A big power plant goes down and other plants have to pick up the load. Load exceeds capacity of remaining plants and they shut down (or breakers blow, etc). Repeat.
Power plants also need energy to start up (black start), and if there’s no grid energy to power those ancillary systems, or if the power plant doesn’t have on-site auxiliary generators to provide black start capability, they’re down until they can get power again from elsewhere.
Base load plants (coal, nuclear) don’t throttle up and down quickly for changing loads. For quick response, we use peaker plants which are typically natural gas powered turbines and can respond quicker (grid batteries are, thankfully, replacing these in some cases).
That’s grossly over-simplified but it’s more or less the gist of it.
Power plants also need energy to start up (black start), and if there’s no grid energy to power those ancillary systems, or if the power plant doesn’t have on-site auxiliary generators to provide black start capability, they’re down until they can get power again from elsewhere.
This is huge, we have massive drills to make sure we can do this, and idle black start plants for just this purpose alongside almost an entire secondary grid for bootstrapping.
It trips me out that many of these plants don’t have APUs for starting themselves up, or that they were designed in such a way that they require utility power to boot up. Like I understand that black starts could have problems with frequency sync with no point of reference, but I can’t imagine that their control system circuits don’t have any form of self-powering redundancy built in to their design. Is there any reason for this?
We didn’t have digital controls when they were designed, so you couldn’t use GPS or atomic clocks to synchronize frequencies, you just needed to have a single source of coordination. Those coils were manually controlled till not long ago.
Now we should be able to use some kind of small gas turbine with a igbt rig for synchronization, much like they do with wind turbines.
People often don’t appreciate how far we’ve come over the past 2 decades, and how utterly manual and brute force we were until very, very recently.
It’s why “smart grids” are talked about a lot recently, even though it doesn’t mean much for the layperson. But for the people actually working in the industry, it matters a lot and can be a huge benefit
They could trivially save billions a year just by adding intelligent load coordination and shifting for evs, you sign up and you save a few pennies a kwh, and in exchange they can steer load away from shortages and towards surplus. It’s 1990s technology.
But management at places like Pge are just jobs programs for Newsom to sell to sew up the 2028 primary, which hilariously backfired for him (though im sure he cut a decent deal).
Most Ev loads are almost instantly dispatchable, it’s an absolute no brainer, even if we aren’t trying v2g, at least control when they charge.
Ah makes sense, I forget how old a lot of that infrastructure is. I do a lot of work for our local ski resort, they have a 12mw generation plant for their snowmaking system that can backfeed the utility if needed, and they’re sync setup was basically two blinking lights that you had to visually time just right to close the switch and pray you didn’t screw up lol.
That is hilarious, but also I can 100% picture it in my head, and it’s not that far off from what the big machines did until recently, and pretty sure some still do.
BTW, there was money both in 2010 and 2021-2022 that was literally supposed to replace all those linkages with something less ancient, even for small things like yours.
My understanding is the money ran out halfway or so, maybe less, and even a lot of the big stuff didn’t get done, they kept fighting on how to do it and the details, everybody wanred standardized on their system, and in the end very little actually happened.
Article doesn’t say, but one plausible scenario:
A big power plant goes down and other plants have to pick up the load. Load exceeds capacity of remaining plants and they shut down (or breakers blow, etc). Repeat.
Power plants also need energy to start up (black start), and if there’s no grid energy to power those ancillary systems, or if the power plant doesn’t have on-site auxiliary generators to provide black start capability, they’re down until they can get power again from elsewhere.
Base load plants (coal, nuclear) don’t throttle up and down quickly for changing loads. For quick response, we use peaker plants which are typically natural gas powered turbines and can respond quicker (grid batteries are, thankfully, replacing these in some cases).
That’s grossly over-simplified but it’s more or less the gist of it.
This is huge, we have massive drills to make sure we can do this, and idle black start plants for just this purpose alongside almost an entire secondary grid for bootstrapping.
Electricity is expensive and hard as hell.
It trips me out that many of these plants don’t have APUs for starting themselves up, or that they were designed in such a way that they require utility power to boot up. Like I understand that black starts could have problems with frequency sync with no point of reference, but I can’t imagine that their control system circuits don’t have any form of self-powering redundancy built in to their design. Is there any reason for this?
We didn’t have digital controls when they were designed, so you couldn’t use GPS or atomic clocks to synchronize frequencies, you just needed to have a single source of coordination. Those coils were manually controlled till not long ago.
Now we should be able to use some kind of small gas turbine with a igbt rig for synchronization, much like they do with wind turbines.
People often don’t appreciate how far we’ve come over the past 2 decades, and how utterly manual and brute force we were until very, very recently.
It’s why “smart grids” are talked about a lot recently, even though it doesn’t mean much for the layperson. But for the people actually working in the industry, it matters a lot and can be a huge benefit
Yeah, but the utility’s management… Holy shit.
They could trivially save billions a year just by adding intelligent load coordination and shifting for evs, you sign up and you save a few pennies a kwh, and in exchange they can steer load away from shortages and towards surplus. It’s 1990s technology.
But management at places like Pge are just jobs programs for Newsom to sell to sew up the 2028 primary, which hilariously backfired for him (though im sure he cut a decent deal).
Most Ev loads are almost instantly dispatchable, it’s an absolute no brainer, even if we aren’t trying v2g, at least control when they charge.
Ah makes sense, I forget how old a lot of that infrastructure is. I do a lot of work for our local ski resort, they have a 12mw generation plant for their snowmaking system that can backfeed the utility if needed, and they’re sync setup was basically two blinking lights that you had to visually time just right to close the switch and pray you didn’t screw up lol.
That is hilarious, but also I can 100% picture it in my head, and it’s not that far off from what the big machines did until recently, and pretty sure some still do.
BTW, there was money both in 2010 and 2021-2022 that was literally supposed to replace all those linkages with something less ancient, even for small things like yours.
My understanding is the money ran out halfway or so, maybe less, and even a lot of the big stuff didn’t get done, they kept fighting on how to do it and the details, everybody wanred standardized on their system, and in the end very little actually happened.
And where would Cuba get this tech from and how would they pay for it?