Posts by Omicron

Quote from Peppe

Sounds like the new gregtech is maybe top end uranium numbers without the high running costs. Something in the 50 million per ingot range, but with some fuel usage scenarios not actually being much better than uranium.

Ignoring cost to build for a moment would you get more out of each ingot running plutonium+thorium or eff 7 iridium reflector quad uranium?

To be relevant plutonium+thorium should beat that usage scenario, but i don't think it does, which is why i left it off my personal 1.5.1 pack.

That's pretty much what my math is trying to answer, yes - the efficiency number it computes is directly comparable with uranium. So the hybrid reactor in post #288 with its efficiency of 7.76 does beat the 7.00 of quad uranium surrounded by reflectors.

Except for the fact that that reactor doesn't work, as I just discovered in my test world today. The cooling system relies on the fact that the overclocked vent in the lower right corner cannot pull enough heat from the reactor to melt itself down, because no more heat is left to be pulled. Unfortunately the actual game doesn't work that way. You still get one vent that isn't under full load, but it's not the one in the lower right corner. Instead it's one randomly somewhere in the middle or top. I don't begin to understand how and why the reactor planner is simulating things wrong, but the net result is that the design as posted is not stable. If anyone wants to try their hand at making it stable, please share your results.

However, you definitely do beat uranium surrounded by reflectors if you simply use pure plutonium or thorium and surround that with reflectors. Quad plutonium is completely uncoolable with internal vents in all scenarios where it has more than two neighbours (and only very few work with two), but you can get dual plutonium in a 4-neighbour situation just fine. If you look into my spreadsheet, you can see some theoretical comparisons between uranium and plutonium/thorium centrifuge output in different neighbour configurations. Overall you can get more output if you centrifuge your isotopes and run them in the same reflector configurations as you would if you used uranium.

Case in point, let's do some math: one dual plutonium cell with 4 reflectors, and two quad thorium cells with 4 reflectors each (you said to ignore build cost, after all :P). You'll get 64 million + 14 million + 14 million EU, equals 92 million. For the same 10 isotopes, you could have two and a half quad uranium cells surrounded by reflectors, which would pull 28 + 28 + 14 = 70 million. So a 22 million EU advantage in a scenario that could conceivably happen in practical application. Plutonium doesn't do as good as expected here because you cannot use quad cells for heat reasons and thus it isn't running at maximum efficiency. Still, efficiency 9.2 isn't half bad.

Also you can reach reflector-like efficiencies without paying the huge cost associated with reflectors. The multi-reactor system I am planning for my next world will do efficiency 6.84 without using a single reflector. Granted, all those reactors aren't exactly cheap either, but they won't cost me emeralds and they do output a total of 1,368 EU/t all together And there is no CRCS involved, just internal vents. I bet if you tried to beat that kind of EU/t and efficiency with just uranium and reflectors, you'd be paying a lot more than I will be.

EDIT: And that's going to be a system built to look cool. If you aim for maximum efficiency you can go higher with a multi-reactor system, even without reflectors.

Ah yes, then of course it's entirely possible to build a couple of them. In fact, as far as UU-matter is concerned, disabling the need for berylium drops the cost down by more than half.

The thing is that the thick reflectors will expire. Each time you insert a new quad uranium cell, you also need to insert 4 new thick reflectors. That's what makes the reactor so incredibly expensive to run. Also, nothing against you but I highly doubt that you "easily" have the resources required for the reflectors from one quarry cycle GregTech changes the recipe and you now need roughly 10 emeralds per thick reflector. Even a 64x64 quarry in an extreme hills biome won't usually turn up 40 emeralds per full cycle.

The iridium reflectors never expire. You build 4, you can use them forever. However they are unbelievably expensive. 310 emeralds for only one of them, 1240 for the whole set. And that's not counting the other costs either.

A possible alternative here is to invest resources and time/effort into managing a breeder reactor, and turning your uranium into GregTech's plutonium and thorium that way. This can be more efficient than reflector setups, without the cost of the reflectors - and depending on which version of GregTech you run, the advantage could be small, or it could be massive. It adds another layer of complexity to the whole thing though.

Alright, I think I've found a satisfactory method to describe the efficiency of hybrid reactors, although I have yet to come up with a good name for it. See how in post #288 I talked about how much EU the reactor pulls out of a given number of isotopes? We'll just use that, and compare it to what we'd get if all the isotopes were used for uranium instead. It is convenient because uranium outputs 1 million EU per efficiency per cell, which gives extremely simple math with small numbers.

Let's take the plutonium / thorium hybrid out of post #288. It's a thorium neutral design, which makes things easy.

First, we'll figure out how much energy it produces per cycle. The cycle is 25,000 seconds long, so 25,000 (seconds) * 20 (ticks per second) * 194 (EU per tick) = 97,000,000

Second, we need to figure out how many isotopes we're using. Looking into the reactor, we see that we have two plutonium cells and 10 thorium cells. Gotta be careful here: 12 is not the correct number. A centrifuge will produce plutonium and thorium in a 1:4 ratio, and this reactor is neutral. So looking at the 2 plutonium, we should see 8 thorium cells. Why are there 10 in the reactor? Because thorium runs longer, meaning that the dual plutonium cell gets replaced partway through the cycle. That's where the "extra" 0.5 plutonium is hiding that we need to get the 1:4 ratio the centrifuge gives us. 2.5 : 10 equals 1:4. Thus the amount of isotopes consumed per cycle is 12.5. Of course, there are no half isotopes ingame, but for the purposes of math this is irrelevant.

Third, we're going to pretend that we took 12.5 isotopes - the same number we determined in step 2 - and turned them into uranium cells. And we're going to ask the question: how efficient would this uranium have to be, hypothetically, to produce 97 million EU (calculated in step 1)? Because 1 uranium cell is worth 1 million EU per efficiency, we can just directly divide the two figures: 97 / 12.5 = 7.76. And the number we get automatically equals efficiency.

This approach is a modified cell value efficiency calculation, only that it's based on isotopes and not the fuel cells themselves. This neatly avoids the problem of each fuel type having different base values and (in the case of plutonium) different scaling factors, and instead allows all reactors to be described by one reference fuel (uranium in this case, because the math is the easiest, and because standard IC2 reactors are the foundation of it all). But how to call it? Isotope value efficiency / IV efficiency? I'm not sold yet.

Now obviously, this begs the question how to handle designs that are not neutral. If you look at the 4-chamber design in post #282, that's a grossly unbalanced one. +18 thorium positive - that means it's almost entirely plutonium. Specifically, we're looking at 5 plutonium and 20 thorium cells per cycle, the latter of which only 2 are actually used. It would seem that there are only 7 isotopes at work here, and if we divided the 103 million energy output by 7, that would result in an efficiency of over 14.71!

But I think that's the wrong approach. After all, in order to get these 5 plutonium cells, you had to process 25 isotopes, even if 18 of them are never seeing the inside of the reactor and are now lying in your storage unused as thorium. Thus to be fair, you'll need to divide the 103 million EU by 25 instead, for an efficiency of 4.12. Harsh! Even pure uranium posts better numbers than that, yet this reactor is almost pure plutonium!

Doesn't that undervalue the reactor though? After all you still have 18 thorium cells that you could potentially get energy out of. How can we represent their value? Basically: we can't express it in the efficiency of the unbalanced hybrid. The only thing you can do is write the efficiency and the thorium unbalance down together: 'efficiency 4.12 +18T'. That way everyone knows that there are unused fuel cells. The same goes for thorium negative designs; those too will have to post a lower than expected efficiency figure, since they cause unused plutonium cells to pile up in your storage. I'm of half a mind here to drop the 'thorium negative' moniker altogether in favor of a system that only expresses positive values. Like something that is '-2 thorium negative' could also be written as '+0.5 plutonium positive' or 'efficiency x.xx +0.5P', because the 1:4 ratio describes that for every 2 thorium cells there will be half a plutonium cell. An added advantage is that this also works for uranium / thorium hybrids and uranium / plutonium hybrids.

What you can do, however, is rate entire reactor systems together, provided the system as a whole is neutral, or can be made quasi-neutral by controlling which reactors cycle how often.

For example, we have the efficiency 4.12 +18T hybrid discussed above. We could pair it with something that consumes all that thorium. Like, the cheap 0-chamber thorium sink reactor in post #287. Let's profile it quickly: 12 isotopes per cycle, all thorium. It outputs 26 million EU per cycle. To produce 12 thorium, we have to process 15 isotopes and generate 3 plutonium on the side, so the reactor's efficiency is 1.73 +3P. Abysmal!

But the thing is, you already have the thorium from the other hybrid. You don't need to centrifuge any isotopes at all. Run the 4-chamber hybrid for 2 cycles, and you have 36 unused thorium. This can also run the thorium reactor for 3 cycles, with no extra isotopes required - the system of two reactors is quasi-neutral. So 50 isotopes for two cycles of 103 million EU and three cycles of 26 million EU: 103 + 103 + 26 + 26 + 26 = 284. Dividing that by our 50 isotopes yields an efficiency score of 5.68. Much better than the 4.12 and 1.73 these two had on their own.

So anyway, that was a whole lot of text and math and numbers. Consider it a post worth of being my 100th on this forum If anyone actually bothered to read it, tell me what you think!

.

CRCS breaks the rules, you can't count that

I mean, even then you'll get more EU/t per heat from uranium, but you just don't care about the ratio because you can just build extra cooling towers.

EDIT - This design is not stable ingame because the planner simulates the heat distribution wrong. See post #296.

Okay, one last and then I'll stop, I promise Feel free to critique/improve by the way, I still consider myself an amateur.

[GregTech 3.04+] 6 chamber, maximum efficiency, neutral plutonium / thorium hybrid
+/- 0 thorium neutral, 194 EU/t, plutonium efficiency 8 / 9, thorium efficiency 3.4 / 7
Intended Cell Layout: http://www.talonfiremage.pwp.b…1nxk7ef4uwh9u99b8vfbsj6kg (668.8 heat output)
Cooling System Mockup: http://www.talonfiremage.pwp.b…9adr94pxpenrc7jgjhucgr280 (672 cooling capacity)

Cost Without Cells: 700 copper, 104 tin, 387 iron, 38 gold, 6 diamonds.

Advantage: Amazing efficiency without a single reflector. This thing pulls 388 million EU out of 50 re-enriched isotopes! The highest efficiency uranium reactor on the front page, at 4.67 efficiency, would give you just 233.5 million for the same isotopes. This hybrid is also thorium neutral, meaning you can run it by itself off the output ratio of a centrifuge without accruing an excess of either fuel type.

Disadvantage: Slow EU/t compared to uranium, especially considering the price of the reactor. But that's just how plutonium rolls - it rules efficiency, while uranium remains the output king.

@ skavier: Aaaah, I see now. That actually makes a lot of sense. And... hrm, that'll be annoying to work around.

In the meantime, to complement post 282:

[GregTech 3.04+] 6 chamber, high efficiency, medium output thorium sink reactor
-40 thorium negative, 216 EU/t, thorium efficiency 5.4 / 7
Intended Cell Layout: http://www.talonfiremage.pwp.b…rljkvsxyu0ua6sgdcot06g16o (566.4 heat output)
Cooling System Mockup: http://www.talonfiremage.pwp.b…mcas5t5b0fylq0cz33kpu0em8 (568 cooling capacity)

Cost Without Cells: 652 copper, 100 tin, 350 iron, 32 gold, 5 diamonds.

Advantage: Efficiency and output are good for a pure thorium reactor. Only a single cell type means it can be fully automated by the simplest of methods.

Disadvantage: Unsuitable for solo operation, but rather meant to go with pure plutonium reactors or thorium positive hybrids as a fuel consumption balancer. Running cost of 50 copper per (7-hour) cycle.

[GregTech 3.04+] 0-chamber, cheap thorium sink reactor
-12 thorium negative, 52 EU/t, thorium efficiency 4.33 / 7
Intended Cell Layout: http://www.talonfiremage.pwp.b…bcdq4g1e5iqgaexmvbbp63w8w (112 heat output)
Cooling System Mockup: http://www.talonfiremage.pwp.b…ickaqql54gf59a833zzi4cnwg (118 cooling capacity)

Cost Without Cells: 224 copper, 16 tin, 84 iron, 18 gold.

Advantage: Decent efficiency, cheap to build, negligible running costs (less than 1 copper per hour). Only a single cell type means it can be fully automated by the simplest of methods.

Disadvantage: Unsuitable for solo operation, but rather meant to go with pure plutonium reactors or thorium positive hybrids as a fuel consumption balancer.

I need to learn more about how cooling systems work.

My primary question right now is: http://www.talonfiremage.pwp.b…uo7ew8zev4w4oe6r36ve8hudc
How come these four overclocked vents at the bottom aren't drawing any heat? It's an issue I've been running into constantly the last couple of days, when it never troubled me before (and I don't think the planner was updated, or was it?)

True, good point. Though I normally don't run Factorization in my worlds because there's little reason to... I suppose this case begs a re-evaluation of that.

Okay, as a result of tonight's research, I abandoned my previous idea, and looked for some alternatives with less than 4 neighbours for the plutonium (I still might make a 6-chamber with that config someday):

[GregTech 3.04+] 3-chamber plutonium/thorium hybrid
+5 thorium positive, 144 EU/t, plutonium efficiency 6.5 / 9, thorium efficiency 2.8 / 7.
Intended Cell Layout: http://www.talonfiremage.pwp.b…njccle9jt9l7lndir1n4zyqyo (429.6 heat output)
Cooling System Mockup: http://www.talonfiremage.pwp.b…ggc6lktkiuy5jfdl9tcivw9og (432 cooling capacity)

Interesting quirk: in the GregTech computercube, this is fully stable. In the reactor planner, the normal heat exchanger at the bottom right needs to be replaced by a component (gold) heat exchanger. Probably due to the minimal heat output profile differences between the intended cell layout and the "stunt double".

Cost Without Cells: 478 copper, 70 tin, 250 iron, 26 gold, 3 diamonds.

Advantage: can be tightly packed with others of its kind. Fairly high efficiency and not too expensive.

Disadvantage: Still needs a thorium sink somewhere to balance it out. Also, each 3-chamber system needs an advanced regulator of its own if full automation is intended, which can get expensive very quickly. The alternative is of course to use pure plutonium, partnered with reflectors instead of thorium cells. Which also gets expensive very quickly :p

[GregTech 3.04+] 4 chamber plutonium/thorium hybrid
+18 thorium positive, 206 EU/t, plutonium efficiency 5 / 9, thorium efficiency 3 / 7
Intended Cell Layout: http://www.talonfiremage.pwp.b…on2insg1anlogcq75v3d15rls (505.6 heat output)
Cooling System Mockup: http://www.talonfiremage.pwp.b…1t1mj6o4a4ego9tw63ryq6800 (506 cooling capacity)

Cost Without Cells: 602 copper, 82 tin, 296 iron, 30 gold, 6 diamonds.

Advantage: more EU/t than the other design, can be tightly packed with others of its kind, would need less reflectors or advanced regulators per EU/t which drives costs down

Disadvantage: less efficient, very strongly unbalanced, slightly bigger, whether the saved regulators or reflectors make up for the more expensive reactors is unknown

Well, the best I managed so far is this: http://www.talonfiremage.pwp.b…06u1rbey64bmw0jsnfr6ccvsw

Especially vexing about this is the fact that the lower left corner melts off while the upper right corner is running below capacity (reduce the time slider to see what I mean).

It's in my spreadsheet. Dual plutonium 4-neighbour, 608. Dual thorium 1-neighbour, 9.6. Single thorium 1-neighbour, 2.4.

I've tried a few configs so far but I'm always short a few points of heat =/

Yeah, unfortunately I am stuck at work (and it's a slow day) so the reactor planner and my imagination is the only thing I have at my disposal :p

Because I have a building layout in mind where it would be visually appealing to have a few five-chamber reactors, whereas six-chamber ones wouldn't really fit.

If it can't be done I'll have to come up with something else, but it would be cool if it could fit this way.

Okay, would somebody like a design challenge? http://www.talonfiremage.pwp.b…nqcrvacz7dkk2a3c33bvd7ri8

I'm not sure this is even possible, but I'm not that good at cooling systems, so I'll defer to the cracks in this case. Can you cool the above using just internal vents in five chambers?

Mind you, this is for 1.5.x, meaning the heat output is not 399, but rather 625. The corners around the cross can be safely occupied by heat vents to siphon off some heat from the thorium (dual cell: 9.6, single cells: 2.4), so that leaves roughly 610 to take care of. You could use something like this to model the heat load in the current planner. Feel free to move things around and/or use other components in place of the heat vents if it helps.

The advantage of running a hybrid reactor is, for example, that it consumes both plutonium and thorium in a single reactor. Then you only need to build that one instead of multiple in order to utilize all of the centrifuge output. It may also allow you to hit a specific heat or power target better if you use thorium or uranium cells for the fine-tuning since plutonium cells would likely overshoot it.

If you're the kind of guy who builds dozens of reactors anyway, then it probably has less appeal since it's more tricky to automate (pure reactors don't need an advanced regulator, for example).

Updated spreadsheet: https://docs.google.com/spread…VHhEZzUtX0dTZ3E2aVE#gid=7

New tab for GregTech 3.04c for 1.5.2, with new numbers for plutonium, which Greg added in one of the recent updates. Makes plutonium much better. (thanks, Greg!) Although the scaling looks a bit odd. As if the double-pulsing feature only adds half output on the second pulse somehow. But hey, if it works...

Added centrifuge calculation for new plutonium numbers.

Also toyed with some different efficiency measurements in the new tab. Cell efficiencies compare each fuel type to itself, while uranium efficiencies compare all fuel types to uranium. Value efficiencies compare total output (and thus factor in runtime), while power efficiencies compare EU/t (and thus do not factor in rutime). Unsure if these metrics are of any use (cell power and cell value are always identical, for instance).

Ask yourself though if you really need that much breeding speed. The quad thorium breeder will give you one re-enriched isotope every 28.5 seconds. Considering the shortest possible burntime you can have in a nuclear reactor is 10,000 seconds, you would need many dozen reactors running side by side to consume fuel at anywhere near the speed this breeder outputs it.

And of course you'd need more than three stacks of uranium (220 ingots) to provide even one single cycle's worth of isotopes for the breeder.

I did, but it's not in the recommended listing and I was too lazy to check every post in a 14 page thread :p

Regarding breeders and GregTech 3.x... what do you think of this desing? http://www.talonfiremage.pwp.b…00r2chc4oghtvi6mtltb1h81s

The reactor planner shows it to overheat, but single thorium cells with four neighbours dropped from 15 to 12 heat in the update, so it should be stable at 64k heat regardless of whether or not any breeding is happening (take one isotope away to check). Just 4 chambers for the maximum stack of heating cells.

I also briefly entertained the thought of using plutonium for breeding, since it pulses twice per tick and therefore should charge isotopes like nobody's business. But then I looked at the 240 heat for the single cell, lol'd IRL, and went back to thorium. If you used a double thorium cell, which costs just 1 copper per cycle, you'd have the same breeding speed as single plutonium for just 34 heat (down from currently 42).