You are not logged in.

IC2 Experimental builds (jenkins):
v2.0/2.1/2.2 / 2.3 / 2.5 / 2.6 (For Minecraft 1.6.4/1.7.2/1.7.10 / 1.8.9 / 1.9.4 / 1.10)
IndustrialCraft² recent version: v1.117! (For Minecraft 1.5.2 → topic)

Dear visitor, welcome to IC² Forum. If this is your first visit here, please read the Help. It explains in detail how this page works. To use all features of this page, you should consider registering. Please use the registration form, to register here or read more information about the registration process. If you are already registered, please login here.

601

Thursday, September 8th 2016, 5:25pm

Huh? Can you explain that further? The only way I can think of that those "0-chamber" reactors can be more efficient at producing plutonium is in terms of space - they have 3 columns in one block space, as opposed to 1 extra column per block for added chambers.
That was the impression I got: the economy of scale in favor of raw plutonium output. I'm not personally convinced either that a bunch of smaller reactors will produce more plutonium than 6-chamber reactors will for the same amount of time, but that was the impression I got from his argument.

602

Thursday, September 8th 2016, 6:16pm

That design uses heat exchangers, which complicate things - the descriptions on the wiki don't adequately explain them, based on what I found when I decompiled the mod a while back, so I got special permission to adapt the decompiled code for my planner. However, none of them are adjacent to the failing overclocked heat vent.


They're not really that arcane. They just look at the heat level of themselves and all components they are able to interact with (which may or may not include the reactor core), and then move heat around between all of these reservoirs with the aim of balancing them out percentage-wise, up to the limits of their transfer capacity.

I started to explain to you here why I specifically used component exchangers here - because they don't interact with the core. In this whole design, only the OC vents draw heat from the core. No other part does. Then I did some math, and it occurred to me what the problem might be. Namely, there are 18 OC vents in this design. Each draws 36 heat. 18x36 = 648. The reactor also outputs 648 heat. So, contrary to my initial assumption, the corner vent is actually forced to draw its full load. There is no bug with IC2 here.

However, the output of your planner is not clear here. If I remove the bottom right vent, the planner says that there is 23 excess heating. However, what it actually should say is that there is 36 excess heating, because that is the unhandled amount that accumulates in the core every second. Additionally, the parts claim that they are cooling more than they actually do. It says that 625 out of 628 cooling is happening. Which makes sense at first glance, because one OC vent was removed, and therefore 20 vent cooling went away: 648-20 = 628. However, the 17 OC vents only manage to draw 612 heat from the core each second. How can the cooling system run only 3 points below capacity, when each turn it is fed 16 heat less than its capacity? I went and clicked the little "i" buttons for each active component, and indeed: all of them except two claim to be fully loaded. One is a component vent at 5:7 (row:column), which claims 14 of 16 capacity; the other the component vent at 3:5, which claims 15 of 16 capacity. Three short of maximum. But there's not enough heat for that!

So the actual thing that the planner should display in this case is both 36 excess heating and 19 excess (or better, "unused") cooling, at the same time. And the components should adequately reflect if their turn in the tick order comes up and there is no heat for them to interact with.

So, umm. Yeah. Consider it a bug report? :)

MauveCloud

Coal Miner

  • "MauveCloud" is male

Posts: 1,048

Location: California

  • Send private message

603

Thursday, September 8th 2016, 7:25pm

They're not really that arcane. They just look at the heat level of themselves and all components they are able to interact with (which may or may not include the reactor core), and then move heat around between all of these reservoirs with the aim of balancing them out percentage-wise, up to the limits of their transfer capacity.

I beg to differ. I've actually looked at decompiled code, and there are some conditional statements that I still don't fully understand that make the finer details more complicated than that. Admittedly, that might have changed in recent builds - it's been a while since I've checked.

I started to explain to you here why I specifically used component exchangers here - because they don't interact with the core. In this whole design, only the OC vents draw heat from the core. No other part does. Then I did some math, and it occurred to me what the problem might be. Namely, there are 18 OC vents in this design. Each draws 36 heat. 18x36 = 648. The reactor also outputs 648 heat. So, contrary to my initial assumption, the corner vent is actually forced to draw its full load. There is no bug with IC2 here.

However, the output of your planner is not clear here. If I remove the bottom right vent, the planner says that there is 23 excess heating. However, what it actually should say is that there is 36 excess heating, because that is the unhandled amount that accumulates in the core every second. Additionally, the parts claim that they are cooling more than they actually do. It says that 625 out of 628 cooling is happening. Which makes sense at first glance, because one OC vent was removed, and therefore 20 vent cooling went away: 648-20 = 628. However, the 17 OC vents only manage to draw 612 heat from the core each second. How can the cooling system run only 3 points below capacity, when each turn it is fed 16 heat less than its capacity? I went and clicked the little "i" buttons for each active component, and indeed: all of them except two claim to be fully loaded. One is a component vent at 5:7 (row:column), which claims 14 of 16 capacity; the other the component vent at 3:5, which claims 15 of 16 capacity. Three short of maximum. But there's not enough heat for that!

So the actual thing that the planner should display in this case is both 36 excess heating and 19 excess (or better, "unused") cooling, at the same time. And the components should adequately reflect if their turn in the tick order comes up and there is no heat for them to interact with.

So, umm. Yeah. Consider it a bug report? :)


That does make it seem like my planner is calculating excess heating and cooling improperly, though perhaps they should be called "unhandled heat" and "unused cooling" respectively. However, I think your logic is off on one point: the reactor generates 648 heat each tick, but with 18 overclocked heat vents, none of that heat is accumulating in the core while those last. The heat is building up in that corner OC heat vent, because with only 2 adjacent component heat vents, that's only 28 heat dissipated from it per tick.

I've got a computer upgrade pending (there are a few parts that haven't arrived quite yet), so it may take take me a few days to get to it. In the meantime, perhaps you could file a formal bug report on GitHub.

604

Thursday, September 8th 2016, 8:19pm

However, I think your logic is off on one point: the reactor generates 648 heat each tick, but with 18 overclocked heat vents, none of that heat is accumulating in the core while those last. The heat is building up in that corner OC heat vent, because with only 2 adjacent component heat vents, that's only 28 heat dissipated from it per tick.


My logic is not off - I completely agree with you on the above point. :P I now accept that this setup is bug-free. If I didn't make myself clear enough, I apologize.

When I talked about the planner showing the wrong thing, I chose to examine the case in which the corner vent was removed (as stated), because I thought it would make a better example.

I'll raise an issue on github tomorrow, sure.

MauveCloud

Coal Miner

  • "MauveCloud" is male

Posts: 1,048

Location: California

  • Send private message

605

Thursday, September 8th 2016, 9:15pm

However, I think your logic is off on one point: the reactor generates 648 heat each tick, but with 18 overclocked heat vents, none of that heat is accumulating in the core while those last. The heat is building up in that corner OC heat vent, because with only 2 adjacent component heat vents, that's only 28 heat dissipated from it per tick.


My logic is not off - I completely agree with you on the above point. :P I now accept that this setup is bug-free. If I didn't make myself clear enough, I apologize.

When I talked about the planner showing the wrong thing, I chose to examine the case in which the corner vent was removed (as stated), because I thought it would make a better example.

I'll raise an issue on github tomorrow, sure.

Apologies. Sometimes I get annoyed when people respond in a way that indicates they haven't read my posts well enough, but it seems I was guilty of that this time.

I have a feeling that tracking down why it says it only has 23 excess heating with the 17 OC vent setup instead of 36 will be difficult, though.

albijoe

Technophile

  • "albijoe" is male
  • United States

Posts: 51

Location: San Antonio, TX

  • Send private message

606

Friday, September 9th 2016, 4:14pm

Omicron, I found the reflector reactor to be the hottest stable reactor for fluid reactor heat generation. I have not seen a reactor generate more than 672 heat per tick, and I would love to have a more powerful one. The subsets of the main one are because I don't like rebuilding reactors, this allows plug-and-play of only key components.
On the flip side, I've found the 420 reactor to be the highest-output reactor EU-wise.
I submitted these because it looked like Korlus was asking about high-efficiency reactors, and I use these for their efficiency. Using reflectors makes them even more efficient, so I do.
If you have a higher heat reactor that only uses IC2, I want to know!

607

Friday, September 9th 2016, 5:49pm

If you have a higher heat reactor that only uses IC2, I want to know!


Hmm! Well, I am fairly rusty, but if memory serves... Unless things have changed more than I am aware of, you should be able to safely exceed 700 heat per second.

*goes to fiddle with the planner for a while*

Okay, here's one variant that does it. 0303030C0D120D0C0D0C020C0D0C0A0C0A140D0C0D0C0D0C0D0C0D120A0C0D0C0D0C0D0C0D0C0A0C0D0C0D0C0D0C0D120D0C0D110A11

At efficiency 4.43, 310 EU/t and 14 fuel rods, it isn't all that different from the "Highest Overall Efficiency" reactor at 4.67 / 280 / 12. But with two extra uranium input for worse efficiency and a power output increase barely worth mentioning, people who are concerned about efficiency would never use it - and those who wanted raw output, or plutonium, have much better options too. So a design like this was never really good for anything, and for good reason. There are various ways in which you can arrange fuel rods; some configurations can have the same efficiency and the same power output, but differ in the amount of cooling required. Obviously, those with the higher cooling requirements for the same stats never see use. This reactor uses such a less optimal arrangement.

However, if it's exclusively heat that you want, the less effective fuel rod design can become the better one. In this case, you get a fluid reactor running at a blazing 704 heat per second - or rather, 1408, after the fluid reactor bonus. Yields a total output of 28.155.058 heat per full cycle, compared to 26.876.784 for the hottest among your designs. Of course, it uses almost twice as much uranium per cycle... but, on the upside, it does not require a single reflector. In fact, adding a reflector would only lower the heat output.

albijoe

Technophile

  • "albijoe" is male
  • United States

Posts: 51

Location: San Antonio, TX

  • Send private message

608

Friday, September 9th 2016, 6:25pm

I bow to your superior reactor engineering!

Ok, two more challenges and one more question.

Question: Do you think there is any more to be gained beyond 704?

Challenges:
1. Create a reactor that generates base 690-700 heat, so it will not need an additional fluid exchange but is still significantly better than 672.
2. Create a reactor that requires only one type of fuel (Only Quad Rod, or Only Double rod) so it can be better automated

609

Friday, September 9th 2016, 7:09pm

Those probably won't work out, I'm afraid. The smallest step down is converting the dual rod into a single, and that will drop heat output to 668. Upgrading to all quad rods would increase the heat load to 816. Placing the dual rod in the middle of the T-junction and having quads all around would drop the heat down to 600, even.

You are fundamentally limited in how much heat you can dissipate by the number of OC vents you have in any given design, since those are what primarily pulls the heat from the core (where the fuel rods dump it, since there are no active adjacent components). Each vent pulls 36 heat. However, it can cool only 20, so it needs to be surrounded on all four sides by component vents to avoid meltdown. That's how you get the checkerboard pattern you see in almost any reactor design. At the edges of the the field, there is no more room to fully enclose the OC vents, and so you have to remove some OC vents and some component vents to add heat exchangers and advanced vents instead, which handle the excess load from the not completely stabilized OC vents. However, doing so costs you both cooling capacity and core transfer capacity.

This design here has room for 19 OC vents: 19x36 = 684. Additionally, there are three advanced and two basic heat exchangers in the grid. With the basic ones able to do 4 and the advanced ones able to do 8, they are what makes up the 20 missing core transfer not handled by OC vents. In fact, they could make up as much as 24. Which happens to match the excess vent cooling capacity. (For that reason I was able to use a component exchanger without core transfer in one spot, to save cost. The extra transfer of an advanced vent would not translate into extra cooling because the vents are already maxed out.)

The fuel rods need to be surrounded on all sides by passive components (things that don't accept heat, like component vents), so they dump their heat straight into the core. If an active part was adjacent to a fuel rod, it would receive all of the heat output of that rod and just melt, because these are big quad rods running at elevated efficiencies. Now, we have already established that this T-shape can't produce other useful heat values (see the start of this post). However, if you go and start allocating additional slots to fuel rods - or even try to change their shape into a square - you have to break up the cooling system. The square is particularly bad because you want that square surrounded by passive components, but you want those passive components in a checkerboard pattern, not in straight walls. Any change in the fuel rod configuration, no matter how you twist and turn it, results in the loss of at least one OC vent.

And that lost vent will drop the cooling by at least 20 (likely more, since you are likely to lose a component vent as well), and what's more important, it will drop core transfer by 36. So instead of 708 core transfer, you will end up with... drumroll... 672, and you're back where you started. It's no coincidence that precisely that number showed up, because the number of OC vents in any given design is just that defining of its performance.

TL;DR: there's no room (literally) to tweak this design further, especially not by small heat steps. And since this is one of the very few configurations that can safely fit 19 OC vents without any one of them melting, you're not going to be able to find many (if any) different setups of similar potency.

This post has been edited 1 times, last edit by "Omicron" (Sep 9th 2016, 7:15pm)


albijoe

Technophile

  • "albijoe" is male
  • United States

Posts: 51

Location: San Antonio, TX

  • Send private message

610

Friday, September 9th 2016, 8:57pm

The discussion and results are all much appreciated. Thank you.

Outlaw2101

Tree Cutter

  • "Outlaw2101" is male
  • Australia

Posts: 2

Location: Australia

  • Send private message

611

Thursday, September 22nd 2016, 9:25am

Hey fellas.

Thought I would dive in and share our setup that we run on our server. I'm no nuclear engineer - it has been a long time since I last used IC2 Nuclear Power. Its probably been about 3-4 years since I last touched nuclear power? It was when Ice and water could be used to cool the reactors anyway. But I believe this is my first stable IC2 Experimental setup. (I really should say ours since it was a joint effort involving many players and months of work to set this up).

We run four Liquid Cooled Reactors. Each reactor has its own Coolant System set up which consists of 22 Liquid Heat Exchangers and 22 Stirling Generators. The reactor and its cooling system are connected together with Thermal Dynamics Fluid-ducts. The reactors rely on a Redpower circuit. Its quite difficult to explain in words, but I will try explain it as best I can. What this circuit does is it automatically regulates the running time. Each reactor has its own individual redstone circuit. When you turn the reactor on, it will trip a Thermal Monitor set to 300, however a Toggle Latch overrides this signal and the reactor will continue to function. When the reactor reaches 30% Core Heat, another Thermal Monitor set to 3000 will trip and override the Main control, switching the reactor into "Cooldown Mode" (off). Once the reactor cools down past the 300 Mark on the Thermal Monitor, the reactor will once again re-start and heat up again. This process continues over and over and over again. We have run the reactors for two and a half hours on this automated set up and thus far we have not had any catastrophic failures. However, we have yet to see how long it can run for. Each Reactor outputs 456 HU/s, except for Reactor 3. Reactor 3 is a very tempermental one and I don't know why. It only outputs 432 HU/s despite having the same setup as its siblings.

Unfortunately our setup is very expensive to make. We have four of them. So go figure. We wanted the best of the best in this project.

For just one reactor you need.

16 * Coal
704 * Copper
21 * Diamond
666.500 * Iron
96 * Lapis Lazuli
192 * Redstone
576 * Rubber
202 * Tin
32 * Uranium Fuel

Now multiply those statistics by four :P

Here is the code for the IC2 Reactor planner: 0A120A0A0A0A0A120A1203120A120A1203120A120A1203120A120A0A121203080312120A1203121203121203120A120A0A120A0A120A

I do not quite know how to read the new Planner so please, if there is anything that should be brought to my attention, please do not hesitate to tell me.

For this setup, you need IC2 (obviously), Thermal Dynamics, Project Red and Nuclear Control. We use Wireless Redstone to control our reactors as it allows for clean and tidy circuitry. I recommend it, but it is not necessary for this setup to function.

If anyone has any suggestions on how we could better improve on our setup, please let me know. I am eager to expand and optimize this setup as best I can!

I have shared some screenshots below.

http://hostthenpost.org/uploads/cf81c860…37e9c766769.png

http://hostthenpost.org/uploads/9fde3c79…81451fbe000.png

http://hostthenpost.org/uploads/fbd836fc…98e99498bb5.png
"Give a Man a Fish, and You Feed Him for a Day. Teach a Man To Fish, and You Feed Him for a Lifetime" - Anne Isabella Thackeray Ritchie

612

Friday, September 23rd 2016, 12:00am

You know you get more EU for your HU using steam turbines instead of stirlings right? I just run a single quad fuel rod surrounded by 4 neutron reflectors and get 840 HU/t. 4 superheated steam boilers consume 800 and the two turbine/kinetic generator pairs put out a combined 600 EU/t. It is cycled on and off based on cold coolant tank level. Tanks and pipes are from enderio.

Outlaw2101

Tree Cutter

  • "Outlaw2101" is male
  • Australia

Posts: 2

Location: Australia

  • Send private message

613

Friday, September 23rd 2016, 7:26am

Thanks for your reply!

I did experiment with the whole turbine jazz but the constant explosions of steam just did my head in. I couldn't build a setup that would have no steam explosions so I ruled that out. I was going to have a drone of explosions 24/7 that would drive everyone batty.
"Give a Man a Fish, and You Feed Him for a Day. Teach a Man To Fish, and You Feed Him for a Lifetime" - Anne Isabella Thackeray Ritchie

614

Saturday, September 24th 2016, 3:58am

If you do it right you only get explosions during startup/shutdown. You need to keep hot coolant feeding into two fully populated heat exchangers per boiler for 200 hu/t, and set them to 221 bars of pressure and 1 mb/t.

615

Sunday, September 25th 2016, 5:54pm

If anyone has any suggestions on how we could better improve on our setup, please let me know. I am eager to expand and optimize this setup as best I can!


Hi Outlaw,

Your setup raises a few questions in my head.

For starters: 22 liquid heat exchangers per reactor? That's a wee bit excessive. Those are capable of 2,200 hU/t in terms of heat throughput. But each of your reactors only outputs 456 hU/t. You should therefore be able to get by on 5 exchangers per reactor. Heck, you could take the 22 exchanger bank from one reactor, remove three exchangers, and use the remaining 19 to cool all four reactors at once!

Second thought: if reactor #3 has consistently less heat output than the others, but uses the same blueprint, then you made a mistake in inserting the components. One or two of the heat vents aren't getting any heat shuffled to them because something is misarranged compared to the blueprint.

Third thought: your reactor blueprint contains two advanced heat vents that literally do nothing, even when everything is inserted correctly. They are in the top row, in the fourth and sixth slot. Since advanced heat vents cannot draw heat from the core, and there are no heat exchangers touching these two vents, they sit idle even when the reactor is under load. You can remove them with no adverse effect on reactor operations, which saves you eight advanced vents worth of construction materials across your four reactors.

Fourth thought: your reactor blueprint as a whole is no good. I don't mean to be disrespectful to whoever came up with it, but viewed objectively, it performs very poorly for its build cost; there are reactors half the price with more throughput. And I'm going to try and explain to you why that is.

You have a fuel rod setup that produces 2048 hU/t, but your cooling setup only transfers 456 of that - barely a quarter - out of the reactor and towards the power production. Which is the reason you have to run it pulsed. But then why do you have such a massive fuel rod setup in the first place? It even has active running costs, due to the quad rods and especially the thick reflector. Like, you could remove the central cross of four quad rods and the reflector... just remove them outright, and the reactor would still faithfully output the same 456 hU/t. In fact, it would still need to be pulsed, because the four outer quad rods on their own still output 768 hU/t.

As for pulse configurations in general: people use these because they want to run a fuel rod setup that is impossible to stabilize on internal cooling without runtime pauses. But this does not increase your power production output. A reactor that runs on 500 hU/t with 100% average uptime produces exactly the same power output as one that runs on 1000 hU/t with 50% average uptime. Therefore, you don't use pulse configurations to maximize power throughput. Instead, you use them to maximize fuel rod efficiency. As efficiency increases linearly, heat output climbs exponentially, so if you really push your fuel rods to those 5's, 6's, or 7's in efficiency, keeping them cool may become impossible. At that point, you start pulsing. You accept a penalty in power throughput in order to maintain the highest possible efficiencies for your fuel rods.

However, your setup isn't doing this. Your setup runs at efficiency 3.5, and uses lots of fuel rods instead. Which means: your setup is a power throughput configuration. In other words, the kind of configuration you should never be pulsing. Now consider the fact that there are existing reactor designs that transfer over three times the heat of your setup towards power production... with a higher fuel rod efficiency than the build you have here... while also running continuously, no pulse automation required... and you should start to see why I am unfortunately not impressed with your setup ;)

Additionally, your setup uses extremely expensive components without regard for whether less expensive ones might have done the same job, and it arranges them poorly. You have heat exchangers next to heat exchangers in multiple places, which do nothing but shuttle heat back and forth between them, instead of actually providing cooling. Also, the two heat vents that don't do anything (see point three, above).



I'll gladly suggest some alternative designs to you, but you'll have to tell me what you're actually after. Do you want to maximize efficiency, because you are super short on uranium, which should never happen with normal IC2 ore spawn rates? Do you want as much power throughput as possible in order to feed your base's energy needs? Or do you want to burn through as many fuel rods as possible in order to rapidly obtain useful amounts of plutonium?


EDIT: for what it's worth: I really dig your reactor room! It totally looks the part, and everything is clean and organized. :)

This post has been edited 2 times, last edit by "Omicron" (Sep 25th 2016, 6:13pm)


616

Monday, September 26th 2016, 6:15pm

Or do you want to burn through as many fuel rods as possible in order to rapidly obtain useful amounts of plutonium?
I'm actually interested in this myself. I posted a setup earlier in this thread for a 43 rod setup, which under use of MOX, should be the fastest output of the most plutonium (assuming constant supply of uranium.) Have you found differently?

617

Monday, September 26th 2016, 10:53pm

43 is pretty good, not sure if anyone can beat that.

618

Thursday, March 2nd 2017, 9:56pm

This is my first post on this forum with 2 of my first save setups

1:Rods 6 dual fule rods(uranium) with 240EU/t and an Efficiency of 4: 0D140D0D0C0D0C0D0C0C130C1413140D14130D140D0D0C0D0C0D000D0C0D0C0D000000021413140D14130002020D0C0D0C0D00020202
1.1 pretty much the same with one thick neutron deflector: 0D140D0D0C0D0C0D0C0C130C1413140D14130D140D0D0C0D0C0D0C0D0C0D0C0D0C0000021413140D14130002020D0C0D0C0D08020202
2: 6 dual fule rod (MOX) with 200EU/t at 0; 600EU/t at 5.000; 760 at 7.000; 880EU/t at 8500.
000000140A000000000C0A0C0A0C0A0C0A0C0A1405140A140A140A0C0A050A0C0505050C0A1405140A140A140A0C0A0C0A0C0A0C0A0C
2.1 pretty much the same with one duel rod more and 968,00EU/t at 8.500
0000000A050A00140A0C0A0C0A0C0A0C0A0C0A1405140A140A140A0C0A050A0C0505050C0A1405140A140A140A0C0A0C0A0C0A0C0A0C
i hope i didn´t invent the weel new.
both setups are easy to automate becouse they both use only dual fule rods.

Counter:

Hits today: 14,427 | Hits yesterday: 1,368 | Hits record: 152,331 | Hits total: 65,310,518