Path: utzoo!utgpu!trantor.umd.edu!ntp-relay Message-ID: <9002271735.AA03948@trantor.umd.edu> To: ntp@trantor.umd.edu Cc: winter@apple.com, clements@BBN.COM Subject: More on WWV/WWVH timecode generators Date: Tue, 27 Feb 90 12:06:39 EST From: clements@BBN.COM Newsgroups: list.ntp Distribution: ut Approved: devnull@gpu.utcs.toronto.edu My call for interest in building the new WWV/WWVH time code generators has produced a total of one other enthusiast from the bushes. He's on the left coast and I'm on the right coast, but we're talking about it. I got the spec from the procurement office. It has 58 pages of legalistic boiler plate, with many more pages included by reference, and 15 pages of technical spec. This has convinced me that it would be silly to try to form the "Friends of WWV/WWVH, Inc." that I suggested to build the TCGs. The legalisms are far too intimidating for me. But we might just build them anyway as a gift to the government if there is no bidder that they can afford. It turns out the deadline for bidders to reply is this Thursday. Anyhow, here is a portion of the spec of direct relevance to NTP-ers. It describes how the WWV/WWVH time code will be enhanced to deal better with Daylight Saving Time and with Leap Seconds. FYI. [Excerpt from spec, Attachment 1 to Department of Commerce Solicitation number 52RANB00C059, issued 1-23-90] Attachment 1 Page 1 of 15 Revised 11/22/89 SPECIFICATIONS TIME CODE GENERATOR/PROGRAMMER 1.0 Scope This specification establishes the requirements for two time code generator/programmer systems that will be used to generate precision timing signals for use by NIST standard time and frequency stations WWV and WWVH. Each of these systems will consist of: (1) Three (3) time code generator/programmer units, and (2) Software and interfacing hardware to provide for entering commands to the time code generators through a computer. [...] 2.5.6 Time Code Generator - BCD Format Table Second Data 00 low subcarrier 01 0 02 0 03 0 04 0 05 0 06 0 07 0 08 0 09 position marker 1 10 minute 1 11 minute 2 12 minute 4 13 minute 8 14 0 15 minute 10 16 minute 20 17 minute 40 18 0 19 position marker 2 20 hour 1 21 hour 2 22 hour 4 23 hour 8 24 0 25 hour 10 26 hour 20 27 0 28 0 29 position marker 3 30 day 1 31 day 2 32 day 4 33 day 8 34 0 35 day 10 36 day 20 37 day 40 38 day 80 39 position marker 4 40 day 100 41 day 200 42 0 43 (new) leap second warning 44 (new) Daylight Saving Time indicator #2 45 (new) year 1 46 (new) year 2 47 (new) year 4 48 (new) year 8 49 position marker 5 50 UT1 sign 51 (new) year 10 52 (new) year 20 53 (new) year 40 54 (new) year 80 55 Daylight Saving Time indicator #1 56 UT1 .1 57 UT1 .2 58 UT1 .4 59 position marker 0 Note: Daylight Saving Time indicator bits #1 and #2 are scheduled for change by the operator. The normal procedure is for DST indicator bit #1 to change at 0000 UTC on the day of a change to or from Daylight Saving Time and DST indicator bit #2 to change in the same way 24 hours later at 0000 UTC on the day after the time change. When both bits are set to "0", standard time is in effect and it is not a day of time change; when both bits are set to "1", Daylight Saving Time is in effect and it is not a day of time change. Each receiver can then interpret these bits unambiguously so as to change onto or off of Daylight Saving Time at 2:00 AM receiver local time. Note: Leap seconds are always inserted/deleted at the last second of the UTC month. Leap second warning is scheduled to be turned on at any time during the month to which a leap second is being added, and is automatically turned off immediately after the leap second. The UT1 bits also automatically change after a leap second. Incremental UT1 changes may be scheduled by the operator at any time. [End of excerpt from spec.] Again, the above is the spec for the NEW generators, not what exists today. Today the DUT1 stuff doesn't change automatically, for example. And there is no leap second warning or year info. Not clear how a single leap second warning bit helps you when the leap second is negative. Does anyone know whether negative leap seconds are expected to be needed in the next century or so? /Rcc Bob Clements, K1BC, clements@bbn.com