Hi If you take a look at the CMOS switches the direct conversion radio guys use, that's probably the easiest way to do the inversion. If they work well enough for a 30 MHz receiver, they probably will do just fine at 60 KHz. You probably would need a transformer in and a transformer out to keep things from getting to crazy noise wise. Simple couple of windings on a ferrite core should do just fine. Mini Circuits probably would be happy to sell you some as well.
I suspect you still need some sort of smarts involved simply to decide when to flip and when not to flip the signal. There's not enough to it to tax a cheap micro. Until they publish the "final" description of the modulation, I see no reason to go crazy writing code. Bob On Jun 15, 2013, at 12:33 PM, paul swed <[email protected]> wrote: > Bob no idea actually. But I do have the MSA 8160 rcvr on the bench and > simply can add the ne602. Also John Lowe from NIST actually > suggested/alluded to it over a year ago before he ducked out of time-nuts. > Its one of those fairly obvious answers. Definately not high on my priority > list. > Its as simple as I can imagine and doesn't use a FPGA or a uProc. > > As a discussion point on "simple" folks that work armchairs always have > something to say. But they rarely contribute to actual results. Far to easy > to debate simple. > > I know that you do help and I appreciate that. > Regards > Paul. > > > > On Sat, Jun 15, 2013 at 10:35 AM, Bob Camp <[email protected]> wrote: > >> Hi >> >> I believe I suggested this approach back about 6 months ago. There was a >> significant controversy about my use of the term "seems simple". >> >> Bob >> >> On Jun 15, 2013, at 9:51 AM, paul swed <[email protected]> wrote: >> >>> As I have been working on the d-psk-r its always been clear to me that a >>> simple non invasive approach to fix the 8170 class clocks existed. Since >>> the d-psk-r has consumed far more time then I ever would have guessed, I >>> would like to offer the approach up to the group. Maybe someone else has >>> some time to do this and it is easy. However with easy there are trade >> offs >>> as always. >>> >>> Here is teh approach >>> Using a wwvb clock chip like the MSA 8160 still obtainable (And no doubt >>> others say from an atomic clock) >>> Take the am time code out. >>> Feed that to a NE 602, SA602 or 612 series balanced modulator there all >> the >>> same as a AM modulator. >>> The 60Khz comes from a tuning fork crystal using the SA602 built in >>> oscillator. >>> Attenuate the output to 100 uv or so. I don't think you really need any >>> filtering. >>> >>> There you go. No real magic. >>> Possible downsides >>> Drift as the tuning fork crystals are OK. If true use a better source >>> I expect the tick to be slightly delayed by this method. >>> 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.
