Shutoff of WWV-WWVH: two days left to the end of the subscription

Yes on backup/redundancy and the curators of the GPS system agree...
https://www.gps.gov/support/faq/#jamming

Like all radio-based services, GPS is subject to interference from both natural and human-made sources. A GPS unit can lose reception in the presence of devices designed for intentional radio jamming. Solar flares can also disrupt GPS equipment. For this reason, the U.S. government strongly encourages all GPS users to maintain backup/alternative positioning, navigation, and timing capabilities. In addition, the government is currently fielding new GPS signals that are more resistant to interference.

Commercial aircraft that use GPS are required to maintain alternative means of navigation. If intentional jamming were directed against aircraft, the pilots would revert to other sensors and ground-based navigation aids. Air traffic control would continue to provide surveillance services and ensure aircraft separation.​

Emphasis added.

These articles are interesting as well...

• https://arstechnica.com/information...ise-ushers-in-gps-jamming-season-across-west/
• https://arstechnica.com/gadgets/201...ke-global-return-as-gps-backup-because-cyber/
  
• https://arstechnica.com/information...research-measures-growing-gps-jamming-threat/

Keeping alternative time broadcasts is prudent and, thanks to the already sunk costs into the WWV/H/B systems, relatively cheap.

 

WWV is still running. What is the deal?

 

It's overall concern regarding NIST putting the future of WWV/H/B on the budgetary chopping block to makes ends meet.

https://www.hamradio.me/news/nist-takes-aim-wwvb-wwv-and-wwvh.html

 

I believe every ham on the planet knows about the GPS option, Ive been experimenting with one for quite a while.
I worked for Motorola for more years than I care to remember, but we used the master clocks that referenced off the 60 Khz signal for many public safety agencies.
Are they just supposed to junk them now? They were extremely expensive. The master clock ran/referenced everything from 911 recorders, to console clocks.
I'm retired now, I guess they will get upgraded.

 

The biggest money drain from the government is the military and defense budget.

 

This was a poorly titled thread. The "two days left" referred to the deadline to sign a petition. The proposed NIST budget suggested cutting these programs as part of a cost savings measure - key words are PROPOSED.

 

I loaded up an I5 at 3.5 Ghz, with lots of memory, and three physical serial ports with Fedora 28. Removed chronyd, installed ntpd and gpsd. Had an epiphany. I think type 28 (shared memory) local clock is the way of the future. That removes a lot of interoperability problems and makes it easier to write a driver for your device. I was trying to do things the old fashioned way.

Plugged in a ublock GPS USB receiver and along side the default fedora time server pool, I get this:


remote refid st t when poll reach delay offset jitter
==============================================================================
2.fedora.pool.n .POOL. 16 p - 64 0 0.000 0.000 0.000
*SHM(0) .GPS. 0 l 14 16 377 0.000 5.386 4.168
SHM(1) .PPS. 0 l - 16 0 0.000 0.000 0.000
-144.202.41.122 132.163.96.3 2 u 38 64 377 39.337 65.234 7.898
-utility0.ncssm. 152.2.253.126 2 u 50 64 377 39.106 64.743 8.057
-mailhost.netads 209.51.161.238 2 u 31 64 377 49.144 64.179 7.650
+23.134.96.254 129.7.1.66 2 u 43 64 377 12.810 64.421 8.618
+198.167.142.149 140.142.2.8 3 u 56 64 377 62.139 47.440 18.266
-pegasus.latt.ne 130.207.244.240 2 u 17 64 377 35.964 58.165 5.423
-time.tritn.com 66.220.9.122 2 u 11 64 377 43.467 33.060 4.851

PPS is working. GPS NMEA strings are working. According to cgps I'm seeing 7 satellites and the receiver's about 10 feet away from a window.

Another funny thing. This implementation is choosing the GPS as "the clock" to use even though all the other fedora pool internet servers should say GPS is the false ticker.

Hmmmm...

I'm looking at cgps and I can't tell the difference between the WWV ticks and the clock readout from cgps.

Now I'll work on the audio driver which reads the one-minute BCD frame from WWV. I should have my WWVB receiver in the next few days.

 

In response to "Does anybody know what time it is? Does anybody really care?"

We will all one day not care about this linear time we are currently experiencing.

 

Actually, PPS is not working, according to the ntpq output you posted. Your reach is 0, where it should be 377. Also, the when column is '-', but it should show the count of seconds since the clock was last checked.

 

I switched one of my Ubuntu boxes over to chrony. It found a stratum 1 server in 2.debian.pool.ntp.org.

 

Ah! You're right! I missed that. I was assuming since it was a 1 pulse per second input, it was a special case. I'll look into it. It may be that I'm using one of the cheapest of the USB GPS receivers and it may not produce a 1 PPS.

I'm also seeing as I'm searching the internet that I may not have much luck with the WWV audio decoder or the WWVB decoder because it looks like those driver types have been deprecated.

 

For 1PPS to be accurate, it really need to be acted on by a kernel level hardware interrupt - and I believe there is a custom version of Raspian for the Pi to do this. Otherwise, acting on the interrupt gets scheduled. All depends oh how accurate you really want to be.

 

GPS isn't free to those missing GPS equipment. Yes, I know many hams who know how to use it, but don't trust it when they end up driving down a road that isn't pictured on the map -- in some exurban area! In some places, we still don't have electricity here.

Plus, "stick in the water" gaseous diffraction, tropospheric delay (ducting), multi-path (Rayleigh fading) and other distorting propagation to/from these satellites makes them less than fool proof. It happens more than you notice.

Without a doubt, the maintenance cost is enormous just for timekeeping.

Internet time keeping: don't get me started. I used to deal with delay over 5k miles of fiber and microwave network links. Even my experts called it "pin point accuracy guesswork."

Propagation models are just that: models. And you know how skinny models are. I rarely see one that works in this area.

WWV & WWVB are in-your-face time clocks, ZERO analysis, instantaneous propagation verification and don't use skinny [runway] models. Yeah, they're modified by propagation, but in a wholly defined and still usable way.

Sorry you guys missed using them. They answer many questions that most other systems are still scratching their heads over.

I guess the days of radio below 800 MHz are over!

 

Yes, it's kernel level in Fedora 28 also (lsmod: pps_core). And I've given up on these el-cheapo USB GPS receivers. They're good for location and only good for within a second of time since they don't produce a 1 PPS.

I'll soon have a Novus GPSDO. It has a BNC connector on it labeld "PPS". I'll finally have a GPS PPS to work with.

And the latest version of NTP still has the drivers for WWV (audio codec) and WWVB (serial port). I think all I have to do is create a soft link from /dev/audio to /dev/pulse and a soft link from /dev/wwvb0 to /dev/ttySxx and I'll be able to compare them (along with adding in fixed propagation delay). And I'll also be able to tell precisely when WWV and WWVB go off the air (probably in a date/time stamped log file somewhere).

 

Well said, Ryan. Please don't let this forum jade you.