If anything, this is the most complex issue. The question "what is the correct voltage?" is almost both the beginning and what is to be accomplished. A good start would be a specified voltage, but that relies on a universally agreed specification. What is helpful is an older spec and a newer one seem to arrive at roughly the same result and thus proves a good place to kick off.
The older spec said "240V -10..+6%". This translates into limits of 216..254.4V. The newer spec (EN50160) says "230V -6..+10%" and translates into 216.2..253V. Hang on! We have just quoted two nominal voltages with very similar limits making it a range that appears to be an acceptable voltage, rather than a fixed point. The only reason a nominal exists is it forms a target. Imagine a dart board with no segments, there's not much fun in saying "if you get the dart in the board, you've won!".
However, having a range has a purpose. It means the network does not have to be perfect (lossless) and allows it to 'give' when under strain. When everyone is drawing current (example dinner time) then the network will, owing to simple physics identifying itself as resistance, have voltage drop across it meaning the final users not receiving everything that was pumped in.
But herein lies a tiny rub. The network operators know this acceptable voltage range exists and instead of aiming at the humanly devised nominal, treat it like a motorway speed limit. Although there is a minimum allowable speed as well as a maximum, it's yet to be seen where people are driving at some 'nominal'. Most aim to stay around the upper limit.
The positive argument is there are technical advantages to having a slightly higher voltage than slightly low. Motors are quicker to start therefore reduce starting currents and time therefore reduce the drain on a system therefore the amount of noticeable flicker, which is also reduced by the higher voltage because the lamps are burning fairly white. There are other such convincing arguments, and they would prefer you believe these.
The real reason behind the voltage being kept around the 240-245V mark is purely financial, and has a two pronged approach. The demand on the network is growing and it is hoped that by raising the voltage at the start it can deliver sufficient voltage to the clients at the end of the line. This is usually achieved by simply changing a tap on a transformer and no money being spent on upgrading the network as such. This has a slight flaw being the losses go up, but....
What also goes up is the overall current draw. Now as power is related to Vē this means the consumption rises in proportion to the change in voltage, squared! What this means is the consumers are, unbeknown to them, forced to part with more of their hard earned cash. Although the above mentioned losses are higher, they are proportional to the energy consumed. The simple equation is the more electricity the network operating companies sell, the higher are their profits in a defined period, and the faster they pay off their debts. I know this appears cynical, but the sad fact is it is generally true.
The mechanics of this has proven to not be too clear for some. What is being referred to is the fact that even though the standard voltage (e.g. 230VAC) is meant to be fed to houses, electricity suppliers push this up to 240V and even as high as 245V to 250V (especially during "quiet" hours). Some stuff (such as switch mode power supplies) tend to lower the current as the voltage goes up, most other devices around the home don't. The current rises with the voltage.
Such items are lamps, washing machines, fridges (refrigerators), etc. Items such as electric kettles are not as prone to this problem (although the current does go up, the water boils quicker so the actual energy usage is pretty much the same).
A fridge motor is a prime example of wastage; The motor is usually a squirrel cage type meaning it runs at a speed determined by the frequency, not the voltage. If the voltage goes up the motor simply runs warmer. With the motor running warmer the fridge not only has to work to keep the food cool, but also needs to work against the warming effect of the motor that is now straining (i.e. getting hotter) because of the extra voltage.
The net result is a rise in the wasted energy within the domestic home. Although the network losses are higher owing to the higher current draw, the electricity company is not worried because these losses are proportional to the final usage. I don't want to go into great maths here, but suffice to say they buy in bulk and sell in bits. The more trading they do the more profit they make (basic business sense!), and this happens when the voltage goes up.
All would be well if the demand for power were stable as the supply companies could make adjustments along the path from beginning to end. Sadly this does not happen and the attitude of the bean-counters has left us with the real problem of fluctuating voltage. In the short term this is not going to go away and if high voltage is a problem then the supply company must simply be asked to bring it down.
It is not all doom and gloom and there is a ray of hope in the use of modern technology. This uses efficiency techniques that adjust the current such that the same amount of power is consumed, regardless of the voltage. In short, the higher the voltage the lower the current. This has two advantages being the networks are slowly being given a bit of breathing space as well as providing less of a change in total demand. In the long term we can also look forward to a better environment as less fuel has to be burned off. We are still a long way off and we need to deal with the present.
Because the networks (right down ot house wiring) are not keeping pace with demand, they are becoming 'soft' which leads us to a further problem. It is mainly domestic properties that are suffering with this effect. In times-gone-by the load was a few lamps but now, as more appliances are used in a home, the load has become extremely sporadic leading to quick changes in voltage. This is reflected in a measurement known as Feed Impedance. We investigate this next.