By Phil Kreveld
Imagine that while solar and wind generators as well as batteries make steady incursions replacing synchronous generators, more synchronous condensers (syncons) are being installed to provide grid strength and inertia.
Let’s take stock. There is still synchronously generated voltage at constant frequency being provided by traditional energy sources. The synchronous condensers are motors—synchronous motors. Their stators provide a synchronously rotating magnetic field; their rotors are magnetic poles which by means magnetic attraction forces, rotate in synchronism with the stator’s rotating magnetic field. To start a syncon, we’ll use the convenient acronym, power needs to be applied to the rotor to get its speed up to that of the synchronously rotating magnetic field—that is the classical way although these days there are alternatives such as the static frequency converter. Thereafter the rotor having locked onto the stator field, the syncon can continue to run without the need of an auxiliary power source. As the motor is not connected to a mechanical load it only draws a small amount of power.
Related article: Power and reactive power
We have established the initial conditions of the thought experiment. Before we move on, let’s imagine a grid event. A short circuit somewhere in the network will do. It causes a momentary drop in voltage and a rapid rate of change of frequency. A nearby syncon experiences the change in rotational speed of the stator magnetic field but the heavy, high inertia rotor cannot speed up instantaneously. Under stable conditions, before the short circuit, the syncon was spinning under no load, basically stirring the air separating the rotor from the stator. The rotating voltage ‘vector’ applied to the stator is also the one being applied to solar panel inverters. As the short circuit occurs, the heavy rotor momentarily assists in restraining the ‘speeding’ voltage ‘vector’ by adding a counter rotation ‘voltage vector’, i.e., think of it as being at right angles to the rotating voltage vector, bringing the voltage vector back towards its position before the transient conditions caused by the short circuit. If the above is just gobbledygook, take a look at Figure 1 below. Hopefully that will make the above clearer.
The usefulness of the syncon has been established. Most solar, wind, and battery generators feature grid following inverters, although battery inverters can also be grid forming, more about this later. If we stick to the grid followers for the moment, we see that the role of the syncon, replacing synchronous generators, is rather important. The grid followers lock on to the ‘rotating line voltage vector’, a useful image for the further scenarios sketched here. The picture to be kept in mind is that the complete network, comprising of generators, be they wind, solar, batteries and remaining synchronous generators (coal, gas and hydro) all run off the not quite the same ‘voltage vector’, but of vectors that not only rotate at the same speed, and as importantly maintain, the same phase relationship between themselves. Hopefully Figure 2, below, will make that clear.
We now have the initial conditions of the thought experiment. Just one more thing, although no knowledge of electrical engineering is needed, all the power produced by generators of whatever type has to be used up, at every instant, by heating or cooling stuff, running pumps, escalators, shoving energy into batteries, etc. Anyway, that’s just common sense.
Now for the experiment. Things are humming along with a mixture of syncons, synchronous generators, batteries, and voltage following inverters. Surreptitiously, with not a word to the Australian Energy Market Operator, we start turning off a few synchronous generators. The syncons take over the job and life goes on. Power is being generated by solar and wind, some of it going into batteries and the rest doing useful stuff—and, of course, a little bit of power to allow the syncons to continue ‘stirring the air’.
Every time there is some disturbance, the ‘voltage vectors’ are kept within bounds. One day the last synchronous generator is quietly turned off. The syncons haven’t been ‘informed’ and keep spinning happily. The little bit of power they need to stir the air is coming from the renewables, the rest of the power doing useful stuff and charging batteries. It’s bit like riding a bike without holding onto the handlebars. At this juncture, someone tells AEMO, that the system is 100% renewable. The cold sweat breaks out. Why? Because every kilowatt has to be ‘accounted’ for. If there is more power being generated than required for useful stuff or that can be stored in batteries, the syncons are going to stir the air a little faster—frequency is increasing—and with the synchronous generators in retirement, there is no mechanism to directly control frequency.
No matter, because while there are no disturbances, we can put up with a higher frequency provided that the individual voltage vectors stay in a more or less stable relationship. Renew Economy already has a headline out for its next edition; “Australia, first in the world with a 100% renewable grid”! But then it happens. A single solar farm and its nearby syncon lose their load to which energy was being supplied. All the power from the solar farm now goes to the nearby syncon. The voltage vector accelerates; its rotational speed rapidly trying to absorb excess energy by stirring the air faster—if only it could get that sluggish rotor to keep up. It can’t and ‘slips’ a (magnetic) pole. The solar farm inverter experiencing a rapidly increasing frequency disconnects, and unstable voltage vectors, shut down the rest of our grid. We have just fallen off our bike!
Related article: How to bake an electrical mille-feuille
Alright, it was only a thought experiment, and AEMO is very much in charge of stability. However, what we learn from the foregoing is that an all-renewables grid is a totally different thing from what we have at the moment—a mixture of synchronous and renewable sources, which happily keeps ‘all the balls in the air’! The answer we are given to assure us of the viability of an all-renewable grid is that we replace the synchronous generators with grid forming inverters, sourced by batteries. It is an excellent proposition, yet to be proven in the field but it stacks up theoretically. In particular, because unlike synchronous generators (remember Callide C), the grid-forming inverters can almost instantaneously become converters and charge their batteries (provided these are not already fully charged). That said, the insertion of grid-forming inverters, replacing synchronous generators in a complete, 3000-odd busbar system representing the south east grid is also a thought experiment!
We’ve played with ideas, no nightmares for AEMO, but a thought experiment that provides food for thought. If we are guilty of anything it is the glossing over of what a huge experiment the renewable transition is. There are folk who would take these scenarios as proof that we are on a hiding to nothing. But actually, nothing could be further from an unavoidable calamitous failure. All we need is electricity system engineering, not just connection approvals but ‘whole of grid’ engineering—and an unhurried, cautious approach to the replacement of synchronous generators by renewable sources, not forced along by government-set targets.
