Atomic science power generation. Tips and tricks.

[weirleader]

my main problem is that I haven't yet reached the 'atomic age' (I started from scratch with 1.1.8 and don't have nearly the time to play that I'd like); and I'm hesitant to even touch the wiki for things that I have absolutely no experience with. But my AE network is ALMOST fully up and running and at that point perhaps I'll dip into atomics. So hopefully within the next couple of weeks I will try to get a handle on things and transfer over all your guys' helpful posts (I've been keeping tabs on several really good threads). And then you can clean it up for me with better info.

[Flextt]

With this thread and the thread "atomic science fuel life config" atomic science is a lot more accessible. The most important ground rules are summed up in jakalths posts. Be careful though: atomic science tends to a bit of randomness. What worked for jakalth awesomely was a major disappointment in my case. But using a single reactor, his torus design is pretty powerful (5-10 mio. Mj per Cell). You will have to try and modify on your own and dismantling atomic science contraptions is a huge pain in the butt

Addendum: What always worked better for me: Not using a multi-tiered layout of steam funnels, but stacking funnels on top of each other, until I have an even layer of funnels. That is to say, I still use multiple levels of water

edit: jakalth, I think, for posterity and being a major source, referencing the "Atomic Science Fuel Life Config" thread would be a good thing.

edit2: jakalth, thanks for your quick LUA program in computercraft. If you set the first sleep period to 0,5 (it takes double values), it will only send one pulse. This should be taken with a grain of salt however, as I play in SP. Not reliable

[Steve From Marketing]

Some of this might be repeating jakalth has already mentioned but I did some more singleplayer creative mode testing with a fission reactor, liquiducts, steam storage, redstone energy conduits and redstone energy cells.

1. Storing steam is, in my opinion after my tests, a futile attempt at best. I had 18x4 Liquid Managers (128 buckets per Liquid Manager) and a fission reactor running at peak will fill all of these in a matter of minutes. Then considering that when you have to pipe the steam out it will pipe out at max rate of a liquiduct and even handful (approx. 20) will consume the stored steam fast. So a direct feed to your turbine liquiduct line, in my opinion, is the best solution although there might be some loss of steam. Also consider the amount of liquiducts, liquid manager and rednet cabling you have to make for this storage.

2. A turbine has no "spin up" time if it is fed steam at maximum rate be it small or large steam turbine. It will instantly generate peak rate power even though the block animation might make it look otherwise.

3. A small steam turbine generates 2,25 MJ/t. A large steam turbine generates 20,25 MJ/t. 9x2,25MJ/t = 20,25MJ/t (!!!). So the only plus side of greating a large steam tubine seems to be the possible savings in conduit and liquiduct lines. And perhaps less server load? This was tested by small and large turbines separately connected by liquiducts from stored steam and power outputted to separate empty redstone energy cells and measured at cell input with a multimeter. Average for 20 ticks were 2,25 MJ/t and 20,25 MJ/t.

4. Only a single input of steam from any side expect dead center is required for a turbine. There seems to be no gain/loss if a turbine has more than 1 input of a liquiduct.

5. A turbine, small or large, has only one power output and that is dead center above.

6. Most "efficient" layout for large steam turbines would seem to be 3x7 and 3 high as "described" below: (T turbine, L liquiduct, P power conduit, A air)

1. level

TTTLTTT

TTTLTTT

TTTLTTT

2. level

AATTTAA

APTTTPA

AATTTAA

3. level

AAAAAAA

APPPPPA

AAAAAAA

With this setup a 3x7x3 space has 3 large steam turbines. Extending this along the liquiduct at first level center makes it nice and extendable along the liquiduct axis. Also it is easy to extend it parallel like this:

1. level

TTTLTTTTTTLTTT

TTTLTTTTTTLTTT

TTTLTTTTTTLTTT

With this kind of setup you can place 30 large steam turbines with the confines of a single chunk (16x16) and 3 high. There is 1 block wide space left on one side (axis) of this layout and 2 block wide space left on the other side (axis) to run for example liquiduct lines.

7. Not knowing the mechanics and properties of liquiducts (how much they can transfer and is it per "face" and so on) there is the possibility of bottlenecks and optimization improvements.

[Digdug83]

Liquiducts can transfer 100 mB/t (multimeter says that anyway). They can hold variable amounts of fluid depending on what's being stored, with the default being 500 mB.

[phazeonphoenix]

I recently figured liquiducts out. Liquiducts transfer 100mb/t PER CONNECTOR. If you have 4 inbound connections to a single tank you'll get 400 mb/t. And how much liquid it can hold is dependent on how many segments are in the run of liquiducts. Each segment adds 500mb to the total. Liquiducts don't function like pipes. They are basically giant tanks that take any shape. The data a multimeter gives you when used on a liquiduct is actually not for the segment but the whole duct. The liquid level represented in it's graphics are also for the whole duct. On the other hand, if you connect a BC waterproof pipe to a very low flow liquid source like a sewer you'll see the liquid running through the pipe as it goes. BC pipes have the notion of "where" the liquid is in the pipe, liquiducts are just big tanks.

[Steve From Marketing]

Slight sidenote concerning the power storage of the aforementioned steam turbine layout running at max speed.

The produced power of 30x20,25MJ/t required 7 redstone energy cells to store it without the loss of 7,5MJ/t.

Also note that the input cap of a redstone energy cell is NOT per face. Even if the cell has 6 inputs the maximum amount it can take in is the 100 MJ/t (defaults to max).

[Digdug83]

Actually REC's can take in more than 100 MJ/t per face if you hook the REC directly up to a tesseract/power source that can pump greater than 100 MJ/t. It's only output that stymies them. I ran timing tests with tesseracts and confirmed this. Conduits may or may not have an input bottleneck when hooked to an REC though.

[Steve From Marketing]

Actually REC's can take in more than 100 MJ/t per face if you hook the REC directly up to a tesseract/power source that can pump greater than 100 MJ/t. It's only output that stymies them. I ran timing tests with tesseracts and confirmed this. Conduits may or may not have an input bottleneck when hooked to an REC though.

Interesting!

I tested this just now in singleplayer creative mode and indeed an REC connected directly from 5 sides to a Energy Tesseract (per side) recharged the cell to 600k MJ in approx. 60 seconds. This is 10k MJ/s which is 500 MJ/t which is 100MJ/t/Energy Tesseract.

Now this could be a bug or a feature and without being able to see the source code of the cell and conduit it is hard to say which it is.

[Steve From Marketing]

Concerning your Ring Type Fusion Reactor setup. I tested this in singleplayer creative and it produced enough steam to run 96 large steam turbines. After checking the liquiducts with my multimeter after adding 24 large steam turbines at a time it would most likely produce enough steam for another additional 24-48 large steam turbines.

This amount of energy production is starting to push the limits of Redstone Energy Conduits, afaik. IIRC someone tested that they would go +++2600 MJ/t.

[Digdug83]

Behold, my completed single flood type fusion reactor! It would be a lot cleaner inside if autarchic gates' single pulse would work like it should but *shrug*

http://imgur.com/a/Swd1P

[phazeonphoenix]

Me thinks you have an excess of emerald there Diggy... LOL

[Digdug83]

I only used 6.5K emeralds to make that I think. What's sad is that the outer chamber only consumed about half of my supply heh. All hail the laser drill! The other ore blocks are ugly IMHO and I wanted something neat looking and uncommon.

[jakalth]

Oooo! Pretty! But, can it make coffee? *pushes the big green button and a cup of cappuccino comes out* Ok, you win...

Seriously, that is a nice looking reactor. Lots of fun gizmos and doodads inside the main housing.

[Daemon_Eleuel]

http://imgur.com/a/IZ1xB#0

Time to share mine I guess

It's a 5x flood type fusion reactor.

Using about 500 liquid managers to hold the steam (enough buffer for a single cell run) and still working on the turbines.

Energy is stored in a REC cluster of another 500 RECS.

Still wondering how to control the input of cells. I think I should use computer craft (sending a redstone signal to the reactor when the liquiducts are empty) but I still have to figure that out :)

[Tuxmelv]

Concerning the Conduits max transfer, I seem to remember in another thread to be determined at 3000 mj/t. Is the 3000 mj/t the total network saturation or just the output limit? As this limit can be accomplished with just 30 RECs outputting at 100 mj/t each, segregating RECs into banks of 30 and network segments of 3000 mj/t sounds like a nightmare when using multiple power sources (biofuel and atomic science).

[phazeonphoenix]

I'm not sure it will be an issue. Are you saying your power draw is more than 3000 mj/t? If that's the case maybe a single source of power isn't a feasible design to utilize. Granted I'm just one player and I don't always design the most power efficient set ups but I can't imagine ever needing that much power. And if you do, large amounts of mining lasers for instance, it'd be easier to give it a dedicated power generation source.

[Digdug83]

I believe conduits only contain power if there's power connected to them, although I haven't tested that. I was the one fiddling with the conduit rates and I was basing it off of using TE machines with set power consumption rates and then testing network saturation. So in answer to your question I'm pretty sure it's the output or transfer limit.

[weirleader]

I do think conduit stores some power, though I have no numbers or testing to back that up.

[Tuxmelv]

I just like to have as much power as possible and future proof my builds. Been wondering on how to bridge different power sources for a while now. I'll have to test this out. I'll just set up a bank of 30 RECs then a bank of 60, then count the time till they drain under the same load.

[Digdug83]

According to my testing (one full REC and one empty REC hooked up with 6 connections each, one all input and one all output) the conduits only draw energy if it has somewhere to go other than the conduit. In my experiment when I switched the faces of the full REC to output the conduit network did take some power, but then gave it right back to that same REC when I switched its faces back to input mode. While it does appear that the conduits can store some (I couldn't get a reliable number) residual energy if you suddenly yank the object receiving power, for the most part they are nothing more than conduits.

Network saturation is nothing more than a function of how much of the conduits inputted energy is being used at that particular moment. So if you were inputting 20K MJ/t but only using 300 MJ/t across all the conduit lines it would say you have a network saturation of 90%.

Honestly my big hope is that TE can somehow make tesseracts function the same for accepting power as they can for spitting it out. Right now a tesseract can easily feed as much energy into an REC as you can output to it (ie it's not limited by the 100 MJ/t limit the REC says in the GUI). Outputting from an REC on the other hand creates a bottleneck, with a maximum of 5 (assuming one for input) connections or 500 MJ/t per REC.

My fusion reactor runs 4 banks of 5 REC's each with one input face and 4 output faces (one face is used to determine when it's running low), all tesseracts, so my reactor battery can output up to 8000 MJ/t. I honestly can't imagine needing that much power like....ever, but I built it more for the entire server than just my needs anyway :P

[Flextt]

Digdug, I don't get that part about the transfer rates of RECs and ETs. I had 4 faces of a single ET connected to an Energetic Infuser. The device worked with roughly 100 MJ / t no matter what I did. Now I have 4 ETs connected with 1 face each to an energetic infuser which runs with 360 MJ / t which is close enough to the approximated optimum of 400 MJ / t. Applying these observations to your logic however, I would need 80 ETs being fed by my reactor...

oh shit wait. I have a modular setup of 4 turbines feeding into 4 faces of a single ET.... does this limit my available MJ / t so much?!

[Digdug83]

I know I explained this earlier somewhere but I'll reiterate it now. I am running with some modified values in terms of TE and AS files. In TE, my tesseract output has been pushed to 10K with 0 loss. In AS, my turbine efficiency has been increased to 10 from 4.5.

Now that that's out of the way, let's see if I can puzzle this out lol. ET's have an apparent limitation of 100 MJ/t per input face (ie where the power goes into it to be transferred). Whether or not the power loss is factored before or after transfer is beyond my knowledge as I'm lazy and have always played with 0 loss. The output face (ie where the power comes out from) is not so limited, although by default I believe it's set to 500. That would mean that, assuming a vanilla loss of 20% a tesseract can only output 400 MJ/t. What I'm not sure on is whether that's a face limit or a total limit (again, I play with a large output limit for laziness reasons).

A vanilla large turbine should produce somewhere around 23ish MJ/t (I think someone said 22.5). If you have only 4 turbines you couldn't possibly be feeding more than 100 MJ/t anyway so now you have me confused.

[Flextt]

Okay, but I previously had the energetic infuser being fed by 1 ET with 4 connected faces. However it never pulled more than approx. 100 MJ / t, making me think I have bottleneck somewhere.

[Digdug83]

I wouldn't connect it to more than one side of the tesseract since I believe the EI can only use 400 MJ/t anyway, which is close to the maximum transfer limit for the tesseract. It sounds like you have an input bottleneck somewhere to me but honestly it's hard for me to tell without looking at it heh.

[Flextt]

So with my modular turbine makeup, my ETs only get fed approx. 80-90 MJ / t each. To boil it down to a single point: Since I have 20 of these, is their voltage not cumulated?

edit: i don't get the dynamics between conduits and Energy Tesseracts... I just adjusted my large turbine setup, so that 5 large turbines are connected to one tesseract face. i got that idea out of digdug83's insights. however, now the laser drill prechargers perform subpar, even though I got 80+ turbines...

so 20 turbines connect to one tesseract in total and create approx. 400 MJ / t

edit2: how can i turn off tesseract loss? cant seem to find it in the cofh config