Re: [time-nuts] GPS for ntp
On 21 Oct, 2014, at 08:58 , Simon Marsh subscripti...@burble.com wrote: How do you map the timer counter value in to a PPS timestamp ? (that is, how do you turn the HW counter value in to what the OS thought the time was when the event occured ?) On the NetBSD prototype I have the clock adjustment system call interface is expanded to deal with multiple clocks, only one of which is the system clock. The HW counter becomes its own clock, which is the clock in which PPS measurements are expressed and which is adjusted into alignment with the PPS data. The system clock is adjusted into alignment with the HW counter clock using offset data from PIO polling of the clock pair. The IEEE1588 timestamp counter becomes a third clock, which gets adjusted into alignment with the system clock for use as a PTP server, or which is used to adjust the system clock when operating as a client. For the beaglebone this is probably overkill; since the clocks are all synchronous the system-peripheral clock polling essentially determines a constant offset, after which you can keep them in sync by making the same relative rate adjustments to all clocks. For the general IEEE1588 case, however, the counter being sampled at the ethernet interface is often clocked by a different crystal then the clock you would prefer to use as the system clock, and the process of steering one clock into synchronization with another needs to be more complex. I should note that none of these clock adjustments really requires a PLL or other feedback control loop, nor does NTP, since no clock hardware is actually adjusted. The crystals are all free running and are unaffected by the adjustments. What is adjusted is instead a paper clock, that is the adjustment is to the arithmetic done to convert each free running counter to a time of day, and this can be done open loop, with perfectly predictable results and with no feedback control, by just doing the adjustment arithmetic accurately and transparently. The thing the PLL does for ntpd, then, is to allow it to deal with (paper) clock adjustment interfaces which don't do the arithmetic accurately, or at least don't tell you what they actually did, so that the arithmetic done can only be determined by further measurement. This is unavoidable if you need to deal with a big variety of operating systems, I guess, but it does make the problem harder than if the adjustment interface is fixed and feedback loop eliminated, leaving just the measurement problem. Dennis Ferguson ___ 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] Ublox neo-7M GPS
On 21 Aug, 2014, at 16:27 , Tony tn...@toneh.demon.co.uk wrote: David, No problem, its still set-up. As you'd expect its rock solid at 8MHz with no visible jitter. Can you point me to the datasheet you're referring to? The MAX-7 and NEO-7 datasheets don't provide any information on clocking. In the 'u-blox 7 Receiver Description Including Protocol Specification V14' the only clue as to the clocking characteristics is that the timepulse output must be configured with a minimum high or low time of 50ns or pulses may be lost. Make of that what you will! Tony It sounds like this part is similar to the LEA-6T (including the 48 MHz reference oscillator). This white paper has some information about the frequency output of the latter: http://www.u-blox.com/images/downloads/Product_Docs/Timing_AppNote_%28GPS.G6-X-11007%29.pdf See, e.g., figures 11 and 12. I think the cleanliness of the 8 MHz phase is a little bit misleading, however, since the clean part isn't actually 8 MHz. It is instead the frequency of the free running 48 MHz reference divided by 6, and it will still be throwing in the occasional short or long cycle to correct the 48 MHz oscillator frequency error and make the long term average come out at a true 8 Mhz. In some sense this is the high frequency equivalent of the 1 PPS hanging bridge case. The output at any frequency has a phase error of +/- 10.5 ns but at 8 MHz the phase error of the output changes very slowly, and can hang near one of the extremes for long periods, so it requires very long integration times to reduce that to zero. If you were using this output to drive a cleanup PLL (which I would call the DO in a GPSDO) I think you would actually be better off using the output at 10 MHz since with that integrating over just a few microseconds of the jitter reduces the short term average phase error by a factor of 5, to +/- 2.1 ns, and an odder divider might be better still. Dennis Ferguson ___ 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] Synergy-GPS SSR-6tru problems
On 15 Jul, 2014, at 16:32 , Bob Stewart b...@evoria.net wrote: I do not know what the problem is. The UT+ works (with the appropriate software). The SSR-6tru is deaf using all the software I have available to me. I am at my wits end, and I have nothing else that I can think of to try. I've got several of these boards and they've worked okay for me without doing anything special. I have not tried the software you are using, however, nor do I use the adapter you mention (does that adapt the connector on the SSR-6tru to a connector that fits your UT+ board?). I made a connector to talk to mine directly from a 3 volt serial port in a BBB SOIC. Since you can see the NMEA sentences the transmit side of the serial port from the LEA-6T is clearly working and you have the baud rate right, but your symptoms suggest the module doesn't hear you. Have you tried looking at the basic connection, i.e. that the serial port receive pin on the module wiggles at the right voltage and polarity when the software tries to send stuff (maybe there are two ways to plug in the adapter, only one of which works)? If that looks okay then the only other guess I can think of is that the software is trying to talk to the board with u-Blox binary messages but that protocol has been turned off for input on the port (the PUBX,41 NMEA sentence can turn it on and off), but that seems unlikely since, no matter how I reconfigure mine, a power-on reset always sees the serial port come up willing to receive either protocol. Dennis Ferguson ___ 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] PI Math question
On 16 Apr, 2014, at 09:50 , WarrenS warrensjmail-...@yahoo.com wrote: With the values of K1, K2 K3 constant, and the initial state of I#1, I#2 and Last_Input all zero assuming there is no rounding, clipping or overflow in the math and that if I've made any obvious dumb typo errors that they are corrected, If we assume that your 'Input' value is a real-valued measurement with an unlimited range then I think your algebra is correct. All those rearrangements will produce the same value of 'Output'. Note, though, that 'Input' doesn't have to be a value like that, and overflow in the math may be unavoidable. It depends on the nature of the sensor producing the value. For the particular case that might be relevant here, suppose 'Input' is the output of a phase error detector with an output limited to a range proportional to [-180, 180] degrees (i.e. is truncated to a fraction of one cycle) and the job of the controller is to try to keep that value at 0. Because the output of the sensor wraps around at +/- 180 you will want to do certain computations (in this case, differences between 'Input' values) with modular arithmetic. For an example of the difference this makes, assume that your first 8 'Input' values are 40 80 120 160 -160 -120 -80 -40 noting that (((-160) - (160)) mod 180) == 40. If you run these values through your first and last set of equations I think you'll find the value of 'Output' diverges at the wrap-around. I think the practical issue here is that if the basic PI phase-locked loop, as expressed by your last set of equations, has a long time constant it may fail to lock if the phase detector output wraps around like that and the initial frequency error is large enough to make the wrap-arounds occur frequently. The addition of the FLL term with its modular difference, as your first set of equations has it, will widen the capture bandwidth of the loop by keeping the integral term moving in the right direction until you get to a point where the frequency is close enough that the PLL becomes effective, at which point the behaviour of the loop becomes that of the PLL alone. The latter is what you've demonstrated. Dennis Ferguson ___ 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] First success with very simple, very low cost GPSDO
On 10 Apr, 2014, at 22:06 , Chris Albertson albertson.ch...@gmail.com wrote: You originally described a system that counts to 5M every second. Tom and others pointed out that you do not need the complete 5M count, all you need is the remainder of a modulo count. The question then is, how much of a remainder do you need to be sure that it spans all anticipated errors in both the PPS and the oscillator? Yes this right. I'm sure that under limited conditions I can get by looking only at the remainder. It is harder than was said because the overflows per second rate is a non-interger but there are till only two flavors of seconds: Those with N overflows and those with N+1 overflows. I think you are seeing a complexity that isn't there. 5,000,000 is 19531 * 256 + 64, so if you get an 8-bit counter sample t0 in one second and t1 in the next then the difference t1 - t0 will be equal to 64 if the oscillator is on frequency, and t1 - t0 - 64 will be equal to 0, when computed with 8 bit arithmetic. If the result of the latter, interpreted as a signed value, is less than zero your oscillator is going too slow, if greater it is going too fast. This will be true whether there have been 19531 or 19532 counter overflows in the second; the (t1-t0) subtraction will set the borrow bit in the latter case if you want to know that but there is no reason to care about it. 8 bits is enough to measure a +/- 25 ppm error in this case, which seems sufficient for any oscillator you are likely to want to discipline with this. All keeping the full count seems to do is require you to subtract two much larger numbers to find the same small difference. Dennis Ferguson ___ 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] Another Arduino GPSDO
On 28 Mar, 2014, at 20:53 , Bob Camp li...@rtty.us wrote: The problem is that it can / might / could / will create a “dc bias” in the noise. When you filter it, you get a bump rather than zero. If your GPSDO has a 47 ns wide sawtooth, that could be a pretty big bump. There’s no way to know if the bridge is seconds, minutes, or hours wide. You can make a good guess that hours are a *lot* less common than seconds…. Yes, and you can get long term dc bias not only in the hanging bridge case but, in diminishing amounts, when the sawtooth frequency passes through 1/2 Hz, 1/3 Hz, 1/4 Hz and so on, i.e. where the sawtooth period is an exact integer number of seconds. I think the 4th graph here http://www.leapsecond.com/pages/m12/sawtooth.htm shows a 1/2 Hz sawtooth with a non-zero mean error for an extended period. I guess, viewed this way, the 5th graph would be the 1/1 Hz sawtooth (which becomes 0 Hz when folded into the Nyquist bandwidth), which is the worst case but not the only case. Dennis Ferguson ___ 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] PLL Math Question
On 12 Mar, 2014, at 23:08 , Hal Murray hmur...@megapathdsl.net wrote: b...@evoria.net said: In the moving averages I'm doing, I'm saving the last bit to be shifted out and if it's a 1 (i.e. 0.5) I increase the result by 1. That's just rounding up at an important place. It's probably a good idea, but doesn't cover the area I was trying to point out. Let me try again... Suppose you are doing: x_avg = x_avg + (x - x_avg) * a_avg; For exponential smoothing, a_avg will be a fraction. Let's pick a_avg to be 1/8. That's a right shift by 3 bits. I don't think there is anything magic about shifting, but that makes a particular case easy to spot and discuss. Suppose x_avg is 0 and x has been 0 for a while. Everything is stable. Now change x to 2. (x - x_avg) is 2, the shift kicks it off the edge, so x_avg doesn't change. (It went 2 bits off, so your round up doesn't catch it.) The response to small steps is to ignore them. Note that you can't do fixed-point computations exactly the same way you would do it in floating point, you often need to rearrange the equations a bit. You can usually find a rearrangement which provides equivalent results, however. Let's define an extra variable, x_sum, where x_avg = x_sum * a_avg; The equation above can then be rewritten in terms of x_sum, i.e. x_sum = x_sum * (1 - a_avg) + x; With an a_avg of 1/8 you'll instead be multiplying x_sum by 7, shifting it right 3 bits (you might want to round before the shift) and adding x. The new value of x_avg can be computed from the new value of x_sum with a shift (you might want to round that too), or you could pretend that x_sum is a fixed-point number with the decimal point 3 bits from the right. In either case x_sum carries enough bits that you don't lose precision. Dennis Ferguson ___ 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] GPS W/10KHz
On 10 Feb, 2014, at 00:48 , Bruce Griffiths bruce.griffi...@xtra.co.nz wrote: Dennis Ferguson wrote: On 8 Feb, 2014, at 14:50 , ewkeh...@aol.com wrote: The problem with the PLL analog version is the same as with any digital GPSDO. The saw tooth is present at 10 KHz just like 1 Hz. To the best of my knowledge there is no GPS receivers out there for less than $ 1000 with out saw tooth. Timing receivers output the correction value and you can either with software or a variable delay do correction. This is very true, though the sawtooth at a 10 kpps sample rate is going to a little different than the sawtooth at a 1 pps sample rate. The frequency of the sawtooth noise will lie somewhere in the Nyquist bandwidth. At a 1 pps sample rate the frequency of the sawtooth noise will hence be somewhere between 0 Hz and 0.5 Hz, while at 10 kpps the sawtooth frequency will range from 0 Hz to 5 kHz. Noise at less than 0.5 Hz is not easy to filter, so you are going to require the correction from the receiver and/or an integrator with a time constant that can only be realized digitally. Sawtooth noise over most of a 0 Hz to 5 kHz range, on the other hand, should be eliminated by the analog low pass filter after the phase detector in the PLL, giving you something nice and clean coming out. It is only if you get unlucky and the beat frequency between GPS time and the receiver's oscillator ends up very close to an integer multiple of 10 kHz that you'll see noise at a low enough frequency to leak through into the control response. This is interesting because it suggests that very simple GPSDOs using 10 kHz from the receiver might at times work worse than you are likely to observe in a single bench measurement as aging (or something) moves the receiver's oscillator frequency through one of the bad frequency errors. Or is there a way to avoid that altogether (maybe if the receiver does dithering)? Dennis Ferguson Instead of speculating try reading the specifications. 1Hz phase modulation of the 10kHz output is present. The receiver sawtooth error sample rate is 1Hz not 10kHz. The 10kHz output signal phase is adjusted at a 1Hz rate by the receiver. Bruce, I'm not sure which equipment you want me to read the specifications for, though I'd be very interested in knowing. What I'm describing is the behaviour of the timepulse output of the LEA-6T, which can be configured to output edges at any rate from 1 Hz to 10 MHz. There the only relevant specification I see is the timepulse output quantization error, which is a constant 21 ns on every output edge independent of the rate at which the receiver is configured to generate edges. This should cause exactly the behaviour described above, and as best I can measure by comparing 1 pps and 50 pps outputs to the divided-down 10 MHz output of a GPSDO is consistent with how the receiver actually behaves (though my best measurement is none too good; I need to get a TIC with a resolution better than the 10 ns my Beaglebone has). If you run the output at 10 kpps you get 10,000 samples of the quantization error every second and can average it out a lot faster than if you only get one sample of the quantization error every second. I don't know what a sawtooth error sample rate is if not this. You seem to be describing a piece of equipment where the sawtooth error is not a simple consequence of pulse output quantization caused by generating edges with the receiver's internal free-running clock. I'd be really curious to know what equipment this is. This page http://gpsdo.i2phd.com says he looked for but failed to find any sub-Hz sidebands in the Navman Jupiter 10 kHz output so it doesn't seem like that receiver is the one you are thinking of either. Dennis Ferguson ___ 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] GPS W/10KHz
On 10 Feb, 2014, at 00:48 , Bruce Griffiths bruce.griffi...@xtra.co.nz wrote: Instead of speculating try reading the specifications. 1Hz phase modulation of the 10kHz output is present. The receiver sawtooth error sample rate is 1Hz not 10kHz. The 10kHz output signal phase is adjusted at a 1Hz rate by the receiver. Ah, as soon as I pressed send for the last note I realized what you were likely telling me. Yes, the LEA-6T only provides you with a quantization (sawtooth) correction for 1 pps and no higher rate. At 1 pps you should pay attention to the digital correction (implying no analog-only implementation is possible; you minimally need the delay line thing) since the frequency of the saw tooth is often low enough to leak into the control response and the correction should make the sawtooth go away. All I was pointing out is that at a higher output frequency, like 10 kpps, the frequency of the quantization saw tooth error will almost always be much higher as well. There's no need for the digital correction since averaging over a relatively short period, like in the loop filter of an appropriate analog PLL, will almost always be sufficient to smooth the sawtooth. Dennis Ferguson ___ 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] GPS W/10KHz
Bjorn, All I was pointing out is that at a higher output frequency, like 10 kpps, the frequency of the quantization saw tooth error will almost always be much higher as well. There's no need for the digital correction since averaging over a relatively short period, like in the loop filter of an appropriate analog PLL, will almost always be sufficient to smooth the sawtooth. The sawtooth correction is the difference between where the receiver would wish to place the edge and where its known limited resolution electronics lets it put the edge. Yes, exactly. The range of the error in edge placement will be a constant related to (probably equal to) the period of the internal clock it is using to generate the edges. For an LEA-6T this is 21 ns so, assuming it rounds off, any edge it places will be in error by +/- 10.5 ns. The receiver wish is based on the timesolution from the last measurement. In the Jupiter this is done at 1Hz maximum. The sawtooth correction will apply the same for all 10k (pos or neg) edges in the 10kHz signal during that one second. Yes, this is true. Note, however, that the time solution produces not only a phase error of the receiver's internal clock with respect to GPS but also the frequency error of the receiver's clock with respect to GPS. More than this, since the time solution takes time to compute it will be telling the receiver what the phase error was at some point in the past rather than what it is now, let alone what it will at the point in the future when you want to assert a 1 pps signal. It can place that future edge because it knows the actual frequency of its clock with some precision from the time solution, and that plus knowing a phase offset at some past time is sufficient to allow it to extrapolate to a future edge placement. With an LEA-6T the precision of the edge placement will be +/- (10.5 + epsilon) ns, with the epsilon occurring because it is extrapolating a past measurement to place a future edge. Note, however, that the rate at which it can compute time solutions doesn't change any of this very much. The fact that an LEA-6 can compute 5 solutions per second rather than just one will at best just make epsilon a bit smaller, and this matters not at all since epsilon should be pretty small already. If the receiver instead only computed a time solution once every 3 seconds it also wouldn't make a difference, it could still place a 1 pps edge every second by extrapolating from whatever the last solution was that it managed to complete. More than this, if you told the receiver to place 10,000 edges per second instead of just 1, the placement error of each one of those edges, individually, would still be +/- (10.5 + epsilon) ns. The rate at which the receiver computes new solutions has about epsilon to do with the precision of edge placement. The sawtooth doesn't come from the epsilon, it comes from the +/- 10.5 ns. There are effects that are not easily filtered away in the analog domain. See the archives and http://www.leapsecond.com/pages/m12/sawtooth.htm This is good. Notice the amplitude and the frequency of those sawtooths. The amplitude is the period of the internal clock placing the edges, i.e. 21 ns for an LEA-6T and what looks like 30 ns for the receiver above. Then there's the frequency. It varies widely but the highest frequency seen, in the 4th graph down, looks to be about 0.5 Hz. It isn't an accident that there is no higher frequency, and I'll just assert that this maximum frequency has nothing to do with the time solution update rate of the receiver. It would be not change if you looked at the 1 pps output of a 5-update-per-second LEA-6. Instead the highest sawtooth frequency is 0.5 Hz because he's looking at a 1 pps output, getting one sample per second, and if you sample a signal at one sample per second then the frequencies you see in the samples are always going to be in the range 0-0.5 Hz. Essentially this is integrating a beat frequency between the receiver's oscillator and GPS time, which could be very high in frequency, but by sampling at 1 pps the difference, whatever it is, gets folded into the 0-0.5 Hz Nyquist bandwidth. The low frequency of the sawtooth observed at 1 pps makes it a problem for analog filters. All I'm pointing out, then, is that if you increase the pulse rate output by the receiver from 1 pps to 10 kpps you will still get a sawtooth, like 1 pps, and the amplitude of the sawtooth will be unchanged from 1 pps, but the frequency of the sawtooth won't be limited to 0-0.5 Hz and will instead be folded into the 10 kpps Nyquist bandwidth of 0 - 5 kHz. Unless you are very unlucky this will give you the same sawtooth error at a much higher frequency, making it much more amenable to analog domain filtering. Dennis Ferguson ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo
Re: [time-nuts] GPS W/10KHz
On 8 Feb, 2014, at 14:50 , ewkeh...@aol.com wrote: The problem with the PLL analog version is the same as with any digital GPSDO. The saw tooth is present at 10 KHz just like 1 Hz. To the best of my knowledge there is no GPS receivers out there for less than $ 1000 with out saw tooth. Timing receivers output the correction value and you can either with software or a variable delay do correction. This is very true, though the sawtooth at a 10 kpps sample rate is going to a little different than the sawtooth at a 1 pps sample rate. The frequency of the sawtooth noise will lie somewhere in the Nyquist bandwidth. At a 1 pps sample rate the frequency of the sawtooth noise will hence be somewhere between 0 Hz and 0.5 Hz, while at 10 kpps the sawtooth frequency will range from 0 Hz to 5 kHz. Noise at less than 0.5 Hz is not easy to filter, so you are going to require the correction from the receiver and/or an integrator with a time constant that can only be realized digitally. Sawtooth noise over most of a 0 Hz to 5 kHz range, on the other hand, should be eliminated by the analog low pass filter after the phase detector in the PLL, giving you something nice and clean coming out. It is only if you get unlucky and the beat frequency between GPS time and the receiver's oscillator ends up very close to an integer multiple of 10 kHz that you'll see noise at a low enough frequency to leak through into the control response. This is interesting because it suggests that very simple GPSDOs using 10 kHz from the receiver might at times work worse than you are likely to observe in a single bench measurement as aging (or something) moves the receiver's oscillator frequency through one of the bad frequency errors. Or is there a way to avoid that altogether (maybe if the receiver does dithering)? Dennis Ferguson ___ 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] Arduino Frequency Accuracy
On 18 Jan, 2014, at 20:11 , Jim Lux jim...@earthlink.net wrote: Just in case you want to build a clock with an Arduino.. http://jorisvr.nl/arduino_frequency.html ADEV measurements, etc. take home message.. absolute accuracy is a few kHz out of 16 MHz... probably a 100 ppm crystal. On some Arduinos (or Teensy3's which is what I use) there's a provision for a 32kHz clock crystal.. that might be a bit better as a time base. That's a sloppy crystal if what you are looking for is an oscillator with an output frequency which is very close to the number written on the crystal's package, but it might not be so bad if you measure its frequency under actual operating conditions and use the calibrated value instead. I think the ADEV of the crystal is in fact rather good judged by PC standards. I interpret the floor of near 10^-9 at 100 seconds as meaning some (mythical?) optimal synchronization software might keep a clock based on that within 100 ns of the GPS receiver at an adjustment rate of about one every 100 seconds. This is quite good compared to other hardware I've recently been looking at. The ceramic resonator, on the other hand, is pretty awful. Dennis Ferguson ___ 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] L1/L2 GPS Receiver
On 17 Jan, 2014, at 11:43 , Michael Perrett mkperr...@gmail.com wrote: Magnus, I believe that he is referencing the the new L2 C/A code, which is not protected. Reference http://www.gps.gov/systems/gps/modernization/civilsignals/ It would be nice to have a receiver for that when they turn it on, but I don't think that's what he wants. The observables used for PPP processing are L1 and L2 carrier phase. You don't need a receiver capable of decoding the P(Y) code but you do need a receiver capable of receiving the full bandwidth of its carrier on both L1 and L2 and tracking the phase. The commercial receivers which do this seem to cost dearly. Dennis Ferguson ___ 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] Loran
On 15 Nov, 2013, at 19:12 , Bob Camp li...@rtty.us wrote: We probably could agree on Seneca NY since that’s about equal distance to the pair of us. Does anybody know the proposed ERP on the new system? Some of the master’s on the China chains are pretty high power if I remember correctly. This powerpoint presentation http://www.tinyurl.com/l2humtb says the NL40 transmitter they just bought is 300 kW. I thought a lot of the Asian chains, including China, used Megapulse equipment like the US. I think Megapulse did use to say their transmitters were multi-Megawatt, but I can't check that since http://www.megapulse.com now goes someplace else. I wonder whether the last bit means that Megapulse is now out of the transmitter business for good, or if Ursanav's infatuation for the Nautel transmitters is just a passing fancy while they complete their vertically integrated monopoly. Dennis Ferguson ___ 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] Time stamping with a PICPET
On 26 Oct, 2013, at 23:53 , Hal Murray hmur...@megapathdsl.net wrote: dennis.c.fergu...@gmail.com said: That's perfect if it works like it seems it should. The problem with modern CPUs is finding an instruction sequence that does the read-write-read in that order, allowing each to complete before doing the next. The write is the biggest problem. Writes are often buffered, and even when you can find an instruction which stalls until it clears ... I'm far from a wizard in this area, but I used to work with people who were. The rules for things like PCI cover that case. If you do something like write to a register to clear an interrupt request, you have to follow it by a read to that register or one close to it. As you hippity-hop through bridges and such, the read gets trapped behind the write and doesn't happen until the write finishes. When using the CPU cycle counter as a system clock source it is common to find that the two reads in a read-write-read sequence are only a cycle or two different even when you know the write is crossing an interconnect with 10's of nanoseconds of latency (not that 10's of nanoseconds is bad...). That's reasonable if the read-write-read were to cycle-counter, someplace-else, and cycle-counter. The write has been started. It's in the piepline, but you haven't told the memory system that you need it to finish. Try read-write-read-read where the outer reads are to the cycle counter and the inner write-read both go to the same IO device. Note that you've turned a read-write-read into a read-read-read with an additional write wart. As I mentioned you can often find instructions to do a read-read-read correctly, so this will likely work too. Putting some numbers to this might help get a handle on the cost, though. One reason for doing the before- and after- reads is to get a measurement of the ambiguity of the sample (which also provides a basis for filtering damaged samples). Cycle counter reads hardly cost anything but on a 166 MHz, 64 bit PCI-X bus, the last, highest performance PCI bus that was a real bus (PCIe is a packet protocol running on a network of point-to-point links) a single register read takes about 74 ns to complete. I'll guess the write adds about 40 ns to that. Since the write increases the ambiguity from +/- 37 ns (i.e. read-read-read only) to +/- 57 ns, finding a way to do it with a read-read-read alone provides a useful improvement. For PCIe the write is probably cheaper but the read is likely to be even more expensive due to the packetization and (de)serialization logic that bus requires. It is the case, however, that if you do a naive implementation of the read-read-read, or the read-write-read-read, you may end up finding the first and last read of the cycle counter are still only a few cycles apart. The reason is that while most modern CPUs will execute the instructions more-or-less in order (most will do 2 instructions per cycle if they can now, so the order may not be exact) the CPU won't have a reason to actually wait for the 74 ns it takes that middle read to complete and will go barrelling along executing additional instructions until it finds something that actually uses the result it hasn't got yet. The cycle counter read doesn't depend on the previous read so there's no reason to wait. To get these operations serialized correctly you need to find the magic instructions that force that to happen. On a recent x86 you might be able to use the serializing rdtscp instructions, on older ones you might need to separate the operations with cpuid instructions. On other CPUs it could be barrier instructions or something else entirely. A posted write may only be serializable by adding an extraneous and expensive read to the same device after it, as you suggest. Dennis Ferguson ___ 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] Time stamping with a PICPET
On 26 Oct, 2013, at 18:21 , Tom Van Baak t...@leapsecond.com wrote: Right. The key is not to use count-down timers and interrupts at all. The key is no interrupts; nothing asynchronous. What you do is: 1) About once a second (it doesn't really matter), with hot cache and interrupts disabled, record the s/w time before and after you output a pulse. If your PC/SBC has a low-latency DTR or GPIO pin code path, you're golden. That's perfect if it works like it seems it should. The problem with modern CPUs is finding an instruction sequence that does the read-write-read in that order, allowing each to complete before doing the next. The write is the biggest problem. Writes are often buffered, and even when you can find an instruction which stalls until it clears the buffer the write will also often be posted across the interconnect bus so there's no way for the CPU to know when the write makes it to the device, let alone make it wait until that happens. When using the CPU cycle counter as a system clock source it is common to find that the two reads in a read-write-read sequence are only a cycle or two different even when you know the write is crossing an interconnect with 10's of nanoseconds of latency (not that 10's of nanoseconds is bad...). It is usually easier to find the magic instructions to make a read-read-read work the way one expects, though even that can be a challenge. It is possible to do the same output pulse thing with a read-read-read if there is a PWM peripheral to generate the pulses. The PWM is programmed to output pulses at whatever frequency is convenient while the read-read-read sampling is used to determine the relationship between the PWM counter and the system clock. Of course, this requires a peripheral which legacy PCs often don't have. If the CPU/PC/SBC has h/w counter/capture latches, you're all set. Then there's no jitter and NTP should be as accurate as the ADEV(tau 1s) of the LO that clocks the CPU and the ADEV(tau) of the external (GPS) 1PPS source. But h/w counter/capture is something no legacy PC has had AFAIK. If the new breed of SBC have this capability, NTP should show a one or two orders of magnitude jump in precision on those platforms. The TI CPU used for the Beaglebone (Black) has three. The counter being sampled is 100 MHz. Dennis Ferguson ___ 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] Speaking of Costas loops (WAAS)
On 10 Jul, 2013, at 14:08 , David I. Emery d...@dieconsulting.com wrote: It seems completely inconceivable to me that either the antenna system (particularly feeds) or transponder RF hardware on any commercial Ku or C or Ka or X band satellite could possibly be frequency agile enough to tune to 1575.42 MHz unless it was purpose designed to radiate on that frequency from the start. So any hosted WAAS payload is completely application specific. If you look at the pictures here http://www.orbital.com/NewsInfo/Publications/Galaxy_Fact.pdf the satellite on the right has things sticking out the bottom, in the back corner, that are missing on the others and that look a lot like the antennas on GPS satellites. The WAAS satellite is also 350 pounds heavier than the other two even though the C-band payload is identical on all three, so it seems like there could be a fair amount of extra stuff added for WAAS support. Dennis Ferguson ___ 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] Speaking of Costas loops
On 5 Jul, 2013, at 09:18 , Jim Lux jim...@earthlink.net wrote: I believe that the original WAAS repurposed transponders intended for other L-band satellite signals (e.g. Sirius/XM/LightSquared). I'm not sure. The original WAAS satellites, I think these http://nssdc.gsfc.nasa.gov/nmc/spacecraftDisplay.do?id=1996-070A http://nssdc.gsfc.nasa.gov/nmc/spacecraftDisplay.do?id=1997-027A indeed were commercial L-band satellites, as you suggest, but they still note that Each INMARSAT-3 also carried a navigation transponder designed to enhance the accuracy, availability and integrity of the GPS and Glonass satellite navigation systems. which leaves the impression that particular bit of hardware might be special-purpose. Dennis Ferguson ___ 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] Speaking of Costas loops
On 5 Jul, 2013, at 08:33 , Bob Camp li...@rtty.us wrote: The sat needs to transmit at the GPS frequencies and have an uplink that works exclusively with those frequencies. (or at least that sub band). A normal transponder probably would not radiate at the GPS allocation, simply to be a good citizen. I believe the specialization is simply a frequency mod to allow WAAS to pass through. There is no mention of a space qualified Cs and / or Rb flying on those birds and no indication that the ground segment is controlling such a payload. If all that *was* present, then including them in the normal navigation solutions would be a zero cost next step. Addressing the last sentence I found a government WAAS reference which indicates that the WAAS satellites are indeed interchangeable with GPS satellites in navigation solutions. It is is on page 7 of this http://www.gps.gov/technical/ps/2008-WAAS-performance-standard.pdf where it says The WAAS GEO broadcast also provides an additional ranging source for improved availability of navigation services. When a WAAS receiver is using the corrections and integrity messages broadcast by the GEO, only four GPS or GEO satellites are needed, which increases the availability of service versus RAIM or RAIM/FDE. While what is or isn't required in the satellite to support this is still a mystery it seems like the timing accuracy coming back must end up being equivalent to a real GPS satellite. What this is good for is interesting to think about. Dennis Ferguson ___ 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] Speaking of Costas loops
On 3 Jul, 2013, at 21:05 , Bob Camp li...@rtty.us wrote: If the WAAS sats were purpose designed to provide a high accuracy carrier, then yes there are ways to do it. The fundamental design concept of a bent pipe is that you don't do any of that. You do not care what's going through the bird, it just maps the input frequencies to the output and amplifies them (a lot). Again, the WAAS signal is simply piggybacking on existing hardware. The conversion oscillator is not locked to the GPS carrier (or to any other carrier). It's simply a free running quartz based oscillator, running into a synthesizer to get the appropriate microwave frequency. I'm not sure about the Again, ... part. All three WAAS satellites are commercial satellites but they were all launched recently enough (2 in 2005, 1 in 2008) to have had WAAS-specific payload added. The solicitation for the 2008 satellite is here https://www.fbo.gov/index?s=opportunitymode=formid=f5aacd4bba2ef67b0c59b586900499b6tab=core_cview=1 and is dated 2002; this isn't looking for service on a satellite already in orbit. For the 2005 satellites, the Telesat one is mentioned here http://www.telesat.com/services/government-services which says Telesat’s Anik F1R includes a specialized payload for the Wide Area Augmentation System while you look at the Orbital Sciences blurb on the last three satellites it built for PanAmSat, here http://www.orbital.com/newsinfo/publications/galaxy_fact.pdf you'll see that they are all exclusively satellite TV things, with 24 active C-band transponders and 8 spares, except for Galaxy 15 which weighs 350 pounds more than the other two and about which it says: The Galaxy 15 satellite, which features a unique hybrid payload configuration, was launched on October 13, 2005. In addition to C-band commercial communications, the spacecraft also broadcasts Global Positioning System (GPS) navigation data using L-band frequencies as part of the Geostationary Communications and Control Segment (GCCS) implemented by Lockheed Martin for the U.S. Federal Aviation Administration (FAA). I don't think they can use any old satellite for WAAS, they added payload for it. Note that when Galaxy 15 went awol it took the WAAS service with it for most of a year even though it was replaced in its orbital slot for TV service by a spare within a week or so (though Wikipedia says the replacement was Galaxy 12 so I guess that's predictable from the blurb above). So I've been assuming that while the WAAS satellites are commercial the WAAS transmitters are specialized to the service and included for its exclusive use. I hence guess they could have been designed to work however they needed to. Dennis Ferguson ___ 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] Speaking of Costas loops
On 3 Jul, 2013, at 10:48 , Attila Kinali att...@kinali.ch wrote: On Wed, 3 Jul 2013 08:29:02 -0400 Bob Camp li...@rtty.us wrote: There are two batches of GPS / WAAS sats up there: 1) The ones with numbers above 100 that are geosync and that only do WAAS 2) The ones with numbers = 32 that do nav. These are not geosync. I believe the only ones with corrected / high stab clocks on board are those in the second group. The stuff in the first group aren't dedicated sats, just leased transponders on conventional multipurpose geosync birds. I don't know about WAAS, but AFAIK the EGNOS signals are generated on ground using Cs references and retransmitted by the satelites using a bend pipe. Ie. the signals should be of time-nut quality even without high accuracy frequency standards in the birds themselves. (Sorry, i'm not able to find where i read about that, so no references today) I have also read that WAAS satellites can be usefully included in the GPS solution, so they aren't necessarily inferior, but I also don't have a reference. There is this: http://tf.boulder.nist.gov/general/pdf/2299.pdf The clocks are indeed ground based and good quality. The advantage of using them as an alternative to GPS CV (which is what the paper is about) is that they transmit unencrypted code on two frequencies to allow computing ionospheric corrections and they don't move (much) so you can track them continuously with a dish to get a big signal-to-noise improvement and multipath insensitivity. The last bit seems like a mixed blessing, though, since the dish means you depend on only the one satellite it is pointed at and hence suffer from whatever bad things happen to it. The paper notes events that it characterises as an increasing problem with the broadcast WAAS ephemeris, followed by an outage and clock jump, which I interpret as maybe being an adjustment made to the satellite orbit which can't be represented properly in the ephemeris. I assume that could happen with regular GPS satellites too, but if you are tracking a lot of them at once it is easy to detect and toss out a solution outlier. Dennis Ferguson ___ 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] Speaking of Costas loops
On 3 Jul, 2013, at 11:47 , Bob Camp li...@rtty.us wrote: The pipe in this case is up on one frequency and down on another. The conversion oscillator on satellite that's the weak link, no matter how good the signal from the ground happens to be. That's certainly true but it doesn't seem like a problem that the presence of a high stability free-running oscillator, like a rubidium, would help. The oscillator on a geostationary satellite has a continuous frequency reference to lock to (the uplink carrier) and hence only needs short term stability sufficient to track this and transfer it accurately to the downlink. It seems like this is the kind of problem that quartz excels at. Dennis Ferguson ___ 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] BPSK decoder for WWVB
On 3 Jul, 2013, at 14:03 , Tim Shoppa tsho...@gmail.com wrote: I have also heard YVTO on 5MHz underneath both WWV and WWVH, strangely off-kilter by half a second or so. When BPM does that, like maybe at the beginning of this http://www.youtube.com/watch?v=WRaRB-x84xg I think it is because they are transmitting UT1 pips rather than UTC pips. I assume this might be convenient for celestial navigation users, though I can't imagine that there are a whole lot of those left. Dennis Ferguson ___ 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] looking for low-power system for gps ntp timekeeping
On 30 Jun, 2013, at 08:50 , David J Taylor david-tay...@blueyonder.co.uk wrote: From: Attila Kinali [] Oh.. and if you want to go the linux way and use a Raspberry Pi.. just dont! Use a Beaglebone black instead. It uses less power and is easier to deal with. Not to mention that you dont have all those USB related problems. [] Attila Kinali === I've built three Raspberry Pi stratum-1 NTP servers: http://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html one of which has both a Wi-Fi dongle and a DVB TV receiver stick attached, and on none of these have I seen any USB problems. I'm using a 5.25 V 2A power supply from ModMyPi.com. You can view the timekeeping accuracy here: http://www.satsignal.eu/mrtg/performance_ntp.php You aren't necessarily showing the part where the Raspberry Pi is a bit weak, though. How well do clients which receive their time via the USB ethernet interface do? The Beaglebone Black has about three advantages going for it in this application: - The ARM CPU is about twice as fast as the Raspberry Pi's for about the same power consumption (I'm not sure this is a particular advantage for NTP, however, so I won't count it). - The Ethernet MAC core is built into the SOIC, and tightly coupled to it, so packet traffic doesn't have to sit waiting for the USB scheduler to get around to doing something with it. - The Ethernet MAC core also provides fairly good, complete IEEE1588 support. This is not of direct use to NTP but does provide a way to calibrate the software timestamps which NTP produces and consumes to better match when the packets arrive from and are transmitted by the actual hardware. I.e. you can measure the typical difference between hardware and software inbound timestamps (measuring interrupt latency), and hardware and software outbound timestamps (measuring the processing time spent in the outbound network stack) for PTP UDP packets, and then use these results to improve the symmetry of software timestamps for NTP UDP packets. There is no way I know of to measure this without the IEEE 1588 support (and the outbound number in particular is often big enough to deserve correction). - The TI SOIC also has a hardware timestamp capture peripheral (look for eCap in the documentation) which can capture PPS edge times with single-digit-nanosecond accuracy. That's a couple of orders of magnitude better than interrupt sampling and eliminates the jitter of the latter measurements. For a $5-$10 difference in price for the board I think these are worth it. The RPI makes a fine, low-power replacement for Intel hardware for this, but the Beaglebone Black has the raw material to do significantly better at this than either of them. The only problem with the Beaglebone is that it is not as popular as the Raspberry Pi, so making use of the former is going to require one to do more work on one's own to take advantage of it. Dennis Ferguson ___ 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] clock-block any need ?
so, with the quality of the result depending on the population and nature of variability of the cluster but hardly at all on the outliers, and with the lack of a measurable cluster telling you when you might be better off relying on your local clock rather than the network. The approach relies on the things we do know about networks and networking equipment while avoiding reliance on things we can't know: it mostly avoids making gaussian statistical assumptions about distributions that may not be gaussian. The field of robust statistics provides tools addressing this which might be of use. I guess it is worth completing this by mentioning what it says about ntpd. First, ntpd knows all of the above, probably much, much better than I do, though it might not put it in quite the same terms. If you make the assumption that the stochastic delays experienced by samples are evenly distributed between the outbound and inbound paths (this is not a good match for the real world, by the way, but there are constraints...) then round trip delay becomes a stand-in measure of cluster, and ntpd does what it can with this. The fundamental constraint that limits what ntpd can do, in a couple of ways, is the fact that the final stage of its filter is a PLL. The integrator in a PLL assumes that the errors in the samples it is being fed are zero-mean and normally distributed, and will fail to arrive at a correct answer if this is not the case, so if you want to filter samples for which this is unlikely to be the case you need to do it before they get to the PLL. The problem with doing this well, however, is that a PLL is also destabilised by adding delays to its feedback path, causing errors of a different nature, so anything done before the PLL is severely limited in the amount of time it can spend doing that, and hence the number of samples it can look at to do that. Doing better probably requires replacing the PLL; the replace it with what? question is truly interesting. I suspect I've gone well off topic for this list, however, and for that I apologize. I just wanted to make sure it was understood that there is an argument for the view that we do not yet know of any fundamental limits on the precision that NTP, or a network time protocol like NTP, might achieve, so any effort to build NTP servers and clients which can make their measurements more precisely is not a waste of time. It instead is what is required to make progress in understanding how to do this better. Dennis Ferguson ___ 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] clock-block any need ?
On 27 Dec, 2012, at 11:28 , Attila Kinali att...@kinali.ch wrote: On Thu, 27 Dec 2012 10:55:12 -0800 Dennis Ferguson dennis.c.fergu...@gmail.com wrote: I don't think I buy this. It takes 70 milliseconds for a signal transmitted from a GPS satellite to be received on the ground, but we don't use this fact to argue that sub-70 ms timing from GPS is not possible. If you have a network of high-bandwidth routers and switches doing forwarding in hardware, and carrying no traffic other than the packets you are timing (I have access to lab setups that can meet this description) you can observe packet delivery times that are stable at well under the microsecond level even though the total time required to deliver a packet is much larger. I'm not sure about this. Knowing about how switches work internally, i'd guess they have jitter of something in the range of 1-10us, but i've never done any measurements. Have you any hard numbers? I've measured it for large routers, but the numbers are not mine. In a former life I helped design forwarding path ASICs. I'm interested in what that guess is based on, however, since I can't imagine where 1-10us of self-generated jitter from an ethernet switch would come from, if not from competing traffic. A well-spec'd piece of silicon to handle 20 Gbps of full-duplex bandwidth needs to be capable of processing about 40 million packet arrivals per second, or about one packet every 25 ns. That's pretty much what is needed to build a good ~$200, 24 port gigabit ethernet switch. The cheapest hardware forwarding path to implement, which is generally what you'll find in there, is a fixed processing pipeline (or pipelines) that takes packets in at the required rate and spits out the results at that rate delayed by N chip clock cycles; N might be large (but not too large; N tells you how many packets it needs to be able to have in process simultaneously and it is cheaper in logic if you can minimize that number) but it is a constant. Your jitter estimate implies that such a switch, even when not occupied with other traffic, will either sometimes leave a packet sitting around for between 40 and 400 packet arrival times before getting around to doing something with it, or else will sometimes do between 40 and 400 packet arrival times worth of extra work to forward the thing. My experience with this suggests that it is actually easier to build if it doesn't work like that. The switch I recently bought for my house, this one http://www.netgear.com/business/products/switches/prosafe-plus-switches/JGS524E.aspx# specifies the total latency (that's total time, not jitter) through the switch at 4.1 us for 64 byte packets, a precision I expect they arrived at by just adding up the store-and-forward and fixed pipeline delays. Nearly all of the variation in delay is from competing traffic Even if 1-10us was observed for individual samples, however, that is still missing the point. The interesting number is not the variability of individual samples, it is the stability of the measure of central tendency derived from many such samples (e.g. the average, if the variation were gaussian) that is the interesting number. If you add competing traffic, like real life networks, the packet-to-packet variability becomes much worse, but this is sample noise that can be addressed by taking larger numbers of samples and filtering based on the expected statistics of that noise. Here lies the big problem. While with GPS we pretty much know what the time is that the signal takes to reach earth, we have no clue with network packets in a loaded network. We don't even have an idea what the packet transmit distribution is in the moment we are doing our measurements. Neither the queue length in the router/switch nor anything else. And the loading of a switch changes momentarily and this in turn changes the delay of our packets. You can of course apply math and try to get rid of quite a bit of noise, but you will never get rid of it down to ns levels. ?? NTP is a two-way time transfer. We directly measure how long the cumulative queue lengths are for the round trip for each sample, and we hence directly measure how this changes from sample to sample. There are also good statistical models for the average behaviour of such queues when operating at traffic levels where packet losses are rare and where the bandwidth is not being significantly consumed by a small number of large, correlated, flows, which is the usual operating state for both local networks and Internet backbones (it is usually access circuits that are the problem) and there are heuristics one can use to determine when the statistics are not likely to be so nice; these are of use when designing the thing which has the queues. What we haven't had is hosts and servers capable of making precise measurements either of packet arrivals and departures (why is a ping round trip reported to be 200 us or 400 us when
Re: [time-nuts] clock-block any need ?
On 27 Dec, 2012, at 08:05 , Chris Albertson albertson.ch...@gmail.com wrote: You do not need to use something like the Clock-Block to build a very good NTP server, but if you want to build the *ultimate* server it is part of the mix. Yes this is true. The server can be very good, meaning that if it were better the clients that it servers could not know the difference. A simple is if a wall clock moved the hands with millisecond precision, it would not serve the clients (human eyeballs) any better if it moved with nanosecond precision because human perception is measured in mS not nS. Same with the time server, it communicates with its clients over a network that has someuncertainy in th delay and ultra-presision is lost. So nanosecond level timekeeping in the server is not required. You can do uSec level time keeping with the standard TTL can on most mother boards. However this list is for nuts and you might think it is fun to try and do 1000 times better time keeping than is needed, in that case you will need some kind of specialized clock hardware. I don't think I buy this. It takes 70 milliseconds for a signal transmitted from a GPS satellite to be received on the ground, but we don't use this fact to argue that sub-70 ms timing from GPS is not possible. If you have a network of high-bandwidth routers and switches doing forwarding in hardware, and carrying no traffic other than the packets you are timing (I have access to lab setups that can meet this description) you can observe packet delivery times that are stable at well under the microsecond level even though the total time required to deliver a packet is much larger. If you add competing traffic, like real life networks, the packet-to-packet variability becomes much worse, but this is sample noise that can be addressed by taking larger numbers of samples and filtering based on the expected statistics of that noise. That is, the level of noise effecting each individual sample entering the filter does not alone predict the noise level of the result coming out, the latter also depends on the number of samples and the quality of the model of the noise employed by the filter. Note that I often see claims of time synchronization with PTP at the 10 ns level or better. As this level of synchronization is usually achieved by the brute force method of measuring transit times across every network device on the path from source to destination I have no doubt that what NTP can do will necessarily be worse than this, but I don't know of a basis that would predict whether NTP's worse is necessarily going to be 10,000x worse or can be just 10x worse. Knowing that would require actually trying it to measure what can be done. What is certain, however, that if you want to measure this at the levels that might be possible you probably want nanosecond-level clock hardware in both the server, so that it can produce time of this quality, and in the clients, so that you can measure how well they do directly rather than attempting to have the NTP implementation grade its own homework. I don't think this is a waste of time at all. The best case is that the ability to measure at this level would lead to an understanding of what it would take to transfer time with NTP at this level, but even the worst case would be that one would be able to support one's assertions about what can't usefully be done with data, and that's not bad either. If no one tries then no one will ever know. Dennis Ferguson ___ 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] Cell timing error
GSM cell sites in the US have GPS because it is required to support E911 positioning. I'm not sure if it is used for anything other than this, but it doesn't have to be. In some other parts of the world it has been considered bad taste to let the operation of telecommunications infrastructure become dependent on a facility owned by the US military, so the standards that are popular there often try to avoid that. Dennis Ferguson On 15 Dec, 2012, at 18:59 , li...@lazygranch.com wrote: I can assure you the GSM shacks have GPS timing in them. I can dig up the photos if you want. -Original Message- From: Joseph Orsak jor...@nc.rr.com Sender: time-nuts-boun...@febo.com Date: Sat, 15 Dec 2012 18:24:20 To: Discussion of precise time and frequency measurementtime-nuts@febo.com Reply-To: Discussion of precise time and frequency measurement time-nuts@febo.com Subject: Re: [time-nuts] Cell timing error ATT uses UMTS in most areas which is a self-synchronizing modulation scheme. Supposedly one of the selling points is no dependence on GPS. All the extra sync channels and sync messaging is a capacity hog, not a very spectrally efficient standard in my opinion. About 85 maximum simultaneous voice calls in a 5Mhz UL / 5 Mhz DL sector/carrier before it starts to fall apart. A big step backwards from good old CDMA2000 (also just my opinion). But hey, you can surf the web while you talk on the same device. -Joe W4WN - Original Message - From: Jim Lux jim...@earthlink.net To: Discussion of precise time and frequency measurement time-nuts@febo.com Sent: Saturday, December 15, 2012 5:43 PM Subject: Re: [time-nuts] Cell timing error On 12/15/12 2:16 PM, Scott McGrath wrote: In a prior life we had a CDMA timing receiver for NTP which used VZ for its source On Dec 15, 2012, at 12:18 PM, Graham / KE9H time...@austin.rr.com wrote: You should switch to Verizon. They are inherently accurate to milliseconds. Sub micro-seconds inside the base stations. On 12/15/2012 12:51 PM, Greg Troxel wrote: In central mass, ATT and tracfone (? carrier) are showing phone times very close to 1 min slow. Virgin/sprint is ok. I've never seen this before - usually it's a few s slow. The time *displayed* on the phone might not reflect the time from the network. ___ 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 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 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 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] Fw: Cell timing error
On 15 Dec, 2012, at 21:30 , gary li...@lazygranch.com wrote: This is a shot of the GPS timing rack in an ATT shack. http://www.lazygranch.com/images/att/att_3.jpg Yes, the TruePosition box helps compute handset locations for E911 and whomever else wants to know where your phone is. This isn't a unit you would necessarily see in cell closets in other countries. Dennis Ferguson ___ 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] Cell timing error
On 15 Dec, 2012, at 22:38 , Hal Murray hmur...@megapathdsl.net wrote: GSM cell sites in the US have GPS because it is required to support E911 positioning. I'm not sure if it is used for anything other than this, but it doesn't have to be. So it's cheaper to install and maintain GPS rather than make one measurement and tell the setup where it is? E911 requires the carrier to be able to figure out where the handsets are. I think GPS is used as a common timing reference so they can triangulate to locate the phone using time-of-arrival measurements of the handset's transmissions made at several cell towers. GSM/UMTS carriers do it this way, at least. CDMA2000 carriers instead rely on the handsets to make the time-of-arrival measurements, both of signals from cell towers and of GPS signals the handset can hear. Dennis Ferguson ___ 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] GPS DO Alternatives
On 9 Dec, 2012, at 02:14 , Hal Murray hmur...@megapathdsl.net wrote: Which brings up another worm for the can. How are you going to get a PPS from the OCXO? That's going to be hard without some soldering. My straw man would be to use the OCXO to clock whatever uP you end up using and generate the PPS with a counter/timer. I think I'd do that too, and pick a processor with enough of the right peripherals to avoid having to fart around with an external phase comparison or interrupt timestamping. The TI Stellaris LaunchPad, mentioned earlier, is pretty cheap, the processor can generate its internal clocks directly from a 10 MHz reference input, and it has a set of 32 bit timer/counters that can either capture a timestamp for an input edge event or generate a timed outgoing pulse, both with a resolution of 12.5 ns. The capture could be used to time the GPS PPS (in hardware, no external comparator, no interrupt timestamps) while I think the PWM generation mode could be used to generate a PPS output synchronous with the 10 MHz oscillator. The only thing that is missing for a GPSDO is an internal DAC (it has a bunch of ADCs) so the 10 MHz oscillator would need a digital frequency adjustment for the processor to be sufficient for the job without any external peripherals. The processor core is a fully functional ARM, including floating point, which might make it easier to, say, do a Kalman filter implementation that wasn't write-only. The thing is, I still don't find this all that compelling. Personally I think it might be better to maybe aim higher on the software value chain, find a processor board with the same on-chip processor peripherals as above along with an on-chip Ethernet MAC and enough external peripherals to run a real operating system and use that to build a combination GPSDO and very accurate NTP/PTP server. You generally can't just buy one of those on eBay (at least they are much rarer than a GPSDO alone), and it is a lot easier to do something which is attractive both in price and in quality compared to what you can buy. Dennis Ferguson ___ 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] Z3805 incident
On 29 Nov, 2012, at 02:32 , Charles P. Steinmetz charles_steinm...@lavabit.com wrote: This is a classic crystal jump. The crystal changed its frequency magically from one second to the next and the software compensated for it Here is another example of a 3805 having a bad moment. For just about two minutes, it reported a phase jump of nearly 3 uS and then immediately fell back nearly to its previous baseline, settling to the baseline in about an hour and not requiring any longer-term change of the EFC voltage. This does not look like a typical crystal frequency shift to me, but I cannot rule that out. It looks more like what I'd expect to see if I set the cable delay to 3 uS for 2 minutes, then back to 0. I think I would be more likely to call this one, where the crystal jumps to another frequency for a while and then jumps back to about what it was, a classic crystal jump. I've seen this before, though not as large as the change you show. I hear these raise hell when they try to use PTP to transmit telecom-quality timing over asynchronous ethernet because it is hard to run a PTP control loop tight enough (i.e. at a high enough data rate) to correct that before it does damage. I think the other problem, with the crystal jumping to another frequency and apparently staying there (I'm assuming it hasn't jumped back), could have a broader range of causes. Dennis Ferguson ___ 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] eBay Ublox
On 21 Nov, 2012, at 10:34 , Michael Tharp g...@partiallystapled.com wrote: With respect to interrupt latency, the PPS driver is the best you're going to get without a custom add-in card that provides input capture (timestamping). I considered going that route but making PCI or PCI-e cards essentially means using FPGAs which are expensive and fussy, and sort of overkill. A I2C or SPI based card for use with Raspberry Pi or other single-board computers is also a possibility. I did a prototype PCI-X card like that some years ago. It was expensive but worked pretty well; the high-speed part of the FPGA ran at 320 MHz, so timestamps were taken with a resolution of just over 3 ns. Having a card plugged into your computer which knows what time it is to within 3 ns, but which is on the far side of a bus across which a single PIO read takes 80 ns to complete, is of limited direct use to the computer itself, however, so the hardest part of the design involved minimizing the uncertainty of the time transfer from the card's clock to the computer's system clock. I'm not quite sure that it always requires custom hardware to get a few 10's of nanoseconds precision for PPS sampling, though. Some off-the-shelf hardware may also work for this. In particular the on-chip clock peripherals included in the TI CPU used for the BeagleBone board have hardware timestamp-capture inputs (the TIMER4-7 i/o pins on one of the cape connectors) which seem to do the right thing. That, plus the fact that the TI chip also has an onboard Ethernet mac core (as opposed to the USB Ethernet device the Raspberry PI uses) is making me think that that the BeagleBone might be a bit better base on which to build a good NTP server than the RPI is. Dennis Ferguson ___ 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] Timing performance of servers
On 26 Oct, 2012, at 08:06 , shali...@gmail.com wrote: If you cannot see the horizon because of obstructions (what else?), these obstructions are likely to be a source of multipath. So while technically you do not need to see the horizon, any obstruction above the horizon could cause problems. Of course, distant trees or a hill are less likely to be a problem than your neighbor's garden shed with a tin roof. Though, as I understand it, typical low-end GPS antennas are quite sensitive to multipath arriving from below the horizon as well. I think getting a sharp antenna cutoff at the horizon is the reason that high-end antennas have choke rings. Also, some antennas are better at rejecting low angle signals than others. While the software can reject some undesired signals, it can only do so if the software can identify them as separate. If the multipath signal destructively interferes with the desired signal, there is not much the software can do. Given that the transmitted C/A bandwidth is greater than 1 MHz, however, I'm not sure that it is possible for multipath signals to destructively interfere across the entire bandwidth; I think the issue is distortion, with some frequencies in the bandwidth suffering destructive interference while others are constructively interfered with. This can be compensated for in software, though it is much better not to have to. Dennis Ferguson ___ 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] Adjusting HP 5065A frequency
On 22 Oct, 2012, at 12:48 , Bob Camp li...@rtty.us wrote: The big bucks approach probably is to run a good RF ADC on the input and then do all the offset stuff as DSP math. The VCXO just sits at it's magic frequency and never moves. More money / no pops. A somewhat cheaper way might be to use a DDS. That is, lock the clock driving the DDS to the input frequency and then program the DDS to correct the measured error of the input clock. A DDS with a 48 bit control word will have an effective resolution of about 4e-15, if my arithmetic is right, which seems adequate for the purpose. The DDS also gives you the option of generating any (corrected) output frequency you want. The output frequency could even be programmable if you don't mind looking at the DDS digital noise in the output, though that could be cleaned up by picking a fixed output frequency ahead of time and adding a cleanup PLL for the chosen frequency following the DDS. I'm not sure why this problem isn't always dealt with this way, actually. Since the corrections are applied in digital arithmetic the precision with which they can be made is limited only by the bit-width of the adders you use to compute each cycle's update and, given that the D/A converter it is driving is probably going to be limited to 300 or 400 MSPS, even an FPGA (let alone semi-custom logic) could carry more bits through the computation than are useful to have. There is probably some catch to this that I don't understand. Dennis Ferguson ___ 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] Are serial port headers standardized?
On 20 Oct, 2012, at 02:05 , Sarah White kuze...@gmail.com wrote: Page 15, there is a yellow 10 (9) pin header, and page 26 was what I quoted. Really wish there was more information... I've had this motherboard for something like 5 years at this point, and am fairly certain I lost or outright tossed the serial port headers. Are they fairly standard? I think there are two standard variants, this one http://www.pccables.com/07120.htm and this one http://www.pccables.com/07121.htm For Intel motherboards I've only seen the first one used, but I don't know about anything else. Dennis Ferguson ___ 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] To use or not to use transmission line splitters for GPS receivers
On 9 Oct, 2012, at 12:48 , Bob Camp li...@rtty.us wrote: If you are after sub ns level timing, things are a bit different than if you are happy with tens of ns error. Few of us have an adequate survey of our location to *really* worry about sub ns numbers. If you are one of those lucky few that can worry about sub-ns, yes mismatch and voltage and a whole long list of things matter. The temperature coefficient of your antenna also gets onto that list at some point. I think you can get sub-nanosecond time (if you can arrange for a proper equipment calibration) and sub-centimeter positioning on your own using the IGS products and GPS Precise Point Positioning techniques. The gotchas are that you need to have a high-priced dual-frequency, carrier phase tracking receiver and the software you need seems to only be available to the very rich (though there are free online services which will process your data to determine the location for you). The antenna temperature thing is kind of indicative of just how much lore and black art seems to be involved in arranging equipment for fine timing, however. I have the ITU 2010 Handbook for Satellite Time and Frequency Transfer and Dissemination. In Chapter 12, when discussing GPS Common View techniques, the document says this about antenna temperature 12.5.2 Temperature stabilized antennas It is now well documented, and generally admitted, that GPS time-receiving equipment, and more specifically its antenna, is sensitive to environmental conditions [Lewandowski and Tourde, 1990]. For conventional GPS time-receiving system this sensitivity could be expressed by a coefficient of about 0,2 ns/°C and can approach 2 ns/°C. This was a major precluding obstacle, as it did, the goal of 1 ns accuracy announced earlier for GPS time transfer. and goes on to recommend using an antenna with an oven keeping the temperature of the electronics constant. In Chapter 13, on the other hand, when discussing GPS PPP, it says this: There have been some poorly supported claims of strong variations of geodetic clock estimates with temperature changes in some GPS antennas, together with recommendations to use temperature-stabilized units. While this might apply to certain low-end, single-frequency units, direct tests of a standard AOA Dorne Margolin choke ring antenna have failed to detect any sensitivity of the clock estimates to antenna temperature variations. Ray and Senior [2001] placed an upper limit of 2 ps/°C on the short-term (diurnal) temperature sensitivity and later extended this to 10.1 ps/°C for any possible long-term component [Ray and Senior, 2003]. Even smaller sensitivities, 0.17 ps/°C or less, were determined by [Rieck et al., 2003] for an Ashtech choke ring model. So Chapter 13 says that what Chapter 12 said is bogus. It appears that Chapter 12 may have written been written by a European while Chapter 13 is an American effort, so this may be some sort of cultural thing. Chapter 13 does later go on to point out how crappy the Canadian IGS stations are in the winter and blames this on snow and ice in the near field below the antenna, so even Chapter 13 does find a use for heating at the antenna. Both chapters do agree that keeping the temperature of the receiver constant is good. I think the antenna splitter thing is probably the same kind of issue. Someone, somewhere, may have had a problem with an antenna splitter and published a paper on that, and this in turn reinforces the conservative assumption that you should leave anything out that doesn't absolutely need to be there, so it has become common wisdom that you should avoid splitters. Or something. Dennis Ferguson ___ 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 / Xtendwave patents
On 27 Sep, 2012, at 15:40 , Jim Lux wrote: On 9/27/12 2:58 PM, Bob Camp wrote: Hi It would be interesting to hear what the patent lawyers on the list think about the patents. Given a quick read, they appear to cover any use of the specific transmitted format for receiving time information. IANAL, but.. Me neither. reading Claim 1.. a key aspects are the combination of PSK and ASK, with different data. This is somewhat unusual, and may not have been done exactly like that. said phase modulation *is independent* of the information represented by said pulse width modulation/amplitude shift keyed modulation is a phrase that occurs in ALL of the independent claims. (my emphasis added) QAM is, of course, simultaneous PSK and ASK, but it's a single data stream that is being encoded. Is there prior art for transmitting one kind of data using ASK and something else PSK? For instance, is WWV (which is primarily ASK) has a subcarrier, but the subcarrier is also AM. Another possible source of prior art might be a PSK encoded digital squelch on a AM or FM modulated signal (if such a system exists). I wouldn't mind knowing a legal definition of information since to me most of what is carried in the PSK is the same information as is carried in the ASK, just formatted with different bits: known markers to find minute alignment, minute-of-the-century time, leap second warning and daylight savings information. The ASK alone encodes UT1 while the PSK has expanded DST information, but most of it is not what I would call independent information even if the bit encodings are different. I assume a lawyer's definition is not the same as mine. Beyond that, though, it really does seem like they are attempting to patent all receivers of the new WWVB format which use the phase modulation, while pruning the claims enough to avoid existing DCF77 and BPC receivers. The timing information based on a known sequence spanning multiple seconds avoids DCF77's 0.8 second (and BPC's 0.6 second) known sequence, and independent information thing seems to exist to distinguish it from DCF77 which sends the same bits with both its PSK and ASK (BPC likely does too). The not-yet-granted patent actually seems more odious, since it seems to be attempting to claim the idea of using past time measurements to compute the frequency error of a clock's oscillator so that future timekeeping can be improved by correcting that. That's too bad. One can only hope the patent they got is defensive and they don't plan on generally enforcing it, and that none of the claims in the other one survive the obviousness test. Dennis Ferguson ___ 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 Now a Monopoly
On 26 Sep, 2012, at 10:03 , J. Forster wrote: You go after everything. Soup to nuts, including the contract agreements. IMO, this is potentially very, very big money, because Xtendwave may also have patent protection, and henceforth control all the precise digital clock market. This is tens of millions of units, at least. They claim to have applied for patents on something but I would be surprised if they could patent anything that would prevent anyone from designing their own receiver. What would annoy me is less-than-full disclosure of the transmitted signal and its properties. For example, there's a claim in the paper that the (31 26) Hamming code used can detect double-bit errors in the encoded time. I think detecting double-bit errors would require an additional parity bit, and that the assertion in the paper is just a boo-boo, but I also keep wondering if the claim might in fact be true, that there might be a really clever way to use that with something else in the signal to detect double-bit errors, and the paper just isn't pointing that out. That would be annoying. Dennis Ferguson ___ 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] New WWVB format...
On 26 Sep, 2012, at 11:19 , Majdi S. Abbas wrote: On Wed, Sep 26, 2012 at 10:13:22AM -0700, Tom Van Baak wrote: My reading of the document(s) is that the new format will in fact allow WWVB to be used as a frequency standard with even greater precision then before, though not with unmodified legacy WWVB carrier receivers. My hope is that one of you will produce a clever reference design for such a TF receiver make it available to the group. It sounds like a very fun DSP project; one that we can all learn from. Bonus points for making it an open-source Arduino shield. Making it work with both DCF77 and WWVB would also be a plus. DSP would be good, although I also think an microcontroller implementation could be interesting. Atmel's ARM MCUs look like they'd be good candidates for this sort of thing. (Pretty fast, enough storage to do interesting things with it, and a fast enough ADC for 60 KHz.) This is fine, though to make it maximally useful for time and frequency purposes I believe the hardware might need to provide a way to synchronize the ADC clock to an external reference, and likely some way to time-mark the incoming data (e.g. a quick-and-dirty version might feed a PPS signal to the second channel of a stereo ADC, if no more elegant solution is available). A control loop to discipline an oscillator's output might use that oscillator to clock the ADC and adjust the oscillator to zero the ADC's phase alignment with the input signal, if that can be made to work. A system to measure WWVB propagation delays and signal levels might instead clock the ADC and the time marker with a known-accurate frequency and PPS (e.g. a GPSDO). RFSpace makes commercial LF/MF/HF SDR equipment with almost the right inputs for this (an external frequency input and a timing trigger). What I'd like is a tiny-budget version of this just for LF stations. I've got a couple of these that I might use as a development platform: https://www.olimex.com/Products/ARM/Atmel/SAM7-P256/ Has anyone come up with a reasonable algorithm to implement in a microcontroller? (DSP development kits are a bit more spendy than I'd like to invest in a prototype. :) I guess the trouble with this is only that the availability of brute force can sometimes make it unnecessary to deal with a lot of complexity. If your job is to do a convolution of a model of what you know was transmitted against the incoming signal to measure the time alignment then using a platform where you can store big blocks of data and do Fourier transforms with wild abandon can provide really good results without having to spend a lot of time thinking about it. Even quite modest modern PC hardware comes with a boatload of memory and is exceedingly speedy, and for some purposes it can save a lot of time and effort just to make use of that compared to trying to do without. I have a quick-hack DCF77 PM detector which runs on PC hardware and makes use of one of the above-mentioned RFSpace receivers for the data acquisition. While it is now in boxes being moved, when I get it back up I would love to lose the RFSpace receiver in favor of something much less costly, but would hate trying to make this work with something less capable than the PC. Using a microcontroller like that to do the A/D conversions and send the data collected out (say) an ethernet port to a PC which does the heavy computational lifting (that's what the RFSpace receiver does) would appeal to me, but trying to do without the PC would not. Dennis Ferguson ___ 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] New WWVB format...
On 26 Sep, 2012, at 14:43 , paul swed wrote: Might be a bit of a cost. The SDR runs $1495. Regards Paul The ones with the clock input options (the SDR-IP and the NetSDR, I think) are significantly more than that. But they are also huge overkill if all you want is a digital LF receiver. That's why I'd like to replace it with something cheap, but that something wouldn't be nearly as useful without the clock and timing edge inputs. Dennis Ferguson ___ 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] Hi Power LED Light power supply...
On 18 Sep, 2012, at 12:42 , Chris Albertson wrote: On Tue, Sep 18, 2012 at 5:21 AM, Bob Camp li...@rtty.us wrote: Hi I suspect those same 120Hz sensitive people would not be able to watch TV or a movie :) In the old CRT type TV sets, the phosphor has some persistence. Movies are modulated with a square waves, the frame blinks off and goes dark then blinks on. But the LED's brightness is fast enough to track the sine wave and would be bright only for an instant with quick pulses of light. Just to add to this... Ontario, Canada originally ran its power grid at 25 Hz. When they switched the grid to 60 Hz in the 1930's some of the industrial power users, particularly in northern Ontario where private (usually hydroelectric) power generators were common, never got around to changing their plants over. Mine and paper mills using 25 Hz power were common as recently as the 1980's, and might still be there for all I know. Standard incandescent light bulbs don't have a lot of persistence when run on 25 Hz power (I assume there might have been a time when you could buy incandescent bulbs designed for 25 Hz, but not in my lifetime). They don't go entirely off, but they get significantly dimmer in the visible spectrum in the dips as the output red-shifts towards the infrared; they follow the sine pretty well. In my teens, when visiting a place using 25 Hz power for lighting, I could initially see an incredibly annoying flicker when I first got there but after a minute or two this would fade and I'd no longer notice it. Some other people would also see the flicker but others, including my parents, couldn't see it at all so there seemed to be variation (maybe age-related, maybe not) among individual abilities to see this. I would hence believe that a 50 Hz flicker must be pretty close to the edge of what can be perceived, so I'm having trouble believing that a flicker at more than twice that rate would be perceptible at all by anyone. Dennis Ferguson ___ 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] Hi Power LED Light power supply...
On 18 Sep, 2012, at 15:06 , John Lofgren wrote: snip I would hence believe that a 50 Hz flicker must be pretty close to the edge of what can be perceived, so I'm having trouble believing that a flicker at more than twice that rate would be perceptible at all by anyone. snip Oh, but it is. A couple of years ago I bought one of the Chinese 30 LED spot light bulbs for about $8 on ebay. I thought I'd give it a try for a workbench light. When I plugged it in at work (60 Hz power, here) the two guys standing behind me yelled gaahhh at the same time I did. The flicker was horrendous. The earlier comment about peripheral vision also applies, though. It's worse in the periphery than in direct view. The power supply is nothing more than a bridge rectifier, two current limiting resistors, and a filter capacitor. The capacitor obviously wasn't big enough, though, because it flcikered plenty. Or could the problem have instead been that one side of the bridge wasn't working, so you were getting a 60 Hz flicker rather than 120 Hz? Having seen what I am sure was a 50 Hz flicker, I'd believe that 60 Hz might look awful but I still have some doubt about 120 Hz. Dennis Ferguson ___ 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] GPSDO control loops and correcting quantizationerror
On 16 Sep, 2012, at 00:40 , Tom Van Baak wrote: I worry in your example about the long cross-over time. This may be ideal for frequency stability, but probably is not good for time accuracy. If one is using the GPSDO as a timing reference, I would think a shorter time constant will keep the rms time error down. Has anyone on the list done work optimizing the timing accuracy rather than the frequency stability? I'm not sure there could be a difference between the goals of frequency accuracy and time accuracy that would effect that time constant. The time error is the time integral of the frequency error, so anything which manages to minimize the frequency error of the oscillator (both the magnitude of the error and its duration) will also minimize the time error. The time constant is selected to be the minimum value which makes it probable that the frequency or time error you have measured (for a PLL the data are time errors) is in fact an error that the oscillator has made rather than an artifact of the noise in the measurement system. There might be a difference in the best control action to take to optimally achieve each of those goals. In particular if your goal is frequency accuracy the best control action in response to the measurement of a frequency error might be to correct that error, i.e. to minimize the frequency error once you know you have one. If your goal is time accuracy, however, then the response to a measured frequency error is going to be to intentionally make a frequency error in the other direction for a while to correct the accumulated time error. In this case, though, it seems to me that by selecting a PLL as the control discipline (rather than, say, a FLL) you've already made the decision to take control actions which ensure time accuracy. Dennis Ferguson ___ 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] PP2S
On 16 Sep, 2012, at 17:11 , Tom Van Baak wrote: Some GPSDO have both a 1PPS and a PP2S (pulse per 2 second) output. I have two questions for one of you telecom experts: 1) What is the history, and the purpose of that PP2S signal? 2) What is the official spec for which second the PP2S lands on? Is it odd seconds or even seconds? Is it GPS time (easy) or UTC (problematic)? If UTC, what happens after a leap second? The PP2S signal is a US CDMA (i.e. CDMA2000) thing. It is aligned to the even seconds in GPS time. My memory is dim but I think that the choice relates to the fact that the CDMA spreading code LFSR rolls over every 26.666 ms (it is a 15 bit LFSR, so dividing 32767 by 26.666 ms should be the 1.228 MHz chip rate), so it rolls over 75 times every 2 seconds. The goal is to align the code sequence transmitted by every station, and a 1 PPS timing reference wouldn't guarantee that since 1 second isn't an integral multiple of the roll over time. Dennis Ferguson ___ 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] GPSDO control loops and correcting quantizationerror
On 16 Sep, 2012, at 16:30 , Poul-Henning Kamp wrote: In message 34d5c3ce-6b3d-4944-996a-7637373b2...@gmail.com, Dennis Ferguson wr ites: I'm not sure there could be a difference between the goals of frequency accuracy and time accuracy that would effect that time constant. Note that the that time constant referred to here, the topic of the message I was responding to, was explicitly a PLL time constant. If you have decided to use a PLL as your control discipline I think you end up with the same time constant whether your goal is accurate frequency or accurate time since, with a PLL, these end up being the same problem. It does. A PLL more or less corresponds to an PI regulation, where a FLL only needs to have the I term. Because you don't have the interaction between the P and I terms, the I-timeconstant can be longer. This sounds right. As I said, if you pick a control discipline other than a PLL, as might be advantageous to do if your concern is solely with accurate frequency, then the optimum might be different. If you are using a PLL in both cases, however, then the problems are essentially the same. Dennis Ferguson ___ 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] oscillators
On 30 Aug, 2012, at 13:14 , Rick Karlquist wrote: The other area where a uP is useful is in an environment with high vibration. It can correct for acceleration as well as temperature. There I've never heard of this being done. Do you have a reference? I'm not sure how that would be done with a single crystal but I heard a talk by David Allan about a (carefully oriented) array of 6 crystals for which that was done. A Kalman filter which understood the physics of acceleration effects was applied to the output of the 6 crystals to produce a composite clock and, as a side effect, inertial navigation information. This allowed the composite clock to be corrected for g-effects as well as providing the same data that would be output by a conventional, mechanical inertial navigation unit. I think the target application was a drone aircraft, with the clock output ending up at a GPS receiver while the inertial data was used as a fallback if the GPS was jammed and as a sanity check to detect GPS spoofing. This seemed like a nice one-stone-several-birds solution. I have a copy of the powerpoint somewhere, but I've not seen this written down anywhere else. Dennis Ferguson ___ 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] Time-nut gettogether in Boulder
On 6 Jun, 2012, at 16:44 , Alan Melia wrote: Then there is the NPL Time Frequency Club at Teddington..oh sorry that is the wrong side of the Atlantic :-)) Alan G3NYK Is there still? Google comes up with this from 2004 http://www.npl.co.uk/content/conWebDoc/2054 but the link to the club home page at the bottom goes no where. I would attend if it weren't defunct. Dennis Ferguson ___ 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] thunderbolt no UTC offset
On 2 May, 2012, at 04:30 , Tom Van Baak wrote: The UTC offset is in words 6-10, page 18, subframe 4 -- every 12.5 minutes. /tvb The leap second warning is also in the same place. How the unit could know that a leap second is pending but not know the UTC offset is a mystery. Dennis Ferguson ___ 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] PICTIC II ready-made?
On 27 Apr, 2012, at 02:57 , Chris Albertson wrote: Closed source drivers and binary blob firmware.I'd have nothing to do with a project that includes either of those. I'd require a open source platforms with a 100% free tool chain. Also, it is a bit of overkill after all a bare PIC works fine for this application. The only thing binary blob about the Raspberry Pi that I can see seems to be the stuff associated with the graphics accelerator. This is unfortunate, but unless you are into bare-metal programming of high resolution graphics/video (you can do that with an Arduino?) I'm not quite sure how that is relevant. The programming interfaces for all peripherals on the SoC not associated with the GPU seem to be well documented in the Broadcom data sheet: http://www.designspark.com/files/ds/supporting_materials/Broadcom%20BCM2835.pdf I don't know of anything useful which is missing, and I know a port of a non-Linux operating system (NetBSD) to the board is being done using nothing that is non-public. You can leave the graphics binary blob out if you find that offensive and have no use for it anyway. I think the most annoying thing about the Raspberry Pi is that a lot of the GPIO signals aren't brought out to connectors, perhaps to save money on the board. Dennis Ferguson ___ 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 phase modulation test April 15-16
The WWVB web page at NIST, here http://www.nist.gov/pml/div688/grp40/wwvb.cfm has a notice about another phase modulation test on Sunday and Monday. Dennis Ferguson ___ 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] NTP jitter with Linux
On 5 Apr, 2012, at 13:03 , shali...@gmail.com wrote: An older laptop (Pentium M for instance) can be had for $80 or so any day of the week, won't take much space, is completely standalone (built-in keyboard and display, built-in battery backup) and sips power when idle, which it will be most of the time. The only issue is that you might be tempted to run more things on it and affect NTP performance. But if you load it with BSD and use it just for that, it will be a dandy solution. I'll take some issue with that. The best clock source for software running on a computer, particularly when the applications might be expected to take a lot of time stamps, is one which (a) works, (b) has reasonable precision, and (c) is as cheap as possible to sample. On Intel processors the most precise and inexpensive-to-read counter available (i.e. conditions (b) and (c)) is the TSC, so you want to use this if at all possible. The problem with using the TSC is that it sometimes violates condition (a), that is the works part. On some older processors it does not necessarily increment at a constant rate, and on boards with multiple CPUs there can be multiple TSCs with different times, so in these cases you may have to use something else which isn't as good as the TSC would be if it worked. For some random computer whose clock you want to set this is all fine. It should use a counter known to work, and if the TSC doesn't it should just use something else. For something you are buying to be a dedicated NTP server, however, it is worth while adding working TSC very high on the list of desirable attributes. The problem with some older CPUs, like the Pentium M, is they are of a vintage which did not guarantee the TSC would increment at a constant rate, and a variable rate makes it useless as a clock. It is possible you could work around that at some cost (maybe the sips power when idle part) but I'm of the opinion that life is too short for that and it would be better to pick something with a known-to-work TSC for this application. The Atom processors aren't bad. Dennis Ferguson ___ 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] NTP jitter with Linux
On 4 Apr, 2012, at 16:10 , Mike S wrote: On 4/4/2012 6:51 PM, Eric Williams wrote: Could the CPU be reducing its clock rate when it's not being loaded? Just a guess, most multi-core processors these days have power saving features like that. On Wed, Apr 4, 2012 at 3:22 PM, Mike Smi...@flatsurface.com wrote: I've played around with different cpufreq setting, thinking it might be related to the processor speed during an IRQ varying, but that seems to have minimal impact (performance vs. conservative vs. ondemand). Setting /sys/devices/system/cpu/cpuX/cpufreq/scaling_governor to performance should lock that core to the max clock rate. In looking that up, I found that the script I made to set this was just doing cpu0 (i.e. one of four cores). Doh! I've changed it to do all 4 cores, and am trying that again to see if that's it. I don't know much about Linux but if that doesn't help try to find out what the operating system does in its idle loop. If it is ending up in some power-saving state when it is idle it may be volunteering to do this by executing some magic `wait' instruction which does the power-saving thing as a side effect, and if you can find where it does this you might be able to work around it. Dennis Ferguson ___ 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] LEA6-T Group Buy
On 2 Apr, 2012, at 13:11 , lstosk...@cox.net lstosk...@cox.net wrote: Note that there should soon be a LEA6T eval board available from sysmocom http://laforge.gnumonks.org/weblog/2012/03/16/#20120316-osmo_lea6t_gps_timing Estimated price is 90 EUR excl VAT in the EU. Anyone know if these will have the RAW output for use with RTKLib? Also assume can program for 10 MHz out? Had the door slammed in my face when trying to buy an eval board for the 6-T from the mfgr. ?? I bought an EVK-6T from the manufacturer a little while ago without trouble. The only thing associated with the transaction that I wasn't perfectly happy with was the price. I should say, though, that the manufacturer's eval kit puts the board in a rather nice case, with the two programmable pulse outputs only being available on the DE-9 RS232 connector, at RS232 voltage levels. Not only are RS232 voltages inconvenient for many purposes, but the MAX3232 converter they used adds 100 ns of delay. The high speed versions of the signals are available on the board, but you need to take the board out of the case to use them and they use those teeny, tiny MMCX connectors. Given a choice I'd rather have a board without paying for a case I have to throw away, with better connectors for the timing signals. I think this is a very good 50 channel GPS receiver, though, and the manufacturer's board isn't bad if you want to play with it from a computer since both PPS outputs and the Extint input are conveniently tied to RS232 control pins. Dennis Ferguson ___ 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] LEA-6T Group buy
On 26 Mar, 2012, at 12:56 , Chris Albertson wrote: On Mon, Mar 26, 2012 at 12:50 PM, Attila Kinali att...@kinali.ch wrote: Moin, I have the numbers together for the group buy of the u-Blox LEA-6T. I have not been able to locate a spec sheet for there. Do you have a link? Or maybe you could say how these are improved over the M12M? Apart from other issues it may potentially provide significantly more accurate time for computer timekeeping via the Time Mark input, depending on the accuracy of measurements on that input. That is, rather than connecting the PPS output to a computer input pin and then trying to timestamp when the interrupts occur, one can instead tie an output pin from the computer to the Time Mark input and poll for timestamps measured by the LEA-6T, say with a programmed sequence like get computer clock timestamp toggle Time Mark pin get computer clock timestamp . . . ask LEA-6T for Time Mark timestamp If done well that may get the time ambiguity at the computer end down from the microsecond level of interrupt latency to the 100 nanoseconds it should take for a PIO write to a hardware register to complete. If the LEA-6T takes Time Mark timestamps with that precision then this may be a significant improvement. Dennis Ferguson ___ 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 phase plots
On 18 Mar, 2012, at 10:52 , John Seamons wrote: They do talk about using the 11-bit Barker code for autocorrelation. But the sync bits transmitted only match the Barker code if you interpret them a little bit out-of-order. The part of the paper that talked about the Barker code confused me somewhat since I couldn't quite figure out how it was relevant. The autocorrelation property of the Barker code is only interesting if the Barker code is the only thing being sent (over and over), but in this case the concerns are more about spurious correlations with the variable data, something for which no solution seems to be possible. It is the case, however, that (non-circular) autocorrelations of the fixed sequence are relevant at small offsets. In your data the fixed sequence seems to be -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, -1, -1 which, ignoring the contribution of the variable data (which increases with increasing offset), gives this basic result for offsets from 0 to 13 seconds: 14, 1, 2, 1, -6, 1, -2, -3, 0, -3, 0, 1, 0, 1 So there is a quite large autocorrelation at 4 seconds offset. If I weight the search pattern by the fixed pulse widths (there are 3 0.2 second pulses and 3 0.8 second pulses in the fixed sequence; I gave the rest a weight of 0.5) that gets a little better, i.e. 7.0, 1.4, 0.4, 0.5, -2.1, -0.4, -0.4, -0.9, -0.3, -0.9, 0.0, 0.5, 0.0, 0.2 if I did that correctly, though at the apparent cost of making the autocorrelation at a 1 second offset a bit worse. In any case, if this is the pattern they selected I really would have liked to have seen a discussion of the tradeoffs involved in picking it, along with the assumptions they made about how it would be detected. And I kind of hope I don't have to read about that in someone's patent since technical descriptions written by lawyers are really boring. In any case, I think the paper left out the good parts. Dennis Ferguson ___ 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 phase plots
DCF77's AM modulation is a much better fit for what they did, and a much better design in general. All the useful phase modulation needs to be carried by the carrier at full power. DCF77's AM modulation drops the carrier power for only 100 ms or 200 ms at the beginning of the second, which gives them a full 0.8 seconds in every second at full power (if I'm remembering right the minute marker has no carrier reduction, so the very longest carrier reduction is only 0.2 seconds). Their chip sequence is just under 0.8 seconds long and sits in the full power part of each second. WWVB is not nearly so convenient. The carrier reductions for WWVB are deeper than DCF77, making it even more imperative that the information be carried in the high power segments only, but WWVB's carrier drops are 0.2, 0.5 or 0.8 seconds long, so in many seconds they only have 0.5 seconds of high power and in 7 seconds per minute there is only 0.2 high power seconds. I think there's no good way to make DCF77's silk purse out of the WWVB sow's ear. It is also the case the DCF77's phase modulation probably isn't as good as it could be if the goal is to find it in the noise since it only swings +/- 15 degrees rather than +/- 90. Its big advantage might be that it is high speed, with lots of transitions, so you can probably measure phase alignment pretty accurately with that. As a national time service, however, it only needs to serve a fairly compact country relative to WWVB's intended coverage area, so that plus WWVB's crappy AM format probably pushed them to forget about trying to match DCF77 and to just concentration on doing the best they could to improve coverage. That would be my guess, anyway. Dennis Ferguson On 19 Mar, 2012, at 19:47 , ehydra wrote: Hm. I had a quick look at http://en.wikipedia.org/wiki/WWVB I cannot see why it won't work with the DCF77 scheme. The carrier is always on-air. Do I miss something? To low bandwidth of the transmitting antenna? Sorry, I didn't followed the thread in whole. - Henry Brooke Clarke schrieb: Hi Henry: There are millions of WWVB clocks in use and the new signal must be fully compatible with them. -- 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. ___ 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 BPSK Receiver Project?
On 14 Mar, 2012, at 18:08 , Brooke Clarke wrote: The WWB paper New Improved System for WWVB Broadcast given at the 43rd PTTI November 2011 is at: http://jks.com/wwvb.pdf Part of the processing gain comes directly from the BPSK modulation and that amounts to a little over 10 dB improvement, but there's a further 18 dB gain to be had by accumulating an hours worth of data and processing that. It is a little interesting that the PTTI paper left out some of the interesting details one would need to actually decode the new signal, in particular the specification of the 14 second Sync sequence, which is necessary to know to find the alignment of minutes, and the value of the 60-bit hour-synchronization code, which defines the sequence of phase reversals in each minute's modulation in an hour and, as I understand it, is necessary to know to take full advantage of the hour-averaging thing. I assume this might have been done to allow the company which participated in the design of the signal to complete a receiver for it before they start transmitting that way while keeping anyone else from starting a receiver project until after the transmissions start? Dennis Ferguson ___ 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] Loran in the US
On 8 Mar, 2012, at 02:58 , Poul-Henning Kamp wrote: Has anybody asked them how good timefreq they're trying to deliver ? I would assume that they are aiming for a backup for GPS in telecom-GPSDO context. If so, frequency stability is priority number one and time is probably just better than 100msec or so I could swear I saw something that said 50 ns, though I can no longer find it and that sounds like science fiction. I note, though, that the Federal Register publication for the project, here: http://www.gpo.gov/fdsys/pkg/FR-2012-01-11/html/2012-307.htm indicates they aren't just looking at Loran by itself. The MF dGPS bands and 500 kHz are also included in whatever they are doing. Dennis Ferguson ___ 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] Loran transmitters back on the air.
The publication in the federal register, here http://www.gpo.gov/fdsys/pkg/FR-2012-01-11/html/2012-307.htm says they are playing with more than Loran. There are several MF bands they are playing with as well, in particular the dGPS bands and 500 kHz. I noticed a while ago that UrsaNav's UN-151 receiver was advertised as being capable of processing multiple signals in the LF and MF bands, and wondered what the MF part was about. That is a bit clearer now. Dennis Ferguson On 1 Mar, 2012, at 21:04 , paul swed wrote: Hmmm did find a paper that suggests various goals and such and the old loran gear might not work. Depends on what modes they try. Would be great to find some form of updated news. Regards Paul. On Thu, Mar 1, 2012 at 8:25 PM, paul swed paulsw...@gmail.com wrote: Eloran is compatible with the older timing rcvrs. Or at least it was supposed to be. Now the message suggests that they will try other modulation modes. I couldn't find anything really further then what was sent. I did hook the longwire directly to the austron so far no lock and I am less then 70 miles from the Nantucket site. Will keep trying Regards Paul On Thu, Mar 1, 2012 at 8:20 PM, Bob Camp li...@rtty.us wrote: Hi The obvious advantage to backwards compatibility would be much greater coverage area. It is a bit tough to envision them getting a reasonable user population with a 100% from scratch approach. Indeed that may be wishful thinking. Bob On Mar 1, 2012, at 8:09 PM, Charles P. Steinmetz charles_steinm...@lavabit.com wrote: Greg wrote: A friend in Texas has confirmed that Loran signals are now up and receivers are showing position. I am including a note from UrsaNav regarding this event. What are the odds that any long-term deployment would be backward-compatible with legacy Loran receivers (not the same as the initial tests being backward-compatible)? The primary revenue stream would appear to be from sales of new receivers that use patented technology (unless the government wants to get back into the business of subsidizing Loran, which it just vacated -- not very likely). Cynical, maybe, but it is always a good idea to keep an eye on the money. I suppose they could make the enhancements transparent to legacy receivers, so you would buy new receivers if you needed the enhancements but could also use older receivers if you didn't. But would they? There does not appear to be an incentive to do so, absent a government subsidy. Best regards, Charles ___ 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 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 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 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] Low-long-term-drift clock for board level integration?
On 19 Feb, 2012, at 15:56 , Bill Woodcock wrote: Hi. This is my first posting to this list, and I'm not a timekeeping engineer, so my apologies in advance for my ignorance in this area. I'm building a small device to do one-way delay measurements through network. Once I'm done with prototyping, I'm planning a production run of several hundred of the devices. They'll have a GPS receiver, probably a Trimble Resolution SMT, and they have a bit of battery so they can initially go outdoors for ~30 minutes to get a good fix, but then they get taken indoors and plugged into the network, and probably never get a clear view of a GPS or GLONASS satellite again. - From that point forward (and we hope the devices will have an operational life of at least ten years) they'll be dependent on their internal clock and NTP, but we really need them to stay synchronized to within 100 microseconds. 10 microseconds would be ideal, but 100 would be acceptable. And in order to be useful, they need to stay synchronized at that level of precision essentially forever. My plan, such as it is, was just to get the best clock I could find within budget, integrate it onto the motherboard we're laying out as the system clock, and depend on NTPd to do the right thing with it. 10, or even 100, microseconds is tough with NTP. I don't think it is impossible, but it requires a good, reliable network connection and a bunch of work to identify and reduce the systematic errors. And if NTP == ntpd I'm not sure putting a better oscillator on the board is likely to help all by itself since ntpd's magic internal constants are organized to work with the class of oscillators you typically find in computers, and this would need to be redone to do anything useful with something better. I think making use of NTP at the 10-100 microsecond level might require doing your own software, the generic reference implementation probably won't cut it. Before doing that you might consider some alternatives: - If you are deploying this stuff in the US, and if cell phones (particularly Verizon or Sprint phones) work where you are installing the stuff, you might look at this for a time source: http://www.endruntechnologies.com/time-frequency-reference-cdma.htm This is good if it works everywhere you need it, and assuming CDMA networks continue to operate for another 10 years. - Failing that, look at IEEE 1588. The trouble with this is that it severely constrains the kind of network the equipment is attached to, and the gear used to build that network, but if this is in your control you can buy stuff for this without having to build it. If none of the above works, and you just can't get GPS antennas installed, then you may be stuck with NTP, but getting a reliable 10-100 microseconds out of that is a lot closer to the research part of RD then the development part. I don't think running the generic reference implementation, ntpd, will deliver this. Dennis Ferguson ___ 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] Low-long-term-drift clock for board level integration?
On 19 Feb, 2012, at 21:08 , Bill Woodcock wrote: It's my assumption that some of them will be able to get enough GPS signal (or GPS via a GSM BTS, as we also have a Sierra Wireless GSM chipset onboard) and would thus be able to act as Stratum 1 servers for the others. In the US I suspect all GSM base stations have GPS available (E911 support generally requires it), but I think you may find that in many (most?) other countries the GSM BTS gear has no idea what time it is. GSM doesn't require the time synchronization (it requires frequency, but they can often recover that from the tail circuit connecting the BTS to the network), so in many places they do without GPS either to save money or because the carriers are subject to regulatory requirements to avoid allowing the country's telecommunications facilities to become dependent on GPS (I assume because their regulators don't fully trust the owners of GPS)… GPS can sometimes work under non-optimum circumstances, but it sounds like you have run out of fallback options apart from trying to advance the state of the NTP art. Dennis Ferguson ___ 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] 32768 Hz from 10 MHz
On 3 Feb, 2012, at 14:15 , Orin Eman wrote: On Thu, Feb 2, 2012 at 7:16 PM, Hal Murray hmur...@megapathdsl.net wrote: It's possible to use Bresenham with two integers 10,000,000 and 32,768 but I found no way to perform all the 24-bit calculations on an 8-bit PIC quick enough. Removing the GCD often helps but in this case the accumulator remains 3-bytes wide. To generate 32 kHz you have to toggle a pin and calculate if the next toggle must be 38 or 39 instructions in the future; all the math must occur within 37 instructions. That's why I came up with the binary leap year kind of algorithm; it's as close to math-less as you can get. You missed the simple way. Table lookup. :) The table is only 256 slots long. That's toggling between 305 and 306 cycles. If your CPU uses N clocks per instruction, multiply the table size by N. Well, I thought table lookup too, but I figured a 2048 x 1 table. Easily done with a rotating bit and 256 byte table. Assuming clocking a PIC at 10MHz, you have 2,500,000 instructions per second. Since there was talk about time to the next toggle, we have 2,500,000/65536 instructions between toggles, ie 38.1470... instructions. The fraction turns out to be 301/2048, so you have to distribute 301 extra instructions over every 2048 half-periods of the 32768Hz waveform. I only barely know the instruction set on those processors, but it seems like it should be way easier than that. You know it is going to be 38 or 39 instructions, so that only question is when it should be 39. The value of 250/65536 is 38.1470… in decimal, but in hex it is exactly 26.25a; that is the 0x26 is 38 decimal while the fractional part is only 10 bits long. This means you should be able to compute when the extra cycle is required by keeping a 16 bit accumulator to which the fractional part 0x25a0 is added at every change and executing the extra instruction when there is a carry out of that. The seems straight forward. If `lo' and `hi' are the two halves of the accumulator then the working part of this becomes something like (excusing my PIC assembler, which I mostly forget): movl0xa0,w // low byte of increment into w add w,lo// add w to lo, may set carry movl0x25,w // high byte of increment into w btfsc 3,0 // skip next if carry clear add one,w // increment w by one; I'm not sure how to do that add w,hi// add w to hi, may set carry // if carry set here need extra instruction. Maybe this does it? btfss 3,0 // skip if carry set gotoblorp // carry clear, don't execute next instruction nop // the extra instruction blorp: // enough instructions more to make 38/39 Maybe someone who knows what they're doing can interpret that? Dennis Ferguson ___ 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] GPS receiver vs local oscillator
On 3 Feb, 2012, at 05:07 , Hal Murray wrote: I thought the 4th satellite was needed to determine the time. Wouldn't it take a 5th satellite to also determine the frequency of the local clock? Not really. There are two ways to get the postion and time derivatives. One is to either use two fixes which give you each a (x,y,z,t) tuple, while you know what your expected delta-t is, you can calculate the real delta-t and get from that your frequency offset. That's the sort of thing I'm looking for, but I don't quite get it yet. I have 4 satellites. If I know f, I can solve for x, y, z, and t. If I don't know f, I'm short an equation. If you are using an undisciplined free-running oscillator, as most cheap receivers do, you never know f. What you know is the frequency written on the oscillator's package (call it fn, the nominal frequency), but the actual f is a mystery. Whatever f is, however, you assume f=fn and use that oscillator to generate a local timescale to measure signal phases against. When you solve for x, y, z and t from data generated by measuring the phase of the incoming signals against your oscillator, the `t' you compute is actually a delta_t with respect to the local time scale generated from that oscillator. The value of delta_t tells you the phase error of your local timescale, so the rate of change of delta_t from sample to sample tells you the error in the fn you assumed, that is (f/fn) integrated over the sample interval. If I get two samples, I have 8 equations and I need to solve for: x0, y0, z0, t0, and f0 x1, y1, z1, t1, and f1 That's 10 unknowns with 8 equations. I get a 9th equation by setting t1 = t0 + 1. I'm still short one equation. Can I do something like assume f0 = f1? That would make sense if the change in frequency is small relative to the noise/error in all the other calculations. I suspect that if the local oscillator does not exhibit fairly good short term stability there is no hope of any of this working. That doesn't matter, though, since the GPS `t' you compute is actually a delta_t from whatever your local time scale is, so (delta_t1 - delta_t0) directly tells you how the rate of your local time scale differs from the rate of the GPS timescale. The GPS receiver in fact has no knowledge of the GPS `t' other than as a function of the local time scale. The GPS time scale is purely a paper time scale from the receiver's point of view unless the receiver does the additional work of somehow using that information to generate a real timescale out of the paper. Dennis Ferguson ___ 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] Another Trimble Thunderbolt-like GPS?
On 19 Dec, 2011, at 11:25 , Rob Kimberley wrote: There was a spec issued many years ago to the industry from Lucent I believe to come up with a GPS product for base station requirements. 10 MHz, 1PPS, OCXO, RS-232 port, and a certain holdover spec. The Thunderbolt was one, Starloc another, NanoSync (from Odetics/Zyfer). There were others. Rob Kimberley I think it was actually Qualcomm. The requirement for GPS time and 10 microsecond time synchronization (which informs the holdover spec) came from Qualcomm's CDMA specification and are unique to it. GSM, UMTS and (I think) LTE base stations can get by without GPS at all. Dennis Ferguson ___ 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] Another Trimble Thunderbolt-like GPS?
Yes, USA GSM/UMTS operators also use GPS at their base stations. They may use it for timing, but I think the primary requirement for it comes from E911 support, and maybe to provide AGPS and/or clock-setting support for phones. European and Asian operators often do without it (they typically won't set the time on your phone either). I know of more than a few countries which in fact have national regulatory constraints against relying on GPS for anything important. The base stations still need a frequency reference, but they can generally get that by recovering the clock from the transmission circuit which connects the base station to the rest of the network. Dennis Ferguson On 19 Dec, 2011, at 16:22 , li...@lazygranch.com wrote: I was inside a AT$T shack about a month ago. They have GPS timing inside. I took some photographs, so I will dig up later what timing they use. -Original Message- From: Dennis Ferguson dennis.c.fergu...@gmail.com Sender: time-nuts-boun...@febo.com Date: Mon, 19 Dec 2011 14:00:29 To: Discussion of precise time and frequency measurementtime-nuts@febo.com Reply-To: Discussion of precise time and frequency measurement time-nuts@febo.com Subject: Re: [time-nuts] Another Trimble Thunderbolt-like GPS? On 19 Dec, 2011, at 11:25 , Rob Kimberley wrote: There was a spec issued many years ago to the industry from Lucent I believe to come up with a GPS product for base station requirements. 10 MHz, 1PPS, OCXO, RS-232 port, and a certain holdover spec. The Thunderbolt was one, Starloc another, NanoSync (from Odetics/Zyfer). There were others. Rob Kimberley I think it was actually Qualcomm. The requirement for GPS time and 10 microsecond time synchronization (which informs the holdover spec) came from Qualcomm's CDMA specification and are unique to it. GSM, UMTS and (I think) LTE base stations can get by without GPS at all. Dennis Ferguson ___ 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 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 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] Symmetricom TimeSource 2700
Yes, you are correct. 10 microseconds comes directly from the CDMA spec, it is the amount of time the reference at a base station is allowed to drift when it is in holdover before it is out of spec and needs to be removed from service. I still don't know what they do about path delay since (as you point out) I believe this can be measured only after a handset has registered with the tower, and the timing receivers never register. And the path delay can be quite large if you live far enough away from civilization. When I take my Verizon phone to Toronto it often registers with a Verizon tower which must be at least 20 miles away (i.e. the width of the lake). If that was the distance to the only tower the timing receiver had to listen to that would be more than 100 microseconds of delay, and I don't see how it could correct that. Dennis Ferguson On 30 Nov, 2011, at 02:42 , Peter Bell wrote: It's been a while, but from what I remember the sync channel message does indeed include the system time (which is the same as GPS time with a UTC offset) and also the PN code offset that this cell is using. This leaves the only remaining unknown as the path delay to the cell and the possible error in the local clock on the BTS. The other possible source of error is that if one of the sites loses GPS lock, it will flywheel - this will generate a yellow alarm, but this is not communicated over the air interface - I suspect that the largest component of that stated 10uS maximum timing error is based on worse-case accumulated phase error. I also suspect this is why that Symmetricom box is tracking multiple pilots, so it can isolate and discard any that appear to be significantly out. Regards, Pete On Wed, Nov 30, 2011 at 1:37 PM, Dennis Ferguson dennis.c.fergu...@gmail.com wrote: I think they track both the CDMA pilot and sync channels. The latter channel sends a message which tells the phone about the cell, and gives gives the phone enough information to figure out the time of day. I'm pretty sure CDMA phones have to know what time it is before they register with the cell. To receive the paging channel and negotiate a registration the phone has to receive and send the long code chip sequence, which I think is 2^40 bits long and takes more than a month to repeat. The phone has to know what time it is before it has any hope of tracking that. I don't know how (or if) they deal with the distance from the cell. The accuracy of the PPS signal from CDMA time receivers is usually specified as no better than 10 microseconds or so, so they may just assume the cell tower is close enough not to make it worse than 10 microseconds. Dennis Ferguson On 29 Nov, 2011, at 18:54 , Peter Bell wrote: Assuming it's just tracking the CDMA pilots, the 1PPS output is likely not aligned with UTC. The problem is that the pilot channel is just a PN sequence with no modulating data - so when you lock to it you can know that your local clock is 19200Hz * 64 chips/bit (1.228MHz) - but that's all you know. Even the code phase doesn't tell you anything, since there are two unknowns - the first is the distance to the cell and the second is the code phase offset on this specific pilot (each BTS has it's modulating sequence offset by an integer multiple of 64 chips to reduce mutual interference) - the second piece of information you can obtain by reading one of the overhead channels, but the first is basically not available just using a receiver (your phone can do it, since it can ask transmit back to the BTS and measure the round trip timing offset). ___ 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 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 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] Symmetricom TimeSource 2700
I think they track both the CDMA pilot and sync channels. The latter channel sends a message which tells the phone about the cell, and gives gives the phone enough information to figure out the time of day. I'm pretty sure CDMA phones have to know what time it is before they register with the cell. To receive the paging channel and negotiate a registration the phone has to receive and send the long code chip sequence, which I think is 2^40 bits long and takes more than a month to repeat. The phone has to know what time it is before it has any hope of tracking that. I don't know how (or if) they deal with the distance from the cell. The accuracy of the PPS signal from CDMA time receivers is usually specified as no better than 10 microseconds or so, so they may just assume the cell tower is close enough not to make it worse than 10 microseconds. Dennis Ferguson On 29 Nov, 2011, at 18:54 , Peter Bell wrote: Assuming it's just tracking the CDMA pilots, the 1PPS output is likely not aligned with UTC. The problem is that the pilot channel is just a PN sequence with no modulating data - so when you lock to it you can know that your local clock is 19200Hz * 64 chips/bit (1.228MHz) - but that's all you know. Even the code phase doesn't tell you anything, since there are two unknowns - the first is the distance to the cell and the second is the code phase offset on this specific pilot (each BTS has it's modulating sequence offset by an integer multiple of 64 chips to reduce mutual interference) - the second piece of information you can obtain by reading one of the overhead channels, but the first is basically not available just using a receiver (your phone can do it, since it can ask transmit back to the BTS and measure the round trip timing offset). ___ 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] PC time app
On 25 Nov, 2011, at 21:56 , Steve . wrote: I'm curious as to what folks are doing with PC's that require micro second accuracy for days or weeks or what have you. Any examples? Curious, Steve I have a PCI-X board with an FPGA which implements a clock running at 320 MHz. The 320 MHz can be phase-locked to an external 5 or 10 MHz frequency input, and the card also has 4 PPS inputs. A transition on a PPS input causes the FPGA to record a timestamp, with a precision of not quite 3 ns, and deliver it to software via an interrupt. The 10 MHz and PPS outputs from my GPS receiver are synchronous, so once the board clock is set it keeps the time of the GPS receiver without any further adjustment. The system (the OS is NetBSD, but with the kernel timekeeping replaced) computes its time as a linear function of the CPU's cycle counter, which on my machines seems to run at a constant 2.4 GHz. I can get a sample timestamp (actually a pair for them, the board-computer time comparison mechanism is the trickiest part of the design) from the FPGA by doing a load from a card register, so an 'rdtsc; load; rdtsc' gives me a sample offset between the computer's clock and the card's clock with a constant systematic error which (arguably) should be less than +/- 10 ns and with the board's precision of about 3 ns. I get sample offsets at randomly jittered intervals which average to about 0.25 seconds, so I get about 4 offsets per second with about 3 ns of round-off noise. The processing of these reduces to a linear least squares fit (the y-value is the offset, the x-value is the time of the sample with respect to the computer's clock) after some sanity filtering. The least squares fit gives me a frequency error and a time offset error, along with confidence intervals for each. I adjust the computer's clock when either the frequency error or the time offset becomes non-zero with 80% confidence. Typically I find the result of this to be, very roughly, a clock adjustment every 10 seconds, with a frequency adjustment on the order of 10^-9 and a time adjustment on the order of 10 ns. This is not perfectly reliable, of course; if I leave the cover off the computer and cold-spray the computer's innards I can drive the clock crazy, so it depends on temperature variations inside the case being modest, or at least occurring relatively slowly compared to my offset sample rate. When left alone in a rack in a quiet room, however, I seldom see anything bad happening, so I think it isn't dangerous to assert that the arrangement is typically keeping the computer's clock within +/- 20 ns of the GPS receiver, with worst case excursions being no worse than maybe +/- 50 ns. This has a number of uses, but is particularly good for NTP and PTP development. You can use a board in server synchronize the server's system clock to a GPS receiver, and then use a board tracking the same GPS receiver in a client machine to independently measure how well the software is managing the client machine's clock. This avoids having the NTP or PTP software grade its own homework. Dennis Ferguson ___ 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] Epoch rollover ?
On 21 Oct, 2011, at 11:53 , k4...@aol.com wrote: Bruce, the most common cause of a GPS receiver getting the date incorrect is due to cross-correlation. And cross-correlation is usually the result of too much gain in the GPS antenna's LNA. And depending on the make of the receiver, some will clear the date information with a power cycling and others require erasing the flash memory where those parameters are stored. Do you have a way to see the relative C/No reading for your receiver? Most receivers start experiencing cross-correlation when this reading exceeds 50. Regards, Doug, K4CLE… That might be true in general, but seems exceedingly unlikely to be the problem when the date error has a magnitude of exactly 1024 weeks. The problem is that while GPS tells you the time within a 1024 week era it provides no information about which 1024 week era we're in so even perfectly received GPS signals don't tell the unit what is needed to fix that. Either the receiver needs a separate source of time information to determine the era, or it needs to implement a heuristic to guess at it from the data it does have. Some of the heuristics I've heard of are these (or maybe combinations of these): - Assume the date must be more recent than when the firmware was compiled. By itself this leaves the device with a 1024 week rollover problem, but the 1024 weeks are counted from the date the firmware was compiled rather than from the GPS epoch. - Assume the date must be more recent than the last date on which the unit saved restart information (ephemerides, etc.) in its flash memory, or whatever non-volatile storage it uses for this. This may fail if the unit's flash memory fails. - Guess at the era based on the leap second count (i.e. the UTC offset) informed by some expectation of how many leap seconds are likely to occur in each 1024 week era. This would be an excellent heuristic if the rate of leap second insertion was relatively predicable over long periods, but as it has turned out the relative dearth of leap seconds in the past dozen years might cause particular implementations of this guestimate to fail now. I don't know if any of this matters in the current case. The last (which I think was considered the best practice, or at least the best that could be made of a bad situation, at the time the GPS rollover occurred) is of particular interest right now since it suggests that ending UTC leap seconds might eventually have some unintended consequences for boat anchor GPS timing equipment. Dennis Ferguson ___ 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] Epoch rollover?
On 19 Oct, 2011, at 14:18 , Bruce Lane wrote: Did we just have another GPS epoch rollover? My trusty old Odetics 425 seems to believe the date is March 4th, 1992. I could probably correct it in firmware, if I looked hard and long enough, but the ToD is still correct and the frequency standard is staying nicely locked. Not sure if recovering the correct Julian date is worth the effort. That's neat. I understand that the typical heuristics used to determine the GPS epoch included one or more of: a) Assume the time must be more recent then the date the firmware was compiled; b) Assume the time must be more recent than the last time it dumped the ephemerides out to flash to speed reacquisition across a reboot (assuming it does that); or c) Take a guess based on the leap second count (i.e. the UTC offset) and some expectation of the number of leap seconds per epoch. If your unit does a) it may be that the units firmware just passed 1024 weeks of age. If it does b) it may be your flash memory (or whatever non-volatile storage it keeps that stuff in) has died. It it does c) it may be that the relative dearth of leap seconds in the past dozen years has fooled that algorithm. Or it could be something else entirely. It would be interesting to know which of those it is. Dennis Ferguson ___ 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] Cable delay correction for Tbolt Cs substitude
On 16 Oct, 2011, at 11:21 , WarrenS wrote: Does anyone ever add a temperature controller on the antenna? Maybe that should be my next test. I've seen commercial temperature-controlled antennas. Here's one: http://adriang.com/AACE-Industries/products.htm Dennis Ferguson ___ 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] DCF77 question
On 12 Oct, 2011, at 16:03 , asma...@fc.up.pt wrote: QUOTE: DCF77 marks seconds by reducing carrier power for an interval beginning on the second. UNQUOTE This is not a good (Heaviside step) time marker. For what it is worth, the amplitude modulated on-time second pulse isn't the only time marker DCF77 transmits, it also transmits a phase modulated 512 bit pseudo-random sequence in the remainder of the second. See http://http://en.wikipedia.org/wiki/DCF77 or, for the best technical description I've seen, google Zeit- und Normalfrequenzverbreitung mit DCF77 The phase modulation is much, much better than the amplitude modulation for a pile of reasons. Is it possible to decode that signal by any very simple on-my-shelf gear, as for instance a PC sound card,to recover a good seconds time mark? Do you know some software to perform the above task? I don't yet. I've been playing with DCF77 with a high-zoot RFspace DSP receiver located in Toronto, however, and have found that even though the waterfall display shows only noise at 77.5 kHz, so there's no chance of measuring the amplitude modulated time marker, I can usually detect the phase modulation sequence in the noise for quite a few hours per day using the brute force approach of convolution with an 800 ms long filter matched to the sequence. When I find the time to actually finish this I want to begin recording propagation delays when the signal is detectable. If you end up wanting to build your own detector I'd highly recommend chasing the phase modulated sequence rather than the amplitude modulated second marker. BPC on 68.5 kHz transmits a similar phase modulated code (according to a PTTI paper they presented) which there may be some hope of detecting on the North America west coast. While they don't publish the details, I've been planning to take the receiver and an antenna with me when I next go to Hong Kong and see if I can record enough good quality signal to eventually reverse engineer the encoding out of it. Dennis Ferguson ___ 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] The future of UTC
On 18 Jul 2011, at 05:23 , Tony Finch wrote: Jose Camara camar...@quantacorp.com wrote: I think before adding to the fire of UTC1, UTC7 etc. why not just abolish this silliness called Daylight Savings Time? If there is any benefit to it, just change business operating hours instead. If you want to know why your suggestion doesn't work, David Prerau has collected many many examples. http://www.seizethedaylight.com/ Yet most of the people on the planet live in a place where DST is not observed now, and that includes people living as far north as 65 degrees latitude and as far south as 55 degrees. Should they all be told this doesn't work? Dennis Ferguson ___ 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.