Precision Standard Time Model 1020 WWV Receiver

© Brooke Clarke, N6GCE 2003 - 2019

Background

Heathkit offered a WWV & WWVH synchronized clock (GC-1000) that was very sophisticated in that it also adjusted the frequency of it's crystal when the signal was strong.  See U.S. patent  4582434 Time corrected, continuously updated clock Apr., 1986 Heath Company for more about this.  The RS-232 capability of the Heath clock was minimal and it had a 1 hour error twice a year when displaying local time because it switched to/from daylight savings time when Colorado switched, not my local time.

The PST Model 1020 is a larger (11 1/2 w x 3 h x 11 d) unit that has more computer interface capability including NTP capability.  The main PCB has a notch that's about 7" x 7" with a daughter board plugged into the main board.  The receiver proper is on the daughter board so maybe they also offered different versions for WWVB or time stations in other countries? It also has built in provision for not only external 12 Volt main power but also for seperate 6 V Gell Cell backup power with trickle charge. The display includes both month & day as well as hh:mm in 1" digits and ss in 3/4" digits.  The PST1020 uses software tracking, rather than crystal frequency adjustment, so that the clock keeps good time during those times when the signal is not strong enough to read the data.

28 Oct 2015 Update Received the following email from David S. Brooke-- ...you write:     "The receiver proper is on the daughter board so maybe they also offered different versions for WWVB or time stations in other countries?" The reason for that design was that the same receiver was used for the "OEM 10" which was later rebranded the Model 1010, if memory serves. What you call the receiver is the analog radio receiver and the digital processor. The other board is mostly logic for the digital display. I wrote the software. David

The PST 1020 has the same bug as the Heathkit GC-100, it changes to/from daylight savings when Colorado changes, not in my time zone.  I have upgraded the ROM chip and as of 31 Octoer 2004 the transition is correct.  The 12 hour time format is illogical.  At midnight the DATE changes to from 30 Oct to 31 Oct, and the time changes from 11:59:59 to 12:00:00, but 12:00:00 is not a further hour on 30 Oct.! It should change to 00:00:00, which is what a 24 hour clock would do.

 A few days ago (today is 9 Apr 2003) the 1020 lost lock and could not re-establish lock, but did after a power down and restart. It did this for two or three days in a row.  But it now seems able to work 24 hours a day which includes changing frequencies as the HF propagation conditions change so maybe the problems of the last few days are related to poor HF propagation.  Note that this problem of loosing lock for a couple of days is the main reason why HF was replaced with low frequency time transfer methods. (powered by an Agilent E3617A bench supply pulling maybe 400 ma that fluctuates with the time being displayed.)

10 Apr 2003 - Connected my house 12 V storage battery to the PST1020 as it's main power and this caused two strange behaviours, (1) every now and then (maybe 30 to 60 minutes apart, but not on a schedule) the sould would be "click" . . . "click"... then a sound like a ratchet being turned in the free direction that lasted for maybe a second (very much like cocking a double action revolver and then spinning the cylinder), and (2) the dispaly would go blank for a second or two then come back with the correct time.  My house HF antenna is fed from a multicoupler and the antenna feeding it has the coax shield grounded outside as it goes through a lightning protector so the coax to the PST1020 is has a grounded shield.  The 12 V storage battery has it's negative terminal connected to a copper ground rod.  The input DC power connection shows -12 above +12 implying a positive ground system.  I used a couple of BNC-5-way adapters with a 0.1 uF cap in series with both the shield and center conductor to DC isolate the antenna connection.  This has not fixed the problem, now the cock&spin noise goes along with the display turning off.

The company name Traconex and/or Multisonics also shows in relation to this receiver.  These are companies that are involved with traffic signal light controllers where time synchronization would be a good thing.

U.S. Traffic Corp now called  Quixote may or may not offer the receivers.

Model 1010 - is the receiver board from a 1020 with external power supply and display
Model 1030 - rackmount version of 1020

U.S. patent 4768178 (Google patents)High precision radio signal controlled continuously updated digital clock August 30, 1988

"Prior efforts have been made to provide a clock signal receiver that receives the radio broadcast signals described above, such as the system disclosed in U.S. Pat. No. 4582434. However, the system in that patent is subject to locking onto a noise-heavy signal rather than the desired signal; this is a major fault because the NBS (now NIST) signal transmissions are subject to serious noise problems. Further, the '434 patent lacks a sufficiently reliable method of verifying accurate data reception."

When I connected the 1020 to my house HF antenna system and power it locked up within a couple of minutes.  Much faster than the Heathkit clock that would take a long time.

Like the Heathkit, the PST uses DIP switches for setting options, but unlike the Heathkit the PST switches have no labels making setting them much more difficult.  It was displaying Eastern Time not Pacific Time.

U114 has a paper sticker labeled: 008-0002-000, V03.06.000, 0788.

The PTS1020 makes a ticking sound each second.  This is probably due to a now dry capacitor.  I've disconnected the speaker.

7 March 2005 - I've been using this clock to mark exactly noon in order to make a ceiling sundial.   On March 6 this clock displayed 12:00:00 but was a full minute fast.  This is very apparent by the distance of my mark from the line of other "noon marks".  So this clock is not 100% reliable, even with the latest PROMs.

Bad Caps

DIP Switches

From Tim K4RA:

DIP Switch Bank #2 is the time zone bank.  I’m not sure what switch #1 and 3 do but I think one of them is a DST correction.  Switch #2 is the 12/24 hour time display format.  Here is what I figured out for time zones:
 
Atlantic Time = Switches 1, 3 & 6 set to ON
Eastern Time = Switches 1, 3, 6 & 8 set to ON
Central Time = Switches 1, 3, 6 & 7 set to ON
Mountain Time = Switches 1, 3, 6, 7 & 8 set to ON
Pacific Time = Switches 1, 3, & 5 set to ON
Alaska Time = Switches 1, 3, 5 & 8
 
Given this progression, it would be pretty easy to use the binary code to get other time zone corrections.

Data Format

The following data is from the Network Time Protocol (NTP) web page describing some radio clock data formats.

PST/Traconex 1020 Time Source (WWV/H) (firmware revision V4.01)
WWV Broadcast Format
WWVH Broadcast Format
WWV/H Timecode Format
"ahh:mm:ss.fffs<cr>" "yy/dd/mm/ddd<cr>" "frdzycchhSSFTttttuuxx<cr>"
  T            C                D                   A   Q

     poll: "QTQDQM"; offsets: T = 1, D = 24, Q = 41, A = 37
     a = AM/PM indicator (" " for 24-hour mode)
     hh:mm:ss.fff = hours, minutes, seconds, milliseconds
     s = daylight-saving indicator (" " for 24-hour mode)
     yy = year (from DIPswitches)
     dd/mm/ddd = day of month, month, day of year
     f = frequency enable (O = all frequencies enabled)
     r = baud rate (3 = 1200, 6 = 9600)
     d = features indicator (@ = month/day display enabled)
     z = time zone (0 = UTC)
     y = year (5 = 91)
     cc = WWV propagation delay (52 = 22 ms)
     hh = WWVH propagation delay (81 = 33 ms)
     SS = status (80 or 82 = operating correctly)
     F = current receive frequency (4 = 15 MHz)
     T = transmitter (C = WWV, H = WWVH)
     tttt = time since last update (0000 = minutes)
     uu = flush character (03 = ^c)
     xx = 94 (unknown) (firmware revision X4.01.999 only)

     note: alarm condition is indicated by other than "8" at A, which occurs during initial synchronization and when received signal is
     lost for an extended period unlock condition is indicated by other than "0000" in the tttt subfield at Q

     example: " 15:36:43.640  91/08/04/216 O3@055281824C00000394"

Links

Patents related to Disciplined  Oscillators -