“Experiment” on the US power grid will change the way some clocks and other equipment function.
Story submitted by Joe Ryan
The AP has released an “exclusive” story concerning the nationwide “experiment” that will be conducted on the US power grid. The experiment will relieve the power providers from the duty of regulating the frequency of power on the line.
Normally the power stations condition their power to a frequency of 60 cycles a second, a frequency that many old clocks use to maintain their time. With the new standard, or lack of standard, these clocks will stop keeping time properly.
But the problem is more than that.
First, we have this gem from Joe McLelland who heads the Federal Energy Regulatory Commission (from AP article):
“Is anyone using the grid to keep track of time?” McClelland said. “Let’s see if anyone complains if we eliminate it.”
… forgive me for not getting warm fuzzies from this. Likewise, Demetrios Matsakis, head of the time service department at the U.S. Naval Observatory, had this to say (AP Article again):
“A lot of people are going to have things break and they’re not going to know why,”
So, we have what appears to be an untested, for the hell of it, “experimental” major change to the US electrical grid coming in a few weeks and those in charge aren’t really sure how it will work or if it may break something?
Not only is this what a LAB is for, but it is also something that the Federal Government should be TELLING people about in advance, and not in an AP “exclusive” press release.
This reminds me of the hype over the Y2K “issue”
Althos says:
Spinning reserve costs money, the generators have to be coupled to the system but not be actively producing power…
Spinning reserve has three functions.
First there is fast spinning reserve which must react very quickly to arrest the fall in frequency to levels that will cause plant to trip. It is not required to tremain in place for more than a few seconds to a minute. It is generally provided by plant that is connected to the system and generating, but not at full load. Conventional thermal stations and hydro stations are great at providing this. Even with a relaxation in the frequency keeping standards this fast PLSR (Partly Loaded Spinning Reserve) will still be needed to prevent cascade failures.
Second is Sustained PLSR. This is supplied by plant that is connected to the system and generating but not at full looad, and is slower to react than Thermal or Hydro. CCGTS often come into this category, as do hydro plant on Tail Water Depressed or Synchronous Condensor mode. This is probably the area where savings can be made.
Third is Standby reserve. This is plant that is available, but not spinning. It must be plant than can be started quickly like hydro or Gas Turbine plant. This type of reserve will still be needed.
I do hope they have done a full cost benefit analysis of Murphy’s Law, which seems to be prevalent in all things connected to green. The laws of unintended consequences are pussy when compared to Murphy’s Law.
Bet you don’t notice the difference!
The only computers I have that use 60 Hz from the power line for their clocks are my PDP-11’s and my Commodore 64’s. I doubt that many C64’s are now used in process control, but one can still buy PDP-11 CPU’s from non-DEC sources (well actually DEC no longer exists) and I suspect there are a good number of them hidden away in systems where they have been working flawlessly for decades. I guess we’ll find out soon enough where some of these ancient PDP-11’s are being used that employ a power line 60 Hz frequency standard. Most of the applications I wrote for the PDP11 used a clock board for the sub-msec precision timing I needed but I remember being amazed at how good the power line frequency was at keeping time in the 1980’s.
It was this type of ‘I wonder what will happen if—‘ thinking that caused the Chernobyl problem. I hope that the American nuclear stations are not frequency sensitive.
But why would they do it?
Here’s something for you to chew on:
I’ve been thinking about this, and if the reason given – only a money saving issue – why was it not shouted from the rooftops, and why has there been no notification in our monthly statements? This doesn’t add up.
Given all the concern about Cyberwar lately (Stuxnet, etc ), I wonder if there is any connection to that? I’m retired military, and one of the things that is commonly done is to test systems for defensive capability: What is the effect of such and such kind of attack on our systems and how do we defend against it?
Now that’s what I call good thinking!
Deja Vu. In a faraway land, a long time ago, the aroma of an open air fresh meat vendor(1-2 day) displayed on wood planks created a lasting sensory memory.
Smells like that hired gunfighter that sits at the round table with his double barreled blue ink pen – Cass Sunstein the regulatory tsar. A nudge here, a nudge there. Cattle prod even when the cattle are moving along just fine.
OLD clocks don’t have AC power cords. They have mainsprings and a key to wind them.
NEW clocks may or may not have power cords but they do not rely on AC line frequency but rather have quartz crystal oscillators to establish a time base.
SOME clocks that are neither old nor new use AC line frequency as a timebase. These are about as important to keep working properly as rotary dial telephones.
bob sykes says:
June 25, 2011 at 3:18 pm
“Could this actually damage electronic devices?”
No.
“Should I disconnect all my devices from electrical outlets during the test?”
No.
They aren’t talking about radical changes to line frequency. They are talking about not making fine adjustments so that the total number of cycles each day adds up to precisely the right number. Line frequency varies by a tiny fraction from moment to momentfor routine and largely unavoidable reasons. Power companies monitor the variation and then purposely cause variations the serve to balance the net variation to zero each day. Nothing but a small and diminishing number of clocks will be effected and those would only be effected such that they might randomly gain or lose a few seconds each day. Even then they won’t be any less accurate over longer periods of time because on average the days where they lose a few seconds will be nullified by days when they gain a few seconds.
Talk about making a mountain of a molehill… this is a total non-concern that isn’t worth a moment’s time, trouble, or effort to warn or prepare the public for it.
My understanding of spinning reserve is slightly different. The reserve is the difference between the rated capacity of a power generation unit and the load it is actually operating at. It is spinning because the unit is operating and on-line. Thus, a 600 MW unit operating at 450MW has 150 MW in spinning reserve. It doesn’t take very long for the plant to increase the load to 600 because the boiler and steam turbine are already in operation, compared to bringing a unit up from cold shutdown. Some units can be shut down, but keep a fire in the boiler so it doesn’t get cold, and then come back up pretty quickly when they are needed. At least this was my experience when I worked in a power plant 30 years ago.
Nonegatives says:
June 25, 2011 at 11:57 pm
“This reminds me of the hype over the Y2K “issue””
On a tiny scale, yes. 1998-2000 saw a huge boom in computer sales as companies and individuals rushed into early replacement. It also resulted in a computer sales depression right afterward. The road to Y2K was a road to fortune for some of us as we invested heavily in the benefactors (Microsoft, Intel, Dell, Compaq, etc.) before the boom years and sold off before the downturn in sales happened after Y2K.
The chance of repeating that windfall by investing in clock companies is nil.
starzmom says:
June 26, 2011 at 7:19 am
They have some pretty fancy predictive software now they didn’t have 30 years ago that takes all sorts of factors into account to predict demand and schedule production capacity accordingly days and weeks in advance. Big bucks are involved in making substantial improvements in load prediction and scheduling so it gets a lot of attention by really smart people. It’s sort of like communications companies sparing no effort in software development to squeeze more data through the same transmission hardware. Tiny improvements have disproportionately large payoffs.
Anyhow, with the improved prediction in demand less spinning reserve is scheduled. The less spinning reserve that’s needed and the more profitable the power plant. When prediction fails a prime example is what happened in Texas this past winter. Weather forecasts were predicting much warmer low temperatures and in actuality record low temperatures across a lot of the state happened. There wasn’t enough spinning reserve and simutaneously a couple of natural gas fired plants fell offline due to freeze damage. These plants supplied the power to some electrical booster pumps along natural gas pipelines. Simultaneous with that was increased demand for natural gas heating in businesses and residences which heat with gas instead of electricity. This resulted in a large pressure drop in some gas supply pipelines which forced even more gas fired plants to reduce production due to insufficient fuel. It cascaded fast and planned rolling blackouts for as long as it took to get non-spinning capacity back online. Several hours as I recall.
Frequency modulation will just be a symptom of the intermittant nature of wind and solar. They are using this as an excuse to not have sufficient standby ready for when the wind stops blowing and the sun stops shineing. Load response is the issue. Frequency modulation is a way to respond to fluctuating loads and supply. Frequency modulation also helps ensure that voltage remains constant. Each is dependant on the other. Personally I don’t give a **** whether my clock loses or gains time. I am concerned about the effects this MAY have on industrial equipment if it is not regulated properly. As I said earlier I suspect I am already seeing the results of adding more wind and solar onto the grid in Ontario and it ain’t pretty.
um –
1stly – David Thomson says:
June 25, 2011 at 5:57 pm 3 phase – all the phases are generated by the same generator – they are not separate; so all 3 phases are as solidly ties together as the guts of the machinary. So no problems there; each phase will remain + or – 120.0000….. degrees different from the other two.
I hope someone will get you clarification; as the potential damage is a lot more subtle and invidious than any so far discussed.
For instance; I was helping some plant in Vermont on Friday set up a control to maintain 80 deg C water temperature; using a thing called a PID loop (or three term controller) to control some electronic switching on the heater. A PID uses three variables to maintain control; all of which run in the time domain. In some plants this control is sensitive to millisecond changes in execution time over a 1 second base – so it will be sensitive to small changes in frequency.
OK – so what ? Well with a heating control; that is optimised for financial efficiency – change the 60Hz and that’s no longer working efficiently. If it’s controlling via relays those relays are going to work harder for frequency drops as the PID will become less responsive – potentially reducing service life as well.
Um; I can think of an abattoir in Scandinavia where the chiller cabinets are cascaded (10 cabinets- all joined up to make a single run). No doubt you have similar set ups in the US – that installation was horribly sensitive to tiny changes in the time domain (the previous error gets multiplied in each loop) – anything unstable in the time domain is going to cause vast amounts of problems.
This is why the VFD (variable frequency drives) are not running so well – the control algorythm is shot if you are using PI or PID control (think of it this way; PID works in the time domain – change of the working frequency of the controlled device is the same as a change of the executions per second of the PID equation; therefore all terms (PI or PID) will need re-tuning at the new frequency).
I think what you need to do is demand a full and public disclosure about what laxity is being planned – and how long for; and what are the long term requirements to maintain 60Hz average.
In the meantime – note to self – warn all the engineers in employer’s US tech support and global tech support to consider making US customers monitor line frequencies as part of any investigation to VFD problems; solid state AC switch problems and complaints about useless PID controllers …..
We are going BACKWARDS –
no more super sonic flight
no more space flight (sorry; earth orbit mini habitat is NOT real space flight)
no more Moon flights
soon – no more dependable mains supply
soon – no reliable mains supply
soon – no 20th century infrastructure leading to the return of the middle ages with high child mortality; short and brutal lives for all bar the elite (and guess who they are going to be…)
Don’t worry about keeping time. It seems most of our alarm clocks these days are made in China and use a 555 rc multivibrator to generate the clock pulses (or its equivalent on a chip along with the rest of the junk). I have one that gains 20 minutes a day. No way to rubber the r or the c either.
As an electrical engineer, I can understand this: this is crazy. If it is an enviro inspired “regulation” to enable the hydro distribution to cater for more unreliable electricity production (wind and solar) then it will be counter-productive as the various machines on the supply will operate more inefficiently, nullifying any imaginary gain due to “renewable” energy.
If there is no enivro-plot behind this, then it is simply stupid.
Showing my age, but I remember during and post WW II when the power system was struggling to keep up in Sydney, electric clocks had a very bad reputation. It was so bad that the hobby magazine of the time followed its construction project for a synchronous clock with a project for a pendulum electric clock. Now, clocks are in plain sight, and get reset if the error is too much. But time switches are installed just because they can work unattended, so they can accumulate errors over months and be a long way off before they get noticed and corrected. I was shopping last week, and noticed that the synchronous time switches had been replaced on display with quartz versions. But there will be a lot of synchronous time switches still in service, and probably some large clocks in public places that will suffer. Keep an eye open for radio masts and TV towers that don’t switch their obstruction lights at the right time, or have strobes at day level (20,000 candelas) at night, when they should be at night level (2,000 candelas). That will annoy the neighbors, to say nothing of the pilots!
Dave Springer is right, mountain out of molehill. The frequency variations during the day won’t change, just the make-up cycles late at night. During the day, under heavy load, the line frequency usually drops a bit, 59.7-59.9 Hertz. At night, it usually runs high to make up for it. Which probably costs a few cents more, spinning those generators a tad faster. So the utilities will raise your rates to make up for their slightly reduced costs.
What I want to know is how they swap power with other utilities. Phase shifting, at the same frequency is no biggie, but how to deal with frequency shift? Converting to DC and back to AC could work, very expensive.
I still have a few clocks that use the 60Hz line for a reference, long term far better than any cheap quartz oscillator. Maybe I’ll have to ‘pipe’ the 10 MHz Rubidium standard all over the house after this. (right after I make the final tweaks on my perpetual motion machine.)
Much adu about nothing.
Robert of Ottawa
Have you noticed any unexplained failures in the last few years? Does Ottawa get power from Hydro One, I assume it does so let me rephrase, does power get fed through from Hydro Quebec which isn’t as subject to the variances of wind and solar?
A nice description of an experiment conducted in the Netherlands in 2005 (with some background):
http://wwwhome.cs.utwente.nl/~ptdeboer/misc/mains.html
More or less the same from the US, done in 2004:
http://www.leapsecond.com/pages/mains/
Real time mains frequency monitoring (UK):
current: http://www.dynamicdemand.co.uk/grid.htm
past 60 minutes: http://www.nationalgrid.com/uk/Electricity/Data/Realtime/Frequency/Freq60.htm
Also, look up the Dynamic Demand wiki:
http://en.wikipedia.org/wiki/Dynamic_demand_(electric_power)
To measure mains frequency and analyse its statistical properties in real time you only need a small transformer, some pretty simple intermediate electronics and an NTP synchronized computer with appropriate software tools. I can’t believe there is no such equipment on the market.
Anthony, you may consider looking up such a gadget and add it to the collection at your shop. With this current grid de-regulation I am sure there would be a lot of willing citizen scientists out there who would like to monitor what’s going on and some would even put their results online.
Good software is of course paramount, but one can always start an open source project for that.
There are aviation test sets that are calibrated at 60hz using light bulbs. Must have missed the NOTAM on this?
Pure BS!
I’ve been in a power plant while they run a “frequency drill”. Listen to the T&D department communicating with the plant. “Ah, we are down to 59.99 CPS (Hz, it was a LONG time ago!)…
Still dropping, 59.97.. still dropping.. 59.95 at 59.92 they would KICK THE PLANT OFF LINE because of the “line losses” from the plant sub station to the GRID!
Now the fact of the matter is the grid is held at 60.00+/- .02 HZ for GOOD REASON. Having nothing to do with the “convenience” of the consumers.
File this under “Y2K deja vue”.
Max
Everyone is missing the point (some have got close).
There are few ways of ensuring long term (>1 month) timing accuracy in the home.
1. Counting grid cycles 60/50 Hz
2. Internet time
3. GPS time
4. Broadcast time – Fort Collins WWVB on 60kHz etc
All other methods of time keeping suffer from long term uncorrectable (except manually) drift.
Your government wants to take away the simplest way of you achieving accurate very long term (> many years) timekeeping.
Conspiracy theorists might think that this ‘experiment’ is way to drive you toward internet connected devices over which they may exert control remotely.