[hpsdr] FMT differences between Magister and Metis?

[Ron Mozer]

Hello Phil,

Clearly you are the wizzard, and I think it may be just the luck of the 
draw depending on variable things like propagation for which I have no 
controls.  Although I ran it several times last night with results 
appearing to be related to the board, today they seem to be the same.

I am using the K1JT FMT that is part of the WSPR package.  Please see 
http://physics.princeton.edu/pulsar/K1JT/wspr.html .  All I am doing is 
running the FMT program, stopping PowerSDR, swapping between Magister 
and Metis, powering up again and repeating.

I'll see if I can get some more empirical results with better controls.  
I'm just a piker when it comes to trying to zero in on an unknown 
frequency with sub Hz resolution, but I'm learning....

Thanks,
Ron


On 3/4/2011 9:58 PM, Phil Harman wrote:
> Hi Ron,
>
> I can't think of any obvious way that Metis and Ozy/Magister can have 
> an effect on the receive frequency.
>
> Both send a 32 bit word to the CORDIC on Mercury.  If we were dropping 
> some of the lower 32 bits when using Metis then I would have expected 
> the frequency error to be 2^n based which it is not.  I guess we could 
> be altering the bits in some manner but that would be odd this late in 
> the development.
>
> The rate that  Metis or Ozy/Magister sends data to the PC is 
> determined by the 122.88MHz clock on Mercury since neither boards have 
> any timing knowledge.
>
> If you could please describe in a little more detail how you are 
> measuring the frequency error then perhaps I can reproduce it here.
>
> 73 Phil...VK6APH
>
>
> ----- Original Message ----- From: "Ron Mozer" <ron at wy3d.org>
> To: <hpsdr at openhpsdr.org>
> Sent: Saturday, March 05, 2011 8:56 AM
> Subject: [hpsdr] FMT differences between Magister and Metis?
>
>
>> ***** High Performance Software Defined Radio Discussion List *****
>>
>> Very curious, using the PowerSDR v1.19.3.3 (W5WC 2/25/11) I did a 
>> little experiment using K1JT's FMT (Frequency Measuring Test) 
>> utility.   My setup includes an Excalibur feeding a 10 MHz reference 
>> to the Atlas buss from a Thunderbolt GPS with a short term error in 
>> the realm of 0.03 PPB.
>>
>> By only changing the physical interface and matching it in the 
>> PowerSDR setup, I see about 0.35 Hz difference in the measured 
>> frequency between the USB interface on Magister and the IP interface 
>> on Metis when referencing WWV on 10MHz.
>>
>> In fact Metis has a repeatable error of  about -0.35 Hz compared to a 
>> more respectable error of less than 0.007 Hz when using Magister.
>>
>> If it makes a difference, I still have Metis V 1.0 installed and am 
>> running on a Windows 7 x64 platform.
>>
>> Perhaps those playing with FMT can provide some insight as to tricks 
>> they do with the HPSDR to minimize error.  I was thinking of trying 
>> to use ghpsdr3 running multiple receivers.  Have a couple receivers 
>> locked to different WWV frequencies and one measuring the unknown 
>> frequency.  This might allow one to zero out apparent frequency 
>> changes due to the Doppler effect.
>>
>> 73 de WY3D
>> Ron
>>
>> _______________________________________________
>> 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/
>>
>>
>>
>> -----
>> No virus found in this message.
>> Checked by AVG - www.avg.com
>> Version: 10.0.1204 / Virus Database: 1435/3482 - Release Date: 03/04/11
>>
>
>
>



 1299350576.0