Hi Ok, well then, here’s a thought - 10 MHz divided by 164 is 60.975.xxxx Hz. 165, 166, 167 and 168 are also “close enough” to give you a reasonable tone out of the converter. That assumes you are after a 10 MHz standard at the end of all this. The fact that the tone frequency is something really weird isn’t as big a deal once you convert it to digital bits.
More or less - make the LO something easy, lock it up, and do the rest with software. Bob On Sep 22, 2014, at 7:59 PM, paul swed <[email protected]> wrote: > Bob > You missed nothing really. 100 Hz because thats what the NAA radio used > with the tracor doubler. So when I moved it to wwvb that was the baseband > signal. > Now that said I liked the 1 second phase flips at 100 Hz delay. Dog slow > and perfectly OK to resolve the phase flip and I can use my favored uproc. > > Whats funny is that any crystal close to anything existed. > Digikey says they have a 6.1Mhz xtal but when selected goes to 6.144 MHz. > I was amazed that both mouser and digikey had the 5.9904 xtals and both > stocked a variety of cases. Must be some story about whatever that > frequency is. > Did try lots of frequencies and divider math to come up with a simple LO > scheme for 61 or 59 KHz. Messy. > Regards > Paul > WB8TSL > > > On Mon, Sep 22, 2014 at 7:42 PM, Bob Camp <[email protected]> wrote: > >> Hi >> >> Is it really that much harder to process a 1 KHz note out of a direct >> conversion receiver than it is to process a 100 Hz note? >> >> Put another way - you already have a crystal filter in the front end of >> the “radio” right? It will do a better job knocking down an “adjacent >> channel” 2 KHz away than one 200 Hz away. Even without the crystal filter, >> the only likely thing you are trying to reject is noise. Effectively the >> “no filter” approach degrades your signal to noise by 3 to 6 db depending >> on the type of noise you are dealing with. >> >> I’m asking because I may have missed something that makes 100 Hz “magic” …. >> >> Bob >> >> On Sep 22, 2014, at 6:15 PM, paul swed <[email protected]> wrote: >> >>> One of the methods suggested for working with the wwvb psk is down >>> converting to a lower frequency. When experimenting with NAA I had built >> a >>> 24 Khz receiver that down converted that signal to a 100 Hz base band. >>> It was a minor effort to shift it to 60 Khz by changing the LO. I liked >>> what I saw. Essentially a simple direct conversion receiver with about 42 >>> DB of gain. >>> The PSK flips at every 1 second depending on the data and is 1/100Hz >>> delayed from the wwvb flip due to the down conversion process. This makes >>> the further processing very reasonable in just about any micro. >>> But the LO is a mess. Not bad if you have an HP3335a as I did till last >>> week. (Another project to repair) for 100 Hz the LO must be at 60.1 or >> 59.9 >>> Khz. Or synthesize it etc. Starting to get seriously ugly. >>> Or jump to a 1 Khz with 59 or 61 KHz. But I hated the idea of mixers and >>> filters. >>> >>> Any Time-Nuts seen those crystals or oscillators around lately? Seriously >>> missing from the general haunts. >>> >>> But turns out and heavens knows why there are plentiful 5.9904 MHz xtals >>> around. I ordered 5 from Mouser $ .46 ea. >>> 5.9904/100 = 59.904 as a LO and suspect it can be moved to the correct >>> frequency not that it has to be. >>> Not sure there is a real need to lock the LO at all. >>> So an exciting bit of news. >>> Regards >>> Paul >>> WB8TSL >>> _______________________________________________ >>> 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. _______________________________________________ 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.
