Nutty time-nuttery with WWVB
On 11/10/2016 4:52 PM, Ruslan Nabioullin wrote:
On 11/10/2016 07:18 AM, Peter Reilley wrote:
I have a few of those "atomic" clocks that receive WWVB to set the time.
However since I live on the east coast they may only pick up the signal
once or twice per year.
Could I implement my own personal WWVB transmitter that would
be powerful enough to be picked up by the clocks in my house?
The signal at 60 KHz might be able to be produced directly by some
sound cards. With that and a ferrite rod antenna I might get
reliable time elsewhere in my house outside of my lab.
Has anyone tried this?
Pete.
To be honest, this is very impractical and backward-thinking. I would
suggest instead upgrading to the Internet-of-things paradigm, replacing
these time-of-day displays with full computers running NTP and connected
to your LAN (Android smartwatches; repurposed old smartphones, tablets,
laptops, etc.; and smartclocks [I'm certain that some Silicon Valley
genius'' has already come out with such an invention'' and the
Chinese are churning out cheap knockoffs]), which will query your home
metrology lab's NTP server(s), and instead using WWVB as an additional
timing signal for diversifying your timing source portfolio (with a good
antenna, of course), if you haven't done so already (though such
products appear to be extremely sparse nowadays, for civilian-minded
users have superficially reasoned that GPS is all that is necessary).
-Ruslan
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.
Ruslan
I think you have missed the point this is not so much about keeping time
down to the nano-second
The WWVB clocks being discussed here come in a plethora of decorator
styles and display varieties ( though most are old analog dials) some up
to 3' in diameter.. They are battery powered so one does not have to
connect them to external power source. They serve a dual purpose ....
for the lady friends who keep us fat and happy they are pleasing to the
their sense of form and feng shui (If momma aint happy ..aint nobody
happy) . For the engineer in us we know that the time is accurate to
the second and we don't have get out the step ladder to crawl up there
and reset the clock twice a year when daylight savings changes over.
When I was a road warrior I had a wrist watch that kept WWVB time and I
always knew down to the second how much time I had before I missed the
next flight or train out of Dodge. ( I haven't worn a watch by the way
since I retired in 2008 !)
Nowadays I suppose you could build/buy one that talks to your WLAN but I
suspect the battery life is not quite as good as the current WWVB
version which can run off a watch battery for 5 years. On the other hand
I know I can sleep better at night knowing my WWVB clock has not joined
the dark side and is mindlessly helping the terrorists bring down the
power grid 8^(
Cheers
Dave
Hi
On Nov 10, 2016, at 4:52 PM, Ruslan Nabioullin rnabioullin@gmail.com wrote:
On 11/10/2016 07:18 AM, Peter Reilley wrote:
I have a few of those "atomic" clocks that receive WWVB to set the time.
However since I live on the east coast they may only pick up the signal
once or twice per year.
Could I implement my own personal WWVB transmitter that would
be powerful enough to be picked up by the clocks in my house?
The signal at 60 KHz might be able to be produced directly by some
sound cards. With that and a ferrite rod antenna I might get
reliable time elsewhere in my house outside of my lab.
Has anyone tried this?
Pete.
To be honest, this is very impractical and backward-thinking. I would suggest instead upgrading to the Internet-of-things paradigm, replacing these time-of-day displays with full computers running NTP and connected to your LAN (Android smartwatches; repurposed old smartphones, tablets, laptops, etc.; and smartclocks
Ok, I have one to two dozen clocks around here that sync to WWVB. I swap out batteries every few years and they all keep chunking along. No wires to any of them. No updates or patching on a regular basis. They each have nice useful displays that make sense for their location. The cost for the batch of them was < $400 and that was over a 10 to 20 year span. I have yet to see one of them die.
If I replace all of them with computers, they all need displays on them. Not small displays, but displays that I can see from across the room. The display on a phone or an old tablet isn’t going to cut it. I will now be running / patching Linux on all those devices so none of them are exactly small machines attached to the displays. I have a number of NTP servers running around so yes, I have a pretty good idea what the hardware will look like. I need to get power and network to each clock location. Unless I really enjoy WiFI nonsense, I’ll hard wire them. I also need to work out how to mount the displays and the computers in each location.
There is the minor issue of errr … money. If each display is $100 and the Beagle Bone is $50, 24 of them will run me $3600. There will be another ethernet switch in the mix as well. There will be power supplies, mounting boxes and other stuff. They each will pull at least 50 W with the monitor on. That’s 1.2 KW at 10 cents a KW plus tax plus air conditioning in the summer, call it 20 cents a KWH. That is about $2.1K a year. Over 10 years it’s another $21K.
So what do I get with my investment?
I get a bunch of displays that aren’t going to last five years of 24/7/365 operation (let alone 10 or 20 years). I get a bunch of night lights in places they are not useful (or wanted). I get a system that is less reliable than what I had. I get to have even more fun patching and debugging everything on a regular basis. I get a bunch of clocks that are ugly compared to what is there now.
Hmmm ….. why would I do this?
Bob
[I'm certain that some Silicon Valley genius'' has already come out with such an invention'' and the Chinese are churning out cheap knockoffs]), which will query your home metrology lab's NTP server(s), and instead using WWVB as an additional timing signal for diversifying your timing source portfolio (with a good antenna, of course), if you haven't done so already (though such products appear to be extremely sparse nowadays, for civilian-minded users have superficially reasoned that GPS is all that is necessary).
-Ruslan
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.
Hi Pete:
It's been 4 years since WWVB added phase modulation that's supposed to fix the East coast problem. Have you tried using
a clock that makes use of it?
For example the La Crosse Model # 404-1235UA-SS from Home Depot.
I haven't got one yet since they are about $60 and my old style WWVB clocks work fine, but it would be a lot easier to
get one as opposed to building something.
--
Have Fun,
Brooke Clarke
http://www.PRC68.com
http://www.end2partygovernment.com/2012Issues.html
The lesser of evils is still evil.
-------- Original Message --------
I have a few of those "atomic" clocks that receive WWVB to set the time.
However since I live on the east coast they may only pick up the signal
once or twice per year.
Could I implement my own personal WWVB transmitter that would
be powerful enough to be picked up by the clocks in my house?
The signal at 60 KHz might be able to be produced directly by some
sound cards. With that and a ferrite rod antenna I might get
reliable time elsewhere in my house outside of my lab.
Has anyone tried this?
Pete.
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.
To be honest, this is very impractical and backward-thinking.
I would suggest instead upgrading to the Internet-of-things paradigm, replacing
these time-of-day displays with full computers running NTP and connected
to your LAN (Android smartwatches; repurposed old smartphones, tablets,
laptops, etc.; and smartclocks
Hi Ruslan,
Please tell me this comment is humor or maybe just trolling.
Or are you actually serious?
Thanks
/tvb
Moderator, http://leapsecond.com/time-nuts.htm
On 11/10/2016 05:10 PM, Artek Manuals wrote:
I think you have missed the point this is not so much about keeping time
down to the nano-second
I was suggesting NTP, not PTP.
The WWVB clocks being discussed here come in a plethora of decorator
styles and display varieties ( though most are old analog dials) some up
to 3' in diameter.. They are battery powered so one does not have to
connect them to external power source. They serve a dual purpose ....
for the lady friends who keep us fat and happy they are pleasing to the
their sense of form and feng shui (If momma aint happy ..aint nobody
happy)
Then find a more rationally-minded wife or partner.
. For the engineer in us we know that the time is accurate to
the second and we don't have get out the step ladder to crawl up there
and reset the clock twice a year when daylight savings changes over.
When I was a road warrior I had a wrist watch that kept WWVB time and I
always knew down to the second how much time I had before I missed the
next flight or train out of Dodge. ( I haven't worn a watch by the way
since I retired in 2008 !)
Nowadays I suppose you could build/buy one that talks to your WLAN but I
suspect the battery life is not quite as good as the current WWVB
version which can run off a watch battery for 5 years.
Well you would have to bother with the rather-pointless project of
building a WWVB simulator, taking care not to cause interference.
Stationary clocks do not have to be battery-powered. Yes, battery life
would obviously be a problem for typical smartwatches, but assuming a
rational mind, virtually everyone will opt for a smartwatch over a
WWVB-synchronized dumbwatch, despite the battery life penalty incurred.
On the other hand
I know I can sleep better at night knowing my WWVB clock has not joined
the dark side and is mindlessly helping the terrorists bring down the
power grid 8^(
What a brilliant idea---lets refuse to embrace computerization and the
transition to Internet-of-things (which should've happened in the 80s)
out of fear of the expansion of botnets; furthermore, lets roll back
computerization, and maybe even eliminate all computers? Why have
progress when we can have regress? Of course, we shall disregard the
fact that there are a wide array of all sorts of security
countermeasures (correctness proofs, sourcecode auditing, multifactor
authentication, ASLR, DoD-style MACs, firewalls, IDSs, honeypots,
tarpits, etc.) and the fact that all properly-designed high-security
and/or critical systems are either completely air-gapped (SIPRNet,
JWICS, NSANet, stock exchange internal networks, etc.), or are
unidirectionally interfaced with the Internet (many SCADA sites, etc.)
-Ruslan
On 11/10/2016 05:46 PM, Tom Van Baak wrote:
To be honest, this is very impractical and backward-thinking.
I would suggest instead upgrading to the Internet-of-things paradigm, replacing
these time-of-day displays with full computers running NTP and connected
to your LAN (Android smartwatches; repurposed old smartphones, tablets,
laptops, etc.; and smartclocks
Hi Ruslan,
Please tell me this comment is humor or maybe just trolling.
Or are you actually serious?
I am absolutely serious.
-Ruslan
There are some problems with the Internet of Things and security.
For example, look up the Mirai botnet and the DDoS attack on DNS
provider Dyn last month.
To understand the situation better, read Bruce Schneier's "Secrets and
Lies", available as a PDF.
The book was written in 2000 when things were simpler and easier to
understand, but it still applies.
The man has a sense of humor.
Schneier makes the point that it is difficult to test software
functionality but impossible to test for security.
Buffer overruns were a problem back when Morris wrote the first worm.
They are still a problem because software vendors can write legal
license agreements that say the vendor is not responsible for any
failures caused by the software. Basically, you bought it on an "as-is"
basis. So the software ships, bugs are discovered, and sometimes the
vendor fixes the bugs, especially if they are made public. Vendors don't
spend money doing things they don't have to do.
IoT devices are made as cheaply as possible. They are made user friendly
by not burdening the user with security configuration, so user names and
passwords have well-known defaults. The device has no anti-virus
application. The simple routers offer little protection, as they have
their own issues with default keys to their configuration.
Marketing departments have gotten very good at stampeding buyers for
things of little value. So good that companies that make the control
systems for the nation's manufacturing plants and utilities have
embraced the Industrial IoT. That should have been the Industrial
Distributed IoT (IDIoT). Previously, control systems had no connection
to the Internet, and so there was no need for Internet Security
concerns. Now there are many security services, so the IDIoT has created
jobs, as well as sales of routers. The air gap and the data diode have
been discredited since Stuxnet, which was spread by strewing the parking
lot with USB drives.
Schneier emphasizes that the difficulty of providing security increases
with the complexity of the system. As you may have noticed, each new
revision of an operating system provides about a 4X increase of lines of
code (in the case of Windows). Security is always in a catch-up race
with the ability of criminals to find and exploit faults.
Please pardon this intrusion into the world of precision time, but the
issue was raised here. As a designer of industrial control systems, I've
made it a point to study security, and found Schneier to be a fount of
information.
Perhaps a disclaimer is in order. I do not know Schneier and receive
nothing from plugging his work.
Bill Hawkins
I thought the same at first, but then I thought a little more, it may not
be time nut standards of accuracy but...
It's possible to buy ESP8266 modules for a couple of pounds, they will run
NTP, if you want to go up market then a Pi Zero.
Hooked up to a cheap DS1307 module or one of the higher stability ones,
then your choice of display, cheap SPI TFT, retro VFD or Nixies, you could
build an NTP synchronised clock for under £10.
Rolling it yourself like that would allow you to negate the security risks
associated with IoT as you'd have control of the security aspects.
On 10 November 2016 at 22:59, Ruslan Nabioullin rnabioullin@gmail.com
wrote:
On 11/10/2016 05:46 PM, Tom Van Baak wrote:
To be honest, this is very impractical and backward-thinking.
I would suggest instead upgrading to the Internet-of-things paradigm,
replacing
these time-of-day displays with full computers running NTP and connected
to your LAN (Android smartwatches; repurposed old smartphones, tablets,
laptops, etc.; and smartclocks
Hi Ruslan,
Please tell me this comment is humor or maybe just trolling.
Or are you actually serious?
I am absolutely serious.
-Ruslan
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/m
ailman/listinfo/time-nuts
and follow the instructions there.
--
Clint.
No trees were harmed in the sending of this mail. However, a large number
of electrons were greatly inconvenienced.
Well I guess I am lucky to live only 20km from the DCF77 transmitter. I can receive DCF77 with a wet finger. I can even use a DCF77 controlled OCXO as my 10MHz standard. Probably not as good as a GPSDO but should be within 1x10^-10 or even better, well enough for my hamradio applications which is shortwave to UHF.
If I ever get to it I will try to measure the frequency error of the DCF77 controlled oscillator over the day. I guess not much difference as I receive groundwave all day.
Even at my second home in southern France I have a strong DCF77 signal during the night with an indoor antenna (5cm ferrit rod).
Heinz DH2FA
Von meinem iPhone gesendet
Am 11.11.2016 um 09:14 schrieb Clint Jay cjaysharp@gmail.com:
I thought the same at first, but then I thought a little more, it may not
be time nut standards of accuracy but...
It's possible to buy ESP8266 modules for a couple of pounds, they will run
NTP, if you want to go up market then a Pi Zero.
Hooked up to a cheap DS1307 module or one of the higher stability ones,
then your choice of display, cheap SPI TFT, retro VFD or Nixies, you could
build an NTP synchronised clock for under £10.
Rolling it yourself like that would allow you to negate the security risks
associated with IoT as you'd have control of the security aspects.
On 10 November 2016 at 22:59, Ruslan Nabioullin rnabioullin@gmail.com
wrote:
On 11/10/2016 05:46 PM, Tom Van Baak wrote:
To be honest, this is very impractical and backward-thinking.
I would suggest instead upgrading to the Internet-of-things paradigm,
replacing
these time-of-day displays with full computers running NTP and connected
to your LAN (Android smartwatches; repurposed old smartphones, tablets,
laptops, etc.; and smartclocks
Hi Ruslan,
Please tell me this comment is humor or maybe just trolling.
Or are you actually serious?
I am absolutely serious.
-Ruslan
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/m
ailman/listinfo/time-nuts
and follow the instructions there.
--
Clint.
No trees were harmed in the sending of this mail. However, a large number
of electrons were greatly inconvenienced.
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.
Please see http://support.ntp.org/bin/view/Main/SecurityNotice
or Google "NTP security"
But perhaps you meant to create a local NTP network with no connection
to the Internet.
In that case, SNTP is sufficient for wall clocks.
Bill Hawkins
-----Original Message-----
From: time-nuts [mailto:time-nuts-bounces@febo.com] On Behalf Of Clint
Jay
Sent: Friday, November 11, 2016 2:15 AM
Rolling it yourself like that would allow you to negate the security
risks associated with IoT as you'd have control of the security aspects.