); SAEximRunCond expanded to false Errors-To: [EMAIL PROTECTED] RETRY I took another shot with the analyzer set to 10 kHz RBW, and the signal now looks more like 20 kHz wide.
See http://www.ko4bb.com/Timing/Loran-2.jpg Didier > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Didier Juges > Sent: Monday, November 12, 2007 10:02 PM > To: [EMAIL PROTECTED]; 'Discussion of precise time and frequency > measurement' > Subject: Re: [time-nuts] LORAN-C antenna > > I get this (see picture) with the spectrum analyzer and my > wire antenna. > That looks a lot cleaner than what I hear on the HP 3586. > > The spectrum analyzer was in peak hold, because the signal > has on/off modulation at several Hz. I got the picture after > about one minute. > > The picture is here: > > http://www.ko4bb.com/Timing/Loran.jpg > > (sorry it's 2.2 MB) > > Is that the Loran signal? Seems too narrow, based on your comment (20% > bandwidth) > > If so, I would like to find something smaller and maybe more > portable than my 20m wire up the tower :-) > > Didier KO4BB > > > -----Original Message----- > > From: [EMAIL PROTECTED] > > [mailto:[EMAIL PROTECTED] On Behalf Of Carl Walker > > Sent: Monday, November 12, 2007 8:31 PM > > To: Discussion of precise time and frequency measurement > > Subject: Re: [time-nuts] LORAN-C antenna > > > > You can build your own LORAN-C antenna coupler without too much > > trouble. > > > > A lifetime or so ago, I was one of the analog design team at the > > company that made Northstar LORAN-C receivers for marine > and aircraft > > navigation. The first generation of receivers used an > active coupler > > (MOSFET amplifier) with some high frequency roll-off to > avoid BC band > > overload. These receivers were quite simple, with bandpass > filters and > > a couple tunable notch filters to eliminate interference > close to the > > LORAN-C band - before some hard limiting to allow the uP > and sampler > > logic to process the information. This basic type of > antenna coupler > > is what I'm using at home (with a distribution buffer > > amplifier) for the 2100F, 2000C, and the various WWVB > receivers; this > > has been quite satisfactory - given the low-pass filtering in the > > coupler allows both 60 KHz and 100 KHz signals through quite nicely. > > > > Based on your location, you may or may not have interfering VLF > > signals in the neighborhood of LORAN-C; there's only one > real way to > > find out - have a look with the spectrum analyzer at the output of > > whatever you devise for an antenna coupler amplifier and see what's > > there. Also bear in mind the receiver itself is generally designed > > with filtering of its own (may or may not have internal, > fixed notch > > filters for close in interference in addition to some band-pass > > filtering), and may not require that you do all that much external > > filtering in the coupler itself. I must admit I've not > snooped around > > in either Austron for some time, and the details of the > those receiver > > designs escape me at the moment. > > > > If LORAN-C is all you're interested in receiving, you'd do > well with a > > bit of bandpass filtering before the amplifier stage in the antenna > > coupler to avoid overload and interference both above and below the > > desired signal. The energy in a LORAN-C pulse is very > broadband (a 20% > > bandwidth pulse), so making a filter that's as flat in > amplitude and > > group delay distortion over the 90-110 KHz band helps > preserve pulse > > envelope shape and zero crossings; liner-phase filters work > quite well > > here - although the skirt selectivity might not be all that you'd > > like. > > Preserving pulse fidelity is the key here. Pulse envelope shape is > > often critical - since many receivers use the envelope shape of the > > pulse to determine which zero crossing to track when > cycle-selecting. > > The other thing to bear in mind is that if you'd like to > use a short > > length of wire for your receiving antenna, the impedance of > the input > > bandpass filter needs to be quite high; as an example, we > used 8 foot > > CB-type whips for marine applications - and to approximate this > > antenna length with 50 Ohm signal sources, we used a 20 pF > series cap > > at 100 KHz. > > > > It's also interesting to note the diurnal effects due to sky-wave > > contamination of the pulses that was mentioned earlier. > Depending on > > amplitude and delay of this sky-wave signal, it's quite possible to > > get vector-sum effects that cause the perceived zero > crossings of the > > pulse to shift in time. Since the ionosphere isn't stable > in height, > > and the sky-wave signal often is greater in amplitude than > the ground > > wave signal by 10 to 20 dB, the point at which your receiver is > > tracking may appear to be time displaced in a jittery sort-of way > > (based on delay and amplitude of the sky-wave > > signal) - and the receiver tracking loops will follow this > > displacement early and late in time - making the oscillator > appear to > > be unstable. I believe this to be one possible cause for the > > degradation of stated accuracy by the 2100F for a given oscillator > > during the evening hours. I see the degradation clearly > here - whether > > the receiver is driven by the Austron xtal oscillator, or > the HP5061. > > Changes of two to three orders of reported magnitude are > not uncommon > > for the frequency offset display on my receiver between daytime and > > nighttime operation. > > > > Here we have another reason to maintain pulse fidelity - since too > > narrow a filter selectivity will tend to distort and > suppress the rise > > time of the pulse envelope, causing a receiver to select a zero > > crossing later than desired in the pulse to track - late > enough in the > > pulse to allow the sky-waves to have more effect on the > zero crossing > > its trying to track. > > > > I'd be happy to share some ideas on LORAN antenna couplers > and their > > design if anyone is interested - drop me a line. > > > > -Carl WA1RAJ > > > > > > > > > > > > _______________________________________________ > > time-nuts mailing list -- [email protected] To unsubscribe, go to > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- [email protected] To unsubscribe, > go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
