Matthias Fripp
@matthiasfripp.org
Finding the best ways to integrate all the energy things—wind, solar, geothermal, nuclear, hydrogen, transport, industry, buildings. Creator of Switch model—software to help with this. Global Policy Research at energyinnovation.org. Opinions are my own.
I used to tell my students that fossil fuels are a unique endowment of stored energy. The fact that we can store and move them around to have energy in even the trickiest situations is rare and special. But here we are, just burning them off willy nilly. And wrecking the climate while we’re at it.
October 14, 2025 at 12:29 AM
I used to tell my students that fossil fuels are a unique endowment of stored energy. The fact that we can store and move them around to have energy in even the trickiest situations is rare and special. But here we are, just burning them off willy nilly. And wrecking the climate while we’re at it.
If Florida did actually manage to prevent contrails, that would be pretty good for the climate. But I think that might annoy some of the same people. rmi.org/aviation-con...
Aviation Contrails: What We Know — and What We Don’t — about This Warming Phenomenon - RMI
Our new comprehensive overview report shines a light on the science and potential solutions to aviation’s biggest non-CO2 climate impact.
rmi.org
October 9, 2025 at 12:45 AM
If Florida did actually manage to prevent contrails, that would be pretty good for the climate. But I think that might annoy some of the same people. rmi.org/aviation-con...
I was worried about the case of grid forming inverters downstream of a partial fault in the distribution network. If they deliver a lot of current, could that prevent upstream relays from tripping? (e.g. if the fault runs at low normal voltage, with half the current coming from each side)
September 11, 2025 at 5:29 PM
I was worried about the case of grid forming inverters downstream of a partial fault in the distribution network. If they deliver a lot of current, could that prevent upstream relays from tripping? (e.g. if the fault runs at low normal voltage, with half the current coming from each side)
Thanks for the update! I’m tied up with some other work right now but will check it out as soon as I can.
September 4, 2025 at 9:54 PM
Thanks for the update! I’m tied up with some other work right now but will check it out as soon as I can.
It’s because these work together to create a functioning grid with stable voltage and frequency, rather than needing to hook into a grid controlled by other devices (as grid-following inverters do). This means they can replace spinning generators which have traditionally filled this role.
August 26, 2025 at 5:33 AM
It’s because these work together to create a functioning grid with stable voltage and frequency, rather than needing to hook into a grid controlled by other devices (as grid-following inverters do). This means they can replace spinning generators which have traditionally filled this role.
There’s already a pretty well established technology called power line networking that sends data at higher frequencies using power lines as the carrier. I think it has trouble getting across transformers though. It should work on DC too in principle, but wouldn’t get past the voltage converters.
August 23, 2025 at 7:17 PM
There’s already a pretty well established technology called power line networking that sends data at higher frequencies using power lines as the carrier. I think it has trouble getting across transformers though. It should work on DC too in principle, but wouldn’t get past the voltage converters.
What kind of failures do you mean? This is already done fairly routinely, and as far as I know, it works as advertised: transfers the scheduled amount of power from one grid to the other.
August 23, 2025 at 7:12 PM
What kind of failures do you mean? This is already done fairly routinely, and as far as I know, it works as advertised: transfers the scheduled amount of power from one grid to the other.
It’s basically just a clutch between the turbine and the generator. Pretty cheap when you build the plant, but seems to be expensive to add after the fact—maybe more than a new, freestanding synchronous condenser. Statcoms and batteries are getting so cheap they may win anyway.
August 23, 2025 at 4:49 PM
It’s basically just a clutch between the turbine and the generator. Pretty cheap when you build the plant, but seems to be expensive to add after the fact—maybe more than a new, freestanding synchronous condenser. Statcoms and batteries are getting so cheap they may win anyway.
in an all-DC grid, you’d only have voltage. That might make things more flickery (though DC devices can overcome that). It might also make protection more difficult in the transmission network, which sometimes use voltage-vs-frequency effects to distinguish between faults and high normal loads.
August 23, 2025 at 4:44 PM
in an all-DC grid, you’d only have voltage. That might make things more flickery (though DC devices can overcome that). It might also make protection more difficult in the transmission network, which sometimes use voltage-vs-frequency effects to distinguish between faults and high normal loads.
But I suspect an all-DC grid would lose something by having only one measure of grid health (voltage) instead of two for AC (voltage and frequency). In AC grids, frequency naturally sends power from where it is produced to where it is needed and is a signal to produce more. V signals line faults.
August 23, 2025 at 4:44 PM
But I suspect an all-DC grid would lose something by having only one measure of grid health (voltage) instead of two for AC (voltage and frequency). In AC grids, frequency naturally sends power from where it is produced to where it is needed and is a signal to produce more. V signals line faults.
But a mainly AC grid is actually a really good idea. Partly because it is cheap and reliable to change voltage between dangerous, efficient levels (transmission) and safe, inefficient levels (in homes). Partly because of path dependency (your household wiring and switches are not engineered for DC).
August 23, 2025 at 4:44 PM
But a mainly AC grid is actually a really good idea. Partly because it is cheap and reliable to change voltage between dangerous, efficient levels (transmission) and safe, inefficient levels (in homes). Partly because of path dependency (your household wiring and switches are not engineered for DC).
But HVDC (or other power-electronics cousins) is also coming into some interest for shorter links, if you want to control how much power goes down a particular path. Otherwise AC power follows “the path of least resistance” and may overload one line before making full use of a neighbor.
August 23, 2025 at 4:44 PM
But HVDC (or other power-electronics cousins) is also coming into some interest for shorter links, if you want to control how much power goes down a particular path. Otherwise AC power follows “the path of least resistance” and may overload one line before making full use of a neighbor.
HVDC is the first choice for lots of power over long distances or underground or underwater, where AC behavior of the line prevents using the full capacity of the line. Also for links between separate AC grids (eastern US, western US, Texas).
August 23, 2025 at 4:44 PM
HVDC is the first choice for lots of power over long distances or underground or underwater, where AC behavior of the line prevents using the full capacity of the line. Also for links between separate AC grids (eastern US, western US, Texas).
(The mechanism I had in mind is that the solar sees voltage collapse, so prioritizes V support over real power delivery. That causes a frequency drop. But meanwhile the battery sees the V and freq drop and prioritizes voltage support and reduces charging. So both V and freq are held OK.)
August 23, 2025 at 4:21 PM
(The mechanism I had in mind is that the solar sees voltage collapse, so prioritizes V support over real power delivery. That causes a frequency drop. But meanwhile the battery sees the V and freq drop and prioritizes voltage support and reduces charging. So both V and freq are held OK.)
That’s probably enough to get us through. Or if not, extra voltage control could come from other devices like HVDC stations, flexible AC transmission devices or backup thermal plants split to run as synchronous condensers most of the year.
August 23, 2025 at 4:21 PM
That’s probably enough to get us through. Or if not, extra voltage control could come from other devices like HVDC stations, flexible AC transmission devices or backup thermal plants split to run as synchronous condensers most of the year.
In theory, if a solar array and battery both back off on charging during faults then you could have more current available for voltage control. In a very high-renewable system, solar might be 2.5x load and battery 1.5x so then 4.3x load is avail for voltage control while still carrying normal load.
August 23, 2025 at 4:21 PM
In theory, if a solar array and battery both back off on charging during faults then you could have more current available for voltage control. In a very high-renewable system, solar might be 2.5x load and battery 1.5x so then 4.3x load is avail for voltage control while still carrying normal load.
Grid-forming inverters also provide this voltage control service. But the quadrature sum of their current for real power output and voltage control is capped around 1.2x the inverter rating. So if they are producing 100% real power, they may not be able to produce much current for voltage control.
August 23, 2025 at 4:21 PM
Grid-forming inverters also provide this voltage control service. But the quadrature sum of their current for real power output and voltage control is capped around 1.2x the inverter rating. So if they are producing 100% real power, they may not be able to produce much current for voltage control.
Traditional generators adjust current continuously to maintain voltage at their point of connection. They can also deliver several times their rated current for a few seconds trying to sustain voltage during a fault. That rush is how circuit breakers detect the fault.
August 23, 2025 at 4:21 PM
Traditional generators adjust current continuously to maintain voltage at their point of connection. They can also deliver several times their rated current for a few seconds trying to sustain voltage during a fault. That rush is how circuit breakers detect the fault.
We have known how to reach net-zero for years; lack of AI was not a barrier.
August 23, 2025 at 6:06 AM
We have known how to reach net-zero for years; lack of AI was not a barrier.
- How does black start work for systems with lots of grid-forming inverters? Does the system operator instruct them to come on one by one? Are there communication standards for that?
August 22, 2025 at 8:09 PM
- How does black start work for systems with lots of grid-forming inverters? Does the system operator instruct them to come on one by one? Are there communication standards for that?
- Is anyone working on inverters for batteries that will provide virtual inertia (and possibly droop and AGC response) while _charging_ as well as discharging? This should enable provision of these services for free—with no extra cycling—from batteries that are already there for day-night balancing.
August 22, 2025 at 8:09 PM
- Is anyone working on inverters for batteries that will provide virtual inertia (and possibly droop and AGC response) while _charging_ as well as discharging? This should enable provision of these services for free—with no extra cycling—from batteries that are already there for day-night balancing.