I was just resetting the clocks in my house, and it brought back a memory.
When I was a technician, the ESS switches each kept their own time, and on occasion, they'd be misset and the programming staff would have to run special correction jobs to change all the tape records that had been created during the interval the clock was incorrect.
Is this still the case, or are the C.O.'s synched to an NBS-traceable source now?
Bill Horne Temporary Moderator
(Filter QRM from my address for direct replies)
(Please put "[Telecom]" (without the quotes, but _with_ the brackets) in your subject line, or I might never see your post.)
Time division switches have to have a very accurate clock. Aren't they all tied to some government time source?
***** Moderator's Note *****
Good point: of course, TDM switches need a very accurate _timing source_, but that's a separate issue. Verizon timing is mostly dervied from GPS these days, and I assume other ILECs are using it too.
The question, though, is if the _time of day_ settings in the COs are being derived from a common source, or if they're still being set locally.
From what I can remember they use a common source, at least the GTD5 did in California, but that was 11 years ago, though I don't think much has changed. I remember when we were installing one someone got the 1 and 0 source reversed on several offices, during a precut I was having problems testing so I followed a cable and found the error.
What do they use to control the dial-in time recording-- (215-846-1212)? Do other cities still have this phoneco provided service? (Note--846 was legacy TIme).
Along these lines, how many cities still have phoneco provided weather (215-936-1212)? For a while, they were standardizing 936-1212 as the weather number (936 is legacy WEather). I remember an ad insert in the phone bill touting the weather number of in different cities, encouraging vacationers to call ahead so they could pack accordingly and be prepared.
***** Moderator's Note *****
Another good question, and again I don't know the answer. The Time annunciator in Boston used to be at Bowdoin square, but I don't know how, or even if, it was synched to any higher stratum in the time world. They might have used a service like Western Union's clock synchronizaiton, but that's just a guess: I left there aroud 1979, so I assume it's all computerized now.
The weather number is still going strong, although in Boston it's been changed to an "Accuweather" announcer, so I presume it's been subcontracter out to the jock-in-a-box at Clear Channel in Cleveland.
I doubt anyone calls for the weather in destination cities anymore: the Internet makes it a lot easier to get that information, and many travel sights supply it during the reservation process.
Bill Horne Temporary Moderator
(Please put "[Telecom]" (without the quotes, but _with_ the brackets) in your subject line, or I may never see your post.)
Why wouldn't they have a routine to automatically convert the GPS UTC to local time?
Also, using GPS, the switch always knows where it is. ;-)
***** Moderator's Note *****
I don't know if they do, but _if_ they do, it would probably be from a different source than the GPS or Stratum 1/2 clocks that provide digital pulse timing. The software inside the switches is proprietary to the vendors, and I'm asking if it was ever modified to use GPS or other time references.
This fall my Razr V3xx from Cingular/at&t (AC 630) set itself back _2_ hours. My guess is that the phone received an update from the switch, then executed some internal command and set itself back a second time.
In the "Initial Setup" menu, "Time and Date" is set to "Autoupdate Time & Timezone" FWIW.
I swear I did nothing but plug it into its charger overnight as I always do. The next morning, the alarm failed to go off at 8am. It did go off at 9am, but the phone display said it was 8am. (So I guess in its own universe it DID go off at 8 :-)
I normally leave the phone on all the time except that I always turn it off for church. After the service ended (I was late :-) at 10 I turned the phone back on and it said 9am -- I turned it to show somebody and they said, "No, it's right." And I looked again and sure enough, it now said 10am. So apparently rebooting the phone forced it to update its clock properly.
My Verizon cell phone time display changed on its own. It keeps accurate time. I don't think I even have the ability to manually set the time myself.
My newer electronic devices previously automatically changed the time for DST and back. They failed to work properly after the dates were changed, but fortunately, they had an option to turn off the automatic change; so I do it myself.
I think it was foolish to change the dates. There was an article stating the candy makers lobbied for the fall change.
Your Verizon phone wouldn't work if its clock were not synchronized to within several microseconds of UTC-GPS. Getting correct local time may be slightly more difficult. At the office, I have a timecode receiver that uses Verizon's signal to within about seven microseconds of UTC. (It's actually on the same floor as the ancient Sun workstation that Telecom Digest runs on, but in a different part of the building.)
More accurately, "your Verizon [or other CDMA] phone won't work if its clock is not synchronized to within several microseconds of the time kept at the service provider."
I have been on isolated CDMA networks where the time had no relationship to whatever was on UTC-GPS.
The reason why CDMA time generally is UTC-GPS time is because when there are multiple CDMA networks, there are problems if a phone is handed off to a tower from a different provider with a different time.
But when CDMA consists of isolated islands (in the case I am thinking of, rural villages 50-100 miles apart), you sometimes will see CDMA time set more or less randomly. I've observed one hour and 23 minutes difference.
-- Mark --
formatting link
does not emerge from voting, party politics, or public debate. Si vis pacem, para bellum.
I presume my cell phone will do that if I go to a different time zone.
The other devices do not have an upgrade capability.
The change kept Halloween within DST, so daylight lasted longer than it did under std time. This made it easier for trick-or-treaters to make their rounds and make their parents happy. Thus more candy sales.
When I was a kid Halloween was a simple holiday observed by little kids. Now it seems every TV show has to have a themed episode about it and throughout October people talk about it, culminating in some big party, as if it were Christmas or Thanksgiving. Some adults come into work dressed in full costume which looks ridiculous and unprofessional. (A simple clothes accessory I could tolerate).
Even then the CDMA pilot is locked to a traceable reference signal, even more likely to be GPS in the sticks than any other sort of PRS.
All towers must be in sync with each other for proper soft handoff. I suppose it's theoretically possible to run the towers in freerun mode, but that would mean a hard handoff to every channel, and CDMA would operate very, very suboptimally. You'd also have to accept the unlikely reality that the MTSO is running isochronous in one way or another, as ultimately they need to interconnect to some other provider who IS running in sync to a PRS. And this basically means clock slips - be it at the sonet level, somewhere in the switching systems, or on the T1s running to the cell sites, or even more unlikely, in the tower equipment itself.
Overall I find it highly unlikely that that network was not running in sync. (And I'd find it highly concerning, as the PRS for my switch site is a TimeSource 2700 - which uses CDMA pilot signals from any in service provider it can hear, as well as an internal rubidium holdover oscillator - and that's used to clock our switches as well as our SONET gear.).
Towers? In the CDMA network that I am thinking about, there is a *tower*, and there is no handoff to any other tower because there is no other tower to hand off to. You have to drive 50 or so miles before you get to the next place that there is a tower.
In the CDMA network that I am thinking of, the only interconnect is to landline, and that landline is also isolated to a single village and connects to the rest of the world via microwave. This is *old* technology. Are you sure that microwave links sync time with GPS?
All I know is that the time reported by the phone was 1 hour and 23 minutes later than the actual local time as reported by GPS.
1 hour I could understand, given timezone offsets. But 23 minutes? And IIRC it didn't change the minute at the right time either.
-- Mark --
formatting link
does not emerge from voting, party politics, or public debate. Si vis pacem, para bellum.
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.