From: DL4YHF@aol.com
Message-ID: <cf.157b1523.29e7233a@aol.com>
Date: Thu, 11 Apr 2002 13:34:50 EDT
Subject: Re: LF: Phase meter for propagation experiment
To: rsgb_lf_group@blacksheep.org
MIME-Version: 1.0
Content-Type: text/html; charset=windows-1252
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>
Content-transfer-encoding: 8bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>Hello Jim, Brian, and group,
<BR>
<BR>Jim wrote:
<BR>
<BR>&gt;&gt;Signals like MSF have on-off keying. &nbsp;Feeding this signal into a digital 
<BR>divider , the counting process will come to a stop when the signal level 
<BR>falls below some threshold, so output from the divider will start and stop 
<BR>at points in the cycle which depend on the amplitude of the input. So the 
<BR>effect of dividing this signal to an audio frequency will be to introduce 
<BR>more or less random jumps in phase each time the signal is keyed, even 
<BR>though the original signal is phase coherent. Can the calibration software 
<BR>cope with this?
<BR>&lt;&lt;
<BR>
<BR>No, the calibration software cannot cope with phase jumps. I am not sure if MSF uses 100% on-off keying. If so, an additional PLL and a stable 60kHz oscillator with a narrow tuning range must be added before the digital frequency divider. This PLL then fills out the "gaps" if there are any.
<BR>
<BR>DCF77 does not use 100% on-off keying, the amplitude is only reduced to 25% during the short "gaps". But a PLL help here also as a narrow input filter.The loop filter should have a lowpass filter with &lt; 1 Hz bandwidth so the PLL does not unlock. 
<BR>
<BR>
<BR>
<BR>&gt;&gt; Also, what signals do you plan to monitor? I expect the FSK and PSK signals 
<BR>would be quite difficult, because both receiving stations would have to 
<BR>track all the phase variations at the modulation rate, and store a vast 
<BR>amount of data to allow the relative phase to be compared - a high degree 
<BR>of timing accuracy (&lt;1ms) would also presumably be required. Stations like 
<BR>DCF39 would be much easier since they are CW most of the time, but even 
<BR>here there is a jump in the phase every few seconds, so timing accuracy of 
<BR>the order of 1 second would be required for the two RX stations to compare 
<BR>the phase - still needing some work to achieve.
<BR>&lt;&lt;
<BR>
<BR>Not only difficult but almost impossible to observe FSK and PSK signals this way, because the phase meter uses a very narrow bandwidth (the phase meter's bandwidth can be set between 10 Hz to 41 milliHertz). So it only makes sense for a signal with a 'coherent' carrier. Maybe a well-equipped amateur can provide such a signal at the upper end of the band.
<BR>
<BR>To check the system without the need for such a test signal, observing DCF77, MSF, or one of the AM longwave broadcasters like Droitwich or DLF (Deutschlandfunk) may be a start (though I don't know about DLF's carrier accuracy). 
<BR> 
<BR>To monitor DCF39 for a long time and compare the results from several receiving sites will indeed be complicated because of the random phase jumps. The data would need to be logged in a file (not too complicated) with an electronic timestamp and a program (possibly a spreadsheet) would have to be written to subtract the phase values collected by a receiver close to DCF39 from the logged phase values recorded by "DX" stations. To me, this sounds a bit like overkill. For a start, take some phase plots (with phase jumps) and compare them visually. If they look good and 'say something', a more complex system like the one mentioned above can be attempted... &nbsp;
<BR>
<BR>Regards,
<BR> Wolf.
<BR>
<BR>
<BR>P.S: I visited the Trimble website today looking for the low-cost "Lassen SK II" GPS receiver. There seem to be no distributors in this region, only two in the Frankfurt/Main area. Anyone &nbsp;in northern DL / PA know other sources for this interesting modules, and thinks about buying one ?</FONT></HTML>