Re: [time-nuts] GPS shielding by power lines?
> The thought was that there as interference arcing or corona noise > from the > line insulators, and a receiver (AM) was deployed to listen for > what was > expected to be a substantial wide band noise signalwe didnt > hear one! We > are now confused about what the effect is. The signal could not be > "screened" by the wires which are about 3 feet apart, but they > definite > provide a cone of interference directly under the run. The > experiment was > later repeated with two further different GPS receivers and > produced the > same result. > > Has anyone seen this before? have you any idea of what level noise > we should > be looking for? I believe this is a wide signal so maybe an AM > receiver is > not the best choice The area is a rural, horticultural area (called > market > garden in the UK) We are obviouslt concerned to trace any noise > sources in > the vicinity of the Hydrogen line frequency at 1420MHz. > > Alan G3NYK Alan; I have no idea if this is a cause of your problem, but in some work I have done, I found that the high field strengths (electrostatic and magnetic) found under power lines can adversely effect electronic equipment that should otherwise be unaffected by the presence of power lines and their low frequency signals. I would imagine that the fields could desensitize receivers by overloading the front end amplifiers, and no amount of filtering is going to help because it is virtually impossible to shield against 50/60 Hz signals. A solid steel conduit several mm thick is only good for about 27 dB attenuation at power line frequencies. Fortunately, distance makes things better, so you may just have to move everything a bit away from the power lines. Our solution was to digitize our signals several hundred meters from the power lines and send the results digitally past them. Try getting a field strength meter (gaussmeter) and see what levels you have under the lines. Tom Frank, KA2CDK ___ 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] Power supply for Thunderbolt
- Original Message -
From: "John Miles" <[EMAIL PROTECTED]>
To: "Discussion of precise time and frequency measurement"
Sent: Friday, August 08, 2008 1:21 AM
Subject: Re: [time-nuts] Power supply for Thunderbolt
>
>>
>> It would be interesting to see how the 'red panel' Thunderbolts
>> (the
>> packaged type that run from an included switching supply fed from
>> 24
>> VDC) perform under a test regime such as that TVB used to compare
>> various three rail power supplies. In 'distant past' posts, maybe
>> 18
>> months back, it was suggested that this supply is a particularly
>> good
>> performer, but IIRC no data was supplied to support this statement.
>> Anyone able to oblige?
>
> This (cluttered) graph is indicative of the results I've seen. The
> Lucent
> supply used in the older red-and-black labelled Thunderbolts was not
> very
> clean, and its proximity and orientation adjacent to the
> Thunderbolt's PCB
> didn't help.
>
> Red = the new ("TAPR") Thunderbolt with my old Thunderbolt's Lucent
> supply
> Blue = the old Thunderbolt with its Lucent supply
> Orange = the new Thunderbolt with the supply Tom chose
> Purple = the old Thunderbolt with Tom's supply
>
> The reference for the red trace was somewhat noisier below 100 Hz
> than the
> one used for the others. You should assume that the red trace
> coincides
> with the orange one below 100 Hz.
>
> -- john, KE5FX
Tnx John-looks like the included power supply was not a great
performer in this regard.
DaveB, NZ
___
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] Z3801A
Hi all, I am reviving an old Z3801A that was stored for 5 years. I did the RS232 header mod and I can communicate with the unit. The power LED does not come on, but I can control the enabled and active LED. I am not able to enter a date after 2007 to speed up the adquisition process. Is there a new firmware that I can load? Anyone heard of LED problems? There are no alarms in the status window. Regards, Pieter, N4IP ___ 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 PPS output
The TB's PPS output has something like 5 ohm output impedance. It is very desirable to terminate the coax cable with 50 ohm at the other end, otherwise you will get endless ringing The following web page uses the PPS output of the Thunderbolt to illustrate ringing in a non-matched cable: http://www.ko4bb.com/Test_Equipment/CoaxCableMatching.php And this page shows a minimum, energy efficient pulse stretcher: http://www.ko4bb.com/Test_Equipment/Thunderbolt/PulseStretching/ Didier KO4BB > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Scott Mace > Sent: Saturday, August 09, 2008 5:41 PM > To: Discussion of precise time and frequency measurement > Subject: [time-nuts] Thunderbolt PPS output > > Has anyone used the thunderbolt PPS output with a N8UR > FatPPS? I'm seeing a large (+-100-300us) amount of jitter on > the output of the FatPPS. This same setup works fine with > the input from z3801a (using a PECL/TTL converter) or the > Fury (even though the Fury's pulse is wide enough to begin with). > Is there something special about the thunderbolt PPS output? > I noticed it seemed to require 50ohm termination to get a > clean signal when I connected it to a counter (5334a) with a > 3 foot cable. > > I'm using the FatPPS in conjunction with a soekris 4501 with > the tmrin mod. > > Scott > > ___ > 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] Thunderbolt PPS output
Has anyone used the thunderbolt PPS output with a N8UR FatPPS? I'm seeing a large (+-100-300us) amount of jitter on the output of the FatPPS. This same setup works fine with the input from z3801a (using a PECL/TTL converter) or the Fury (even though the Fury's pulse is wide enough to begin with). Is there something special about the thunderbolt PPS output? I noticed it seemed to require 50ohm termination to get a clean signal when I connected it to a counter (5334a) with a 3 foot cable. I'm using the FatPPS in conjunction with a soekris 4501 with the tmrin mod. Scott ___ 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] TBolt and Z3801A performance data
> Hi John, > > On Sat, 2008-08-09 at 17:48 -0400, John Ackermann N8UR wrote: > > > The longer tau readings are interesting mainly to see how the GPS loop > > kicks in; tau below about 1000 seconds seem related to OCXO performance > > on the TBolts, while the Z3801A seems to have a longer time constant, > > and somewhat more gradual attack. (By the way -- the TBolts were > > running with the loop parameters they were set to when I got them). > > > > John > > Would the fact that the bump is that "big" indicate that a more > "aggressive" loop parameter setup would improve those two individual > units performance? It would be an interesting thing to try. The default loop bandwidth is 0.01 Hz, but since the peak of the hump is right there too, I'm not sure if GPS or the OCXO is responsible for it. The OCXO could be relatively stable at a 100-second scale but dragged around inappropriately by the GPS receiver, or the OCXO could be relatively unstable and corrected too slowly by the GPS receiver. It looks like HP uses a much lower loop bandwidth. Even so, it's interesting that we see no hint of convergence between the two GPS-locked HP oscillators at longer timescales. It's unfortunate that this kind of test takes so long to run, because it's the sort of thing you need to repeat several times to make sure you're getting at the truth of the matter. -- john, KE5FX ___ 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] TBolt and Z3801A performance data
Hi John, On Sat, 2008-08-09 at 17:48 -0400, John Ackermann N8UR wrote: > The longer tau readings are interesting mainly to see how the GPS loop > kicks in; tau below about 1000 seconds seem related to OCXO performance > on the TBolts, while the Z3801A seems to have a longer time constant, > and somewhat more gradual attack. (By the way -- the TBolts were > running with the loop parameters they were set to when I got them). > > John Would the fact that the bump is that "big" indicate that a more "aggressive" loop parameter setup would improve those two individual units performance? -- Björn ___ 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] TBolt and Z3801A performance data
Yes, all were made with GPS antenna connected (and running from the same antenna via an 8-port splitter). What I was mainly commenting on is the TBolt's pronounced "hump" that shows up in the measurements against the FTS-1050A but largely disappears in the pairwise measurement. The longer tau readings are interesting mainly to see how the GPS loop kicks in; tau below about 1000 seconds seem related to OCXO performance on the TBolts, while the Z3801A seems to have a longer time constant, and somewhat more gradual attack. (By the way -- the TBolts were running with the loop parameters they were set to when I got them). John John Miles said the following on 08/09/2008 05:41 PM: > These tests were made with the GPS antenna connected? At t >> 100 seconds, > they should all look about the same, because that's where GPS disciplining > comes in, no? They should not be uncorrelated in the long run. > > To the extent one Z3801 looks worse than the other at large values of tau, > I'd expect there to be a good reason, like better GPS reception on one of > them, or a much-worse OCXO. > > Measuring phase noise by comparing the two against each other should be > fine, though, since their short-term drift isn't being corrected at that > timescale. Your noise floor is a couple dB worse than the one I tested, but > they're otherwise about the same. > > - john, KE5FX > > >> -Original Message- >> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] >> Behalf Of John Ackermann N8UR >> Sent: Saturday, August 09, 2008 1:28 PM >> To: Discussion of precise time and frequency measurement >> Subject: [time-nuts] TBolt and Z3801A performance data >> >> >> I just put the results of some tests of 2 "Time-Nuts Special" >> Thunderbolts, as well as 2 Z3801As, at >> http://www.febo.com/pages/gpsdo_comparison/ >> >> I learned an interesting (and important) lesson in doing these >> measurement. I initially measured the pairs of GPSDO against each other >> (e.g., one TBolt as "reference" and the other TBolt as "DUT"). In >> theory, if the two oscillators are identical, and if their noise is >> uncorrelated, the results of the pair can be used to deduce the results >> of the individual units. >> >> However, in this case doing so gave a very optimistic view -- the TBolts >> were better than the pair of Z3801As! Another set of measurements >> comparing the GPSDOs against an independent reference revealed that the >> first measurement was a lie. >> >> I guess you can think of it like this. Picture two OCXOs that both age >> at the same rate and in the same direction. Because they drift >> together, measuring their relative phase hides their actual drift and >> makes them look better than they are. On the other hand, if they were >> drifting in opposite directions, they would look worse than they are. >> An identical aging trend is a "correlation" even though it's external to >> the things we usually think of. >> >> John > > > ___ > 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] TBolt
Hi, Where can I purchase one or two of the time-nuts Tbolt specials? Thanks, Pieter - Original Message - From: "John Ackermann N8UR" <[EMAIL PROTECTED]> To: "Discussion of precise time and frequency measurement" Sent: Saturday, August 09, 2008 4:27 PM Subject: [time-nuts] TBolt and Z3801A performance data >I just put the results of some tests of 2 "Time-Nuts Special" > Thunderbolts, as well as 2 Z3801As, at > http://www.febo.com/pages/gpsdo_comparison/ > > I learned an interesting (and important) lesson in doing these > measurement. I initially measured the pairs of GPSDO against each other > (e.g., one TBolt as "reference" and the other TBolt as "DUT"). In > theory, if the two oscillators are identical, and if their noise is > uncorrelated, the results of the pair can be used to deduce the results > of the individual units. > > However, in this case doing so gave a very optimistic view -- the TBolts > were better than the pair of Z3801As! Another set of measurements > comparing the GPSDOs against an independent reference revealed that the > first measurement was a lie. > > I guess you can think of it like this. Picture two OCXOs that both age > at the same rate and in the same direction. Because they drift > together, measuring their relative phase hides their actual drift and > makes them look better than they are. On the other hand, if they were > drifting in opposite directions, they would look worse than they are. > An identical aging trend is a "correlation" even though it's external to > the things we usually think of. > > John > > ___ > 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] TBolt and Z3801A performance data
These tests were made with the GPS antenna connected? At t >> 100 seconds, they should all look about the same, because that's where GPS disciplining comes in, no? They should not be uncorrelated in the long run. To the extent one Z3801 looks worse than the other at large values of tau, I'd expect there to be a good reason, like better GPS reception on one of them, or a much-worse OCXO. Measuring phase noise by comparing the two against each other should be fine, though, since their short-term drift isn't being corrected at that timescale. Your noise floor is a couple dB worse than the one I tested, but they're otherwise about the same. - john, KE5FX > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of John Ackermann N8UR > Sent: Saturday, August 09, 2008 1:28 PM > To: Discussion of precise time and frequency measurement > Subject: [time-nuts] TBolt and Z3801A performance data > > > I just put the results of some tests of 2 "Time-Nuts Special" > Thunderbolts, as well as 2 Z3801As, at > http://www.febo.com/pages/gpsdo_comparison/ > > I learned an interesting (and important) lesson in doing these > measurement. I initially measured the pairs of GPSDO against each other > (e.g., one TBolt as "reference" and the other TBolt as "DUT"). In > theory, if the two oscillators are identical, and if their noise is > uncorrelated, the results of the pair can be used to deduce the results > of the individual units. > > However, in this case doing so gave a very optimistic view -- the TBolts > were better than the pair of Z3801As! Another set of measurements > comparing the GPSDOs against an independent reference revealed that the > first measurement was a lie. > > I guess you can think of it like this. Picture two OCXOs that both age > at the same rate and in the same direction. Because they drift > together, measuring their relative phase hides their actual drift and > makes them look better than they are. On the other hand, if they were > drifting in opposite directions, they would look worse than they are. > An identical aging trend is a "correlation" even though it's external to > the things we usually think of. > > John ___ 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 shielding by power lines?
Alan Melia wrote: > Hi all, in the process of setting up a GPS time standard for a Radio > Astronomy facility (amateur) we installed a GPS receiver in a small cabin > with a translucent roof, thinking that would not impede the GPS signal. > After a lot of head scratching as to why we were not getting the performane > we got at another site, we realised that the "convenient position" for the > cabin was directly below a three phase 11kV power distribution line ( common > UK rural electricity distribution system). We extended the cable and moved > the antenna about 20 - 30 feet to the side of the line run, which was > mounted on wooden poles at about 25 feet. In this position we immediately > got a reliable fix. The fix and number of usable satellites degrades as we > move nearer the lines. > > The thought was that there as interference arcing or corona noise from the > line insulators, and a receiver (AM) was deployed to listen for what was > expected to be a substantial wide band noise signalwe didnt hear one! We > are now confused about what the effect is. The signal could not be > "screened" by the wires which are about 3 feet apart, but they definite > provide a cone of interference directly under the run. The experiment was > later repeated with two further different GPS receivers and produced the > same result. > > Has anyone seen this before? have you any idea of what level noise we should > be looking for? I believe this is a wide signal so maybe an AM receiver is > not the best choice The area is a rural, horticultural area (called market > garden in the UK) We are obviouslt concerned to trace any noise sources in > the vicinity of the Hydrogen line frequency at 1420MHz. It can also be a very simple case of strong multipath causing cancelation reflections strong enought to obscure the signal. Be sure to plot signal strength and if possible delta distance from solution for each satellite along with elevation and azimuth. See if a clear pattern emerges such that certain angles is problematic. Just to make you look in another direction. Cheers, Magnus ___ 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] GPS shielding by power lines?
Hi all, in the process of setting up a GPS time standard for a Radio Astronomy facility (amateur) we installed a GPS receiver in a small cabin with a translucent roof, thinking that would not impede the GPS signal. After a lot of head scratching as to why we were not getting the performane we got at another site, we realised that the "convenient position" for the cabin was directly below a three phase 11kV power distribution line ( common UK rural electricity distribution system). We extended the cable and moved the antenna about 20 - 30 feet to the side of the line run, which was mounted on wooden poles at about 25 feet. In this position we immediately got a reliable fix. The fix and number of usable satellites degrades as we move nearer the lines. The thought was that there as interference arcing or corona noise from the line insulators, and a receiver (AM) was deployed to listen for what was expected to be a substantial wide band noise signalwe didnt hear one! We are now confused about what the effect is. The signal could not be "screened" by the wires which are about 3 feet apart, but they definite provide a cone of interference directly under the run. The experiment was later repeated with two further different GPS receivers and produced the same result. Has anyone seen this before? have you any idea of what level noise we should be looking for? I believe this is a wide signal so maybe an AM receiver is not the best choice The area is a rural, horticultural area (called market garden in the UK) We are obviouslt concerned to trace any noise sources in the vicinity of the Hydrogen line frequency at 1420MHz. Alan G3NYK ___ 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] TBolt and Z3801A performance data
I just put the results of some tests of 2 "Time-Nuts Special" Thunderbolts, as well as 2 Z3801As, at http://www.febo.com/pages/gpsdo_comparison/ I learned an interesting (and important) lesson in doing these measurement. I initially measured the pairs of GPSDO against each other (e.g., one TBolt as "reference" and the other TBolt as "DUT"). In theory, if the two oscillators are identical, and if their noise is uncorrelated, the results of the pair can be used to deduce the results of the individual units. However, in this case doing so gave a very optimistic view -- the TBolts were better than the pair of Z3801As! Another set of measurements comparing the GPSDOs against an independent reference revealed that the first measurement was a lie. I guess you can think of it like this. Picture two OCXOs that both age at the same rate and in the same direction. Because they drift together, measuring their relative phase hides their actual drift and makes them look better than they are. On the other hand, if they were drifting in opposite directions, they would look worse than they are. An identical aging trend is a "correlation" even though it's external to the things we usually think of. John ___ 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 name "Totally Accurate Clock"
On Aug 5, 2008, at 5:10 PM, Tom Clark, K3IO wrote: > > My simple timing receiver was already producing sub-usec results > using > Motorola's PVT-6 "Six-Pack" 6-channel receiver, even in the face of > Selective Availability (SA). I think it was the "Six-Gun" receiver; I recall at the time while I was at U of MD, and we were doing some early NTP work, we bought a couple of PVT-6 receivers and "Six Gun" seems to ring a bell.. louie wa3ymh ___ 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.
