I have found that the cheap V.KEL SIRF=III modules (I paid $15-$20 for three) have excellent indoor performance with their built-in patch antenna. They don't do GLONASS. I even get indoor tracking with the module sitting on the ground floor of a 2 story hose with the patch antenna facing the floor! The NEO M8 is a decent device. I've seen mine tracking over 24 sats. The module that I have has a U.FL antenna connector with pads for adding an edge-launch SMA connector (I hate U.FL connectors). I seem to remember that they can't track GPS, GLONASS, and BEIDOU at the same time. I have gotten surprisingly good performance with a cheap GPS/GLONASS puck like: http://www.ebay.com/itm/201698154683?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT For NTP applications, multi-path errors will be swamped out by the "noise" in NTP. I don't think that a timing receiver adds much value except possibly the ability to still work with one or two sats visible. Many GPS/GLONAS/BEIDOU capable timing receivers don't do precision timing with more than one sat system type enabled... I think the Venus devices are like that... they only to GPS timing.

Millimeter accurate GPS in smartphones and self driving cars would result from tiny atomic clocks -- http://www.nextbigfuture.com/2016/11/millimeter-accurate-gps-in-smartphones.html https://arxiv.org/abs/0707.4624 including energy harvesting, bio-sensing and quantum nanoelectronics. They are producing designer endohedral fullerene molecules with tailored electronic properties. Designer Carbon Materials Ltd is a spin-out company from the University of Oxford. It is based on research led by Dr Kyriakos Porfyrakis and his academic group of 9 researchers. They have developed technology for the scaled-up production of endohedral fullerenes. Our patented arc-reactor system can bring endohedral metallofullerene production to the gram-scale and beyond, faster and more efficiently than conventional arc reactors. They have access to state-of-the-art facilities for the purification of a range of fullerene molecules, including endohedral metallofullerenes and endohedral nitrogen fullerenes. Endohedral fullerenes, also called endofullerenes, are fullerenes that have additional atoms, ions, or clusters enclosed within their inner spheres. The first lanthanum C60 complex was synthesized in 1985 and called La@C60.[2] The @ (at sign) in the name reflects the notion of a small molecule trapped inside a shell. Two types of endohedral complexes exist: endohedral metallofullerenes and non-metal doped fullerenes. Nitrogen endohedral fullerenes is being used to create a small and portable atomic clock – the most accurate time-keeping system in the world – and could make the GPS navigation on driverless cars accurate to 1 millimeter. "At the moment, atomic clocks are room-sized," said Lucius Cary, a director of the Oxford Technology SEIS fund, which now holds a minority stake. "This endohedral fullerene would make it work on a chip that could go into your mobile phone. In 2007, there was an arxiv paper which described the design of a Micron-Scale Atomic Clock Nitrogen atom is introduced into a fullerene cage. This endohedral fullerene is then coated with an insulating shell and a number of them are deposited as a thin layer on a silicon chip. Next to this layer a GMR sensor is fabricated which is close to the endohedral fullerenes. This GMR sensor measures oscillating magnetic fields on the order of micro-gauss from the nuclear spins varying at the frequency of the hyperfine transition (413 MHz frequency). Given the micron scale and simplicity of this system only a few transistors are needed to control the waveforms and to perform digital clocking. This new form of atomic clock exhibits extremely low power (nano watts), high vibration and shock resistance, stability on the order of 10^-9, and is compatible with MEMS fabrication and chip integration. As GMR sensors continue to improve in sensitivity the stability of this form of atomic clock will increase proportionately. It is possible to separate each endohedral fullerene from its neighbors by coating it with a glass shell. Silica gel, an inorganic polymer, has a three-dimensional network and can easily be synthesized via the sol-gel route. Fullerenes cannot be incorporated into sol-gel glasses homogeneously due to low solubility. This problem can be overcome by functionalization of the fullerenes with such groups as will form some kind of bond (hydrogen, van der Waals, or covalent) with the growing silica network The simple scheme discussed gives us a micron scale atomic clock with 10^−9 accuracy and a power dissipation of a nanowatt (10 nW capacitive drive but we can use resonant circuits to store the energy). This will likely be adequate for many mobile/sensor net applications but not adequate for more demanding situations. What can be done? First, as GMR sensors improve (BMR, etc.), we can use more diluted fullerene stacks to gain a sharper line by a cubic factor in separation as we lose an equal amount of magnetic signal. A nanoscale-precise placing of fullerenes would give us a very well determined perturbation situation that can be exploited for accuracy. In the limit of true nanotechnology the ultimate clock is a single fullerene with considerable shielding. This should be competitive with very good atomic clocks of vastly more volume. 2016-11-30 21:36 GMT+00:00 Mark Sims <holrum@hotmail.com>: > I have found that the cheap V.KEL SIRF=III modules (I paid $15-$20 for three) have excellent indoor performance with their built-in patch antenna. They don't do GLONASS. I even get indoor tracking with the module sitting on the ground floor of a 2 story hose with the patch antenna facing the floor! > > The NEO M8 is a decent device. I've seen mine tracking over 24 sats. The module that I have has a U.FL antenna connector with pads for adding an edge-launch SMA connector (I hate U.FL connectors). I seem to remember that they can't track GPS, GLONASS, and BEIDOU at the same time. > > I have gotten surprisingly good performance with a cheap GPS/GLONASS puck like: > http://www.ebay.com/itm/201698154683?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT > > For NTP applications, multi-path errors will be swamped out by the "noise" in NTP. I don't think that a timing receiver adds much value except possibly the ability to still work with one or two sats visible. Many GPS/GLONAS/BEIDOU capable timing receivers don't do precision timing with more than one sat system type enabled... I think the Venus devices are like that... they only to GPS timing. > _______________________________________________ > 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.

On 2016-11-30 13:45 , André Esteves wrote: > Millimeter accurate GPS in smartphones and self driving cars would > result from tiny atomic clocks I saw a different form of this article a month or so back, and for the life of me I can't figure out how having a tiny atomic clock helps GPS accuracy at all (nor how it would help w/ inertial reckoning, which was a claim I saw in a different article on the same folks). Especially since (AFAIK) the error budget of the GPS system in general far exceeds "millimeter" accuracy, without post-processing all your data... I'm also pretty sure numerous people here would object to the characterization of atomic clocks as "room sized", in general. ;) -j

Hi Most of this is “I have an idea, I can (loosely) connect the idea to big volume, give me *big* piles of money”. Trying to work it out on a technical basis is not going to work very well …. Bob > On Nov 30, 2016, at 6:49 PM, Jay Grizzard <elfchief-timenuts@lupine.org> wrote: > > On 2016-11-30 13:45 , André Esteves wrote: >> Millimeter accurate GPS in smartphones and self driving cars would >> result from tiny atomic clocks > I saw a different form of this article a month or so back, and for the life of me I can't figure out how having a tiny atomic clock helps GPS accuracy at all (nor how it would help w/ inertial reckoning, which was a claim I saw in a different article on the same folks). Especially since (AFAIK) the error budget of the GPS system in general far exceeds "millimeter" accuracy, without post-processing all your data... > > I'm also pretty sure numerous people here would object to the characterization of atomic clocks as "room sized", in general. ;) > > -j > _______________________________________________ > 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.

Before joining time-nuts I had purchased a QLG1 GPS receiver kit from www.qrp-labs.com. It was $23 plus a few bucks for shipping. I installed the patch antenna as it suited my purposes, but there is the means to omit the patch antenna and use an SMA to go to the antenna of your choice. QRP-Labs claims that the ground plane on the side the patch antenna is mounted provides another 4+ DB of gain. The actual GPS chip is the YIC51612. I am running it sub-grade in a “split entry” where the first floor is mostly below ground on a slopes property. It connects to an Arduino programmed to sent the time from the $GPRMC sentence to an LCD. I will ultimately use the 1PPS output from the receiver as well. I use the presence of the 1PPS signal as an indication that the receiver is happy with life at that moment. It has been chugging away for a few months now. Cheers, -rick, K7LOG

On 30/11/2016 4:36 PM, Mark Sims wrote: > ... V.KEL SIRF=III modules (I paid $15-$20 for three)...They don't do GLONASS. I think I should have GLONASS, to maximize my chances. > The NEO M8 is a decent device. I've seen mine tracking over 24 sats. The module that I have has a U.FL antenna connector with pads for adding an edge-launch SMA connector (I hate U.FL connectors). I seem to remember that they can't track GPS, GLONASS, and BEIDOU at the same time. I'm seeing them listed as "Concurrent reception of up to 3 GNSS (GPS, Galileo, GLONASS, BeiDou)". I'm hoping that means multiple sats, but limited to three types. > I have gotten surprisingly good performance with a cheap GPS/GLONASS puck like: > http://www.ebay.com/itm/201698154683?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT Good to know. With my limited view of the sky, I'd be angling it anyway, so perhaps it wouldn't be bad with snow. Thanks!