cyamamot@aludra.usc.edu (Cliff Yamamoto) (01/11/90)
I ask because I'd like to get one of Heathkit's Most Accurate Clock kits. I remember listening to WWV back in elementary school so I know its been here a while. How long has the embedded time format been used in WWV? What year was WWV started? Does anyone think the NBS has any plans or reasons to cancel WWV or WWVH transmissions? I hope it stays around for another 50 years at least. Just wondering... Cliff Yamamoto - KA6JRG
louie@sayshell.umd.edu (Louis A. Mamakos) (01/11/90)
I suspect that WWV, at least, will be around for a little while yet. I spoke an engineer in Boulder a few weeks ago, and he said that they're in the process of procuring a new time-code generator for the WWV transmitters. It will allow the WWV transmitters to transmit the year, along with the day of the year and time as part of the time code. They recently did a similar upgrade to the WWVB 60kHz transmitter at the first of the year, after the leap second. As far as the Heathkit "Most Accurate Clock", well, its not. The RS-232 interface has some severe problems which precludes its effective use to those persons that want accuracy in the millisecond regime. If you're looking for somethat that's accurate to within a second, you'll have no problem with it. There are a number of other WWV based radio clocks on the market, though not "consumer priced" like the Heath clock. If you're a time freak, like me, you'll really want to invest in a WWVB clock. The propagation characteristics at 60kHz are much more predictable than the HF propagate at 5, 10 and 15 MHz. We've got a Spectracom 8170 WWVB clock in use here, and there are a bunch of other scattered around the Internet. louie
jeffc@ncr-fc.FtCollins.NCR.com ( Jeff Cook) (01/12/90)
In article <1990Jan10.175719.8720@haven.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes: > If you're a time freak, like me, you'll really want to invest in a > WWVB clock. The propagation characteristics at 60kHz are much more > predictable than the HF propagate at 5, 10 and 15 MHz. We've got a > Spectracom 8170 WWVB clock in use here, and there are a bunch of other > scattered around the Internet. Sounds interesting, but expensive. How much would something like this cost? -- --------------------------------------------------------------------- Jeffrey L. Cook jeffc%ncr-fc@ncr-sd.SanDiego.NCR.com NCR Microelectronics uunet!ncrlnk!ncr-sd!ncr-fc!jeffc 2001 Danfield Court Fort Collins, CO 80525 These views are entirely mine, etc... "I was gratified to be able to answer promptly, and I did. I said I didn't know." -- Mark Twain ---------------------------------------------------------------------
louie@sayshell.umd.edu (Louis A. Mamakos) (01/13/90)
Well, here's a comment out of the module of ntpd for unix. It seems that
while you can get time samples from the clock, which one do you believe?
Other clocks, like the Spectracom 8170 WWVB clock, provide a seperate
TTL level output which is the on-time pulse. It is driven high for 100 ms,
with the leading edge of the pulse corresponding to the beginning of the
second.
/* read_clock_heath() -- Read the Heath Radio Clock.
*
* Strategy Notes:
*
* When the heath clock is set at 9600 baud and is in continuous mode,
* it will output an alternating set of two or three time strings per tenth
* of a second. For instance:
*
* 22:20:04.2 07/26/89<CR>
* 22:20:04.2 07/26/89<CR>
* 22:20:04.3 07/26/89<CR>
* 22:20:04.3 07/26/89<CR>
* 22:20:04.3 07/26/89<CR>
* 22:20:04.4 07/26/89<CR>
* 22:20:04.4 07/26/89<CR>
* 22:20:04.5 07/26/89<CR>
* 22:20:04.5 07/26/89<CR>
* 22:20:04.5 07/26/89<CR>
*
* We assume that the best place to take a time sample is at the start of
* the first string of a 3 string sequence. This is based solely on the
* guess that fitting 3 strings in 1/10th of a second means that the first
* string is closer to "on-time" than if just two strings fit.
*/
Some other people have asked about sources of WWV/WWVB synchronized and other
reference clocks. Here's a list of clocks and their sources:
Radio timecode receivers and other sources of synchronization
Following is a list of radio timecode receivers currently on the market.
These devices are attached via a serial asynchronous line, which is used to
send poll messages and receive responses in the following example formats:
Spectracom clock: <cr><lf>q ddd hh:mm:ss tz=zz<cr> (22 chars)
Truetime clock: <cr><lf><ctrl-a>ddd:hh:mm:ssq<cr> (13 chars)
Heath clock: <cr>hh:mm:ss.f dd/mm/yr<cr> (23 chars)
PST clock: hh:mm:ss.fff yy/dd/mm/ddd status<cr> (49 chars)
These formats are exhibited for example only. In some clocks an extensive
repitoire of commands and formats are available.
In the Fuzzball implementation the Spectracom and PST clocks are polled,
which return the time-code string shown, while the other clocks run
continuously. In the Truetime clock the start bit of the last <cr>
represents on-time, while in the other clocks the first <cr> represents on-
time. In the time code ddd is day of year, hh:mm:ss is time of day, f is
fractions of seconds, and dd/mm/yr is conventional (US) date. In the
Spectracom clock, q is " " for in-synch or "?" for off-synch and zz is the
time zone (always 00 for UT). In the Truetime clock, q is a precision
estimate in milliseconds and coded as follows: "?" +-500 msec, "#" +-50
msec, "*" +-5 msec, "." +-1 msec, <sp> less than 1 msec. In the Heath clock,
the hh:mm:ss.f field is displayed as 0?:??:??.? before synch is first
established and hh:mm:ss.? once synch is established and then lost again.
Model 8170 WWVB Synchronized Clock ($2500)
Spectracom Corporation
101 Despatch Drive
East Rochester, NY 14445
(716) 381-4827
This is a LF receiver and decoder for the WWVB station. It is supported
in the Fuzzball operating system. For expected accuracies less than 10
ms, a special one-second on-time signal produced by the clock must be
used to fine-tune the indication.
Model 468-DC Satellite Synchronized Clock ($4500)
True Time Division, Kinemetrics
3243 Santa Rosa Avenue, Santa Rosa, CA 95401
(707) 528-1230
This is a UHF receiver and decoder for the GOES satellite. It is
supported in the Fuzzball operating system.
Model 1020 Integrated Time Source ($1500)
Precision Standard Time, Inc.
105 Fourier Avenue, Fremont, CA 94539
(415) 656-4447
This is an HF receiver and decoder for the WWV/WWVH stations. A Unix
driver is included in the NTP daemon ntpd; however, Version 4.01 of the
clock firmware is required. It is supported in the Fuzzball operating
system. This receiver is now being marketed by the Traconex
Corporation.
Model GC-1000 Most Accurate Clock ($300)
Heath Company
Benton Harbor, MI 49022
(616) 982-3411
This is an HF receiver and decoder for the WWV/WWVH stations. It is
supported in the Fuzzball operating system.
8101 Automated Computer Time Service (ACTS)
Time and Frequency Division, National Bureau of Standards
US Department of Commerce, Gaithersburg, MD 20899
This is a package of C software for the IBM Personal Computer. It can
be used to call a special NBS number using either 300-Hz or 1200-Hz
modem and set the computer clock. The latest version has been adapted
to run on Unix workstations.
Model ? MSF Synchronized Clock ($?)
European Electric Systems, Ltd.
Woodham Mortimer Place, Rectory Lane
Woodham Mortimer, Maldon, Essex. CM9 6SW, United Kingdom
+44 24541 5785, (FAX) +44 24541 5785
This is a LF receiver for the MSF (Rugby, UK) station. It is reported
they manufacture a US version for the WWVB station.
Model 5061A/B Cesium Clock ($40,000)
Hewlett Packard, Co., Customer Information Center
19310 Pruneridge Avenue, 49A
Cupertino, CA 95014
This is a cesium-stabilized oscillator which produces various outputs,
including a 1 pulse-per-second output which can be used to stabilize
frequency and provide accurate time when used in conjunction with a
radio clock. It is supported in the Fuzzball operating system.erc@khijol.UUCP (Edwin R. Carp) (01/14/90)
In article <1990Jan12.220458.5088@haven.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes: >8101 Automated Computer Time Service (ACTS) >Time and Frequency Division, National Bureau of Standards >US Department of Commerce, Gaithersburg, MD 20899 > This is a package of C software for the IBM Personal Computer. It can > be used to call a special NBS number using either 300-Hz or 1200-Hz > modem and set the computer clock. The latest version has been adapted > to run on Unix workstations. Any ideas on where I can get the software for this? I need the XENIX version (if it's been ported to XENIX) or the UNIX version (I'll port it myself). Any help would be appreciated! -- Ed Carp N7EKG/5 (28.3-28.5) uunet!cs.utexas.edu!khijol!erc Austin, Texas (512) 832-5884 "Good tea. Nice house." - Worf
toto@tank.uchicago.edu (Sandra Jessica Smyth) (01/15/90)
In article <1990Jan12.220458.5088@haven.umd.edu louie@sayshell.umd.edu writes: : This is a package of C software for the IBM Personal Computer. It can : be used to call a special NBS number using either 300-Hz or 1200-Hz : modem and set the computer clock. The latest version has been adapted : to run on Unix workstations. There is also a PD program for the Mac. It call a a 1-804 number, and if done off peak (from Chicago) costs about 15 cents to set the Mac clock. It is small and I can mail it to anyone that has binhex 4.0. -- Sandra Jessica Smyth Believer in lost causes toto@tank.uchicago.edu ...!mimsy!oddjob!tank!toto
marks@mgse.UUCP (Mark Seiffert) (01/15/90)
In article <1013@khijol.UUCP> erc@khijol.UUCP (Edwin R. Carp) writes: >In article <1990Jan12.220458.5088@haven.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes: > >>8101 Automated Computer Time Service (ACTS) >>Time and Frequency Division, National Bureau of Standards >>US Department of Commerce, Gaithersburg, MD 20899 > >> This is a package of C software for the IBM Personal Computer. It can >> be used to call a special NBS number using either 300-Hz or 1200-Hz >> modem and set the computer clock. The latest version has been adapted >> to run on Unix workstations. > >Any ideas on where I can get the software for this? I need the XENIX version >(if it's been ported to XENIX) or the UNIX version (I'll port it myself). This reminded my of something i found on the alphacm bbs a long time ago. I pulled it off of tape and tried it. It is called nbs_time.tar.Z, and includes a C program and a 286 Xenix binary. I tried it, and it works, it dialed into NBS at 1200 baud on port tty1a (the defaults) got the time and set it for me. My system time is now within 4 seconds of that which is displayed on my cable converter box (assuming that is right). If you are interested in getting this program DO NOT reply to this message, instead send the following message to the archive server on my system, archive@mgse, and you will get the uuencoded compressed tar file back along with the system help file. - cut here -------------------------------------------------- set uuencode set compress send other/xenix/nbs_time.tar.Z help - cut here -------------------------------------------------- >Ed Carp N7EKG/5 (28.3-28.5) uunet!cs.utexas.edu!khijol!erc >Austin, Texas (512) 832-5884 "Good tea. Nice house." - Worf -- Mark Seiffert, Metairie, LA. uucp: rex!mgse!marks bitnet: marks%mgse@REX.CS.TULANE.EDU internet: marks%mgse@rex.cs.tulane.edu
ridder@elvira.enet.dec.com (Hans Ridder) (01/16/90)
In article <JEFFC.90Jan11145835@bach.ncr-fc.FtCollins.NCR.com> jeffc@ncr-fc.FtCollins.NCR.com ( Jeff Cook) writes: >In article <1990Jan10.175719.8720@haven.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes: > >> If you're a time freak, like me, you'll really want to invest in a >> WWVB clock. If your not as much of a time freak as Louis, and all you want is to set the time on your computer, you may be satisfied with the dial-up time service provided by NIST. The number is 303-494-4774, 300 or 1200 baud. It was rumored that they would be adding a 900 number, but I've never heard about it. When you connect, it prints a banner then starts dumping time stamps to you (much like the WWVx synced clocks do). If you echo the time stamps back, it computes the round trip delay and (assuming the one-way delay is half) advances it's time stamps to you so they arrive on time. If you type a question mark when you connect, it will send a help text documenting the format, etc. This method doesn't have the long-term accuracy that a WWVx clock has (since you can't stay connected forever), but it's better than the local phone company! >Sounds interesting, but expensive. How much would something like this >cost? Dunno. But the NIST time service would probably be a local call for you (Jeffrey)! It may be what you're looking for. >Jeffrey L. Cook jeffc%ncr-fc@ncr-sd.SanDiego.NCR.com -hans ======================================================================== Hans-Gabriel Ridder Digital Equipment Corporation ridder@elvira.enet.dec.com Customer Support Center ...decwrl!elvira.enet!ridder Colorado Springs, CO
michaelb@mikebat.UUCP (Michael R. Batchelor) (01/16/90)
> > ... call a special NBS number > > ..... and set the computer clock. The latest version has been adapted > > to run on Unix workstations. > Any ideas on where I can get the software for this? I need the XENIX version > (if it's been ported to XENIX) or the UNIX version (I'll port it myself). If you find this don't be selfish with it. I have a clock at work which does the same thing but I must set the computer's clock by watching the wall clock. Please let me automate if I can. I.e. post the source if you can. Tell me where to buy the binary if I must. -- Michael Batchelor / KA7ZNZ uunet!wshb!mikebat!michaelb Ships don't come in; they're built. -- (I don't know who said it.)
wiml@blake.acs.washington.edu (William Lewis) (01/30/90)
In several articles people say things like: >> > ... call a special NBS number >> > ..... and set the computer clock. The latest version has been adapted > >If you find this don't be selfish with it. I have a clock at work >which does the same thing but I must set the computer's clock by >watching the wall clock. Please let me automate if I can. >I.e. post the source if you can. Well, it's fairly simple. There's a modem line somewhere that will do this. I have a SALT script that will do this (SALT is a C-like script language for the Telix terminal program), but it's at home with my PC. I'll try to remember to post it tonight, along with any docs I may find. I'm not familiar with the ins and outs of dialing out under program control under Unix, so someone else will have to port it ... all the modem has to do is 1. call somewhere 2. get a few lines of ASCII 3. mark precisely what the time was when you got the info 4. decode the info into a time stamp 5. set the clock 6. hang up after step 2 =8) --- phelliax the obscure -- wiml@blake.acs.washington.edu (206)526-5885 Seattle, Washington