4 ms·

There was confusion on Twitter how a frequency drop could cause a power outage. It's the other way around: loss of power/a plant leads to a drop in frequency. H

There was confusion on Twitter how a frequency drop could cause a power outage. It's the other way around: loss of power/a plant leads to a drop in frequency. Here's why: Imagine a generator spinning at 50 rpm (for the sake of simplicity). Now if the load on the generator is larger than the power it can provide, it will be slowed down and the frequency goes down. If the load is lower than the power output of the generator though this surplus goes into spinning it up, raising the frequency. There are plants that can more easily vary their output which are used to manage these fluctuation and providing frequency control. A pretty new and fast one is the Hornsdale Power Reserve in Australia, based on Tesla Powerpacks which can react in milliseconds, instead of seconds and minutes.

matthewmacleod 7y ago
I’m not an expert - but I understand that while this is true (generator loss causes frequency to drop), this drop in frequency can in turn trigger actual “power outages” for users. Automated systems respond to sagging frequency by shedding loads - reducing the demand for power, so that the frequency can increase and a new equilibrium can be achieved. The goal is to avoid an uncontrolled cascade across the power network as demand outstrips supply; this is substantially harder to manage than a short interruption. The UK does have battery-powered frequency management services, but it looks like it was quite a large and sudden drop in generation capacity that caused this deviation, and it took several minutes before it was back to normal. Not totally sure what it was yet, but sounds like a gas turbine generator suddenly tripped, along with a nearby offshore wind farm, taking out about a gigawatt of capacity that normal frequency stabilisation couldn’t handle, resulting in load shedding.
- noir_lord 7y ago
  Weather has been unusual as hell today, sunny one minute 50mph winds and torrential rain the next.
Scoundreller 7y ago
Or you could pay people to put a device on their air conditioner or water heater or car charger that’ll back-off when the frequency dips. Conservation seems way more sensible than building new and advanced capacity.
- ascorbic 7y ago
  They do that too, with large commercial users with loads that can easily be shed for short periods such as blast furnaces, aluminium smelters, refrigeration systems etc. With car charging, they're doing one better, and can feed power back into the grid: https://www.ovoenergy.com/electric-cars/vehicle-to-grid-charger https://www.ovoenergy.com/electric-cars/vehicle-to-grid-char...
  - Scoundreller 7y ago
    Agreed, but we now have the technology to do these at much smaller scales, which, in aggregate, account for a lot of the load. It just bother me when we focus on big industrial solutions. We had the right idea with PV solar: why build a giant farm owned by ??? when we already have these structures everywhere tilted toward the sun and grid-connected.
    - londons_explore 7y ago
      > Agreed, but we now have the technology to do these at much smaller scales, which, in aggregate, account for a lot of the load. A lot of companies have this capacity, but are waiting for the right environment to deploy it. Smart fridges, aircons, nest thermostats, electric cars with Wifi, etc. They have all been built with the capability to load-shed or even power the grid at a moments notice, but the manufacturers refuse to just do that for free - they want to be paid. Many of those products were even sold at a loss because the companies expected to be paid by electricity networks for their fast-response stabiisation services. Today, they aren't allowed to participate in the power markets, and anyway, markets are settled on a half-hourly basis rather than the 1-second basis that would earn them the most money and provide the most reliability for the grid. This is all a politics problem within electricity regulators (who would much rather stick with the status quo, leaving some power stations underutilized, rather than pay these 'unreliable' web 2.0 companies). Nearly none of it is a technical problem.
MisterTea 7y ago
Yup. The synchronous speed of a motor is directly proportional to the frequency and number of poles in the motor. The formula RPM = 120f/p is used to solve for speed given f in Hz and p is the number of poles. Poles being one of the N-S halves in a magnetic pole pair. So the minimal number of poles in an AC induction motor is 2 and increases by 2. Typically, most AC induction motors are of the 2 and 4 pole variety as well as the less common but usually available 6 pole. The same goes for generators. The more poles, the slower it turns but torque increases proportionally. In ye olde days frequency was converted using motor generators. A 50<->60 hz system matches up at 600 RPM so a 10 pole on the 50Hz side and 12 on the 60Hz side (complex and expensive designs). Using that formula, a 2 pole generator running at a synchronous speed of 3000 RPM will produce 50Hz. Using that 50Hz to power a 4 pole squirrel cage motor will cause it to run at just under synchronous speed, usually 4-5% below synchronous speed at full load (it will spin slightly faster, towards synch under no load). So about 1450 RPM. The reduction in speed is called slip and is what generates torque as the rotor has no magnetic field of its own, it depends on slip to generate the counter field to push the rotor along. Synchronous motors avoid slip by having their own magnetic field from permanent magnets or a wound armature on the rotor fed DC from slip rings or rotary transformer. Coincidently, a synchronous motor and AC generator or alternator are almost the same machine. The basic idea is a magnetic field mechanically spun around pushes electrons around in the stator windings. Sort of like a propeller pushing through water, the magnetic field lines push and pull electrons along the wires. If the field is too weak the voltage will sag as load increases. You can visualize this as weak field = propeller made of rubber, flopping about and can't push much, full strength = made of steel. If the mechanical power source is weak or overloaded, the shaft speed will drop as well as voltage. You are literally transmitting the generators rotational speed along the wires and motors follow. Adding generators means they have to be in phase so must be brought up to speed AND phased in before being connected to the grid. This is called synchronizing. So thanks to my long winded motor/generator post you can see how a large load can drag down frequency.

matthewmacleod 7y ago

I’m not an expert - but I understand that while this is true (generator loss causes frequency to drop), this drop in frequency can in turn trigger actual “power outages” for users. Automated systems respond to sagging frequency by shedding loads - reducing the demand for power, so that the frequency can increase and a new equilibrium can be achieved. The goal is to avoid an uncontrolled cascade across the power network as demand outstrips supply; this is substantially harder to manage than a short interruption. The UK does have battery-powered frequency management services, but it looks like it was quite a large and sudden drop in generation capacity that caused this deviation, and it took several minutes before it was back to normal. Not totally sure what it was yet, but sounds like a gas turbine generator suddenly tripped, along with a nearby offshore wind farm, taking out about a gigawatt of capacity that normal frequency stabilisation couldn’t handle, resulting in load shedding.

noir_lord 7y ago
Weather has been unusual as hell today, sunny one minute 50mph winds and torrential rain the next.

Scoundreller 7y ago

Or you could pay people to put a device on their air conditioner or water heater or car charger that’ll back-off when the frequency dips. Conservation seems way more sensible than building new and advanced capacity.

ascorbic 7y ago
They do that too, with large commercial users with loads that can easily be shed for short periods such as blast furnaces, aluminium smelters, refrigeration systems etc. With car charging, they're doing one better, and can feed power back into the grid: https://www.ovoenergy.com/electric-cars/vehicle-to-grid-charger https://www.ovoenergy.com/electric-cars/vehicle-to-grid-char...
- Scoundreller 7y ago
  Agreed, but we now have the technology to do these at much smaller scales, which, in aggregate, account for a lot of the load. It just bother me when we focus on big industrial solutions. We had the right idea with PV solar: why build a giant farm owned by ??? when we already have these structures everywhere tilted toward the sun and grid-connected.
  - londons_explore 7y ago
    > Agreed, but we now have the technology to do these at much smaller scales, which, in aggregate, account for a lot of the load. A lot of companies have this capacity, but are waiting for the right environment to deploy it. Smart fridges, aircons, nest thermostats, electric cars with Wifi, etc. They have all been built with the capability to load-shed or even power the grid at a moments notice, but the manufacturers refuse to just do that for free - they want to be paid. Many of those products were even sold at a loss because the companies expected to be paid by electricity networks for their fast-response stabiisation services. Today, they aren't allowed to participate in the power markets, and anyway, markets are settled on a half-hourly basis rather than the 1-second basis that would earn them the most money and provide the most reliability for the grid. This is all a politics problem within electricity regulators (who would much rather stick with the status quo, leaving some power stations underutilized, rather than pay these 'unreliable' web 2.0 companies). Nearly none of it is a technical problem.

MisterTea 7y ago

Yup. The synchronous speed of a motor is directly proportional to the frequency and number of poles in the motor. The formula RPM = 120f/p is used to solve for speed given f in Hz and p is the number of poles. Poles being one of the N-S halves in a magnetic pole pair. So the minimal number of poles in an AC induction motor is 2 and increases by 2. Typically, most AC induction motors are of the 2 and 4 pole variety as well as the less common but usually available 6 pole. The same goes for generators. The more poles, the slower it turns but torque increases proportionally. In ye olde days frequency was converted using motor generators. A 50<->60 hz system matches up at 600 RPM so a 10 pole on the 50Hz side and 12 on the 60Hz side (complex and expensive designs). Using that formula, a 2 pole generator running at a synchronous speed of 3000 RPM will produce 50Hz. Using that 50Hz to power a 4 pole squirrel cage motor will cause it to run at just under synchronous speed, usually 4-5% below synchronous speed at full load (it will spin slightly faster, towards synch under no load). So about 1450 RPM. The reduction in speed is called slip and is what generates torque as the rotor has no magnetic field of its own, it depends on slip to generate the counter field to push the rotor along. Synchronous motors avoid slip by having their own magnetic field from permanent magnets or a wound armature on the rotor fed DC from slip rings or rotary transformer. Coincidently, a synchronous motor and AC generator or alternator are almost the same machine. The basic idea is a magnetic field mechanically spun around pushes electrons around in the stator windings. Sort of like a propeller pushing through water, the magnetic field lines push and pull electrons along the wires. If the field is too weak the voltage will sag as load increases. You can visualize this as weak field = propeller made of rubber, flopping about and can't push much, full strength = made of steel. If the mechanical power source is weak or overloaded, the shaft speed will drop as well as voltage. You are literally transmitting the generators rotational speed along the wires and motors follow. Adding generators means they have to be in phase so must be brought up to speed AND phased in before being connected to the grid. This is called synchronizing. So thanks to my long winded motor/generator post you can see how a large load can drag down frequency.