Electrical Distribution

I worked in the Australian Power distribution industry for around 14 years in Australia (NSW) as a linesman and other roles.

I thought I would gather a few thoughts here.

A lot of people have no real exposure to the distribution system other than the outlet on the wall and a basic understanding of AC.

But it’s actually not that hard or that different. In fact, a lot of the layout is fairly basic compared to your average electronic circuit.

However the distribution system has its own unique challenges such as fault detection and isolation, as well as high voltage conductors and reactive power.

The high voltages also mean that special principles apply to circuit breaker design and  insulation.

In New South Wales, generators usually producing 11 or 22KV step up to 330 or 500KV and pass this to the transmission operator (transgrid) who marshall the feeds at switching yards before transforming it down to 132KV and sending it out on long span steel towers to our  local companies, essential energy, endeavor energy and ausgrid.

Distribution companies receive the 132kv and generally pass that around their own network on steel towers, although more recently vertically constructed concrete poles are used for 132.

Transmission zone subs pretty much only deal with splitting up the 132 and do not supply any customers, they will generally kick the volts from 132 down to 66kv and send this out to various distribution zone subs who do supply customers.

The general layout for a Transmission zone sub is one or two incoming feeders at the highest voltage will sit on what we call a primary bus bar. The primary bus bars job is pretty much mechanical with some mechanical switches/isolations available. Also connected to this bar will be a couple of transformers that have the 132KV primary connected to this bar and a 66 KV secondary. The output of those transformers will go onto the secondary bar which also contains mechanical switching/isolations.

Each connection of a feeder or a transformer to a bar will go via a breaker. There will also be a current sensor ( CT ) and a voltage transformer (VT) which generally transforms the voltage to a standard of 110 volts.

Relays monitor a proportion of the real current via the CT and a proportion of the real voltage via the VT. Parameters can be set in the relay which has control over the breaker.

When high voltage contacts are opened an arc is formed and this can continue to exist as it ionizes the air around it creating a nice path. Arcs are not actually a dead short and therefore will not induce any further tripping as over current.

 Unfortunately, an arc is hotter than the surface of the sun. In fact, the hottest thing in the universe and if it’s there for too long, it’s going to melt everything including your metal, which by the way is often aluminium as it is almost as good a conductor as copper but much lighter and therefore much more suitable for an aerial conductor.

 High voltage breakers rely on many different techniques to extinguish the arc . High speed opening, air blasts, arc chutes which are layers of metal that become magnetized during an arc and will suck the arc into them, oil to quench the arc, or gas like SF6 sulphur hexafluoride which is an insulating gas meaning the arc has trouble passing through it.

If you look closely, you will see levers that can open mechanical connections across the tubular bar  conductors. These are usually not operated under load. But you can see plenty of YouTube videos of people doing it the wrong way. That’s when you see the big arcs. Generally, breakers are used to de energise the bar in a zone sub and then the mechanical switches are opened, locked off and danger tagged.

Sometimes the bar can be split into one or more sections by a breaker to allow isolating for faults or to switch between say Transformer 1 and Transformer 2 . One may be in use and the other is awaiting to step in If the first one has a fault it will automatically step in or one may be in use and the other is currently undergoing maintenance.

After leaving the secondary bar at 66 KV, via transmission breakers it often leaves on poles and wires to the distribution zone subs. These are the buildings and yards you see scattered around your neighbourhood. There is usually an incoming 66 and an outgoing 66 as the network is set up either daisy chained or in a ring configuration. When set up as a ring this means if an incoming feeder needs to go down, the network can be back fed to provide supply from the other direction.

Distributions zone subs take the 66 and bring the voltage down via 1,2 or several (66/11/22) Transformers and put this onto an 11KV or 22kv bar . Connected to this bar are all the distribution circuit breakers. This is where the real business begins as the 3 wire 3 phase HV conductors leave the last zone generally on poles and wires or possibly underground out into the neighbourhood.

The three conductors travel on wooden poles with ceramic or poly insulators keeping the conductor from arcing to the pole.
In general, dry wood is a pretty good insulator and helps insulate the system from the ground.

Up until this point all the Transmission transformers were most likely delta to delta windings, meaning threee conductors in and three conductors out. But now we need to derive three phases and a neutral. The neutral will be a return point for out of balance currents, and more importantly an MEN earthing system.

The three 11/22kv phases enter the primary windings of a pole mounted transformer or a ground level padmount transformer. These three coil windings are internally connected in delta (triangle). There are always fuses or a breaker on the high voltage side leading into the transformer. (Eg expulsive drop out fuses) these can also be used for isolation to allow working on the transformer.

On the secondary side, there are four phases derived from a star configuration in the windings. The centre of the star becomes the neutral which in Australia is tied to the mass of earth beneath the transformer. Each house receives the phases and a neutral. The consumer installation then ties the neutral to an earth at that location as well , resulting in what we call the Multiple Earth Neutral system or MEN. The nominal voltage between phases in Australia is 400 volts and from phase to neutral or ground is 230 volts.

The MEN serves a few functions. For example, if there is a fault to ground eg a live phase to earth then the current will enter via the local earth rod and from there can travel back to the originating transformer via the neutral conductor, thereby activating protection devices faster and with lower power required to do so, therefore limiting fault currents and the level at which the voltage on the earth can rise.

There are other reasons for earthing the network including preventing the build up of static on equipment.

There are fuses on the secondary side but they are fairly slow acting HRC. There is usually a wiring harness on the Transformer that connects to the conductors in both directions and provides fusing and blade switching options.

The lowest set of 4 wires travels from pole to pole and at each premises a single layer XLPE aerial bundle cable (service mains, private service, private mains) is tapped onto the conductors and sent to the point of attachment on the house.

From there we have the consumer mains which run from the point of attachment into the house and down into the switchboard, usually a galvanised box in Australia on the side of the house.

Inside the switchboard is a service fuse which is a HRC for each phase. This then carries through to the meter and onto the customer’s main switch and hot water switch which go out to the customers light and power CBs and thence into your GPOs and lights etc.

Some things to note.

Transformers are measured in KVA. A typical pole mounted transformer may be  100 – 300 KVA. A padmount transformer would be 500 KVA. To work out power we multiply kva by power factor.

Its important to understand our ratios between capacitive, inductive and resistive loads e.g. power factor.
The best power factor is 1.0 –  inductive loads reduce that number and adding capacitors can raise the power factor back up. In Australia, if you are a private householder you do not need to consider power factor and you will not be charged for poor power factor EG. Lots of inductive loads. However, as a large commercial customer, you are expected to have power factor correction, capacitors etc. It’s not as hard as you think. In fact, most appliances often have power factor correction capacitors installed. (ever wonder why that old metal flouro light had a large cap (PFC) to go along with the large inductive ballast)

So what is power factor? Well, let’s say you have a poor power factor. This means that you will be drawing your current and voltage out of phase because a perfect power factor of one means that the current and voltage are in phase and whatever voltage and current you are taking is all we need to supply. However, let’s imagine a load that’s highly inductive. We know that power is volts times amps so If we think about a voltage sine wave then maximum power will be if both the voltage and current sine waves rise together right? (Most volts, most amps = most power) . Anytime that they are out of sync though, the actual true power delivered will be lower.
So the grid is supplying a lot of extra current that isn’t really making it into true power and although it isn’t used, that extra current still causes losses due to the I squared r rule etc in conductors etc.

So why does inductive current lag behind voltage ?
The best way to imagine inductive power is to imagine that as soon as you apply power, the voltage will be instantly measurable on the far terminals of the coil. However, due to lenzs law, the current buildup will be slowed down because as the magnetic field builds up, it will induce an opposite current back into itself which restricts current flow. Therefore the voltage will get there before (or lead) the current..

Why does voltage lag behind current in a capacitor ? The best way to remember this is that as you apply voltage, the current is drawn straight away to begin to fill the capacitor but the voltage on the terminal of the capacitor only starts to rise as the capacitor fills. Therefore, the current leads the voltage in the cas of a capacitive load.

A lot of people ask or assume that the distribution system is antiquated. The reality is that simple devices made of steel and filled with oil are robust against lightning strikes, heavy loading, faults and are forgiving and easily maintained, cleaned and repaired. A complete rebuild of transformer can be undertaken – drain the oil, remove the winding, repaint the tank, change the insulators, replace the winding and fill the tank.

Early protection systems were inverse time mechanical relays basically looking like a standard spinning electrical meter that was driven by a CT and could only start to spin the one time when the CT produces a very large amount of current due to overload and as that disk spins a quarter turn, then a peg placed by a protection technician at a particular point on the disc will physically strike a set of contacts initiating the breaker.

 For example, lets say we have a transformer connected to a 66 kV bar on the primary side of the transformer and an 11KV bar on the secondary side. Upon the 11kv bar is an 11 kV circuit breaker (one of several on the same bar ) feeding a set of overhead conductors. It has a tree fall across the lines clashing the aluminium alloy  conductors together. A large amount of current is drawn raising the proportional CT current feeding the relay disc meter which begins to spin. The disc turns triggering the contacts which will fire the distribution circuit breaker for that feeder. As a redundancy , At the same time, an additional set of contacts starts a timer to trip the breaker coming out of the Transformer onto the 11KV bar. If the feeder breaker does not trip then the breaker feeding the whole bar will trip. When the first breaker trips a set contacts proving that the breaker opened will cancel the timer. This type of cascade protection exist right the way through the network right up to the generators. By placing sectional breakers on the bar, the effects of bar tripping can be limited if it was to occur. Additionally, Transformer faults may trip breakers that isolate them fro a bar and standby transformers automatically switch in to supply the bar. Sometimes many Transformers are connected in parallel to the bar. This has the advantage that they can carry the load of an isolated transformer realtively easily until it is put back into service.

There are many schemes for protection but the most common is overcurrent. Modern electronic relays use moh diagrams which is a kind of voltage current envelope etc to classify the kind of fault occurring. Point to point transmission, where there are no loads, just conductors between zone subs often have a pilot wire which is used for differential protection. One zone sub will say Hey I sent 800 amps at 66 kV. Did you get it? If you didn’t then something must have happened. You better trip. The way this works is the receiving zone sub CT current is sent backwards along the pilot to the sending zone sub. The sending zone sub CT current along with the received pilot wire current are both sent into a balancing CT transformer. Eg they cancel each other out if equal ( Simply one reversed winding)   the output from this Transformer is zero when both CTs are equal, However, if they are out of balance then the out of balance current drives a relay, which trips the breaker. This, of course only works for transmission because there is no current taken from in between the two transmission zones. Distribution protection generally relies on looking for overcurrent , undervoltage or out of balance current between phases.

The status of all contacts is generally marshalled back to a control system at the zone that is commonly known as SCADA. The VT and CT ratios are known, therefore the voltages and currents on conductors are known. So beyond the use for protection eg tripping,  that information is used to determine how loaded transformers are and other general information. Additionally, scada allows remote control of breakers so that system operators can turn breakers on or off or disconnect or reconfigure transformers. 

Many switching operations are carried out daily across the network. Transformers maybe placed in and out of service along with breakers for maintenance reasons or to isolate conductors in the field. In a lot of cases 11KV distribution feeders are laid out in a ring This means that there is generally one open point in the middle of the ring between the two feeders. To isolate a section of line ,the open point is closed temporarily linking the two feeders as one. Then an open point is created at one end of the section requiring isolation and then the second end is opened off at the required location. This provides the minimum disruption to customers.

Some more thoughts to consider…

Generators have governors which adjust the speed of the generator automatically to feed in more fuel as the load makes the 50hz slow down.

Solar has ohms law, eg you need to float your solar voltage higher than the network if you wish to drive current out on to the grid.

Most data and control systems like scada aren’t on internet, and the control systems mostly don’t make the decisions for the relays/breakers eg overcurrent (which is hardwired directly in to the relay) , control systems mostly allow the control room to switch breakers and portions of the network, report and maybe trigger a reclose after 60 seconds or so.

There is also load shedding, this is so that if a supply feeder is overloading you can systematically de-energize feeders to reduce the load and keep the network from crashing.  Eg a black start.

Most coal gennys are between 600 and 1000 Megwatts.

There are some strange gennys like wind turbines and gas powered converted caterpillar engines in shipping containers that can be switched in and out they are about 1Mw each

Warragamba dam was used as a pumped hydro system. e.g, water could be pumped up to the top of the dam during periods of excess electricity and then upon command add an additional 50Mw of power to the network eg heatwaves and cold snaps or breakdowns.

There are also the odd reactor and capacitor banks spread around, but they are usually to deal with odd feeders that may supply an aluminium works for example, they strike HV arcs to melt metals in giant crucibles! This gives the network many spikes that have to be smoothed.

There are so many more features, if you have questions , ask away!!

Leave a comment

Your email address will not be published. Required fields are marked *