Grid Stability - Frequency Variation

swaltner · Jan 1, 2015

On many posts about the doom-and-gloom of Net Metering run amok, there is talk of "grid stability" being an issue. I frequently see references to the grid frequency changing during times of over/under production. Why is that the case? How much does the frequency vary from true 60 Hz?

I'm assuming I just don't understand how the grid uses the frequency of the AC power. My thinking is that the grid frequency comes from the speed that turbines spin at, which I don't necessarily see how those would be impacted. Does the grid use the frequency of the AC power as a sort of power line communication channel to communicate between various power plants on the power requirements at that time? Why does over/under production change the grid frequency?

Cosmacelf · Jan 1, 2015

electricity - How does load affect frequency on the power grid? - Physics Stack Exchange

nwdiver · Jan 1, 2015

Well... hopefully someone with more experience can correct me if I'm wrong but....

My understanding of how synchronous induction generators work is that there are two moving 'things' that determine frequency; The movement of the stator in the rotor and the movement of the induced magnetic field inside the rotor which CANNOT be the same speed or there would be no relative motion => no current/power generated. Increased load can cause frequency to dip slightly. Grid frequency is such a vital health indicator that it's prominently displayed in most grid control rooms.

The great thing is that this information is obviously available to everyone connected to the grid... why this hasn't been used more is frustrating. It works.

Robert.Boston · Jan 2, 2015

The U.S. grid is complicated by the fact that there are many grid operators, each managing a control area of one of the three interconnections (East, West, Texas). Under-generation in one control area therefore has the first-order effect of sucking power in from neighboring areas to compensate. These unscheduled flows, called the Area Control Error (ACE), are readily measured. The national standards set by the National Electricity Reliability Council (NERC) set a standard for the allowed ACE, expressed on a 12-month rolling basis. Therefore there's technically no instantaneous standard regarding deviations around the 60Hz.

In practice, however, if you go to one of the operations centers at, say, the CAISO headquarters in Folsom, you'll see that the operators are keeping ACE in balance so that frequency is within 0.1 Hz or so. A drop of 0.2 Hz is bad news.

One challenge created by PVs and wind is that they have no inertia. There's a really good article explaining the problem. As it states:

Classical generators are big heavy machines that rotate on a large shaft. These devices store mechanical kinetic energy – much like a heavy old car that is careening down a highway. If something is placed in the way, the car will slow down quickly and release a lot of energy, often resulting in damage. The old generators do much the same, but for different reasons and with different results. In both cases, the release of energy is attempting to maintain the speed of the device or vehicle. If a large source of power to the grid is suddenly cut off, the entire system slows rapidly, and as it does, all generators slow their speed of rotation. As the speed is declining, the generators will deliver more power than they were doing only seconds before since they are releasing their stored energy, known as inertia. This inertia resists the decline in speed, allowing control systems some time to automatically apply more power to the remaining generators.
...
There is bad news here. The steam turbines used in nuclear and coal fired generators cannot be allowed to operate at significant reductions in frequency, so at a pre-set frequency, these generators trip offline to protect their turbines and that makes the problem much worse – sometimes leading to a system blackout.

You might think wind turbines would have inertia, but no. Because the rotational speed of a wind turbine is necessarily unlinked from the system frequency, all wind power goes through inverters, much like PV energy. The result, no inertia, and so no contribution to prevent frequency crashes.

AudubonB · Jan 2, 2015

A story from the Paxson archives....

About fifteen years ago, back when we still had a so-called local utility providing the "downtown" area with electricity, a new fella comes in town to work as mechanic at the DOT maintenance shop.

After less than two weeks of coming to work, he queries the foreman as to how come he's the only one getting to work on time - each day everyone else is showing up later and later than the day before.

You've probably guessed the punchline that also set the entire DOT crew roaring with laughter: our utility's generators were so out of whack that they would run at anywhere between 52 and 67 Hz, but, at that time most frequently way on the higher and. Simple electric clocks are set to keep time as f(Hz=60), and so his clock was zooming ahead of real time. All the other employees knew not to trust those clocks.

True story!

wycolo · Jan 2, 2015

My school had its own oil fired steam plant that provided steam to heat the buildings and a turbine generator to provide electricity. There were two identical clocks by the control panel, one connected to the local grid and the other one to the school's wiring. If the operator noticed any time difference between the clocks he would adjust the speed of the turbine to compensate. In my dorm room I had WWV on shortwave and the two times were almost never in synch. But even a broken clock is correct twice a day.
--

bonaire · Jan 2, 2015

Grid stability, say in Hawaii, is based on renewable intermittancy and the speed at which the grid base-load plants can react to demand spikes caused by partly-cloudy days. It can also happen in smaller areas such as within power companies in So. Cal (SCE for example) if they have a very hot day and clouds move in requiring a boost in power brought in from out of state, such as AZ. If hundreds of MW of solar goes to 20% output due to a cloud front on a hot summer day, things are not that stable. They are considered an unregulated power supply. Everyone says solar is going to do so much for power production, but the base load operators actually don't believe it is helping that much and causing them to operate more frequency-demand response products to compensate.

tga · Jan 2, 2015

AudubonB said:
You've probably guessed the punchline that also set the entire DOT crew roaring with laughter: our utility's generators were so out of whack that they would run at anywhere between 52 and 67 Hz, but, at that time most frequently way on the higher and. Simple electric clocks are set to keep time as f(Hz=60), and so his clock was zooming ahead of real time. All the other employees knew not to trust those clocks.

My wife and I were in Italy back in the 90's. We brought one of those 240-to-120 transformers, primarily to run her hair drier. Without thinking about it, we brought a small plug in electric clock (we didn't have a battery powered alarm clock). I got good at predicting the overnight drift, and adjusting the time appropriately, so that the alarm went off at about the expected time in the AM.

TheTalkingMule · Jan 2, 2015

Prior to the solar boom in Germany, the utility operators claimed that if solar went above 5-10% of supply at any one time, the grid would likely be unable to handle it. Of course, they've now crossed something like 60% on certain days with no massive issues that I've heard of.

ItsNotAboutTheMoney · Jan 2, 2015

Robert.Boston said:
The U.S. grid is complicated by the fact that there are many grid operators, each managing a control area of one of the three interconnections (East, West, Texas). Under-generation in one control area therefore has the first-order effect of sucking power in from neighboring areas to compensate. These unscheduled flows, called the Area Control Error (ACE), are readily measured. The national standards set by the National Electricity Reliability Council (NERC) set a standard for the allowed ACE, expressed on a 12-month rolling basis. Therefore there's technically no instantaneous standard regarding deviations around the 60Hz.

In practice, however, if you go to one of the operations centers at, say, the CAISO headquarters in Folsom, you'll see that the operators are keeping ACE in balance so that frequency is within 0.1 Hz or so. A drop of 0.2 Hz is bad news.

And batteries can be used for frequency regulation: RES Americas

nwdiver · Jan 2, 2015

TheTalkingMule said:
Prior to the solar boom in Germany, the utility operators claimed that if solar went above 5-10% of supply at any one time, the grid would likely be unable to handle it. Of course, they've now crossed something like 60% on certain days with no massive issues that I've heard of.

5-10% of total 365 day supply or 5-10% at any given time?

TheTalkingMule · Jan 2, 2015

nwdiver said:
5-10% of total 365 day supply or 5-10% at any given time?

5-10% at any given time. LOL....guess they were a bit off. Germany's at about 8% of total demand sourced from solar and has passed 50% on a few occasions.

swaltner · Jan 3, 2015

Thank you for the clarifications. The explanation of a car engine struggling to maintain power will going uphill makes sense as to why additional load on the grid would slow the frequency down.

It's much easier for an end user to see the Voltage on their power feed and that swings so wildly, and not directly in relation to the power I'm consuming in the house. For example, my TED reports that over the course of today my voltage has ranged between 235.1 and 245.8 and since the beginning of the month a few days ago, the range is 232.4 to 246.0 (which is +/- 3.2% from the nominal 240V rating). I've seen even larger swings before, but these don't directly correlate to my consumption. I guess I assumed that the grid frequency jumped around like this, but apparently not. I'm pretty sure I share the pole-mounted transformer with my next-door neighbor, so maybe their usage is what's causing the uncorrelated voltage swings. One of the links in nwdriver's reply above linked to prototype devices that would start shedding load when the frequency dropped to just 59.95 Hz (only a 0.08% drop from 60 Hz). Even with this very tight tolerance, there was generally only one event a day during the test phase they were running. The readout on the grid control room reading the frequency to three decimal places is another sign that this is a very stable value.

I think the only time that I've seen frequency readouts on equipment is the 60 Hz to 50 Hz "conversion box" that we have at work. This allows us to test our hardware running on 50 Hz power to complete the testing that we need for shipping worldwide. This is a cube about 3' in size with that I've seen in the hall feeding one of the test labs. Every time I've walked by it, it's reading 60.0 Hz on the input and 50.0 Hz on the output.

Thanks again for everyone's input.

Search

Grid Stability - Frequency Variation

swaltner

Active Member

Cosmacelf

Well-Known Member

nwdiver

Well-Known Member

Robert.Boston

Model S VIN P01536

AudubonB

One can NOT induce accuracy via precision!

wycolo

Active Member

bonaire

Active Member

tga

Active Member

TheTalkingMule

Distributed Energy Enthusiast

ItsNotAboutTheMoney

Well-Known Member

nwdiver

Well-Known Member

TheTalkingMule

Distributed Energy Enthusiast

swaltner

Active Member

Similar threads