0.15.11 extendable downscaling nuke plant

Power Plants, Energy Storage and Reliable Energy Supply. All about efficient energy production. Turning parts of your factory off. Reliable and self-repairing energy.
vanatteveldt
Filter Inserter
Filter Inserter
Posts: 947
Joined: Wed Nov 25, 2015 11:44 am
Contact:

0.15.11 extendable downscaling nuke plant

Post by vanatteveldt »

So, .15.11 fixed/nerfed heat pipes, which killed my nuke plant :)

Based on the recent discussion here (viewtopic.php?f=5&t=47757), XKnight's great posts here (viewtopic.php?f=18&t=19851) and my earlier simple control design (viewtopic.php?f=208&t=47687) and the nuclear ratios at (https://www.reddit.com/r/factorio/comme ... ar_ratios/) I think I figured out a good set of components for a power plant design again that is .15.11 safe and downscales automatically for reduced loads.

The idea:

Since heat pipes should now be kept short but steam pipes can be arbitrarily long with pumps, I've separated the turbines from the exchangers to keep the exchangers close to the core. So, we we will setup a reactor with n cores feeding into m banks of 8 exchangers, which produce steam that is lead to m banks of 14 turbines+7 storage tanks. The number 8 is chosen (a) because reactor ratios are all in increments of 16/32 for pairs of cores; and (b) because of fluid mechanics: 8 exchangers convert 825 water into the same amount of steam, and a single pipe can transport that over >200 pipe distances without extra pumps. If needed, a pump can be inserted every 125 tiles to maintain enough flow.

Overview screenshot
Components:

1: Core and control

Control is achieved by monitoring steam in one of the storage tanks, probably best take one far away from the core, and only insert new fuel if (a) there is no current fuel in the reactor [measured by depleted fuel in the memory chest] and (b) steam < X:

Image

(see viewtopic.php?f=208&t=47687 for combinator settings)
blueprint
2. Heat banks

Image

The blueprint contains 2x8 heat banks connected to heat pipe at one end and to water and steam pipes at the other.
blueprint
3. Turbines

Turbines are placed in blocks of 14. I place two blocks in a row with the steam pipe for the second block braided between the turbines:

Image

By placing the blocks back to back I can connect the tanks with a steam pipe to feed to a liquification plant if needed.
blueprint for 2x14 turbines
blueprint for 4x14 turbines
Instructions:

1) Place the components: place the reactor cores and the exchangers close to each other. The turbines can be as far as you want, you need to at least leave enough space for plumbing. I would say 2 UG pipes length.
2) Connect the heat exchangers to (1) offshore pumps (1 pump per 8 exchangers) and (2) the turbines. You can use up to 251 pipe segments between offshore pump and exchangers and between exchangers and turbines, so you have plenty of space. If necessary, place pumps just before/after the exchangers and just before the turbines, and every 125 pipe segments as needed.
3) Connect the green wire from a steam tank to the core. It's probably best to use one of the further steam tanks, but it might not matter that much.
4) Make sure all fuel requester chests have plenty of fuel, and feed a single fuel in each reactor manually to start the process.

Expandability

Everything is modular so you can start with a 2 core / 16 exchanger reactor and add cores as needed. There's a limit because of the heat pipes, but I've expanded to a 2x8 core 2.4 GW reactor without a problem, and at some point it is probably more feasible to build another reactor as the plumbing will get pretty horrible.

Observations on use

The 2x8 plant could provide the theoretical max of 2.4GW without problem over long time spans. It might need some time to heat up, though, and if you draw full power immediately I think it sometimes doesn't get to high enough core temperatures to give full yield (?) With a lower load (I tested with 1.2 GW) it will automatically switch off if there is enough steam and switch back again if steam is low.

Note: 1 tank per 2 turbines should be enough to capture the energy in a reactor cycle, but since there also is quite a bit of energy stored in the temperature of the cores, heat pipes, and exchangers, the system will produce steam for a pretty long time after cores shut down. With a 1.2 GW load it fluctuated between 0k and 24k steam in a single tank without brownout, but I think that with a higher load it might be wiser to connect more tanks and set the lower steam level higher.
iceman_1212
Filter Inserter
Filter Inserter
Posts: 256
Joined: Wed Aug 17, 2016 9:49 am
Contact:

Re: 0.15.11 extendable downscaling nuke plant

Post by iceman_1212 »

I like this configuration of heat pipes/heat exchangers very much; also, the conservative pipe budgeting for water/steam is usually a good idea in practice.

I suspect that folks will likely use their own preferred setups for the central controller, depending on whether they want it to be belt-based vs. bot-based, independent power source vs. powered by reactor, and whether they want to force concurrent operation of all nuclear reactors (to maximize neighbor bonus benefit for all fuel cells consumed). The latter should only be an issue if we have ever run low on fuel cells and if there ends up being a situation where only some input inserters have fuel available and a new cycle is initiated.

I agree with a ratio of 2 steam turbines to 1 storage tank (for setups where steam flows from heat exchanger to steam turbine to storage tank, like this one). Since I haven't seen it anywhere else, here is the calc of storage tanks per turbine for others' reference: An 8-core setup has a theoretical output of 1120 MW, i.e., 1,120,000 kJ per second. It takes 97 kJ to heat 1 L of water from 15 degrees C to 500 degrees C (given it takes 1kJ to heat 1L of water by 5 degrees C) and water is converted into steam at a 1:1 ratio, therefore an 8-core setup is capable of producing the following amount of steam per second at max load: 1,120,000 kJ per second / 97 kJ per liter = ~11,546 liters per second. A fuel cell burns for 200 seconds in a nuclear reactor (8,000 MJ per fuel cell / 40 MW fuel consumption rate per nuclear reactor), so an 8 core setup produces the following amount of steam over the course of one fuel cell cycle: ~11,546 liters per second x 200 seconds per fuel cell cycle = ~2,309,278 liters per fuel cell cycle, which is equivalent to ~93 storage tanks (25,000 liters per storage tank). The ideal # of steam turbines for an 8 core setup is equal to 1120 MW / 5.82 MW per steam turbine = ~193 steam turbines, which gives a ratio of ~2.08 steam turbines per storage tank.
vanatteveldt
Filter Inserter
Filter Inserter
Posts: 947
Joined: Wed Nov 25, 2015 11:44 am
Contact:

Re: 0.15.11 extendable downscaling nuke plant

Post by vanatteveldt »

I created a 6x2 reactor / 1.7GW plant based on this design that is symmetric, fairly compact and connects to a 22 tile straight lakefront (where you need to landfill a strip of land for the secondary row of pumps:

Image

Image

The design consists of the modules in the first post. Only change is that I added a wire to hook up all fuel inserters and changed the condition to [anything>0] (rather than depleted cell >0) so you can activate the whole plant by adding a single item in the chests above the reactor, generating a pulse (and you can restart by rotating the inserter, if needed)

Of course, you can change the power regulator by adding a memory/latch, the main design fun is in the spacing and plumbing...

Note: on full power consumption it seems that some don't perform 100%, so power production is probably slightly under what it should be, but I can't see anything really wrong with the setup. Maybe the heatpipes are still too long?
blueprint
Edit: Here is a blueprint for a "starter plant" containing only the top 2x2 plants that can be directed extended to 2x6. It uses all edges of the design already (including the full complement of pumps) so you can see how much space is needed.

Image
2x2 plant
blueprint
User avatar
MisterSpock
Fast Inserter
Fast Inserter
Posts: 102
Joined: Mon Jun 16, 2014 8:11 am
Contact:

Re: 0.15.11 extendable downscaling nuke plant

Post by MisterSpock »

I did a very similar design, but not that expendable. But its still modular.
BTW: where did you get water?


Image
Image
vanatteveldt
Filter Inserter
Filter Inserter
Posts: 947
Joined: Wed Nov 25, 2015 11:44 am
Contact:

Re: 0.15.11 extendable downscaling nuke plant

Post by vanatteveldt »

MisterSpock wrote:I did a very similar design, but not that expendable. But its still modular.
BTW: where did you get water?
Cool, thanks for sharing.

Water in the last post comes from just below the reactors, it assumes you would use landfill to create a straight costline with a second row of pumps, but if you have enough coast it doesn't matter of course.

The OP simply assumes that you need to get water from somewhere, in the pictures it uses the creative mode fluid source. (the point of the post was the design modules, not the overall design, but the latter posts gives and example of a complete system)
User avatar
MisterSpock
Fast Inserter
Fast Inserter
Posts: 102
Joined: Mon Jun 16, 2014 8:11 am
Contact:

Re: 0.15.11 extendable downscaling nuke plant

Post by MisterSpock »

I turned my design by 90° and build 2 of them in a row:
The Idea was to make them more spaceefficient(the space of my coast line is limited). And i also tried another design (no screenshot), but its not much diffrent. The waterpipes are annoying much. But however if you look at the coastline, thats the maximum density of pump i can get. Also tried waterbarrels, but thats very messy.

I have this question: Why do ppl always store they steam?

Image
Image
Post Reply

Return to “Energy Production”