From UK's National Grid:
Figure 1. The graph shows Frequency data to a 15 second resolution over the hour up to 2016-05-21 13:31 BST.
System frequency is a continuously changing variable that is determined and controlled by the second-by-second (real time) balance between system demand and total generation. If demand is greater than generation, the frequency falls while if generation is greater than demand, the frequency rises.
National Grid has a licence obligation to control frequency within the limits specified in the 'Electricity Supply Regulations', i.e. ±1% of nominal system frequency (50.00Hz) save in abnormal or exceptional circumstances. National Grid must therefore ensure that sufficient generation and / or demand is held in automatic readiness to manage all credible circumstances that might result in frequency variations.
There are two types of Frequency Response Dynamic and Non Dynamic Response. Dynamic Frequency Response is a continuously provided service used to manage the normal second by second changes on the system. While Non Dynamic Frequency Response is usually a discrete service triggered at a defined frequency deviation.
To answer your questions:
So I understand that the generators at the power stations need to spin at exactly 3000rpm.
Not constantly. Over a day it averages to exactly 50 Hz. This used to be a requirement to keep all mains powered clocks in synch. I have one on my electrical supply day/night meter.
This is true no matter what source is (nuclear, coal, gas, hydroelectric etc).
Yes, they all run in synch.
Let's use steam as an example here. How do they use the steam to spin the generator at a constant 50 hertz? It doesn't seem plausible to control the speed with the amount to steam passed though to that precise speed. So how do they do it?
It is, in fact, possible, provided there is enough steam being generated to supply the peak load. With a single steam generator supplying an island, for example, the scheduler would plan ahead using weather information, TV schedules (the famous 10 million kettles going on in the big-game half-time), large load user schedules, etc., to have the thermal plant generating enough steam in anticipation. Meanwhile, while demand is low, the plant has to vent the excess steam to atmosphere. Yes, this is wasteful.
In practice, most grids have a mix of base-load thermal plant with other fast response generators such as gas turbine and hydro which can be switched in and out quickly.
The UK also has capability to import power from France and Ireland by underwater DC interconnects. These allow connection of multiple national grids without synchronising problems.
In a coal or nuclear plant, the thermal power changes very slowly, perhaps 10-20% per hour. To have power available for spinning reserve, the steam turbines are run at a lower power than the boilers, the main throttle is set so that there is some steam available but not used. The excess steam bypasses the turbine and its energy is wasted. If more power is required, the steam valve can be carefully opened and the power delivered increases. For a big steam turbine this might still take 30 seconds.
So to a first approximation the slow thermal plants consume fuel for the full total of actual power + spinning reserve.
You have a good question about the timescales.
At the shortest timescales, fractions of a second, the frequency is passively stabilised by the inertia of all the generators (and rotating loads).
At longer timescales it's entirely up to the control systems adjusting the power of each generator, and depends on the transient power response of the generator.
Some time back I found a very good presentation by John Undrill, called "Power Plant / System Dynamics and Control" presented at a NREL / EPRI workshop, May 2013.
I can't find a copy of the document to link to now, see if you can find a cached copy somewhere.
Best Answer
It is very easy to tell what is being asked here. What is needed is specific detail. Two devices are intended to be used in a load sharing arrangement. The 2301A load sharing controller is meant to enforce sharing. Failure to work suggests that the voltage does not droop enough to allow the voltage to fall below that of the second energy source before the load gets too large and trips the overload protection.
The 2301A is a device which allows load sharing between two power generation systems. It creates a sagging load/voltage relationship so that as an alternator etc is loaded it drops below local bus voltage and causes a more lightly loaded and thus higher voltage generator to pick up progressively more load.
Assuming that the equipment is not faulty it seems likely that voltage shaping being applied to one machine is not adequate to drop the voltage enough for the second alternator to load share. The 2301A can have the amount of "droop" under load adjusted with a pot. I'd guess that the pot needs to be adjusted so that the voltage droops more under load than at present.
Recall the old "joke" - What do you call a person who speaks 3 languages / 2 lanhuagrses / one language. A: Trilingual / Bilingual. guar An Ame.... It's not funny, alas.
This question is worth persevering with for at least 2 excellent reasons.
It is a classic "English as a second language" situation where the OP writes far too little, as expressing the question cogently in English is difficult.
AND some English only speakers who are bright technically are not able to get their brain around the language constructs.
The OP's problem is clear - they have adequately explained what they want to do and the equipment used and what is going wrong.
What is required is clear. The details are not.
A small amount of persuasion will fill in the gaps.
(2) This is a vastly more rewarding question than about 90% that get asked here. This is a real world application of a complex and unusual piece of equipment doing a crucial real world task. Having this topic discussed here will be the first introduction that many people have to this piece of equipment.