to:
> <*> Your email settings:
> Individual Email | Traditional
> <*> To change settings online go to:
> (Yahoo! ID required)
> <*> To change settings via email:
> <*> To unsubscribe from this group, send an email to:
> <*> Your use of Yahoo! Groups is subject to:
I'm following this discussion and read about your DSP challenge with
Here just a couple of comments which maybe can add some information
about Perseus, nothing more.
> Surely the buffer amplifier in front of the A/DThe driver full-scale output is of course (much) greater than the A/D
> could be the source of the problem, but I find it hard to
> believe that Nico did not ensure that the analog part has an
> adequate dynamic range.
full-scale input range. I would have not accepted the contrary ;-)
The OIP3 of the Perseus driver, the TI THS4509, is something like +42
dBm, so maybe someway inferior to the IIP3 of the A/D (for the known
reasons I would call it in another way, anyway), but still high to
grant an overall IIP3 of the receiver greater than +30 dBm.
It can be easily computed that with a 110 dB A/D SFDR, a receiver
input clipping level of say -5dBm, a IIP3 of +30 dBm and a -125 dBm
SSB MDS (numbers taken from the Perseus average spec) the limiting
factor of the dynamic range is not the IIP3 itself but the SFDR of
the A/D. With the above numbers, the IMD3 DR is 103 dB, but of course
spurs from the A/D could be as high as -115 dBm (-5-110).
So far,IMHO, it is not really impossible or difficult, as you wrote
in another message, to meet the ADC requirements with the receiver
front-end, at least as long as the receiver is used in its DR
specification. Of course, if you would like that IMD3 products stayed
below the MDS for any input level from a -125 dBm MDS up to the -5
dBm ADC clipping level, the IIP3 of the front-end should be something
like an incredible +55dBm. In this case, however, the IMD3 DR of the
receiver would be 120 dB, as large as the BDR. I believe that this is
something more than actual receiver technology can really afford (but
maybe I'm wrong).
At last, front-end filters in Perseus should not be the reason of any
intermodulation. All the inductors of the filters, even those of
large values, are all air-wound. The only ferrite component is the
transformer which drives the TH4509.
> All the inductors of the filters, even those ofAaah! Well engineered! This box is too good to have a CE mark (China Export) ;-)
> large values, are all air-wound.