Re: [time-nuts] Newbie question about GPS
On Montag, 2. März 2020 22:37:02 CET Björn wrote: > Hi Mattias, > > Then if you lock and control the 48MHz. You can also control the 1PPS edge > to the accuracy of the computed time... add 1s drift and lock uncertainty. > > Read the beginning of the Tbolt manual. The Tbolt lets it’s locked 10MHz > tick the whole receiver. Yep, you can do that. Any chance for getting that scheme to work with one of the stand-alone timing receivers? > > /Björn > > Sent from my iPhone > > > On 2 Mar 2020, at 22:14, Matthias Welwarsky > > wrote: > > > > On Montag, 2. März 2020 13:14:55 CET Anton Moehammad via time-nuts wrote: > >> 2. In the UCenter or almost any software > >> control for GPS module/receiver we can find data for next pps, why if the > >> software can calculate the PPS should be they do not use it to adjust the > >> PPS output so the PPS jitter is minimal ? > > > > There are no clocks in a GPS receiver with picoseconds resolution and > > everything the GPS outputs will be naturally restricted to its internal > > clock granularity. The time pulse signal from a Ublox module can > > therefore only be accurate to tens of nanoseconds (around 20ns assuming > > an internal 48MHz clock). However, the time solution calculated by the > > receiver will be much better. The quantization error reported in the TP5 > > message can simply be added to the timestamp of the received pulse to > > calculate its true arrival. > > > > BR, > > Matthias > > > >> ___ > >> time-nuts mailing list -- time-nuts@lists.febo.com > >> To unsubscribe, go to > >> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and > >> follow > >> the instructions there. > > > > ___ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and > > follow the instructions there. > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow > the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Newbie question about GPS
Hi Mattias, Then if you lock and control the 48MHz. You can also control the 1PPS edge to the accuracy of the computed time... add 1s drift and lock uncertainty. Read the beginning of the Tbolt manual. The Tbolt lets it’s locked 10MHz tick the whole receiver. /Björn Sent from my iPhone > On 2 Mar 2020, at 22:14, Matthias Welwarsky wrote: > > On Montag, 2. März 2020 13:14:55 CET Anton Moehammad via time-nuts wrote: >> 2. In the UCenter or almost any software >> control for GPS module/receiver we can find data for next pps, why if the >> software can calculate the PPS should be they do not use it to adjust the >> PPS output so the PPS jitter is minimal ? > > There are no clocks in a GPS receiver with picoseconds resolution and > everything the GPS outputs will be naturally restricted to its internal clock > granularity. The time pulse signal from a Ublox module can therefore only be > accurate to tens of nanoseconds (around 20ns assuming an internal 48MHz > clock). However, the time solution calculated by the receiver will be much > better. The quantization error reported in the TP5 message can simply be > added > to the timestamp of the received pulse to calculate its true arrival. > > BR, > Matthias > >> ___ >> time-nuts mailing list -- time-nuts@lists.febo.com >> To unsubscribe, go to >> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow >> the instructions there. > > > > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Newbie question about GPS
On Montag, 2. März 2020 13:14:55 CET Anton Moehammad via time-nuts wrote: >2. In the UCenter or almost any software > control for GPS module/receiver we can find data for next pps, why if the > software can calculate the PPS should be they do not use it to adjust the > PPS output so the PPS jitter is minimal ? There are no clocks in a GPS receiver with picoseconds resolution and everything the GPS outputs will be naturally restricted to its internal clock granularity. The time pulse signal from a Ublox module can therefore only be accurate to tens of nanoseconds (around 20ns assuming an internal 48MHz clock). However, the time solution calculated by the receiver will be much better. The quantization error reported in the TP5 message can simply be added to the timestamp of the received pulse to calculate its true arrival. BR, Matthias > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow > the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Newbie question about GPS
On Mon, 2 Mar 2020 09:13:24 -0500 Bob kb8tq wrote: > > Hi All,I hope some one can help me to find the answer.1. Let say some one > > can extract the L1 carrier frequency 15xx MHz and convert it down to let > > say 1MHz and put it to phase comparator and made the PLL to generated the > > LO to convert and extract the carrier. Is this possible ? Have some has do > > it ? Can share any thoughts ? > > The carrier is significantly doppler shifted due to the way the satellites > orbit the earth. Without correcting for that, the carrier makes a poor > reference. This is only the first of *many* corrections applied….. To put a few numbers in here: Doppler shift a stationary GPS receiver sees is about +/-6kHz or about 4ppm. On top of that we have ionospheric delay variation that is in the order of several 10s of ns. Tropospheric delay in several ns. We are not yet talking about the relatively high noise in the signal. Even after decorrelation we are only around 20-50dB SNR (low SNR -> high jitter; an XO has something in the order of 80-120dB SNR an OCXO is around 140-160dB). Or things like multi-path. All things which are fairly detrimental to getting a good frequency standard out of the GPS carrier. So you _need_ the full system solution to get to have a reasonably stable output. And even then, GPS is very noisy. A cheap OCXO is more stable than GPS up to several 10s of seconds. A good OCXO will be more stable than GPS up to several 100s to a few 1000s of seconds. If you go further, a cheap Rb vapor cell standard like the FE-5680A is more stable than GPS up to around 10ks. A good Rb standard like the PRS10 or HP5065 is more stable than GPS for up to 100ks (about a day) If you go for what is curently up in research for Rb standards, we are probably talking about 2-5 days for the cross over (look at the papers published by Mileti/Affolderbach's group at University of Neuchâtel and Micalizio/Calosso/Gozzelino at INRIM in Torino). Attila Kinali -- In science if you know what you are doing you should not be doing it. In engineering if you do not know what you are doing you should not be doing it. -- Richard W. Hamming, The Art of Doing Science and Engineering ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Newbie question about GPS
Hi > On Mar 2, 2020, at 7:14 AM, Anton Moehammad via time-nuts > wrote: > > Hi All,I hope some one can help me to find the answer.1. Let say some one can > extract the L1 carrier frequency 15xx MHz and convert it down to let say 1MHz > and put it to phase comparator and made the PLL to generated the LO to > convert and extract the carrier. Is this possible ? Have some has do it ? Can > share any thoughts ? The carrier is significantly doppler shifted due to the way the satellites orbit the earth. Without correcting for that, the carrier makes a poor reference. This is only the first of *many* corrections applied….. > 2. In the UCenter or almost any software control for GPS module/receiver we > can find data for next pps, why if the software can calculate the PPS should > be they do not use it to adjust the PPS output so the PPS jitter is minimal ? The correction information can get the PPS to error levels dimensioned in picoseconds. Doing that with some sort of external corrector is complex and expensive (if it can be done at all). The way they do it is quite adequate for a GPSDO application. It costs them next to nothing to do. > 3. Why if we use real time RTK for better accuracy in term of position the > PPS is not changed ? ( At least in my setup) ?Are the RTK software only > process data and show the result in the screen without change anything in > module/receiver ?Any software can do a real change to receiver for better PPS > out based from any corrections available (ionosphere, Tidal etc) ? It depends on which RTK streams you are using. Are you bringing in a clock correction stream? In any case, the change in the PPS is going to be very small. You would need a very good reference to compare to to see it happen. Bob > Thank YouAnton > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] Newbie question about GPS
Hi All,I hope some one can help me to find the answer.1. Let say some one can extract the L1 carrier frequency 15xx MHz and convert it down to let say 1MHz and put it to phase comparator and made the PLL to generated the LO to convert and extract the carrier. Is this possible ? Have some has do it ? Can share any thoughts ?2. In the UCenter or almost any software control for GPS module/receiver we can find data for next pps, why if the software can calculate the PPS should be they do not use it to adjust the PPS output so the PPS jitter is minimal ?3. Why if we use real time RTK for better accuracy in term of position the PPS is not changed ? ( At least in my setup) ?Are the RTK software only process data and show the result in the screen without change anything in module/receiver ?Any software can do a real change to receiver for better PPS out based from any corrections available (ionosphere, Tidal etc) ? Thank YouAnton ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] Newbie Question about GPS
Hi All,I hope some one can help me to find the answer.1. Let say some one can extract the L1 carrier frequency 15xx MHz and convert it down to let say 1MHz and put it to phase comparator and made the PLL to generated the LO to convert and extract the carrier. Is this possible ? Have some has do it ? Can share any thoughts ?2. In the UCenter or almost any software control for GPS module/receiver we can find data for next pps, why if the software can calculate the PPS should be they do not use it to adjust the PPS output so the PPS jitter is minimal ?3. Why if we use real time RTK for better accuracy in term of position the PPS is not changed ? ( At least in my setup) ?Are the RTK software only process data and show the result in the screen without change anything in module/receiver ?Any software can do a real change to receiver for better PPS out based from any corrections available ? Thank YouAnton ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
