[hpsdr] Frequency offset using Excalibur

[Graham / KE9H]

Glenn:

There are several things going on here, so be careful not to mix them up.

The specified accuracy of the master oscillator in the Flex-3000 is ± 
1.5 ppm
or ± 15 Hertz at 10 MHz.  You can trim this to essentially zero by using 
the
Setup\General\Hardware Config\Expert\Clock Offset control. You are not
restricted to one Hertz steps when trimming the master oscillator,
and can trim it in much, much finer steps.

The "0-Beat" button is intended for fast tuning of CW signals, and only 
has the
resolution/accuracy of one panadaptor bin width, so an error of ten or 
fifteen Hertz
is typical.  It is NOT appropriate for use for high accuracy frequency 
measurement .

PowerSDR itself, as used with Flex radios, and as it was used with 
Mercury/Ozy,
does not contribute to frequency error. It gets a packet of data, and 
returns a packet of
processed data. As long as the packets are sent and received in time 
with the
clock domain in the radio, and you are getting your receive audio 
through the
speaker/headphone jack on the radio, you should not see any errors from
PowerSDR.

If you want to dig deeper into any of this, as it applies to your 
Flex-3000, it
is probably appropriate to move over to one of the FlexRadio forums.

--- Graham / KE9H

==


On 12/8/2012 4:49 PM, Glenn Thomas wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
> FWIW, I just calibrated my Flex-3000/PSDR 2.4.4 system on 10 MHz WWV. 
> The prompt display of the WWV/10MHz carrier frequency is 9.999 983 MHz 
> - a 17 Hz error. I also tried the phase display. Minimum rotation 
> occurs when the receiver is tuned to 10.000 002 MHz. That's in CW mode 
> with the 25 Hz filter. Since this system has no HPSDR components 
> except for the stock version of PowerSDR, I suggest that the problem 
> may be with PSDR itself and not have anything to do with HPSDR.
>
> If I repeatedly cluck the "0 beat" button I get several different 
> frequencies indicated 10.000 008, 10.000 000, 10.000 014, 10.000 006, 
> 9.999 997, 10.000 011, 10.000.003. Of course, even ionospheric doppler 
> over 60 seconds will not change the frequency of WWV that much that 
> fast. Likewise, the Flex-3000 clock is unlikely to drift that far that 
> fast either. Note that the tuned frequency is steady at 10.000000 MHz 
> - the measured frequency is that displayed on the lower right hand 
> corner of the panafall display.
>
> I also tried the phase display. Minimum rotation occurs when the 
> receiver is tuned to 10.000 002 MHz (1 Hz is the minimum step size). 
> That's in CW mode with the 25 Hz filter.
>
> Thus, even immediantly after calibration, the frequency accuracy is no 
> better than a few Hz, even if the precision is much better than this.
>
> Perhaps someone with better knowledge of PSDR can weigh in on this?
>
> 73 de Glenn WB6W
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: 
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>


 1355026149.0