Jump to content

Atomic science power generation. Tips and tricks.


Recommended Posts

Well from my personal testing with default configs, an Energy Tesseract can only take in a max of 100 MJ/t no matter how many faces you input to. So you'll need multiple tesseracts sending to the same Freq to supply that Network with more than 100 MJ/t. So to feed each precharger 100 MJ/t, you have to set up 5 sending tesseracts being fed from the turbines on the same Freq. (500 MJ/t - 20% loss = 400 MJ/t)

Link to comment
Share on other sites

  • Replies 77
  • Created
  • Last Reply

Top Posters In This Topic

That's a good idea you have there. Stalling the retraction of the control rod to extent fuel rod life. I'll need to add that to the list of tips.

That is also a nice video explaining how to build it. A lot nicer to view then my picture explanation.

I'm glad you found the information useful.

Link to comment
Share on other sites

  • 2 weeks later...

Okay, so I did a bunch of testing. Putting here here to consolidate it, really, but I'll split it into a different thread if you guys think it's best.

The question I asked (and don't remember seeing on the forums yet) is "how much power can you get from a bucket of steam?" I don't have the answer yet because large turbines will happily accept more steam than is optimal for them. From the information I was getting my question changed to "what is the optimal number of large turbines per liquid manager?"

Scenario: fission reactor producing steam bottlenecked by 1 liquid manager. Manually placed/removed liquiducts to ensure exactly 128 buckets per test. 1 output connection from LM and exactly one centrally located steam input from below each LT. Exactly one centrally located power conduit from the top of each LT, merged them all together and input with 6 faces on a redstone energy conduit.

Variable: # of large turbines receiving steam.

LT# = Number of large turbines

REC = total power in MJ of Redstone Energy Conduit after power generation stops

Difference = REC value on that line - REC value down one line

Note: all MJ values +/-1000 MJ. This was figured out by doing the same test 6 different times and getting results within 1000 each time (sorry, don't have data to back that up)

LT#  REC      Difference


21  243009  -121803

20  364812  -33099

19  397911  -10775

18  408686  +12462

17  396184  -12502

16  382270  -13914

15  365984  -16286

14  345998  -19986

My conclusion is that LTs will take more than they absolutely need, reducing steam efficiency. For a single liquid manager, 18 LTs is the magic number. If you have to vary from this, less is better than more, because the dropoff is more significant about 18. This leads me to believe that either there is a small spinup time where there's less steam efficiency or LTs can consume steam without generating power when given small values.

Edit: Whoops, forgot part of it. Specifically, this is useful because an LM with one output pulls 100mB/t. So we can give a rough calculation that the optimal mB of steam/tick to each LT is ~ 5.5 = (100 mB/t from LM) / (18 Large Turbines). This isn't exact. I would wager that the actual number is 5 or 6 mB/t, because even numbers are what I would use when designing a power generator.

Anyway, using 5.5 you can get a rough estimate of how many turbines you need. If someone puts any of this into use, agrees or disagrees with me, or has more data they wanna share I would love to hear it!

Link to comment
Share on other sites

Actually, your research there was invaluable to mine! I used your info that too many connections reduced power output in my design and intentionally limited each LT to a single liquiduct input. I said in my post that I hadn't heard the question asked, what I actually meant was that I hadn't seen it answered for default config, I just didn't say that hehe >.>

Actually, now that I think about it, your single input testing backs up my conclusion - LTs with multiple inputs will accept steam beyond their minimum input, which reduces their efficiency. If liquiducts follow the same style of tank filling as conduit powers RECs, where the first one in the line gets a little more than the second than the third, etc, then that means a perfectly steam efficient turbine setup would actually have to balance the order and number of connections you give each LT. That would take a lot of testing to figure out the ratio each successive input is at, though, and I doubt it would be worth it.

We really need a mod which adds a decent tracker for how much power/fluid is moving through a pipe over time. Something like the rednet historian would be great. Can computercraft sensors get flow/throughput information, out of curiosity? Perhaps we set up a computer to monitor power/liquid networks.

Link to comment
Share on other sites

Things are looking complicated for atomic science once you update things to Minecraft 1.6.4. I've already played around with it a bit. In the 1.6.4 version, fusion reactors now have a heatup process. the amount of heat exposed to the magnets makes a bigger difference. Piping steam works a bit difference. The ballance of steam vs power has been changed quite a bit. There will need to be a whole new set of tips needed once tekkit hits 1.6.4.

But on the bright side, deuterium can be pumped, as a liquid, strait into the reactor. Huge improvement that basicly nulifies any changes to steam production. in 1.6.4 a fusion reactor can be powered by MJ and water alone. MJ + water = deuterium liquid. deuterium liquid pumps strait into the reactor. reactor uses about the same MJ to run. so reactor produces plasma off MJ, water, and a clean setup of pipes and extractors.

Link to comment
Share on other sites

Oh, that's pretty awesome. I look forward to the changes! In the meantime, I had a Eureka! moment a little bit ago: you can have a steam efficient turbine setup without having the ideal number of turbines! This is important for trying to get your reactor setup if you try a little too early in survival.

My testing previously discovered that the ideal number of turbines was about 18 for steam efficiency (copper or uranium/deuterium efficiency look elsewhere ^_^). I was wrong - it's not the ideal number of turbines, it's the ideal number of liquiduct outputs. Obviously at least one of those outputs needs to head to a turbine, but the rest can happily head back into your single liquid manager buffer.

A quick test of this theory shows:

OUTPUTS#  REC      Difference


4        106429

18        389269  +282840


What does this mean? Early on, when it's far easier to get glass and iron than copper (minium shards!) you can make an excess of liquid managers to hold your buffer and run your reactor in small batches and maintain similar efficiency as having the max number of turbines. It's not exact, but it's within a few thousand MJ. And for extremely low numbers of turbines the increase in MJ stored is nearly threefold, which is significant.


How does this work?




        ----------  Buffer

        |              ^

        |              |

        |  Output ------->  Turbines

        |  Buffer

        |    ^

        |    |

Reactor---> Storage


All storage options here are liquid managers to manage the steam. The reason this works is because there the Output Buffer restricts the total input into the Turbines/Recovery Buffer section to 100 mB/t. Since steam engines will consume more steam than they need, we just need to provide enough outputs to lower their actual input down. In theory if you were trying to get as efficient as possible you play with distances between Recovery Buffer and turbines, since the closer outputs from the input get more liquid (and thus steam), but I haven't done that.

To use this:

i = ideal number of turbines = 18 (currently)

c = your current number of turbines

d = liquiduct inputs into Recovery buffer

i - c = d

For my SMP setup, I have 4 turbines, so mine was: 18-4=14 inputs. Remember to have an equal or greater number of outputs from your Recovery Buffer leading into your Storage Bank or you will end up with basically a timer before your setup reverts to poor efficiency again.

A key thing to note is that using this system will require a *lot* more liquid managers to work properly. My setup is designed to be run in small batches. I fill up the storage bank, then pull out the fission rod and let it go until it's depleted. If you plan to run constantly, you need more turbines or it'll back up eventually, so this isn't as useful to you.

And much of this could be useless in the next version, depending on how efficient the water+MJ -> fusion -> steam -> power is, and how steam piping changes. It may be applicable early game still, but we will see ^_^

Edit: I didn't actually explain how to add inputs comparative to turbines. Fixed.

Link to comment
Share on other sites

Hi guys, I ended up building a fusion reactor copying jaklath's design for the ring type, however after getting everything put together nothing turns on. I have deuterium in the reactor, and have a full redstone cell connected in through the top, however no power is taken from the cell, and the reactor block doesn't seem to do anything. Where/how does the reactor block generate plasma, and where does it send it? In the ring design that is laid out on the ground, wouldn't the plasma consume the dirt blocks? Sorry for all the questions, documentation on this mod is very lackluster.

Link to comment
Share on other sites

Do you have the redstone conduit set to output(orange) coming out of the energy cell? If not, it's not sending any power. Just hooking up the cell to the reactor should take roughly 150Mj strait away, even before the reactor turns on.(filling the energy conduits with power) If the conduit is still default(blue), it will not output any power from the energy cell.

Documentation on this mod is nearly non-existent, which is why this topic and the original one were started. :) Any findings are useful for trying to flesh out the understanding of this mod.

Link to comment
Share on other sites

I can't remember if reactors are one of those blocks that need two pieces of conduit (1 each input/output face) or just the one for output. I would try both and see what that gets you. You only need about 50ish MJ from the cell so keep your output around that level.

Link to comment
Share on other sites

Hey guys, thanks for the replies; I finally got everything squared away, including my computercraft issues( apparently cc and rednet cabling aren't enabled to work with each other), but I have one last question. How do I know how much deuterium my reactor needs? I've been feeding it 2 cells every 3 min, and it seems to go through on/off cycles ( the noise of the reactor is always running, but from underneath there's no plasma and the reactor says no deuterium). Should I be shooting for 100% uptime?

Link to comment
Share on other sites

To maxamize output, 100% uptime is not really needed unless you can utilize every drop of steam your reactor can produce. Otherwise you end up with wasted steam. The reactor will still be able to output full steam power, even without deuterium being generated, for a somewhat random amount of time. This time is going to be short, probably no more then 1 minute after the deuterium stops being produced. Steam generation is not always consistant due to many variables in how much plasma is generated per cell and how far the plasma spreads out in your reactor. It is a rather large set of random variables to keep in mind. Only drawback to running your reactor at full power like this is the cost of setting up all the turbines, and the fact that it will be eating through deuterium cells like a mad man. BUT, it should output enough power to more then make up for it's hunger.

If your going for efficiency instead of max power, lowering the number of turbines, slightly, and finding your reactors cooldown time is key. Maximizing steam usage for a given amount of deuterium will give you a lower powered reactor that produces power for a longer period of time, vs: a much higher powered reactor that eats through deuterium quickly. The highest efficiency I have found, requires delaying deuterium insertion until your reactor is nearly cold, or 80% of it's cooldown time. Once you find how long your reactor setup will run off a given amount of deuterium cells(2 for example). Over at least 3 runs for averaging. You can then take this time and use 80%-90% of that length to set your delay. You will be lowering the potential output of your reactor doing this, but you will save on tin and glass for deuterium cells.


I have a reactor setup that is currently capable of outputting 1175 MJ/tick, consistantly(averaging out its total power output over the duration it is capable of outputting power). I have it set to take 2 deuterium cells in a row, with a delay of 1 second between the 2 cells(seems to increase efficiency but is probably just an illusion). It then sits and waits for 10 minutes, completely cooling down before restarting cold. the turbine array of 45 large turbines keeps spinning for 7 minutes after the deuterium cells are inserted. All total it nets me about 9,800,000 MJ in 7 minutes and 0 MJ for the next 3 minutes. This power is buffered into 4 sets of 10 redstone energy cells that output their power to a fully powered MFR laser drill and a 50x50 4 Deriver MFFS forcefield system with 15 speed upgrades in the generator. The reactor is self sustained, powering its own chemical extractor and auto crafting unit to create the dueterium cells from the tin the laser drill produces and sand I have piped in from elsewhere. I have never used more then 10% of the power in my buffers even with drawing extra power for other experiments on top of the constant demand from its usual sources.

If I change things up and run it as a high output reactor, it would require switching to a 2 minute delay and increasing the large turbine count from 45 too 90+. It's power output would be brought up from a measly 1175/tick, to a much more substantial 2600+/tick. the downside is I would loose some of the total output, dropping that down to less then 8,000,000 Mj per 2 cells. not a big loss though considering I would be getting that much power about every 3 minutes.

Reason I don't do that? Simple. I have no use for that much power. My whole base uses less then 1000 MJ/tick even if I have everything running at once.

As for the CC --> Rednet issue. Here is a simple fix. Computer --> Dirt block --> Rednet cable. No more feedback issues, no more conflicts. Simply works. But only one way, from the computer to the cable.

Link to comment
Share on other sites

well, it would require traveling quite a distance from your base(1000+ meters) to mine for it, or creating a mystcraft age specifically to mine in. Uranium ore has a very low spawn rate, between 0 and 2 ores in a 15 x 15 area. it spawns near bedrock

If your mining nearby, or in a place you had been to before you changed the config, there will still be no ores in that area since that part of your world is already logged by minecraft, so no new ores will be in those locations. Creating a new mystcraft age would be the same is starting over from scratch, in terms of ore generation, so it should generate the ores properly. Also traveling far enough from your base should have the same effect. newly generated ores with uranium added to them.

Link to comment
Share on other sites

  • 2 weeks later...

I just wanted to say this is a awesome thread, I have learned so much. I have built a flood type reactor fully automated (picture below) and it turned out way better then I could have hoped, my questions though is there anyway to measure the power output?


Link to comment
Share on other sites

1 reactor can output between 1100Mj/tick and 3000Mj/tick depending on how many turbines you run off it and how fast you refill the deuterium. That setup of yours... ok... 15 time 20.25 = 303.75. so 303.75Mj/tick per row(left to right in this picture) and I'm guessing your turbines are 15x15, so that's 4556.25Mj/tick potentially, as the output from your setup here. Since each Large turbine can output 20.25Mj/tick when fully supplied with steam, you just multiply that by however many turbines you have on your reactor. That's a fairly accurate measurement of it's potential output.

I have heard of people using as many as 80 turbines per reactor. Doing so reduces the run time of the setup from 5-7 minutes per pair of cells, to less then 2 minutes. Trading efficiency for sheer power. But basically the same amount of total power per cell is generated either way.

Well, you could build a Multimeter from Thermal Expansion. That device only measures how much power is going into something though. So you'de have to isolate each and every turbine and measure them individually as they fill a redstone energy cell. or you could build a huge bank of Redstone energy cells and see how many have to be hooked up to the reactor until they are no longer being filled at max rate(for your setup, would probably be at least 46 REC's). Otherwise just doing math gives you a good estimation.

Link to comment
Share on other sites

  • 2 months later...

I am playing around with the automation of a Fission Reactor via Applied Energistics in Tekkit 1.1.10. Feeding the reactor with a ME Export Bus is simple. However, I am wondering if anyone has any experience pulling partially depleted rods from the reactor using ME Import Bus? I want to conserve the mostly depleted (but still intact) rods for use in breeder reactors. I seem to be running into a strange glitch where once the fuel rod gets to about 1/3 the durability meter disappears and it will randomly flicker between perfect health and its true damaged state, until it is finally depleted and destroyed. If the durability state is inconsistent like that, then I probably won't be able to pull the rod out based on its condition.


Has anyone had experience with this glitch? If I can't sort it out, I suppose I will try to extract it with some kind of timer. Ball parking it shouldn't be too hard.


Even if it is not about the glitch, if anyone has had experience extracting rods from the Fission Reactor before they die, any insight into the mechanics would be helpful. Thanks!

Link to comment
Share on other sites

You can use the fuzzy import bus to ignore damage value ranges, or you can use a percision import bus to extract the fuel rod at the precise damage value you desire.


Is that theory, or have you done it? Because the damage values on the rod seems to tick down too quickly for the precision to compare it. Or perhaps it can only be extracted from a particular side of the block? I only tried extracting a 50% damaged rod using Precision from one of the sides. Inputting fresh rods from the top.

Link to comment
Share on other sites

Could you post a tutorial on fission breeder reactor?


A breeder setup operates on mostly the same principles as a regular fission reactor, except that you surround the central reactor with 4 more fission reactors. The outer reactors all use Breeder Fuel Rods, and the center reactor has your partially depleted Fissile Fuel Rod. After they get hot enough, the Fissile Fuel Rod will begin to heal, while the breeder rods get depleted.


For temperature control you will want a Thermometer under EACH (total 5) reactor, all connected to the same circuit, which will raise all control rods (total 4 control rods) that you place in the corners of the "+" shaped reactor design.


Figuring out how to automate it is another story, but I'll get back to the thread with news if I figure it out.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Anything claiming to be official Technic servers are not allowed here, for obvious reasons

  • Create New...