Reactor Design Help Wanted

[Omicron]

I've been looking into nuclear reactors today. You should know this because before today, I have never broached the subject matter at all. So, consider me a newbie. One competent in the use of his brain, but a newbie nonetheless.

I will be building myself a set of four iridium-neutronreflectors in my world soon (as soon as the matter fab spits out enough iridium; the beryllium will be taken care of by then due to a friendly nearby village willing to buy all sorts of useless crap in exchange for emeralds). These babies are neat in the way they never, ever need replacing. So I thought I'd go and see efficient a reactor I can build with them.

The ultimate in efficient reactors is, of course, the quad fuel cell surrounded by reflectors. An efficiency of 7 means, correct me if I'm wrong, that you're getting 7 million EU for every unit of uranium spent. Well, more really, thanks to GregTech's centrifuge giving additional thorium and plutonium as a side product, but you get the idea.

There's just one issue with this... a quad plutonium cell in this configuration generates 1,008 heat per tick. And I have found absolutely no way to cool this. With the amount of space available, it seems physically impossible.

Now, I did come across the concept of microcycle reactors. A timer, a statecell and careful design should yield something that runs for a certain time, then spends some time cooling down, then runs again, cools down again and so on. The best I have managed so far was a running time of 264 seconds until blowup... or, in other words, about 100 seconds of safe operation below the 40% threshold. And that was a very, very, VERY expensive design, with diamond components all over the place.

Can you do better? Can you design a reasonably sane microcycle reactor running a quad plutonium cell surrounded by neutronreflectors, that aligns with the following goals?

- Minimize initial resource investment
- Run without recharging components (no condensators) - everything else is fair game.
- One quad cell is enough; I care about efficiency, not raw output.
- In the interests of safety, I'd like something that runs for a decent amount of time before needing to be shutdown. I don't trust something that spikes its heat up and down in mere seconds.

 

[eculc]

My first suggestion would be to ditch trying to run a 7 EFF plutonium reactor -- like you said, it's nearly impossible to cool off as much heat as it generates. a better option would be to run the same style of reactor, but with thorium; You'd receive the same total amount of energy as a quad-cell 7EFF uranium reactor, but spread over a period of 5 cycles, with a reduction in heat to match, which is much easier to compensate for.

 

[Omicron]

Yeah, that would be a lot easier. In fact, in the meantime I arrived at a design that can handle anything except quad plutonium cells. Quad thorium, quad uranium and dual plutonium all work indefinitely; the final one is just efficiency 6 instead of 7 due to the smaller cell.

But I'd still like to keep puzzling about the quad plutonium one. Not because I'll die of OCD if I can't have it, but rather because it's fun to try I just don't think I quite have all these components figured out yet, so that's why I'm asking for help.

 

[MilConDoin]

Can you please post your current design? The best i got was 113 seconds until melting. (using 26 Overclocked Heat Vent and 23 Component Heat Vent)

 

[Omicron]

I will, in a bit. I'm currently load testing my aforementioned fallback design in actual application, and I'd like to make sure the quad uranium cell (the hottest thing short of a quad plutonium one) will complete its cycle without failing before I step back into my test world.

I kind of like this design - it's pretty cheap for what it does. Only basic heat exchangers needed, 18 of them. In addition, 21 overclocked vents, 4 basic vents and one reactor vent. Will run uranium and thorium at efficiency 7, and plutonium at efficiency 6 (via a dual cell). The quad plutonium cell however would still blow it up in very short order

 

[eculc]

For future reference, you can go to the IC2 forums, and from there you should be able to find an online java app that will simulate the reactor for you, without having to create a test world and power the gregtech machine to do it in-game.

 

[Omicron]

Yeah, I did that, except it didn't give me consistent results. Then I read some on the IC2 forums, where they say that the planner might have minor bugs and people generally recommend using the GregTech computercube as a simulator because it runs the exact same code as the IC2 reactor does ingame

In the end, I used both to get my desired result. The planner to draft up the rough design, and the GregTech simulation to check that it's in fact valid.

 

[Omicron]

Can you please post your current design? The best i got was 113 seconds until melting. (using 26 Overclocked Heat Vent and 23 Component Heat Vent)

Here is the best thing I could come up with after a few moment's further thinkering: https://dl.dropbox.com/u/44754370/2013-01-21_00.10.13.png

491 seconds to overheat; about 196 to 40%. There was another overclocked heat vent in the corner but it melted a split second before the hull went critical. But, you can see what I meant with "expensive", right?

 

[MilConDoin]

Use http://www.talonfiremage.pwp.blueyo...mdjan902pz732ogtn3tz2wlfcw6t8hk6bfdsl8quhni0w for better effect. According to the cube, it has a runtime of 508 seconds.
Alternatively use http://www.talonfiremage.pwp.blueyo...49mqeb4hpowbsiharswg6t3x4dsis20r3wbvxdnswra4g for better average EU/t. The generation time is lower, but the cooldown is even more lower. Here we have a 5:2 active:cooldown time. The one above has almost 1:1 instead. And this design is way cheaper

 

[Omicron]

Wow, I didn't even know that moving around the fuel cell could have an effect on heat (although in hindsight it makes perfect sense, since it alters the way the vents interact).

And, I do like the second design more, as well. You could run it with a 72 second timer and a state cell set to 50 seconds. It would run for 50 seconds, up to 40% heat, then cool down for 22 seconds (21 are needed according to the planner) down to 0 heat, then start up again. Very neat. It's still somewhat costly, with all those upgraded components, but I suppose that's the least you'll need to accept to make the quad cell work. And it doesn't take any diamonds.

I'll bookmark that design and keep it around, maybe I'll upgrade to something like that once I burned through all the other fuel created in the process of stocking up on enough plutonium to make it worthwhile.

 

[Whovian]

Fusion reactors or fission reactors?