Re: [time-nuts] WWVB cheap chip saa6579 RDS decoder back to the chip
Hi Paul - If you can spend time on that: 1. Look for the different modulation spectrum between BPSK and RDS. The phase modulation angle is different, the bits are manchester encoded or such to get a hole on the carrier frequency (For the ARI carrier), the baseband is DBPSK. 2. Carefully adjust the clock generator. Very sensitive to it! 3. The demodulator can lock on two different phase states. But the difference is just where the clock transition is positioned in the data-bit. In locked state this should be constant and can only change the next time the chip switches von search to lock state (with a 50% probability). 4. Yes, often such chips are interesting but poorly documented. Sometimes it helps to look for similar devices like the SAA6588, TDA7330 , etc. Older datasheets often show more details. 5. I don't think you can use the chip with 1bit/sec. It is made for 1200bits/sec. I don't know if the internal digital part of the circuit is fully static. But surely the capacitors of the filter are to small for a 1200:1 dynamic range. Maybe I misunderstood this in your application. Have fun - Henry paul swed schrieb: Henry Its been a while since that thread and I have not done anything with the chip. But to answer your questions. Really good signal to noise. The modulator is 6 from the saa6579. Its a home brew BPSK modulator and the transitions are programmable. But I am following wwvbs 1sec per bit. So the phase is quite stable for long periods of time. Signal level is 1000uv but again that can be adjusted. Were is someone. My comment was someone on time-nuts suggested that you had to use the clock and a flip flop to properly see the data. Thats what you are also saying. As to the wwvb modulation scheme I fully understand that. But I actually do not know a lot about RDS and how it might be different from wwvbs method of transmission. I believe there is plenty of information on RDS. I just don't have a lot of time to figure it out. I will get back to the chip at some point it is really interesting but very poorly documented unfortunately. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB cheap chip saa6579 RDS decoder back to the chip
paul swed schrieb: To the saa6579 As mentioned in the other thread. Simple to hook up. Cheap Requires 1000uv or more so that ends up making things more complicated. S/N is important, not so absolute amplitude. But in my case simply did not really work at 57 or 60 Khz. The chip looks for correct phase for about 50ms (max. 100ms). Did you check carefully that in this time-frame the phase never inversed or amplitude dropped down temporary? Maybe you misunderstood the modulation scheme anyway? It is not native BPSK! As some someone pointed out you may have to take the clock and data to a flip flop to get the correct information. Where is someone? As I understood the datasheet you get maximum processing time (set-up time) for the data-bit if you use a D-FF delaying DATA strobed by CLOCK. - Henry -- ehydra.dyndns.info ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] WWVB cheap chip saa6579 RDS decoder back to the chip
Henry Its been a while since that thread and I have not done anything with the chip. But to answer your questions. Really good signal to noise. The modulator is 6 from the saa6579. Its a home brew BPSK modulator and the transitions are programmable. But I am following wwvbs 1sec per bit. So the phase is quite stable for long periods of time. Signal level is 1000uv but again that can be adjusted. Were is someone. My comment was someone on time-nuts suggested that you had to use the clock and a flip flop to properly see the data. Thats what you are also saying. As to the wwvb modulation scheme I fully understand that. But I actually do not know a lot about RDS and how it might be different from wwvbs method of transmission. I believe there is plenty of information on RDS. I just don't have a lot of time to figure it out. I will get back to the chip at some point it is really interesting but very poorly documented unfortunately. Regards Paul WB8TSL On Thu, Nov 15, 2012 at 7:14 PM, ehydra ehy...@arcor.de wrote: paul swed schrieb: To the saa6579 As mentioned in the other thread. Simple to hook up. Cheap Requires 1000uv or more so that ends up making things more complicated. S/N is important, not so absolute amplitude. But in my case simply did not really work at 57 or 60 Khz. The chip looks for correct phase for about 50ms (max. 100ms). Did you check carefully that in this time-frame the phase never inversed or amplitude dropped down temporary? Maybe you misunderstood the modulation scheme anyway? It is not native BPSK! As some someone pointed out you may have to take the clock and data to a flip flop to get the correct information. Where is someone? As I understood the datasheet you get maximum processing time (set-up time) for the data-bit if you use a D-FF delaying DATA strobed by CLOCK. - Henry -- ehydra.dyndns.info __**_ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/** mailman/listinfo/time-nutshttps://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] WWVB cheap chip saa6579 RDS decoder back to the chip
Hello to the group. I see the thread took a left turn a day or so ago so thought I would fire up a new one for the interested in the phillips saa6579 RDS chip. Several others have purchased this chip. Jameco electronics had them for 10 or 20 cents. I picked up 10. You had order that many as I recall. Its a costas loop but will say the documentations skimpy. My interest Restoring the old style frequency rcvrs to operation. Like the HP vlf117, tracor 599, fluke 207 A byproduct of accomplishing that is you can restore the old time rcvrs also like the spectracom 8170s. Lastly end up with the new wwvb code to do something with. Thats really low on my list of interests actually. Have tried many approaches to date. The doubling and dividing stuff. Fades and noise are a killer for this approach. So stopped work on those. One hack that works is a small fast uproc that detects phase flips using a spectracom 8163 VCO chain as the reference. When a flip occurs you flip an rf amplifier phase to invert it. Its a hack and you need a spectracom. But this voids a favorite requirement of mine KISS. So currently building a discreet costas loop using ad633s see how that works. To the saa6579 As mentioned in the other thread. Simple to hook up. Cheap Requires 1000uv or more so that ends up making things more complicated. But in my case simply did not really work at 57 or 60 Khz. As some someone pointed out you may have to take the clock and data to a flip flop to get the correct information. Regards Paul WB8TSL ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
