2x6 tileable Nuclear Reactor 1.76 GW

Power Plants, Energy Storage and Reliable Energy Supply. All about efficient energy production. Turning parts of your factory off. Reliable and self-repairing energy.
Post Reply
Tertius
Filter Inserter
Filter Inserter
Posts: 667
Joined: Fri Mar 19, 2021 5:58 pm
Contact:

2x6 tileable Nuclear Reactor 1.76 GW

Post by Tertius »

Here is a my tileable 2x6 (xN) nuclear reactor setup. It has to be placed on water plus landfill, because it isn't feasible to cover whole regions with water pipelines.
Reactor 2x6.jpg
Reactor 2x6.jpg (1.64 MiB) Viewed 4774 times

Features:
  • Intelligent roboport coverage to make building as painless as possible
  • Autostart. No manual fuel cell insertion required.
  • Uses logistics network for fuel cell supply. No belts, no wasting of fuel cells lying around on a belt.
  • Fuel cell saving by using steam buffers. Inserting new fuel cells is postponed until buffered steam level is low. The plant was tested with 160 MW (10% of max power output) and no overheating to 1000°C happened.
  • Is able to provide 100% power. Not a single tick reactor downtime if full power is required.
  • Blueprints include landfill for all buildings.
  • Layout prevents unwanted automatic power connections.
  • Two independent power grids to avoid reactor shutdown if the external power grid has a brownout. The internal power grid is solar powered.
  • Audible alarms if the internal power grid is out of power (cannot happen, but just to be sure) and if the fuel cell supply is low.
  • Steam gauge to see current steam level.
  • A stripped down 2x2 version (480 MW) of the 2x6 layout is also provided. Can be used as start, then upgraded to a 2x6 layout by just removing a few pipes and building the 2x6 blueprint over it.
  • Optimal steam flow due to enough (but not too many) pumps.
How to use:
  • The provided blueprints are intended for the south shore of a lake. If you want to start from the north, rotate all by 180°.
  • Find a lake and build the "Reactor 2x6 supply" blueprint first. It sets up the initial roboports and the service infrastructure for fuel cell insertions. You have to build it in multiple takes, with shift-click, since at the beginning there is no full roboport coverage yet. Keep an eye on the 6 rectangular holes left and right. This is where the offshore pumps will go, so these must enclose water.
  • After the supply blueprint has been fully built, set the "Reactor 2x6" blueprint over the supply buildings. Use the 2 reactors as anchor: they must overlap the 2 south reactors from the reactor blueprint.
  • If you want to add another field, add another "Reactor 2x6" field to the north. No supply blueprint. It's only required once for a whole tiled setup.
  • To start the reactors, disable or remove the constant combinator south outside of the reactors. While active, it inhibits the counter that inserts a new fuel cell every 200 seconds. You need to wait 200 seconds till the first insertion. Time to check if everything has been built correctly.
  • While building you will hear 2 alarms: one is the "internal power grid has zero power", which goes away as soon as everything has been built and there's daylight. The other one is the "out of fuel cells" alarm, which goes away as soon as enough fuel cells are found within the logistic network.
How refueling operates:

A decider combinator is used as counter. It counts from 0 to 11996 + 3 = 11999 ticks = 200 seconds. The 3 additional ticks are latency introduced by the inserters, so the counter needs to count 3 ticks less.
3 steam storage tanks are tapped with green wire, so a steam volume between 0 (0%) and 75000 (100%) can be detected.
The inserters for the fuel cells are triggered if the counter reaches 11997, and if the steam level is below 37500 (50% steam level).
To check both conditions simultaneously within the inserters, the counter actually counts backwards from 0..-11996, so by adding 11996 to the counter and -37500 to the steam level, the condition can be EVERYTHING < 0.
If the inserters were triggered, each will insert a single fuel cell. The inserter arm content is read as pulse, and this is the reset condition for the counter, which restarts at 0. So the counter is reset if fuel cells have actually been inserted. All reactors are refilled simultaneously in the same tick.

In the supply blueprint, there is a tiny builder mall with infinity chests and a power consumer grid. I used these for development and debugging. In a real world map, they cannot be built of course, so you might just remove them from the blueprint. At these locations, it's supposed where the logistics network from outside is connected, and where the main power grid is connected.

Smaller 2x2 Reactor blueprint:
I included a stripped down version of the big reactor field with only 4 reactors (480 MW). Useful if you don't want to move your nuclear plant later. To upgrade it to 2x6, remove all underground pipes (apply the included deconstruction planner over the whole plant), then overlay the plant with the 2x6 reactor blueprint. Should fit and exactly add all remaining entities.
Make sure you don't remove any "unused" power poles from the supply blueprint. Their presence prevents automatic unwanted copper cable connections while adding a reactor blueprint. If you do, the internal power grid will be mixed with the external power grid and the reactors cannot autostart any more. If you messed with the power grids, just deconstruct every power pole, then re-add them by first setting the supply blueprint, then the reactor blueprint.

Supply blueprint (only set 1 supply for a whole tiled setup, no matter the size):


2x6 Reactor blueprint:
(See next message due to message size limit)

2x2 Reactor blueprint:


Remove underground pipes for 2x2 -> 2x6 upgrade:


Images:

Build supply first:
Reactor Supply.jpg
Reactor Supply.jpg (1.09 MiB) Viewed 4774 times
2x6 Reactor 1760 MW:
Reactor 2x6.jpg
Reactor 2x6.jpg (1.64 MiB) Viewed 4774 times
2x2 Reactor 480 MW:
Reactor 2x2.jpg
Reactor 2x2.jpg (954.44 KiB) Viewed 4774 times
Tiled 2x6 Reactor:
Reactor 2x6 tiled.jpg
Reactor 2x6 tiled.jpg (1.6 MiB) Viewed 4774 times
Large tiled 2x6 Reactor (tiled 13 times 156 reactors) 24.8 GW:
Reactors 156 tiled.jpg
Reactors 156 tiled.jpg (969.95 KiB) Viewed 4774 times
Reaches full power of 24.8 GW after 12 minutes:
Reactor power info.jpg
Reactor power info.jpg (203.8 KiB) Viewed 4774 times
Power grid:
Reactor power grid tiled.jpg
Reactor power grid tiled.jpg (215.99 KiB) Viewed 4774 times
Being the laziest person in the world, I designed this for maximum automation. The reactor is a self-contained blackbox that is even built automatically if you put down the supply blueprint first. While operating, you feed fuel cells in and get used up fuel cells plus electricity out. It cannot fail, as long as there are enough fuel cells coming in and taken out, and as long as there are no biters starting to eat the thing.
You can add one radar in the empty field opposite of the central roboport. But don't add a radar to every reactor tile - there's not enough energy for this on the internal power grid.
Last edited by Tertius on Mon Mar 06, 2023 10:20 am, edited 3 times in total.

Tertius
Filter Inserter
Filter Inserter
Posts: 667
Joined: Fri Mar 19, 2021 5:58 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by Tertius »

2x6 Reactor blueprint:

gGeorg
Filter Inserter
Filter Inserter
Posts: 392
Joined: Wed Jun 19, 2019 8:06 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by gGeorg »

I like your approach to chase a smart design. I see you adopted ideas and design features I have introduced in my Perfect cloverleaf 2x2 plant viewtopic.php?f=208&t=96233

Here are some feedback for you:
- for better chance of finding useable lake, make design more narrow. Move water pumps next to reactors. Maby you would need few heatpipes more, but in this scale, I guess, is more important be able to build plant as big as possible. Savings of few pipes is secondary priority.
- number of steam tanks is way over need for wasteless&brownout protection. Now you use 30 tanks per 6 reactors, I guess 6 tanks should be ok. For precise math, you can use my xls sheet, for thermal balance. You can found it for download in my cloverleaf topic.
- it could be UPS wise if you manage to separate steam circuits. Make two lines of turbines fed by one pump as a closed system. e.i. make the seam tanks storage dedicated for the dual-line of turbines. It is a sort of variation on my clover leaf concept where each leaf of turbines is separated and closed
- In my design I have found a recommendation of steam flow: tank>heatExchanger>turbine OR heatExchanger>turbine>tank. However, it is valid for free flow. In your design you can enforce whole steam flow by pumps, which could even help speed of start up. Your beast has 12 minutes start, which ... meh. Could you try this trick : Place steam tanks in between. Heat exchanger > steam tank > turbines. add a tank per turbine line. Make it so, that output from Heaters is a pump directly to the tank, then from the tank directly to pump then dual line of turbines fallows. The trick is, the pump which is pumping directly from the tank has MAX speed output, therefore it can deliver more steam. In this configuration, tank actually help the steam flow.
- I like you try a variation on my "zero_cold_interval" inserting method. However, 3 ticks for inserter arm swing is too short, I think. Yellow inserter movement takes about 15. I recall.


some questions:
Is your intention to build it as wasteless ? If not, then you dont need steam tanks at all. If yes, then how it works ?
Roboports&logibots can be quite power hungry, did you try no_power cold start ? I guess that amount of solar panels you used are not enough to power roboports and logibots to make the beast start moving.

---
Include a deconstruction planner to make build modular, that surprised me most. :idea: I Never thought about this possibility. I probably stole this trick :twisted:

Tertius
Filter Inserter
Filter Inserter
Posts: 667
Joined: Fri Mar 19, 2021 5:58 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by Tertius »

I know the pumps require a quite broad area, however this is a result of all the conditions and constraints of the whole design. There is a chain of (my) conditions that can only be resolved the way I resolved them.

The chain of constraint resolution goes like this.
Mandatory requirements: tileable + max density + supply via roboport + fast startup and reaction + good buffering behavior also on larger load changes + never reach 1000°C + cold start ability + integrated roboports for building

This results in:
Tileable + max density -> 2x6 layout
supply via roboport -> roboport in the center
fast startup and reaction -> shortest possible heat pipes
good buffering behavior -> shortest possible heat pipes + storage tanks are all connected
never reach 1000°C -> enough storage tanks (not more heat pipes, because this will collide with the fast reaction and good buffering condition)
cold start abililty -> internal power grid + solar panels + batteries

This results in:
2x6 layout + roboport in the center + shortest possible heat pipes -> roboport location + heat exchanger layout
roboport location + heat exchanger layout + internal power grid -> offshore pump location (otherwise the distinct power grids cannot be built!)

This results in:
2x6 layout + offshore pump location + fluid behavior + integrated roboports for building -> turbine layout
turbine layout + fluid behavior + good buffering behavior -> interconnected steam storage tanks

So it all came together to what it is.

- There is an interesting additional constraint you may not be aware if you didn't build such a large layout yet. Nauvis is sloped. If you run a large horizontal orientated layout like this, you will see the steam volume on the east half will come out different to the steam volume on the west half, even if it's all perfectly symmetric. Nauvis is also vertically sloped. If you don't connect from north to south, some lines will get vastly more or vastly less steam than neighbored lines, even if their steam systems are never connected and get their heat from the same reactors. So to get fast responses at starting and fast responses at cooling down, the steam volume has to be as equal as possible in the whole plant. And so it's required to connect all steam buffers to each other. I explored alternatives (not connecting at all; connecting after heat exchangers; connecting between turbines and tanks), but the published layout turns out to be the best in terms of general steam flow and equal power production.

So to move the pumps to the center (and I saw designs like that), the whole design will change utterly and completely in every detail. If you want this, use these other designs.


- UPS was not a design choice. May be I get to some UPS limit, but so far I never hit a limit. May be this will be a constraint in the future, but currently it's not a requirement for me. I also don't know how much a design change can change the UPS. May be it's so small, its negligible. After all, it's only a few 100 fluid connections more (or less). Fluid floating in circles isn't affecting UPS just because of the circle ; it's the number of fluid connections that affects UPS, because this is what requires CPU power.


- Putting steam tanks between exchangers and turbines increases reaction time. I tried that. It first fills the steam tanks, then reaches the turbines, so the first heat goes into the tanks and not into the turbines. But I want the heat immediately go to the turbines, so startup is as fast as possible. It's not important to reach maximum output as fast as possible, because you actually never need maximum output. But what you (ok, what I) need is a fast response.

- The counter is actually zero downtime. To calibrate, I set the counter to 12000 and single stepped in map editor through the last ticks and counted the number of ticks the reactor was out of fuel. There were 3 ticks, so I reduced the counter by 3. Keep in mind the counter isn't reset after an inserter swing has been completed; it is reset an item has been picked up by the inserter arm at the beginning of the swing. This time is 1 tick, it's independent from the type of the inserter. The 3 ticks are: signal reaching inserter; inserter picking up item; signal from inserter arm reaching counter to reset.

- And yes, the reactor is able to cold start while having a brownout at the external power grid. The power on the internal power grid is enough to power the roboports. Not all roboports are powered by the internal power grid. It's not enough to build the reactor (for this, you need external power), but for starting up a cold reactor the 1st time or after a fuel shortage, or during regular operations, it's enough.
If there is an external brownout/blackout, the global roboport grid will disconnect due to out of power, so robots from far away will not be able to fly to the power plant for recharge, so the internal grid cannot be exhausted, so as long as there is 1 logistics robot left in the plant, it will be able to move fuel cells.

- You might notice the power switch at the left. It's used for the situation too many (external) robots come and recharge at the central roboports only connected to the internal power grid. That's an unusual and not expected situation and should be avoided. It connects the internal with the external power grid, as soon as the battery charge falls below 3%. Usually, the external power grid isn't on brownout, so this is a quite safe circuit. But if the external power grid does brownout, no more robots will come to recharge, so the internal power grid has enough power to support the power plant.

- What do you mean with "wasteless"?

gGeorg
Filter Inserter
Filter Inserter
Posts: 392
Joined: Wed Jun 19, 2019 8:06 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by gGeorg »

Wasteless design makes sure that no energy from fuell cell is wasted.
e.i. that you insert a new fuell cell only when steam reserve is below certain level. > When you insert new cell, & power consumption switch to zero (pwr plant is disconected) then all energy is stored into heat. e.i. thermal balance of plant is designed to hold all the power of the cell in form of heat and steam.
good buffering behavior -> shortest possible heat pipes + storage tanks are all connected
storage tanks all connected is questionable. I would not create such massive system. I dont see any advantage to connect everything to everything. Dont know what means "good buffering behavior" but if you reorganized tanks in the middle with pumps directly connected, you get faster steam distribution > less reserve needed > less tanks needed > less space used.
Your current design has too much steam tanks.

Narrow design would add 1-2 heatpipes which add 1-2 seconds to response time. I think it is acceptable price for more chances to use design.

gGeorg
Filter Inserter
Filter Inserter
Posts: 392
Joined: Wed Jun 19, 2019 8:06 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by gGeorg »

Tertius wrote:
Sat Oct 01, 2022 2:51 pm
- If you run a large horizontal orientated layout like this, you will see the steam volume on the east half will come out different to the steam volume on the west half, even if it's all perfectly symmetric.
Why do I reccomend to separate the steam system for each dual line of turbines ? Becouse of UPS and wierd flow behaviour.
Tertius wrote:
Sat Oct 01, 2022 2:51 pm
- Putting steam tanks between exchangers and turbines increases reaction time. I tried that. It first fills the steam tanks, then reaches the turbines, so the first heat goes into the tanks and not into the turbines. But I want the heat immediately go to the turbines, so startup is as fast as possible. It's not important to reach maximum output as fast as possible, because you actually never need maximum output. But what you (ok, what I) need is a fast response.
No. You need to use pumps as I described. Then you get even faster response than now.

mootsfox
Manual Inserter
Manual Inserter
Posts: 2
Joined: Mon Mar 06, 2023 4:56 am
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by mootsfox »

This design is great, but the Supply blueprint seems to be corrupted. Any chance of a re-upload?

Tertius
Filter Inserter
Filter Inserter
Posts: 667
Joined: Fri Mar 19, 2021 5:58 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by Tertius »

Updated the OP. Please try again.
Was broken for me as well. Strange thing, there was the same defect due to some Chrome extensíon of mine I saw months ago, and I thought I tested every posted blueprint to avoid exactly this.

By the way, I worked on this setup and made a few improvements. Mostly, it's slightly "nicer". I didn't update the OP yet, because that needs new screenshots.

I improved the water and steam flow slightly.
One less roboport on both sides.
It's now more symmetric if it comes to steam flow, and it adds a missing steam turbine. I also tried to reduce the amount of steam tanks, but it's simply not possible. Either the reactors will overhead to 1000°C, or there isn't enough steam to bridge re-heating.

The circuit is updated as well. It has now 2 thresholds for refueling. One for almost full load situations and one for very low load situations, so these are now also handled correctly without slight reduced maximum output for a minute (max load), or wasting fuel cells (min load).
And it's not required any more to remove anything if upgrading from the 2x2 to the 2x6 blueprint.
It also should be slightly better for UPS, because it uses 78 less pipes+underground belts.
Finally, the offshore pump layout is slightly narrower, so it will better fit and better use available space on the lake.

Supply blueprint version 2:


2x2 Precursor of full reactor version 2:
Last edited by Tertius on Mon Mar 06, 2023 10:54 am, edited 6 times in total.

Tertius
Filter Inserter
Filter Inserter
Posts: 667
Joined: Fri Mar 19, 2021 5:58 pm
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by Tertius »

Full 2x6 reactor version 2:

mootsfox
Manual Inserter
Manual Inserter
Posts: 2
Joined: Mon Mar 06, 2023 4:56 am
Contact:

Re: 2x6 tileable Nuclear Reactor 1.76 GW

Post by mootsfox »

Works great! Thanks for re-uploading. :D

Post Reply

Return to “Energy Production”