search

tomdodd4598 / NuclearCraft

NuclearCraft is a Forge mod for Minecraft 1.12.2.
https://www.curseforge.com/minecraft/mc-mods/nuclearcraft-overhauled
MIT License
196 stars 128 forks source link

[1.12.2] Very large fission reactors cause a lot of issues #161

Closed Buggy123 closed 6 years ago

Buggy123 commented 6 years ago

In summary, extremely large fission reactors don't work well for a lot of different reasons. For reference, here's my reactor: 2017-10-27_20 40 28

1: The maximum energy buffer of a reactor is about 1 million RF and the most energy it can output is that buffer every tick. However, a properly designed max-size reactor can produce 100 million RF/t or more. 2: Heat capacity has a similar issue. A reactor which doesn't produce any heat per tick is fine, but at these levels even a little bit of heat generation or a brief fluctuation in active coolant supply will cause a instantaneous meltdown. 3: If the RF/T and heat/t numbers are large enough, they overlap in the controller GUI, making them hard to read.

I'd recommend scaling the storage buffers to the reactor size, or otherwise giving players a way to expand them.

tomdodd4598 commented 6 years ago

Thanks for the heads up - your recommendation is a good one :)

tomdodd4598 commented 6 years ago

Everything on that list is fixed!

However, could I also ask how you managed to get the efficiency so high? 68000% is crazy :P

Buggy123 commented 6 years ago

Ah, it's fairly simple actually. See, while testing I figured out that pretty much all passive coolers have far, far too little cooling to be useful in such a large reactor, and most of the actives weren't great either. However, the default cooling value for glowstone is five times as high as the next-highest liquid, 80,000 for glowstone vs 16,000 for liquid helium. So, I made a sort of cascading grid layout, and filled the entire reactor volume with it.

2017-12-10_16 34 13 2017-12-10_16 33 58

Anything with heat production much higher than Thorium produced far too much heat, but thorium oxide is just about right, and really efficient as well.

I haven't tried turning it on since the fix went out, so I'm not sure how much glowstone it'll require, but I think I'll manage. At worst, I might have to set up a dedicated ET void miner for it.

tomdodd4598 commented 6 years ago

Oh wow, that is one of the most epic NC builds I've seen yet! You should be fine with the cooling - none of those numbers have changed. Hopefully the new version will fix your original issues :)

You know what, I found the issue - I made a slight mistake in the Fission Reactor code for calculating the effects of graphite. That's why the efficiency is effectively unbounded... oh well, I won't change it in case that wrecks some designs people have spent time on such as yours :P

Buggy123 commented 6 years ago

Ha, thanks! I thought it was a little weird that efficiency increased exponentially as you add graphite. Now I just have to find a use for 100m RF/T!

tomdodd4598 commented 6 years ago

100 MRF/t is, well, probably enough for most things, but hopefully you will have a NC-related use for some of it once the particle accelerators are back in... a size-100 one of those takes about 2 MRF/t to run.