If you don't have a GPS/NTP master clock, you might want to get one...

[jhardis]

Saw this lovely news story tonight.

Since the reporting on this was non-technical, let me fill in some of the gaps.

Current practice is to manage the grid for a target a frequency of 60.00 Hz. However, due to continually changing loads and other factors, the instantaneous frequency is not constant. It is, however, generally held within the bounds of 59.95 Hz to 60.05 Hz. Over time, on average, the effective frequency is a smidge higher than 60.00 Hz, so synchronous clocks tend to run fast.

To correct this, under present practice, once the accumulated time error reaches several seconds, the grid operators change the target frequency from 60.00 Hz to 59.98 Hz in order to back out the accumulated time error. For the Eastern Interconnect, a control center in Indiana, near Indianapolis, sends out orders for all power plants within the Interconnect -- from Florida in the southeast to Saskatchewan in the Northwest -- to operate at 59.98 Hz for some period of time. Of course, they all must do this in sync. (10 seconds of time error can be backed out at 59.98 Hz in about 8 hours. The procedure is different and more automated on the Western Interconnect.)

The proposal changes nothing about operations when the target frequency is 60.00 Hz. It only ends the practice of switching to a target frequency of 59.98 Hz when the accumulated time error gets big enough. Under the proposal, the grid would stay at a target frequency of 60.00 Hz all the time, and the time error would continue to accumulate indefinitely.

The procedure for deliberately running slow at 59.98 Hz is called "Time Error Correction" (TEC). In one form or another, the industry has been doing this since the 1930s, originally as an inducement that people should buy electric clocks (many of which were made by GE).

The Federal Energy Regulatory Commission (FERC) is charged with regulating the _reliability_ of the interstate power grid. The industry maintains that this is a "business practice" issue, not a "reliability" issue. That is, "how" TECs are performed is "reliability." Whether or not TECs are performed is a "voluntary business practice." Nonetheless, there is currently a reliability standard (regulation), BAL-004, that might be read to imply a requirement for TECs. The industry has petitioned FERC to amend this standard. That proceeding is currently pending.

For the complete record on the matter before the FERC, start at http://elibrary.ferc.gov/idmws/docket_search.asp and search for all filings on Docket Number RM09-13 . The nine documents on the docket provide a lot of information about the industry's position.

Additional information can also be found on the NERC website, under "TEC Elimination" http://www.nerc.com/page.php?cid=6|386 . Look at the Webinars.

- Jonathan

 

[OKCRadioGuy]

Thank you Jonathan for the REAL story opposed the MSNBC fluff/misinformation piece. I see the industries point and I really now think they are right in trying to elimnate the time correction practice. Trying to target 60hz and forgetting it would make the most sense for the modern world IMHO.

 

[ScottJ]

This strikes me as a non-issue, really. Contrary to what the story reported, very few things these days have clocks that depend on line frequency. Most clocks (for example, the clocks mentioned in the story that flash 12:00 when powered up) have independent quartz timebases completely unrelated to the AC line frequency.

Unless you're using old Telechrons in critical applications (and if you are, you're taking nostalgia too far!), I doubt there's any need to invest in GPS timekeeping technology ... unless you're looking for a way to justify the expense, in which case, try it and see if it flies.

 

[jhardis]

Most clocks (for example, the clocks mentioned in the story that flash 12:00 when powered up) have independent quartz timebases completely unrelated to the AC line frequency.

Don't conclude this too quickly. Traditionally, AC line frequency has been an even more stable reference than quartz crystals.

Over the weekend, as an experiment, I bought a modern clock-radio at K-Mart. I comes with a battery backup, and has the correct time out-of-the-box. An on-board microprocessor makes DST corrections. Yet it STILL relies on the power-line frequency when it's plugged in, according to the description in its patent (6,567,344).

 

[kenglish]

I believe that all of our Leitch clocks use the power line frequency to drive them, and an hourly "re-sync" to keep them all in step.
I wonder how far off they'll be in 60 minutes.

 

[TomZ]

If you have a 100:1 transformer you could record the 'audio' and take readings of the frequency(ies).

Amplitude, too.

 

[Tom Wells]

I'm becoming more disturbed as I consider this.

A couple of years back, I had been actively wondering who, where, controlled and directed this process of
acculated error correction. I knew it had to be something about as was described. I had been admiring this aspect
of electricity for quite a long time, and really wondered how the correction was directed.
Most deregulation is only about money, and I'm getting disgusted at the dropping of standards and expectations.

This is one too many, making me think a tragic disconnect is
happening as continuing generations lose all track of why things develop as they do.

Humans have been trying to "keep time" and do so ever more accurately ever since they have had enough "time" to think about
it all.

The development of a precision, ubiquitous time standard available to all, that requires NO maintenance or concern
on the part of the user was, and is, one of the highest advancements we've taken advantage of.

I fully expect this is just the beginning of relaxation of standards.
Harmonic power components will likely next be permitted to rise, and reactive power will be allowed to drop.
Every relaxtion can only permit cheaper operation of generating facilities, as long as we
can do away with such niceties as maintaining and correcting the time base.

I can think of quite a few drives, motors etc in the plant where I work that use 60 for establishing
ref for what becomes actuator displacement. Most won't see enough error to matter.
I don't expect any perceptible problem to turn up anywhere, except that the most
reliable clocks in the world are now not going to be.

I'll argue that on that basis, it IS a reliability issue. There are a LOT of those "big circle" multi-event timers out
there in this country, and similar traffic light systems.
The whole solar system/planet/sphere/rotation/360 degrees/wheels/clocks/time/timing/time base/frequency thing
is about the 360 degrees and subdivisions. They must add up to the "proper" number of cycles per day,
or the whole notion of controlled power is just a sham.

I have seen and replaced quartz clocks that were worthless new, and I see 70 year electric clocks that go for years or months
and see the second hand still hits the 12 as WGN's top-of-hour tones sound if that's how I set it.

Anytime someone screws with time it's for often for profit or unfair advantage.
Part of my sensitivity comes from working for a company that's on a 7-hour behind offset.
What's that, you say?

It's where the actual day doesn't start until 7 AM in a 24 hour operation.
So if the plant is actually closed "on" July 4th, that only means from 7 AM to the next 7 AM.
So I come in to work at 11 PM on the 3rd, and work until 7 AM on the 4th of July the company can still claim
I had the holdiay off, even though they got the majority of the shift covered on the actual holiday.

Slick, huh? Anytime somebody plays with time it seems to be to my detriment, so I'm not going
beleive this will be anything more than additional headaches somehow for me.

 

[TomZ]

Nice Rant, Tom

Cue old Chicago Transit Authority album

"Does anybody really know what time it is?"

 

[secondchoice]

Is it possible do 'dark start' the grid?

The “grid” would take days to restart. “Cold” starting any large commercial power plant (coal, natural gas, nuclear) is almost impossible. You need power to control which ever fuel source the boiler is using. Some natural gas turbines (like the ones used for back up power at the phone office) still require at least a battery powered circuit to ignite and control the combustion. IIRC most windmill generators hooked up to the grid will not generate electric (to protect the line crews and anyone on the circuit) if there is an interruption. I have seen some old hydro units that have an “exciter” attached to the top of the generator. In theory someone could manually open the penstock valve start the turbine spinning and very skillfully control the current in the field of the generator and run the unit at a very low power assuming there was some way to “load” up the generator to keep the automatic “self protection” from tripping the unit due to lack of load or “seeing” an open. IIRC During the infamous New York / Northeast power outage of the 1960’s there was a small college (I do not recall the name) that had a small electrical generation plant that at the time could be started by “hand” (a couple of strong men could spin the turbine / generator and start the unit). Then that plants power was used to start another power plant, which started another until there were enough power plants running to restore parts of the grind. These large plants can not be “turned” on an off like a car engine. Some of the large units can take several hours to a day to “get to full power”.

 

[satech]

So NERC says ending the time corrections would cause AC line-driven clocks to skew several minutes fast over a month's time. But wouldn't that also cause electric meters to run fast as well, and thus overcharge the customer? I assume they are built and certified to measure your kWh usage based on a dead-accurate 60.000... Hz AC line frequency.

 

[Nostalgia]

Is it possible do 'dark start' the grid?

The “grid” would take days to restart. “Cold” starting any large commercial power plant (coal, natural gas, nuclear) is almost impossible. You need power to control which ever fuel source the boiler is using.

I was told that following a hurricane in Hawaii that knocked out all power on Kauai that a nuclear submarine from Pearl Harbor had to be floated into their harbor and wiring from their nuke power plant hooked up to the local grid (each island generates its own electricity, no interconnections for obvious reasons) to "jump start" their generating system. Truth or urban legend? Cool story though.

 

[PTBoardOp94]

Probably urban legend. A diesel generator would do the job much more cheaply.

 

[jhardis]

I was told that following a hurricane in Hawaii that knocked out all power on Kauai that a nuclear submarine from Pearl Harbor had to be floated into their harbor and wiring from their nuke power plant hooked up to the local grid (each island generates its own electricity, no interconnections for obvious reasons) to "jump start" their generating system. Truth or urban legend?

It's almost certainly an urban legend. Submarines use 400 Hz systems, and I can't imagine what hooking up jumper cables to a submarine would be like.