All the different ways to increase the efficiency of a nuclear power plant

[mrvn]

Which means the burner inserter that refuel the ejection inserter will do a swing every 40 days, which means it will need to confiscate one fuel to refuel itself every:

17 285.7143 x 3 457 142.86= 59 759 183 772.245 sec, 5.9E10 sec, that's 691 657 days, or 1894 years.

make an independent power circuit for internal usage. Power it by one solar panel then backup by one battery. This lasts forever and can feed inserters and pumps for 4-8 block of cores.
It is way cheaper than fill half belt with fuel.

An accumulator can drive 22.4 yellow inserters. At two inserters per reactor minimum, that's 11 reactors you can handle, but it also must be a pump free design. A solar panel provides an averaged power supply of 42kW and therefore can drive, at most, one pump continuously.

If you use a triple inserter setup for a holding chest, you can only put 7 reactors on that same accumulator -- again, assuming no pump. Each pump added to either system reduces the number of inserters that can operate at peak by slightly more than two, i.e. you lose another reactor.

If you use fast inserters, which is my own preference for other reasons, you can only operate 6.45 per accumulator, which is two or three cores tops.

In a 2xN layout those accumulator fields are somewhat awkward to manage, since if you put them internal to the reactor you stretch your material flows (water/steam/heat), which harms throughput, but if you put them outside of the reactor, you need to be very careful about not crossing power grids. My own preference would be to power 'em with a bootstrap plant of some sort, I think, since it'd be more compact.

Use the right number of steam turbines to power the inserters and pumps internally with their own steam storage tank. And then add solar panels next to the reactor to make up the power you loose for the internal steam turbines. Same result.

[quyxkh]

Really, you only need one solar panel to slowwwwwly drive one nuc's feed pumps and loading inserters, everything else will cascade-start.

But 4.4GW in 4 2×4 ploppable plants requires 160 uranium ore per minute unboosted, and I count 6M within 1km of the spawn point. That's a *month's* worth of fuel for a modest 1kspm megabase, and if you're not at 500% mining prod after a week of that and 1000% after a month I'm really not sure what sort of efficiency you're after.

[foamy]

Really, you only need one solar panel to slowwwwwly drive one nuc's feed pumps and loading inserters, everything else will cascade-start.

Only if you do some fancy grid prioritization trickery to make the plant power itself over doing anything else, a simpler method of which is to tap into the plant's own turbines and then control whether that internal grid connects out.

And also to avoid pumps completely because they're 100% unnecessary.

[gGeorg]

Really, you only need one solar panel to slowwwwwly drive one nuc's feed pumps and loading inserters, everything else will cascade-start.

Only if you do some fancy grid prioritization trickery to make the plant power itself over doing anything else, a simpler method of which is to tap into the plant's own turbines and then control whether that internal grid connects out.

And also to avoid pumps completely because they're 100% unnecessary.

In case of stand alone plant look at my Perfect clover leaf plant, two power grids safely feeding inserters and pumps.
In case of 2N, well the edge cores are always speciial. So as quyxkh says, you only need solar safety in the first edge block of cores, when it starts then cascade effect power up everything else. It assumes your book has 2 blueprints, The first block of cores, nTH block of cores.

[mrvn]

Really, you only need one solar panel to slowwwwwly drive one nuc's feed pumps and loading inserters, everything else will cascade-start.

Only if you do some fancy grid prioritization trickery to make the plant power itself over doing anything else, a simpler method of which is to tap into the plant's own turbines and then control whether that internal grid connects out.

And also to avoid pumps completely because they're 100% unnecessary.

I think he mend with an internal power grid for the inserters. 60kW during the day will get a fuel cell loaded eventually even with a large plant. And once you produce steam the internal power grid powers itself. That's how I self start my reactor blueprint too.

[mmmPI]

You want the chests anyway so that fuel insertion happens as rapidly as possible when triggered and doesn't try to pull from a moving belt (and therefore introduce potential desyncs)

The belt has had at least 200s since the last trigger to return to a backed-up state. It isn't moving.

That's a lot of fuel cells to produce to fill the belt before the reactor can run safely.

Hey i tried to combined both world !

Pulling from a static belt, but without buffer.

Doing so i realised one drawback or lack of efficiency that i haven't seen measured yet which surprised me!

We take for granted that a cell burn every 200 seconds right ? that's 12000 ticks. But if we use,as trigger for fuel,the ejection inserter, there is a small amount of time during which the inserter swing to get the used fuel cell from the reactor, during this time the reactor isn't burning anything, even worse, when the used fuel cell is finally grabbed after a 180° swing, then the feeding inserter is triggered and need to also do a 180° swing, to refuel. That's a fuel 360° swing of time loss.

Given that a (blue) fast inserter has a swing speed of 864°/sec, it should take a full 25 ticks to happen, which means every 12025 ticks the reactor is only burny things during 12000, that makes it an efficency of 99.79%.

Interestingly enough for me to mention a 2X2 power plant should produce 480 MW in perfect conditions, which would power exactly 1000 beacon. If we account for the inefficiency just mentionned, over long period of time the 4 cores of a 2x2 power plant powering 997 beacon should heat up to 1000°, and if powering 998 beacon, i'm not patient enough to tell, but my thing lose 0.4° on the coldest point every 12026 ticks.

Here is picture :

bufferlessbelted.jpg (538.34 KiB) Viewed 5585 times

As you can see, no pipes , no tanks, tons of heat pipes. There's too many turbines but given how they are connected it doesn't matter, it was to make it look nicer.

The 4 cells visible at the bottom are dispatch on demand, 12 are stored for safety in a chest that's it.

Combinators and feeding inserters have their own inner solar grid, it was done when i switch away from burner inserter to have faster swing for refuel. And also because the circuitry is very vulnerable to brown-out as i noticed a few times.

The giant belt loop around is the clock, there is a red science in the inner lane somewhere, every 3 lap the refuel occur, which sucks because 12000/3 is 4000, but 12025 or 12026 doesn't divide by 3 meaning my timer can't work with 3 laps. I started with calculations and tricks to achieve a perfect 4000 tick loop i thought it would look good, then fiddle around to adapt and figure out about the inserter swing delay, then stopped developping it so i don't know what would happen if the refuel is set up to happen every 4 or 5 laps. Then is a constant for this, i wanted to have different regime of power available but eh another time.

It still manage to power 997 beacon for hours without the inaccuracies becoming visible i don't know if/when it would start to cause problem, i'm pretty sure i made the loop 1 or 2 tick too slow, and not too fast.

For the record i used yellow belts on the corners because i calculated that it takes an item on the inner curve 13.25 ticks to cross the belt, which is convinient because you get rid of the decimal right away. The outer corner is 36.875 ticks. and the straight belt is 32 ticks for the yellow. 1/2 of those value for red belts i guess . Above the solar pannel are things that stop an item only 1 tick i added to fine tune the timing.

Here is the save :

Efficient central 6.zip

(4.45 MiB) Downloaded 240 times

It's not very efficient and there is no blueprint because i don't want/ can't try and explain the initialization, it took several many tries.

[mrvn]

Most of this thread uses efficiency to mean best MW per fuel cell. The goal is to not waste fuel by overheating the reactor when not all power is used. So the little extra 25 ticks gap between refuels becomes irrelevant because we never run the reactor at 100% output.

If you want a possible maximum output then you can use a timer instead of the spend fuel cell. Set the timer to exactly 12000 ticks and then refuel if steam is low. At max power all the steam produced will get used up. So every 12000 ticks the steam will still be low and a new fuel cell gets inserted. Uninterrupted fuel supply for the reactor cores.

But really, when ever do you have such a smooth power consumption? Normaly you have inserters and assemblers and chemical plants and science labs in there that stutter. Nothing will run 100% of the time. So consumption will have spikes that you have to smooth out with either accumulators or by running the reactor at less than 100% on average.

PS: For every 480 reactor cores those 25 lost ticks mean you need an extra reactor.

[mmmPI]

Most of this thread uses efficiency to mean best MW per fuel cell. The goal is to not waste fuel by overheating the reactor when not all power is used. So the little extra 25 ticks gap between refuels becomes irrelevant because we never run the reactor at 100% output.

That was to put images on the clocked-ejection and also a proof of concept that belt feeding a nuke plant doesn't have to mean loads of buffer since i don't have them anymore for the plant i mentionned earlier!

The little 25 ticks extra are something worth mentionning imo as it impacts how to use the clock while not being something noticeable at all on other designs.

But really, when ever do you have such a smooth power consumption? Normaly you have inserters and assemblers and chemical plants and science labs in there that stutter. Nothing will run 100% of the time. So consumption will have spikes that you have to smooth out with either accumulators or by running the reactor at less than 100% on average.

Testing at full power during hours was the way to test this particular design where i wasn't sure that many pipes could still carry enough heat to the corners, it does if you consume equal or less than 997 beacons, which is due to the delay on inserter, not a heat pipe bottleneck.

Supposedly i was to implement different refuel timer possible based on amount of lap of the clock, in order to improve MW per fuel cell when under max regime but i stopped realising the belt clock wasn't tick-perfect.

Still this power plant can handle spikes and stutter around 480MW, the amount of excess turbines means that if you are not at max power for a while, steam will fill up all 96 turbines, and instead of a steady 480MW you could spike at 558 MW.

On the other hand, if you don't use all 480MW, the heat pipes will store some energy, even if the heat pipes on the far end are not supposed to reach 950° on normal operation due the their distance to the core, if you don't consume electricity, that's the only place heat has to go.

If you want to slow down refuel to save fuel, you can replace red belts around the plant to yellow belts haha, if you are onto 3 laps for refuel you delay it by 48 ticks for every such replacement of straight belt, but i haven't tested it

[mrvn]

Most of this thread uses efficiency to mean best MW per fuel cell. The goal is to not waste fuel by overheating the reactor when not all power is used. So the little extra 25 ticks gap between refuels becomes irrelevant because we never run the reactor at 100% output.

That was to put images on the clocked-ejection and also a proof of concept that belt feeding a nuke plant doesn't have to mean loads of buffer since i don't have them anymore for the plant i mentionned earlier!

The little 25 ticks extra are something worth mentionning imo as it impacts how to use the clock while not being something noticeable at all on other designs.

But really, when ever do you have such a smooth power consumption? Normaly you have inserters and assemblers and chemical plants and science labs in there that stutter. Nothing will run 100% of the time. So consumption will have spikes that you have to smooth out with either accumulators or by running the reactor at less than 100% on average.

Testing at full power during hours was the way to test this particular design where i wasn't sure that many pipes could still carry enough heat to the corners, it does if you consume equal or less than 997 beacons, which is due to the delay on inserter, not a heat pipe bottleneck.

Supposedly i was to implement different refuel timer possible based on amount of lap of the clock, in order to improve MW per fuel cell when under max regime but i stopped realising the belt clock wasn't tick-perfect.

Still this power plant can handle spikes and stutter around 480MW, the amount of excess turbines means that if you are not at max power for a while, steam will fill up all 96 turbines, and instead of a steady 480MW you could spike at 558 MW.

On the other hand, if you don't use all 480MW, the heat pipes will store some energy, even if the heat pipes on the far end are not supposed to reach 950° on normal operation due the their distance to the core, if you don't consume electricity, that's the only place heat has to go.

If you want to slow down refuel to save fuel, you can replace red belts around the plant to yellow belts haha, if you are onto 3 laps for refuel you delay it by 48 ticks for every such replacement of straight belt, but i haven't tested it

Sure. I do that too but with radars. Next time I think I will try beacons too. No point having the map grow huge by radars discovering chunks.

Replacing red belts with yellow ones means the max power is reduced. Better to add steam tanks. Can you stop the belt when the steam tank is full?

[gGeorg]

We take for granted that a cell burn every 200 seconds right ? that's 12000 ticks. But if we use,as trigger for fuel,the ejection inserter, there is a small amount of time during which the inserter swing to get the used fuel cell from the reactor, during this time the reactor isn't burning anything, even worse, when the used fuel cell is finally grabbed after a 180° swing, then the feeding inserter is triggered and need to also do a 180° swing, to refuel. That's a fuel 360° swing of time loss.

Given that a (blue) fast inserter has a swing speed of 864°/sec, it should take a full 25 ticks to happen, which means every 12025 ticks the reactor is only burny things during 12000, that makes it an efficency of 99.79%.

That is valid point. It is probably biggest waste in common designs.

Althou, the used_cell unload inserter set for Read_hand_content + Pulse send signal one tick after the cell is used. e.g. you can safely use yellow inserter. Because swing speed for unload doesnt matter.
Then you need few ticks for logic (3 in my case)
Then you need 14 ticks to load core from box (by fast or filter inserter)
So the cold interval is 17 ticks include logic processing. (Not 25. You counted one more swing but forgot add logic)


You bring me an idea I could improve my "Perfect cloverleaf plant", make a logic which offset this delay.
I think I can add a counter for 200s since load, then insert another cell on time.

[gGeorg]

Testing at full power during hours was the way to test this particular design where i wasn't sure that many pipes could still carry enough heat to the corners, it does if you consume equal or less than 997 beacons, which is due to the delay on inserter, not a heat pipe bottleneck.

Supposedly i was to implement different refuel timer possible based on amount of lap of the clock, in order to improve MW per fuel cell when under max regime but i stopped realising the belt clock wasn't tick-perfect.

Althou heat pipes looks like good heat capacitor, their delay is deadly. I guess it takes about 30s to re-start plant to full output. For this reason I dont like this desing to small reacotrs which need to be quick and efficient. That is why I rather use design with short pipes as possible. On top, heat units has VERY similar UPC requirements as steam. On top, one heat pipe is way more expensive than one storage tank.

[mmmPI]

Replacing red belts with yellow ones means the max power is reduced. Better to add steam tanks. Can you stop the belt when the steam tank is full?

True for max power reduced, your suggestions would allow self-regulation to a belt-clock design power plant. But i'm not going to try and implement it on a 3 lap thing, because it makes it more annoying if you want to reduce delay by a certain precise amount and you would need to stop the belt only one the last lap adding unecessary things to something that has a lot already. I will just redesign a new one with the new information in mind next game

Maybe using 2 different small loops, one with say 100 laps, and another one smaller with only 1 lap to fine tune timings like accounting for the inserter swing time and the logic.

Althou, the used_cell unload inserter set for Read_hand_content + Pulse send signal one tick after the cell is used. e.g. you can safely use yellow inserter. Because swing speed for unload doesnt matter.
Then you need few ticks for logic (3 in my case)
Then you need 14 ticks to load core from box (by fast or filter inserter)
So the cold interval is 17 ticks include logic processing. (Not 25. You counted one more swing but forgot add logic)

I didn't realised this. Using yellow inserter could save energy !

how did you get the 14 ticks to load ? did you measured it ? I mathed with 864°/s as it's the ingame written value , it is 14.4° per tick, and then 180/14.4 would be 12.5 tick. I guess if you have ingame measurement they are more to be trusted than my math, maybe there is a delay and one should account for more than just roation time, i wonder what happens to the decimal, wouldn't it give have alternating results each time with a 0.5 tick ?

I am not sure that the way the fuel is stopped in front of the inserter on the belt gives an equal angle to all 4 inserter swing on my design which now that i think of it, maybe source of another ineficiency, it appears that 1 reactor is fed 1 tick earlier than the others, if not caused by the circuit, it could be that 1 injecting inserter has a shorter swing, say 13 ticks, others 14 ticks. It would then cause the adjacency bonuses to be lost during 1 tick ! Which would impact MW per fuel cell

You bring me an idea I could improve my "Perfect cloverleaf plant", make a logic which offset this delay.
I think I can add a counter for 200s since load, then insert another cell on time.

i don't think there exist one "perfect" plant, i think the design should adapt to the situation in which it's used

you could measure when the injecting inserter doesn't hold fresh fuel cell anymore, that's after the swing time.

Althou heat pipes looks like good heat capacitor, their delay is deadly. I guess it takes about 30s to re-start plant to full output. For this reason I dont like this desing to small reacotrs which need to be quick and efficient.

If you are producing under max capacity, the heatpipes will stay hot, say 750°, it's instant restart, if full capacity, you don't restart, you are always at the lowest temperature it gets when fuel is burning.

It would take those 30 sec to re-start ( probably more tbh ) but only if you have something like a fuel shortage, then the temperature will go down to 500° everywhere, then it woud take time for the heat flow to start reaching the 2 corners again. It's not normal operation, but yes it wouldn't be reactive in bad scenario.

The good thing with heatpipes though is that they glow when hot, and the heat doesn't dissipate over time. Which means they create light without consuming energy, which not only is quite fantastic on the physical plan, but also looks incredible. For this reason i like playing with their shape, which forces me to well manage my fuel supply .

[SoShootMe]

I guess if you have ingame measurement they are more to be trusted than my math, maybe there is a delay and one should account for more than just roation time, i wonder what happens to the decimal, wouldn't it give have alternating results each time with a 0.5 tick ?

FWIW, inserting new cells from a chest triggered by a pulse from the inserter removing a used cell, I measured a cycle time of 12015 ticks using the blueprint from https://wiki.factorio.com/Inserters#Inserter_Throughput. However, I counted (by single-stepping in /editor) 16 ticks with the reactor showing no fuel, from the tick the used cell appears up to (not including) the tick the new cell is dropped; I can't explain the discrepancy.

The 12.5 ticks for a 180 degree rotation at 864 degrees/s seems to round up to 13 ticks each time. That makes sense because the inserter returns to an idle state between each operation (it always starts on a tick, and hasn't got far enough until the 13th tick), but it might be different if it is continuously active.

[mrvn]

The inserter head has to move from the idle position to where the fuel is getting picked up. With belts it might have to chase the fuel costing extra ticks.

Shouldn't happen with a chest though I thought. But maybe there is some delay like 1 tick to evaluate circuit signal, 1 tick to pick thing from chest.

[gGeorg]

FWIW, inserting new cells from a chest triggered by a pulse from the inserter removing a used cell, I measured a cycle time of 12015 ticks using the blueprint from https://wiki.factorio.com/Inserters#Inserter_Throughput. However, I counted (by single-stepping in /editor) 16 ticks with the reactor showing no fuel, from the tick the used cell appears up to (not including) the tick the new cell is dropped; I can't explain the discrepancy.

0. cell is used
1. inserter_unload-er get noticed the source and grab and send signal
2. inserter_loader processing signal
3. inserter_loader grab the new cell
4-15. arm swing
16. inserter_loader loading the new cell

[gGeorg]

I didn't realised this. Using yellow inserter could save energy !

Sure. And build price too.

how did you get the 14 ticks to load ? did you measured it ?

Yes. Use editor, tab Tools, icon Watches - then you get tools for tick measrment. Priceless for any precise or combinator creations.

I am not sure that the way the fuel is stopped in front of the inserter on the belt gives an equal angle to all 4 inserter swing on my design which now that i think of it, maybe source of another inefficiency, it appears that 1 reactor is fed 1 tick earlier than the others, if not caused by the circuit, it could be that 1 injecting inserter has a shorter swing, say 13 ticks, others 14 ticks. It would then cause the adjacency bonuses to be lost during 1 tick ! Which would impact MW per fuel cell

check the ediitor, track the process tick by tick, you will see what is really happening.

If you are producing under max capacity, the heatpipes will stay hot, say 750°, it's instant restart, if full capacity, you don't restart, you are always at the lowest temperature it gets when fuel is burning.

It would take those 30 sec to re-start ( probably more tbh ) but only if you have something like a fuel shortage, then the temperature will go down to 500° everywhere, then it woud take time for the heat flow to start reaching the 2 corners again. It's not normal operation, but yes it wouldn't be reactive in bad scenario.

The good thing with heatpipes though is that they glow when hot, and the heat doesn't dissipate over time. Which means they create light without consuming energy, which not only is quite fantastic on the physical plan, but also looks incredible. For this reason i like playing with their shape, which forces me to well manage my fuel supply .

My idea of Perfect plant is:
1. fuel wastless e.g. in case of plant is not needed (batteries have acceptable level) it can switch off the grid and it can absorb the rest of the heat of all cells.
2. it can provide flexible source of power. e.g. fast re-start to full power output. Otherwise Brownouts are waiting for you.

Becouse of 1. it can often happen that you re-start the plant when storage of heat is under treshold, in tthis moment, heatpipes are 500C. So you need the shortest pipes possible to support plant feature.

Other important features of perfect plant check my design
viewtopic.php?f=208&t=96233

Also, in your case, of heatpipes as capacitor, you dont have a tool too measure amount of heat stored in plant, so you dont know when insert new cell, so , as result, in case of irregular usage, probably with combination of solars and some burners, or simple bug attack repelled by lasers you will feed nuclear cells nonstop. Which is, as you know, punishable by law.

[SoShootMe]

FWIW, inserting new cells from a chest triggered by a pulse from the inserter removing a used cell, I measured a cycle time of 12015 ticks using the blueprint from https://wiki.factorio.com/Inserters#Inserter_Throughput. However, I counted [...] 16 ticks with the reactor showing no fuel

0. cell is used
[...]
16. inserter_loader loading the new cell

Yes, 17 ticks, of which the reactor shows "no fuel" for all but the last => 16 ticks unfueled. But a measured 12015 ticks from one fuel cell being removed (or added) up to the next => 15 ticks unfueled.

[mmmPI]

The inserter head has to move from the idle position to where the fuel is getting picked up. With belts it might have to chase the fuel costing extra ticks.
Shouldn't happen with a chest though I thought. But maybe there is some delay like 1 tick to evaluate circuit signal, 1 tick to pick thing from chest.

Yes but my belts are stopped, so no chasing, timings for swings should always be the same every cycle like for chests. Belts are stopped however the fuel itself isn't stopped at the exact same position on the stopped belt for all 4 inserters. Which may means some swing are 180° but some are only 177° for example, and if the swing time for 180° is 12.5 ticks rounded to 13, it may round to 12 if the angle is just slightly different. Say because one belt is stopped in 3 logic tick while another only 2 after the fuel entered.

The 12.5 ticks for a 180 degree rotation at 864 degrees/s seems to round up to 13 ticks each time. That makes sense because the inserter returns to an idle state between each operation (it always starts on a tick, and hasn't got far enough until the 13th tick), but it might be different if it is continuously active.

True i got confused with measuring timing it takes for an item to go through a loop of belts. Now i want to play more with those when i can haha, makes me wonder if a straight belt that is crossed in 32 ticks act like a reset for the 0.25 rounding that appears when an item goes through a corner belt in 13.25 ticks. I measured timing by holding the read value on a corner belt into an incrementing counter. The number went up 13 13 13 14 13 13 13 14, from which i deduced it was 13.25.

Also I managed to make a 120 tick loop using 4 yellow corner belts( 4x13.25 = 53ticks) and 4 red straight belts (4x16 = 64 ticks ) this represent 117 ticks, then i used circuit on 3 of the red belts, read-pulse-enable when [redscience] not 1 which stops the item 1 tick and i read the value on the last red belt, an item was entering every 120 ticks precisely. (when the item is in the inner side).

the last part of your sentence seem to apply to belts loops, not for inserter that activate once per 200 second i understand.

0. cell is used
1. inserter_unload-er get noticed the source and grab and send signal
2. inserter_loader processing signal
3. inserter_loader grab the new cell
4-15. arm swing
16. inserter_loader loading the new cell

that's only 12 for the swing itself ?
Could it be because the reactor core is large and the fuel is dropped on a diagonal tile reducing the angle for the swing ?
Or maybe it's 12.5 and the tick 16 used for the end of the swing AND the disappearing of the fresh fuel cell, getting rid of any decimal from the inserter.

i used the editor already to realise it was 12025 oor 12026 ticks in my case but what i found i'm here to try and understand it, may be off-topic but also i need it to make a better power plant, more efficient in the details next time

Also, in your case, of heatpipes as capacitor, you dont have a tool too measure amount of heat stored in plant, so you dont know when insert new cell, so , as result, in case of irregular usage, probably with combination of solars and some burners, or simple bug attack repelled by lasers you will feed nuclear cells nonstop. Which is, as you know, punishable by law.

True no tool to measure the heat, that's why it require a clock design, the only way to achieve wastelessness is to control precisely the refuel timer according to energy consumption. That's still one way to do haha

but no feeding nuclear nonstop as a result of consumption spike !

In the case of clocked-fuel-insertion where energy is stored in heat pipes, you do not react automatically to consumption since there is no tank to have a feedback and auto reduce the timing of fueling. That would be the case if @mrvn suggestions were implemented, with a tank whose steam level is used to stop or release the belt. ( which if you do may as well get rid of the clock entirely and add more tanks and it wouldn't be the same design imo)

The way it is, if you have a simple bug attack repelled by lasers, you will consume the steam in the 96 turbines, if you draw 558+ MW you have enough for only a few second of shooting, since turbines store 200 steam but consume only 60. For a bit of time you'll consume 120 steam using the 2 turbines, then the capacity is capped to 103/sec which is what the heat exchanger can sustain. Then you will have brown out. Same as with every nuclear plant when you consume more than you produce at max capacity.

You only feed nuclear non-stop if you set it so manually, if you set it to 80% and have a spike consumption that would require 100% or 110% , this kind of plant will give you brow-out or intermittent black-out if left alone for too long. If the system cools down under the equilibrium temperature at max-load, due to not having any heat buffer anymore, analog to a steam storage plant where all tanks are depleted, then you'll have some power only a certain amount of time then black-out until next refuel.

This means it's law-compliant ! you set it to 220 sec timer of refuel for 10 minutes untill the glow is nice and warm ( the heat dissipate well enough away from the core that they don't reach 999° before the exterior reach 750° or so), then you reduce it to match the consumption you read when you click a power pole say you consumed 240MW the last hour, you can see the graph, and predict the way it will average later on, so you can set up timer to 200x2 if you plan on consuming around 240MW, but if you are adding a new ore patch maybe you want to put the timer to 200x1.8. If you take 2 minutes to add the ore patch you do not plan ahead the same way as if you take 20 minutes or 2 hours to add an other ore patch And when you reach 200+10%, you know you can't expand anymore before building another power plant.
[Edited with correct value]
If ever the glow gets too bright, you just reduce the refuel timer for a bit to prevent waste.

[gGeorg]

You only feed nuclear non-stop if you set it so manually, if you set it to 80% and have a spike consumption that would require 100% or 110% , this kind of plant will give you brow-out or intermittent black-out if left alone for too long. If the system cools down under the equilibrium temperature at max-load, due to not having any heat buffer anymore, analog to a steam storage plant where all tanks are depleted, then you'll have some power only a certain amount of time then black-out until next refuel.

This means it's law-compliant ! you set it to 220 sec timer of refuel for 10 minutes untill the glow is nice and warm ( the heat dissipate well enough away from the core that they don't reach 999° before the exterior reach 750° or so), then you reduce it to match the consumption you read when you click a power pole say you consumed 240MW the last hour, you can see the graph, and predict the way it will average later on, so you can set up timer to 200+50% if you plan on consuming around 240MW, but if you are adding a new ore patch maybe you want to put the timer to 200+40%. If you take 2 minutes to add the ore patch you do not plan ahead the same way as if you take 20 minutes or 2 hours to add an other ore patch And when you reach 200+10%, you know you can't expand anymore before building another power plant.

If ever the glow gets too bright, you just reduce the refuel timer for a bit to prevent waste.

I dont think, you understnad what wasteless design means. It is fully automated system which can handle any spike or power ned drop for any period of time.
Example : You start to make a base redisign so power consumption drops from 500 - 600MW to 4MW. Where you have an solar field which gives 10MW. Meanwhile you redesign the base, bugs come and lasers make energy spike at 400MW. Power plant need to designed so no waste fuel happens. and everything get power when needed.

[mmmPI]

dont think, you understnad what wasteless design means. It is fully automated system which can handle any spike or power ned drop for any period of time.
Example : You start to make a base redisign so power consumption drops from 500 - 600MW to 4MW. Where you have an solar field which gives 10MW. Meanwhile you redesign the base, bugs come and lasers make energy spike at 400MW. Power plant need to designed so no waste fuel happens. and everything get power when needed.

No, what you describe is self-regulating. Meaning the power plant will adjust its consumption and production itself. Thanks to reading steamflow, steam level, side fuel consumption and so on. This is the feedback that allow self-regulating.

Wasteless/wastelessness, means you don't insert fuel when the reactor is 1000°, because this energy is then wasted. It's less complex behavior than self-regulating.