I read somewhere once that a problem with a utility company in Maine could cause power distribution problems in Florida. This was just given as an example that a problem for a utility company in one state could cause problems for a utility company in another state.

All that said, doesn't all the talk about "early warning" for the western states smack of disconnect? More specifically, just because you may live in the Central or Mountain time zones doesn't necessarily mean that you would have any advance warning in the case of a utility co. failure on the east coast.

Any thoughts to add?

-- dan (dbuchner@fdxsupplychain.com), November 09, 1999

Answers

I believe what you read was from a column by Dick Mills. A problem in Maine wont cause a problem in Florida. There is too much impedance between the two locations for one thing. What Dick was referring too was that a problem in Maine would be sensed by generators in Florida. Theoretically this is true and the generators would respond by increasing or decreasing the amount of power generated. In reality, while there are instruments that can sense the change, there is enough generation between the points and so much system impedance that nothing would happen that someone in a house in Florida would notice.

-- The Engineer (The Engineer@tech.com), November 09, 1999.

My brother-in-law works for a major utility (won't say which one, he'd kick my ass). A few years ago there was a system failure which started in Canada and cascaded down the entire East Coast.....nailed 'em cold.

Several months ago my bro-in-law was in a meeting (VPs' and higher), and the end-result of that meeting was they decided to have every sub- station manned New Years Eve (and day) "just in case". This, they assure themselves, will allow them time to disconnect from the main power grid before it takes them out.

So you see, no problem! Everything's under control....

-- Tuan (Stryder X6@aol.com), November 09, 1999.

Tuan: An outage that took out the entire East Coast would be big news. Care to tell us exactly when this happened other then a few years ago? Ill give you a hint, it didnt. The last big outage was the ice storm that affected Canada and upper New England. That was the result of physical damage and did not take out anything like the entire East Coast.

Second many power companies plan to have people in their substations on New Years Eve. However that would not allow them to disconnect before something happens. Human reaction time is much less then the speed of light. The relays would do the disconnecting. The people are there to make it easier to put it back together.

-- The Engineer (The Engineer@tech.com), November 09, 1999.

Dan,

Some embedded systems run on GMT (London) time - aka Universal Time - which is East+6, Central+7, Mountain+8 and West+9. So systems could show Y2K effects starting at 6:00pm on 12/31/1999 in the eastern USA - 12:00 midnight GMT - simultaneously across the globe.

Tuan: I agree with the Engineer, many power companies will be fully staffed at all substations prior to and after the rollover.

Engineer: Local power company - Cinergy in SW Ohio - is expressing concern over large industrial users with non-interruptible contracts. One big user wanted to go dark at 11:45 and power up at 12:15 - Cinergy convinced them that to do a proper start-up they need miniumum 2 hours - so big user now wants to power off at 11:00 and back on at 1:00am. In light of above about possible GMT timing for rollover - it strikes as something is wrong with this plan. If the plan is to protect vital equipment and the environment from possible damages would it not be wiser to go dark at 5:00pm EST and then power up slowly and incrementally well after rollover say 4:00am EST to allow for any spikes out west?

Do you know how other utilities plan to handle big power users' short term swing in demand?

-- Bill P (porterwn@one.net), November 09, 1999.

Bill: ARGH, The chips do not use GMT time. I wish I could put a stake through that myth. Some equipment is tied to GPS or WWVB or GOES or what ever you are using. The time signals are set to GPS but then you use a receiver to change it to local time. However you can change it to anytime you want. Witness daylight savings time which goes into effect in April and is changed back in October.

I dont blame Cinergy for not wanting that. It would put a heck of a bump on the system. Especially if there were more than one company or plant. That the reason they want them to do it over the period of time. Nothing to do with GMT.

-- The Engineer (The Engineer@tech.com), November 09, 1999.

Bill, Lets see if I've got this right.

According to you, some (or most or all) of the embedded systems in our power stations use GMT. Were they set this this way when they were made, when they were installed, or during maintenance?

As we have some data collection systems running on standard time, and some of our higher level systems running on daylight saving time, can we expect to see a conflict between all of these various time settings causing any problems, and if so, when?

For the embeddeds that had their time set at GMT when they were made, what power source kept the timers active untill they were installed?

What degree of accuracy do they have, and how far out (per year) could we expect the date/time to drift?

Malcolm

-- Malcolm Taylor (taylorm@es.co.nz), November 09, 1999.

Malcolm,

I will have to check through other sources as to embedded systems using GMT instead of local time. If I find the supporting link I'll post under a separate thread.

I respect the Engineer's position and knowledge on this and need to check my sources.

-- Bill P (porterwn@one.net), November 09, 1999.

(The Engineer -- Perhaps I can 'stake' this one for you.)

I believe that Unix systems use GMT as the reference for the 'local' time on their systems. Thus, if one is assuming that there are power company systems controlled by Unix boxes, that is where the GMT idea would come in.

-- just another (another@engineer.com), November 09, 1999.

Just another.

Thanks for the info. What I was trying to get across is that time isnt important. Timing is. There are lines that cross time zones. Georgia Power has a line that goes to Alabama Power. So the one end is in the Eastern Time zone and the other in the Central Time zone. Ditto with lines from AEP that cross several states. My own company has lines that cross time zones. But we set the time at one end of the line to its local time zone and the other end is set to its time zone. Therefore the two ends have an hour difference in the way the clocks are set. They both use GPS time (though in the past we have used other time codes) which is decoded by receivers and set to the local time. So the absolute time isnt all that relevant.

-- The Engineer (The Engineer@tech.com), November 10, 1999.