Trying my hand at breeder design

  • So I've fallen in love with nuclear reactors, and I'm trying to create an efficient breeder reactor. I already have one running, fully automated with temperature regulation so it's running at near max temperature (above the hurt limit), but I'm looking to improve on that design. Here's the design I'm using right now: link

    For the record I'm playing the FTB Ultimate pack. I found that design in some thread/list somewhere. I'm running it at 86000 "degrees", and according to the planner tool it should be producing 580 re-enriched cells per cycle, but I'm not too sure that's correct. Seems quite a bit slower, maybe it's designed for a newer versior or something, I don't know. It also cools down with that setup, which is why I had to automate the temperature regulation despite the simulator saying it should be heat neutral. Pull one heat vent out and it heats up, put it back in and it cools down. I'll say it is very convenient though, no need to worry about starting it up or anything.

    So anyway, onto the point. Here are a few designs I've ended up with:


    1711 cells charged per cycle, seems pretty nice. This was the only design I managed to get to 0 in terms of heat, although I assume I would still need a thermal monitor to remove/insert one of the reactor vents to keep the temperature steady.


    Here's another one that I'm not too sure about. It's not temperature neutral in the simulator and I've no idea how to calculate the effectiveness without it, I was hoping someone around here could help me with that. I was thinking I'd just set up a thermal monitor and send in ice to cool it whenever it goes above the desired temperature.

    If any of you more experienced guys have some input or advice for me that'd be much appreciated :) In the mean time I guess I will be testing these in creative to see how they do.

  • Don't boither using thorium in FTB Ultimate. The pack includes a beta version of GregTech that was intended as a first test/preview of 1.5.x features, not for stable play in an 1.4.7 environment. In that version, thorium is heavily bugged.

    I recommend upgrading to Unleashed, or manually downgrading GregTech to a 2.8x release.

  • I refuse going over to unleashed because I'll miss RP2 and gregtech too much, and I play online so unless I switch servers I'm not really able to do anything about that part.

    How exactly is thorium bugged in the ultimate pack? 'Cause as it is the breeder I'm running may not be producing as much as the simulator predicted, but it does produce enough re-enriched cells to run 4 6-cell single chamber uranium reactors, and considering I don't really have anything else to use my thorium on that seems pretty nice :p

    Maybe I can work around the buggedness, unfortunately a google search yielded nothing on the subject so I hope someone can enlighten me as to how it's bugged.

  • For one thing, it ticks only once per 2-3 seconds instead of every second. That's why your breeder is slower.

    Also, if you use thorium for power generation, it does not scale up in output when you place cells next to each other. It does scale up in heat, though, meaning you get a whole ton of extra heat for almost no extra power output.

    EDIT: by the way, your breeder loses heat because heat values for thorium and plutonium have changed and are no longer compatible with the reactor planner. For one single thorium cell with 4 neighbors you're looking at 12 heat generated instead of 15. Try doing this design: http://www.talonfiremage.pwp.b…n0zm7fx61kkf9qs111qs36dtq
    It should be heat neutral while running.

    The post was edited 1 time, last by Omicron ().

  • Thanks for clearing that up, how about the other designs? My train of thought is basically that I want to be able to supply as many power-generating reactors as possible with as few breeders as possible, it seems like a waste to have 5 breeders running with 300 cells per cycle when I could have one pumping out 1500 per cycle, especially since each breeder uses like 1400 copper worth of heat cap platings.

    I guess if the thorium is that badly bugged I should just let my existing reactor eat up the rest of my thorium and focus on uranium for my new breeder(s).

  • I threw this together in a bit of a hurry, to me it looks pretty decent but of course I have no idea if there's a heat difference with uranium too in the ultimate pack. What do you think?

    Edit: link

    Edit2: I'm trying this design in creative right now and all I can say is wow, that's fast! It is slowly losing heat though, so there seems to be a difference here too. I guess I'll try playing around a bit.

  • Basic IC2 components have not changed, only GregTech-added fuels.

    If you are having troubles keeping a breeder heat stable, then you might want to look into heating cells. they are designed for exactly that purpose. They can be stacked up to 64 in one slot, and will emit <stack size> heat per reactor tick, until total reactor heat passes a threshold of <stack size> * 1000. Meaning, if you have a stack of 64 heating cells in your reactor, they will emit 64 heat per tick until total hull heat exceeds 64,000, at which point they stop emitting heat. That way, you have a mechanism that automatically keeps the reactor heated at your desired temperature, even if your cooling system is a little too strong.

    They need to sit next to a reactor heat exchanger in order to work.

  • Oh thanks, I didn't know that :o I thought they were a single-use item that were only used for startup... Anyway I enjoy automating stuff so I'd probably rather spend a little more time tinkering with my automated processes and use that slot for an extra reactor plating instead. Currently I have the last design I posted running perfectly in creative on 75000 heat, It's on fire, both literally and figuratively :p The simulator only goes up to 64k which is 172 cells per cycle, I assume 75k is 200 cells per cycle. Probably way more than I can spend for the time being, so I think I'll settle for this ;)

    If you or anyone else who happens to bounce in here have suggestions for good breeders (or good uranium/plutonium power generator reactors) feel free to post them :)

  • Yeah, I was thinking of making a tower of one of those actually, but I would prefer 4-chamber reactors.

    Looks like the breeder I threw together in literally 4 minutes is better than the "Very high speed breeder" while using almost the same parts, is there something I'm missing there? Mine can be run on much higher heat levels as well.


    Other one

  • Well, then post it as a suggestion in that thread, for peer review. If people agree it's a good design, it'll be put on the front page, either alongside or replacing the existing design. (might take a couple days)

  • I think there was a "rule" that the breeder had to be auto regulated, and yours isn't. (meaning if fuel runs out or you run out of depleted cells it will cool down)

    Also to heat yours up you need to remove the cooling and manually place it back when it's heated up.

  • Since its hard to test out breeder designs in the GT computer cube (due to the inability to stack heating cells), does anyone know how to make a fast breeder in GT 1.5?

  • Since its hard to test out breeder designs in the GT computer cube (due to the inability to stack heating cells), does anyone know how to make a fast breeder in GT 1.5?

    1. Right-click the heating cell in the reactor planner.
    2. Probably not by using thorium. If the changes in GT for 1.5 that I heard about a while back haven't been changed back, plutonium (especially multicells) might work even better than they did in GT for 1.4. Which means that fast GT breeders could be considered hybrid breeders, producing both re-enriched cells and EU.

    If you stare at my avatar hard enough, you'll notice that it consists of three triangular rings, interlocked in such a way that if you were to remove any one of them, the other two would be free to float apart.

  • Plutonium doesn't work, regardless of iteration. I tried it repeatedly.

    While the multipulse system was active, plutonium charged depleted cells twice as fast, true; but it also scaled in efficiency at double the rate per neighbor, and as efficiency rises, heat generation scales exponentially. For one single putonium cell surrounded by 4 depleted cells, you were looking at 240 heat! That required so large of a cooling system to handle that you couldn't fit enough heat plating to get the temperature up, which slowed down the setup. A breeder with two uranium cells and eight depleted cells was running faster because it needed a lot less cooling and thus could reach a higher temperature.

    And now that the multipulse system was retconned*, and we're back to 120 heat for a single plutonium cell with two neighbors, it's still not useful because it only charges at the same rate as uranium again. Or thorium. Which, now that the multipulse system was retconned, stopped being buggy and can once again be used to breed effectively.

    * this was done in GregTech 3.07; I haven't checked the status quo since, based on Greg's statement that he's stopping to mess with nukes entirely until he can get around to implementing his own nuclear reactors.

  • So thorium now breeds at the same rate as uranium, but produces a lot less heat... Interesting...
    So I guess it's back to the old thorium-powered breeders, then? Longer burn time than uranium plus less heat per tick (and thus less cooling, more heat plating) equals more breeding between refuels, correct?

    Also, are thorium multicells more effective than single cells now?

    If you stare at my avatar hard enough, you'll notice that it consists of three triangular rings, interlocked in such a way that if you were to remove any one of them, the other two would be free to float apart.

  • According to the simulation, they should be. But I have never built a breeder with thorium multicells ingame. Usually even one single thorium cell, properly brought up to temperature, can supply all but the most extreme builds.

    You need to keep in mind that GregTech has also stopped allowing the 8x depleted cell crafting recipe from uranium, which basically allowed one to skip using uranium altogether and just mass-breed plutonium and thorium. This no longer works, thus the requirement for fast, high-output breeders has dropped considerably. The current uranium lifecycle starts in the centrifuge, where 4 uranium dust become 4 uranium cells, splitting off 1/4th plutonium and 1 thorium in the process. (Or was it 2 thorium? I forget.) Then you run the uranium through a generic IC2 reactor (you could build an uranium hybrid, but so far I've not found a good design), which yields 25% of the input worth in depleted cells, on average. Those you can breed and run through the centrifuge, splitting off plutonium which can run through a reactor and sometimes leave depleted cells again, which return into the cycle. But eventually, all of the originally inserted uranium burns out, either by not generating depleted cells, or by eventually becoming thorium through the breeding/centrifuging cycle and taking its last stint in a thorium sink reactor, which never returns any depleted cells.

    In my latest GregTech world I have a group of five reactors, with about 1000 EU/t in combined output. Two uranium, one plutonium/thorium hybrid, one thorium sink, and one breeder. To this date I have not even started the breeder once. I have a couple depleted cells from the uranium reactors, but got unlucky and stayed below the average so far. On the other hand, I have a decent enough stockpile of uranium, plutonium and thorium dust from my industrial grinders. Especially thorium dust. I like silktouching coal ore. And now I have 500 of the glowy black stuff lying around. :p It also helps that the reactors are pretty much only running full burn if I need the matter fab for some reason. Otherwise they only do occasional short burns to top off the storage. I don't believe in production for the sake of production.

    High-output breeders are a relic of 1.4.7's golden era of hybrid reactors (and even then, you could argue that you might as well run Pepe's excellent Hybreeder(tm), let all your reactors breed their own fuel and eschew dedicated breeders entirely). Nowadays, outside of creative-mode playground designs, I'd wager few if any people operate reactor setups big enough to need more than a small one, so long as it runs fully automated. The logistics of getting enough uranium in the first place to produce enough depleted cells for a breeder make sure of that.

  • I've noticed that plutonium cells seem to give a lot more depleted cells than before (appears to be about 1.5 per plutonium cell). Though its still not enough to replenish the supply of plutonium without also having loads of uranium reactors going. I guess aside from that, silk-touching uranium ore would be the quickest way to get plutonium now?

    Incidentally, I actually re-enabled the 8 depleted uranium recipe so I could actually make a nice plutonium reactor, something I couldn't do before due to its prohibitively high heat production.

  • I've noticed that plutonium cells seem to give a lot more depleted cells than before (appears to be about 1.5 per plutonium cell).

    Really? I have never seen any kind of cell output more depleted cells than there were active cells.

    How it should work, simplified: The cell rolls a 1d4 on expiring. If it comes up 1, 2 or 3, the cell simply disappears. If it comes up 4, then it gets replaced by a number of depleted cells equal to the number of active cells that expire.

    In other words, a single cell has a 25% chance to generate 1 depleted cell (but never any other number). A dual cell has a 25% chance to generate 2 depleted cells (but never any other number). A quad cell has a 25% chance to generate 4 depleted cells (but never any other number).

    Unless Greg changed something without telling us (again), it should be impossible to get 1.5x the depleted cells from anything.

    Maybe I need to load up Unhinged and test.