[Karsten (OP)]

I assume power plants cannot generate exactly the amount that is consumed at any given time. Generating more or less does not happen quickly. If they do not make enough - we have brown-outs. If they make too much or people decide to use less for an hour, what happens to the "left-overs"?

Your thoughts?

[lyner]

When the electrical load suddenly reduced, the generators would speed up unless action were taken. More steam than necessary is coming from the boiler  and the turbines would just go faster. A normal 'fueled' power station can reduce the oil (or whatever )supply fairly quickly but a nuclear reactor takes longer to control. This can be a problem and power needs to be dumped somewhere until the reactor output is reduced.

[Karsten (OP)]

This can be a problem and power needs to be dumped somewhere until the reactor output is reduced.

How does it get "dumped"?

[techmind]

I don't claim to have any great expertise in power-systems engineering, but I understand the general idea is that you use nuclear to provide the "base load" ie that below which demand rarely ever falls (so you can leave the nuclear on full power almost all the time) then use the more expensive (yet more versatile) coal, oil, gas, diesel to match the load. These types of power station will have regulator or throttle controls which will allow the mechanical energy to the generator to be adjusted to match the load to a large extent. I imagine it's most efficient to have the individual generators either on full power or completely off, within reason though.

In the UK we have a handful of pumped storage schemes where they pump a lake-load of water up a hill when there's an excess of supply, then allow it to run back down again through turbines to meet short-term peak loads (traditionally during during commercial-breaks in popular programmes on national television). These pumped-storage power-stations can be turned on and off within a few 10s of seconds.

Finally, if you really have to dump electricity, you could always just switch on some huge heating-elements immersed in a lake somewhere. I heard that the lake by Transfynyd nuclear station used to be kept warm by the power station cooling when it was operating, but now the power station has been retired they have to artificially heat the lake to maintain the tropical fish stocks  []

[lyner]

I remember talking to a guy, a long time ago, who was looking at the problem of a nuclear power station which was fed by only one National Grid circuit. They needed an emergency way of dumping energy in the event of the single line going down and to give the reactor to wind down. I think a pump storage system was the solution.

[John Chapman]

I happen to live near one such hydro power storage facility and have toured round it a few times. This one is situated at the foot of a welsh mountain and works in much the same way that Techmind describes. When there is a surplus of power, water is pumped from a lake at the bottom of the mountain to one much higher up. In this way surplus electricity is converted to potential energy. When a favourite soap opera breaks for commercials, and half the nation dashes to the kitchen to put the kettle on, the lake water is allowed to flow back down the pipes through one or more of three turbines and the National Grid is topped up. In this way the power stations don't need to get too concerned about the short term peaks and troughs of consumer consumption.

The facility I live near is called Electric Mountain and they have a (rather unimaginative) website at

http://www.electricmountain.co.uk/pumped_storage.htm
 

[chris]

I seem to remember Dave (shorts) suggesting in the past that some energy might be stored in massive flywheels which can be used either to soak up energy excess (when present), or make up shortfalls to help iron out sudden load peaks. Does anyone know whether this is true, however?

C

[Karsten (OP)]

So, does it actually make any difference if a few thousand people turn off their lights for one hour? It seems that unless it is a predicable event, the power generation would not be reduced.

[techmind]

I seem to remember Dave (shorts) suggesting in the past that some energy might be stored in massive flywheels which can be used either to soak up energy excess (when present), or make up shortfalls to help iron out sudden load peaks. Does anyone know whether this is true, however?

I have never heard of such a device in our power infrastructure, but I would have thought that it would not be realistic given the amount of energy you'd have to store to be significant on the scale of (even modest fluctuations in) nationwide demand.

[techmind]

Consider also that the demand is not absolutely rigidly defined: the voltage tolerance at the end user is +/-6% (ish), so if all loads were resistive (a simplification I know) then demand would rise appreciably as the voltage rises, which gives you something to work with!

I suppose a fair amount of the official tolerance is required to cover for drops in the final local distribution network, but there must be some leeway in the overall system voltage?  []


As an aside there still is (or was) some "diplomatic" use of tolerances so that while we in the UK officially harmonised with Europe at 230Vac quite some years back, the UK tolerance is actually 230V -6/+10% while Europe is 230V -10%/+6% which means that in practice the UK can remain at 240V +6/-6% as it always has been and mainland Europe at 220V...  (it would make a difference to the lifetime of tungsten filament bulbs if they changed the voltages for real). Daft, eh?  []

[lyner]

Chris.
I think the flywheel thing was something to do with power factor correction. There's some dim distant memory - it may have been something to do with running generator sets with very low  turbine power. Kinetic storage cannot have a lot of capacity - a few kWh go a long way.

As techmind says, there is provision for quite a high variation in supply voltage before consumers can complain and, when the load gets too too much, the frequency can dip, too. In the nature of things, a voltage dip of 6% will supply 12% less power - that would mean there is capacity to deal with additional loads (not more total power - just more connected customers). The reverse is true when people start switching off - the others get higher supply volts. The down side is that light bulbs can have very short lives (as I know to my cost; I had a good long argument with the supply company before they turned my local voltage down).

[Mazurka]

Large flywheels for storing power - do you mean homopolar generators? (http://en.wikipedia.org/wiki/Homopolar_generator)

I don't think these are used commercially, but are used to store energy for the likes of fusion reactor experiments.

[fontwell]

sophiecentaur is correct. When demand drops, a generator can be designed to respond in two ways (as far as I am aware). One way is to increase in velocity/rotational speed, thereby transferring unused energy into the rotor. This is not desirable for a number of reasons, not least of which is that it alters the frequency of the mains and plenty of electrical kit is designed for a specific frequency only.

The other way is to allow the ouput voltage to rise a little. Since most of our power is consumed directly as heat - kettles, electric heaters, ovens etc, even old light bulbs are nearly all heat rather than light - increasing the voltage a bit just makes all these things run a bit hotter. So the power dump is distributed over the whole grid. You may have noticed sudden changes in the light levels sometimes. That will be a generator coming on/going off line (or a fault!) and causing the voltage to change as a result of more or less power being available.

Also, note that the power used is proportional to voltage squared, so an increase in voltage of 10% will dump around 20% more power. Designing equipment to cope with a wide range in voltage is not a big deal and has always been done, so no problem with that.

Finally, in power plants they have people who predict demand for every minute of every day, to the extent of running power stations up to account for everyone making a cup of tea when Eastenders finishes on the TV. If they do their job correctly then the need to dump power should never be too big.

[lyner]

Did you hear of the latest action in Iran, taken by the opposition? They are organising mass turning on and off of loads in order to disrupt the power industry.
 If there is a sudden cut in load, all that can be done is for the boilers in the biggest power stations to let off steam and cut the fuel supply. Nuclear Power Plant have huge problems because you can't turn down a reactor in a matter of seconds.

[Karsten (OP)]

No, I have not heard of that. But it is similar to what made me think about this problem initially. I was wondering what happens if thousands of people participate in events such as "Earth Hour". Do power stations literally dump power (= throw it away) for those drops in consumption or are they able to adjust and less power is made (and the environment is actually helped for that one hour) or can they even store it for later? I still don't know if "Earth Hour" results in wasting of already manufactured electricity or if it actually results in measurable benefits.

It seems that in Iran this method of on/off consumption is rather disruptive. Kind of impressive that organized consumption/non-consumption can cause such trouble. Must not be great for hospitals and such.

[lyner]

You can't store the stuff except, in a Pump Storage system, as Gravitational Potential Energy. That is of limited use in a major switch off.

It is a headache for the Load Management Engineers. They have to read the Radio Times to spot when they are likely to expect millions of homes brewing tea in the commercial breaks. Now that program audiences are smaller, due to the large number of channels, the effect is more spread out because not everyone watches the same shows any more.
It would only be at the start of Earth Hour type events that the problem exists. I think the effect is limited to minutes of operation.

[Mazurka]

Shows the current state of the UK grid...

http://www.dynamicdemand.co.uk/grid.htm

and has some useful info...

[HankRearden]

I believed that most power companies actually sold the excess electricity to other companies for their customers, who might have a greater load at that time.

[lyner]

Correct.
The DC cross-channel link is used in that way.
The UK is mostly a net importer but, during the hot weather, the Nuclear plant in France couldn't run at maximum because the cooling water was not sufficient. We were selling them Energy!

[LeeE]

Shows the current state of the UK grid...

http://www.dynamicdemand.co.uk/grid.htm

and has some useful info...

Neat link - Ta.