I live in a very small country. Sometimes we are running entirely on green energy, sometimes on generators. Mostly it's a combination.

But, during the daytime, when most people are at work, the engines don't use as much (or produce electricity) fuel, as when people come home.

So, how does the grid know when to up the power supply? Does it also up the supply, when I turn on my night lamp? I suppose voltage regulators have a lot of say here, but is there a hysteresis at work here, so that I'm not in the power of destroying a generator with my night lamp?

 

You use it or lose it.
Max.

 

Forget the power grid, that is much too complicated to explain here.

Think of a little gas powered generator that you might buy. If it is not running anything, it simply hums along at the correct speed to produce the 50 / 60 Hz alternating current. When doing this, it does not required much gas. Now you plug in a load. If nothing else happened, it would slow down because the same amount of gas no longer spin it at the right speed with the increased load. But the generator senses this slowing down and automatically supplies more gas to keep it spinning at the right rate, and the needed power has been added to the system. In fact, with such a small generator, if you suddenly plug in a large load, you can hear it start to slow and then recover.

Bob

 

It all depends on the exact type of generated source you have.
Max.

 

@MaxHeadRoom Don't think I follow..? Let's say I use it
@BobTPH Yes, that I understand, It needs more gas to keep the Hz up. As driving a car uphill. I was more interested in hearing how the grid handles fluctuations, and how big or small they are, before they affect the generator. Maybe there is some kind of buffer at work here to prevent the gas regulator from going up and and down constantly at every light switch?

 

Conservation of energy equation works both ways.

If your system is 90% efficient and you demand 900kW then the system has to generate 1000kW.
If you only ask for 9kW then the system has to generate 10kW.

Same thing happens when you are driving your car.
If your car has a 200hp engine, the engine is not delivering 200hp if you are crusing at 100km/h on the highway.

 

The dynamo generator is no different from a battery in this regard.

A 12V automotive battery can deliver 200A. If you only ask for 1A that is what you get.
If it were a dynamo generator, the current drops from 200A to 1A. The mechanical load on the generator drops accordingly. The generator does not speed up because there are feedback mechanisms that keep the generator at a constant speed.

 

Maybe there is some kind of buffer at work here to prevent the gas regulator from going up and and down constantly at every light switch?

Nope.
Every light switch operation will cause a minute change in the power to the generator to balance the change in output.
But when the generator is generating megawatts, it's pretty hard to detect the change of a few watts.
And note that there are many people switching switches on and off, so the average change is what is followed.

 

@MaxHeadRoom Don't think I follow..? Let's say I use it
@BobTPH Yes, that I understand, It needs more gas to keep the Hz up. As driving a car uphill. I was more interested in hearing how the grid handles fluctuations, and how big or small they are, before they affect the generator. Maybe there is some kind of buffer at work here to prevent the gas regulator from going up and and down constantly at every light switch?

As already pointed out, there are a lot of devils in the details and a lot depends on the specifics of the system being used, but most AC generating systems (of any size) attempt to control both the frequency of output and the voltage of the output. This is often split into to parts. Since one goal is to maintain constant speed, the voltage controller assumes that this will be the case and controls the voltage by increasing or decreasing the excitation of the generator (the strength of the magnetic field the rotors are operating in). It can do this very quickly and so is very responsive to voltage changes due to load variations. The speed of the generator is controlled by the "throttle" of the driving mechanism, whatever that may be (the throttle on a generator, the control rods on a nuclear pile, the valves on a hydro plant, whatever). Those have a bit more lag in them, as they are more "physical" in their operation, but they are still pretty responsive. Also, large generators have a lot of inertia and so they naturally resist changes in speed.

At any given time, most of the light switches being turned of are offset by light switches that are being turned on, so it's only the overall change in net usage that the generator sees and it sees this as first a change in voltage in the system, which the excitation controller attempts to even out. But there is also a change in the torque on the generator shaft as the amount of energy being transferred from the mechanical rotation of the generator to electrical energy changes and that results in the speed of the generator starting to change, which is sensed by the speed controller which adjusts the throttle position.

In theory, if the only thing that happened in the system was you turning on a light, that would be result in a change in the generator system that would change the amount of power being consumed by the generator. In practice this change would be in the noise and would hard to actually measure. But in small systems, such as a small gasoline-powered generator, you can easily detect the change just by listening to the engine's response to turning on and turning of a light.

In most areas the frequency of the power grid slows down in the morning as industrial loads are coming on line and it speeds up in the evening as they are coming off. Instead of trying to control the line frequency to supertight tolerances at all times, they are kept within significantly looser tolerances to make the system more stable, and then they exercise very tight tolerances to the overall average frequency of the system by intentionally speeding it up or slowing it down in order to remove the residual error. They do this by operating a clock that is based on the frequency of the grid (it's basically counting cycles) and then comparing that to a reference clock (usually slaved to one of the atomic time standards) and issuing speedup/slowdown orders several times a day to keep the maximum error between the two under some allowed limit and also driving it to basically zero error (often something like two seconds maximum error since the last time the clock was restarted) at some point during the day.

 

@MrChips I once had to replace a power supply for a dynamic positioning system on a ship I was on. The old supply was at 10 Amp and the new one at 20 Amp. The captain was afraid I was going to fry our system
@crutschow and @WBahn Yes! This was exactly the answer I was looking for. Thank you.

New bonus question (in order to prevent thread spamming): Now we are done cooking our dinner, and all the residents in the entire country turn off their oven at exactly the same time. Is this when our return effect relay's start trippin'?

IIRC, we had a somewhat similar issue on a ship I was on. The one generator started to produce way more than the other, just out of the blue, causing in our return effect relay popping.

 

The residents of Victoria, British Columbia, Canada are known to be more british than the british of the U.K. themselves.
Whenever there is a royal ceremony in the U.K. such as a royal wedding, live broadcast at 10am in the UK translates to 2am in B.C.

Urban legend has it that whenever there is a commercial break in the TV broadcast, city staff at the PUC has to monitor the telly in order to man the floodgates to cope with the deluge when all the loos are flushed at the same time.

 

Now we are done cooking our dinner, and all the residents in the entire country turn off their oven at exactly the same time. Is this when our return effect relay's start trippin'?

What's a "return effect relay?"

 

the grid is a little complicated. Suppose there is;
Hydroelectric
Nuclear
Wind
Diesel generators
Solar (anything with an inverter)
Load shedding

You can't throttle the power of a nuclaer reactor easily. Hydroelectric not quick either. Wind you have to take it when it's present.

If there is lots of excess capacity, we can pump water up a hill and use it later.

Diesel generators can easily be throttled.

Inverters have an interesting property. Locally, they can be used to power factor correct. i.e. generate current out of phase.\

Your looking at +-10% regulation.

 

@MrChips I once had to replace a power supply for a dynamic positioning system on a ship I was on. The old supply was at 10 Amp and the new one at 20 Amp. The captain was afraid I was going to fry our system
@crutschow and @WBahn Yes! This was exactly the answer I was looking for. Thank you.

New bonus question (in order to prevent thread spamming): Now we are done cooking our dinner, and all the residents in the entire country turn off their oven at exactly the same time. Is this when our return effect relay's start trippin'?

IIRC, we had a somewhat similar issue on a ship I was on. The one generator started to produce way more than the other, just out of the blue, causing in our return effect relay popping.

I have no idea what a "return effect relay" is and a quick Google search revealed nothing. Perhaps it's something unique to the ship power industry?

Here's an interesting article about how a massive turnoff can be managed if you know about it ahead of time:

https://economictimes.indiatimes.co...p-in-demand/articleshow/74980196.cms?from=mdr

Of course, the system has to be able to deal with unplanned for events -- such as damage to the distribution system that disconnects a large section of it catastrophically. But under those circumstances you aren't concerned with keeping things operating within normal limits, just within limits that prevent significant damage.

 

Return effect relay, that was a language hiccup on my behalf. The correct term is reverse power relay, where one of the prime power sources goes below grid voltage and starts to consume energy instead of delivering.

 

That's a whole other kettle of fish. When you have multiple power sources on the grid, especially when you have multiple types of power sources in terms of reliability of output and responsiveness -- for instance solar units that change capacity every time a cloud goes by or plants that take a long time to ramp down or, especially, ramp up, you have to have robust means of monitoring and metering the power from each. One tool in accomplishing this is by using phase-shifting transformers to control the power transferred from one point of the grid to another.

 

I have been involved with the utility industry on and off (pun intended) for over 40 years.

In addition to everything else that has been mentioned, utilities have yet another tool in their arsenal: transmission lines.

A simple example:
In New York it becomes dark earlier than Chicago, because of the time zones. It means that the evening's peak power consumption will start earlier in New York. With a transmission line Chicago can send its excess capacity to New York.

As the evening turns into night, New Yorkers will start to go to bed earlier than people in Chicago. This means that New York will have now excess capacity. It can now ship energy back to Chicago.

The major problem with ultra long transmission lines is maintaining synchronicity between all generating plants.
This is solved by ultra-high voltage DC links, the latest which I am aware of is 1.1 million volts.
In China.

 

There is perhaps a little more than self regulation as far as national grids are concerned. Where the demand increases rapidly for some reason, another generating plant my need to be brought on line to meet the demand.

Many years ago where I worked, included a large electro-plating production facility – staff would ring the national grid to advise when the facility was about to start and cease operation, so that the large change in power demand could be accommodated by the local grid.

 

And that is the problem with an integrated system which resulted in a massive blackout across North-Eastern North America on August 14, 2003.

In order to cope with changing power demands changing by the second, more than 2GW of power were flowing one way and then switched to the opposite direction on the same HV transmission lines.

 

In the UK (and possibly other countries) the National Grid has a neat trick whereby they can automatically reduce the total demand.

Supermarkets (and others with large refrigeration systems) are offered an incentive (money) to have their food chiller systems configured to switch off once the mains frequency drops below 49.5Hz.
No food is spoilt as a result of the refrigeration being off for half an hour or so.

There was an EU proposal to fit all new fridges sold within Europe with this technology – but there was resistance to Big Brother controlling house-hold appliances and it would add around £50 to the price of a fridge – and so the proposal has not been in-acted (yet).