The little big reactor that could.

GreenZombie

New Member
Jul 29, 2019
2,402
-1
0
If you are looking for an easy to build - fuel efficient - passively cooled Big Reactor this might help.

This is my Big Reactor, and showing you why it is the way it is might help you build your own Big Reactors.

Basically the reactor is made up of layers. Each layer contributes up to ~5000RF/t - so you simply build it with as many layers as your power requirements need. For example, add two layers for each MFS mining laser you wish to power (using AgS configs).

In terms of width and breadth, I build mine as an 11x11, but ideally it should be a 13x13, and you can probably get away with 9x9:

Here is the layout of the interior of the reactor:

Code:
...ccc...
...ccc...
...ccc...
cccxxxccc
cccxcxccc
cccxxxccc
...ccc...
...ccc...
...ccc...

.'s represent blocks that can be filled with air, or anything.
c's represent coolant. Cryotheum.
x's represent fuel rods.

Comments:

The core is designed for maximum fertilization: The core is NOT an 'x' or a '+' - it is densely packed with as many fuel rods as possible directly adjacent to each other: This ensures the maximum amount of radiation reaches adjacent fuel rods for fertilization. If you look at each fuel rod ('x') you can see that each one has 3-4 other fuel rods lined n/s or e/w with it.

The core is designed to stay cool: The hollow shape means every fuel rod has 2 faces adjacent to a coolant block. This facilitates the rapid flow of heat out the fuel rods into the reactor interior.

The reactor is big: There are at least 3 blocks of coolant between any fuel rod and the edge of the reactor. 3 blocks means that most of the radiation is captured and turned into RF. This also means the volume of the reactor is large, which keeps the casing temperature down.

It does look like it takes a lot of cryotheum to build: 36 per layer. As flowing cryotheum works just as well as a coolant/moderator/absorber, its quite easy to cut that down to 24 per layer by not filling the middle row of each 3x3 coolant filled area. If you are willing to use the glass pillars to support source blocks, you can possibly get by with as few as 4 cryotheum blocks to fill the entire reactor however high by teasing the flow over blocks that you then destroy.

My 11x7x11 reactor gets 25kRF/t with 0% moderation. Rod temp of 1277. Fuel burn of 0.238 for an efficency of 105MRF/ingot.
Moderated to 50% the output is 14kRF/t @ 121MRF/ingot (so there is a lot of benefit to building it 2x higher than you need).

One day, when I am finished playing with this design, I might build an actively cooled reactor.
 

GreenZombie

New Member
Jul 29, 2019
2,402
-1
0
Chris Becke, I hope you take another look at ReactorCraft one day. If you have a knack for reactor design I could use someone to bounce ideas off of :p

As much as I love Reika's work generally, I have removed RotaryCraft (and related) mods from my server, principally because I support players with low spec PCs, and RotaryCraft does impose a performance burden. I also find the bedrock tools unbalancing, they're just too good and players (myself included) stop exploring other tool options. And, strangely, seem to stand against the "realism at all costs" ethos of the rest of the mod. I also do not tolerate mods that introduce machines that *must* be chunkloaded or they will explode. (Which is why I have also stopped considering buildcraft and railcraft as tech options on the server).
 

Pyure

Not Totally Useless
Aug 14, 2013
8,334
7,191
383
Waterloo, Ontario
As much as I love Reika's work generally, I have removed RotaryCraft (and related) mods from my server, principally because I support players with low spec PCs, and RotaryCraft does impose a performance burden. I also find the bedrock tools unbalancing, they're just too good and players (myself included) stop exploring other tool options. And, strangely, seem to stand against the "realism at all costs" ethos of the rest of the mod. I also do not tolerate mods that introduce machines that *must* be chunkloaded or they will explode. (Which is why I have also stopped considering buildcraft and railcraft as tech options on the server).

Basically agree with everything you just said.

Not sure why but I keep liking the darn mods anyway :p
 

Skyqula

New Member
Jul 29, 2019
568
-1
0
@Chris Becke I decided to test this one out and found something that I find rather interesting. This reactor performs the best when set to 50% with 121M RF per ingot. But when switching this reactor to actively cooled it performs best at 0%(!) with 50B of steam per ingot, good for 600M RF per ingot.

The realy interesting part though, my favorite actively cooled reactor is a 15x15x3 (upgraded to a 15x15x7). Passively it performs best at 70% with 137M RF per ingot. Switching it to an actively cooled reactor it performs best at 0% with 79B of steam per ingot, good for 948M RF per ingot.

So while passively my design is only 13.2% better (bigger so expected), actively cooled its 58%(!!!) better o_O

I put them in a new tab on my sheet if your curious, though its still kinda WIP...
 
Last edited:
  • Like
Reactions: Wagon153

GreenZombie

New Member
Jul 29, 2019
2,402
-1
0
I am re-reading the BigReactors source trying to understand the discrepancy between active and passive cooling results. I did discover some interesting things, namely that I had misread the absorbtion method of RF generation.

Basically, RF is generated in a passively cooled reactor as heat moves from the reactors environment into the outside environment - which is at a reference 20C. (Surely this should change depending on biome?) The surface area (of the interior of the reactor) multiplied by the conductivity of iron, specifies the heat transfer co-efficient. rfGenerated is directly proportional to heat that escapes this way.

And Steam is generated as heat moves from the reactors environment, into the active coolant. The heat difference here is the difference between the reactors environment, and the coolants boiling point. 100C. Then, for each "potential" 4RF the passive cooled reactor would have generated, the actively cooled reactor (Tries to) generate 1mb of steam.

--
@Skyqula I dont have access to google docs from where I access the internet. if you could drop your 15x15 layout into a response i'd appreciate it so I can take a look :p
 
Last edited:

Skyqula

New Member
Jul 29, 2019
568
-1
0
Code:
CCCCCCCCCCCCCCC
CAAAAEEEEEAAAAC
CAAAAGGGGGAAAAC
CAAAAGGGGGAAAAC
CAAAAGGGGGAAAAC
CEGGGRRRRRGGGEC
CEGGGRGRGRGGGEC
CEGGGRRRRRGGGEC
CEGGGRGRGRGGGEC
CEGGGRRRRRGGGEC
CAAAAGGGGGAAAAC
CAAAAGGGGGAAAAC
CAAAAGGGGGAAAAC
CAAAAEEEEEAAAAC
CCCCCCCCCCCCCCC

C = casing, A = annything, E = Resonant ender, G = Gelid Cryotheum and R = Fuel Rod
It was late last night, I should note that I upgraded the 15x15x3 to a 15x15x7.
 
  • Like
Reactions: GreenZombie

Skyqula

New Member
Jul 29, 2019
568
-1
0
Yeah, that center rod has me puzzled. On one hand I dont want it as it has no coolant next to it. On the other hand its the most radiated Rod. I should also remove it and see what it does.

I also added a checker vs dotted tab (both passive and active cooling). Currently has a 32x32x3 for checker vs a 31x31x3 for dotted. Dotted wins, dispite the large amount of rods without a coolant.
 

GreenZombie

New Member
Jul 29, 2019
2,402
-1
0
On that center, the reactor does in fact work better with a fuel rod there. At least at 0% moderation.

Ive been looking at the code trying to piece together how heat is transferred from fuel to casing and it seems that there is no modelling of heat from individual fuel rods. Instead the code simply calculates the net exposed surface area of the rods to compute the fuelToReactorHeatTransferCoefficient. So individual "heat" islands are not a problem - its only the net surface area that counts.

Currently however, if I work backwards from the equations Ive derived from the source to try and predict the Casing temperature given the interior surface area and Energy output I get the wrong answer so something is amiss.
 

Skyqula

New Member
Jul 29, 2019
568
-1
0
So looking around reddit for Big Reactor things I came across this. Wich made be curious for a max size reactor. Ofcourse I derped and made it 32x32x42.... I plan to get a proper 32x32x48 for checker and dotted sometime.

Annyway, I made a 32x32x7 using the same pattern as my 15x15x3. Just to see what the effects of upscaling your reactor is.
  • Passive: Controll rod at 10%, 72100 RF/t with a fuel efficiency of 157M RF per fuel ingot.
  • Active: Controll rod at 0%, 40650 mB of steam per tick with a fuel efficiency of 972M RF per fuel ingot.
Then I was like, lets try this max size: 32x32x48.
pbMGy3F.jpg
Dont mind the floadgate "OOPS!"
The results are... crazy.
  • Passive: Controll rod at 69%, 212000 RF/t with a fuel efficiency of 221M RF per fuel ingot.
  • Active: Controll rod at 89%, 50 buckets of steam per tick with a fuel efficiency of 1176M RF per fuel ingot.
So yeah, over a billion of RF per ingot...
 
  • Like
Reactions: ThomazM and E_DM_B

SirDoctorOfTardis

New Member
Jul 29, 2019
166
-3
0
So looking around reddit for Big Reactor things I came across this. Wich made be curious for a max size reactor. Ofcourse I derped and made it 32x32x42.... I plan to get a proper 32x32x48 for checker and dotted sometime.

Annyway, I made a 32x32x7 using the same pattern as my 15x15x3. Just to see what the effects of upscaling your reactor is.
  • Passive: Controll rod at 10%, 72100 RF/t with a fuel efficiency of 157M RF per fuel ingot.
  • Active: Controll rod at 0%, 40650 mB of steam per tick with a fuel efficiency of 972M RF per fuel ingot.
Then I was like, lets try this max size: 32x32x48.
pbMGy3F.jpg
Dont mind the floadgate "OOPS!"
The results are... crazy.
  • Passive: Controll rod at 69%, 212000 RF/t with a fuel efficiency of 221M RF per fuel ingot.
  • Active: Controll rod at 89%, 50 buckets of steam per tick with a fuel efficiency of 1176M RF per fuel ingot.
So yeah, over a billion of RF per ingot...

@Reika Could the fusion reactor match this?