Project: Fusion Reactor - a chronicle (WIP)

Pyure

Not Totally Useless
Aug 14, 2013
8,349
7,216
383
Waterloo, Ontario
I don't think they really mimic real-life resistors in that regard, I'll play with them tonight and see what I come up with. Out of curiosity, what input/output values were you trying to achieve?
I've heard the same. And I can sort of see why: Reika currently maps the shortest route for a current, and that's intensive enough. Parallel inputs would require mapping considerably more.

Trying to achieve: Any precise 3-digit amperage that doesn't end in zero I believe. Say 105amps.
 

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
I've heard the same. And I can sort of see why: Reika currently maps the shortest route for a current, and that's intensive enough. Parallel inputs would require mapping considerably more.

Trying to achieve: Any precise 3-digit amperage that doesn't end in zero I believe. Say 105amps.
Hmm...do things work if you need not be that precise? I understand the Electricraft batteries have only one output, so you need to make parallel systems. Also, where does power go in?
 

Pyure

Not Totally Useless
Aug 14, 2013
8,349
7,216
383
Waterloo, Ontario
Hmm...do things work if you need not be that precise? I understand the Electricraft batteries have only one output, so you need to make parallel systems. Also, where does power go in?
You don't need to be that precise at all; I just sometimes want to send X torque to a machine and it bugs me to send X+5 torque even if power efficiency is basically irrelevant due to fission reactors :p
 

Kirameki

New Member
Jul 29, 2019
310
0
0
OK I've discovered something very odd with ElC:

For some reason the inline output of the shaft junction is sending power but the generator is ignoring it. Because the generator is producing no power, somehow the active generator's output gets nullified (if I break the connection to the inline gen, power flows through.) Not sure how that works...
That aside, I did discover that if you have two induction generators, each forms a separate path through the resistor. If you set the resistor to the expected combined current, all the current goes through. If you half the resistor, you get the expected result as if it were set to the original value. Example: two turbines, resistor set to 3300. if I combine them with a shaft junction and then use a single generator, I get a fixed 1.7GJ output. If I use two induction generators, I get the full fluctuating 1.7-1.9GJ output. However, if I then change the resistor to 1600 (not quite half but close enough), then I get 1.6GJ.
I was trying to use the above image setup to test parallel current, but....it doesn't seem to want to work with the junction involved. Not sure if I should invoke the summoning ritual for this...
(Should this be split off into a different ElC discussion thread elsewhere?)

Addendum: If I move the inline gen sideways to bevel off the junction, it works fine. Wat
Addendum #2: I managed to get 105A with this setup; one resistor on one branch set to 100A, one on the other branch set to 5A. Not the cleanest configuration but it works.
 
Last edited:

Pyure

Not Totally Useless
Aug 14, 2013
8,349
7,216
383
Waterloo, Ontario
I'd probably suggest we quit polluting Iedra's thread, except that he's likely probably a bit interested himself.

I've done your exact same layout to test parallel and came up with the same results. I noticed that if one line is shorter than the other, it picks the shorter one.

Just to clarify, you managed to merge two inputs to get 105A? I wasn't successful there.
 

Wagon153

New Member
Jul 29, 2019
3,150
0
0
I'd probably suggest we quit polluting Iedra's thread, except that he's likely probably a bit interested himself.

I've done your exact same layout to test parallel and came up with the same results. I noticed that if one line is shorter than the other, it picks the shorter one.

Just to clarify, you managed to merge two inputs to get 105A? I wasn't successful there.
This makes sense realistically since electricity prefers the path of least resistance.
 

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
Today.....I went mining. Not with a pickaxe of course, but with a digital miner, a Tesseract and an Ender Chest. I needed Chisel Marble, as you'll understand when you look at some of the screenshots. And glowstone. Because when I switched glowstone dust to my farming storage because of Magical Crops, I forgot to add a destination to my sorting system and as a result, all glowstone produced lately was sent into the ME Condenser. I still had more than 2000 but they were gone fast. This had to with another little project. Believe it or not, I had to build....yet another production facility. What the hell for, you ask? Well, someone upthread convinced me to power my reactor startup through ElectriCraft components. I needed an Auroral Battery for power storage anyway and the only way to deal with the power levels involved was to use superconductive wire. For that, you need liquid nitrogen. This neat little contraption fills uncooled superconducting wire with liquid nitrogen: the Glacial Precipitator makes ice, the Refrigeration unit uses that ice to extract liquid nitrogen from the air, and the filling station fills any container you might want to put in. Put the uncooled wires into the hopper and throw the switch, that's all.



BTW, there may be a bug in this: filling the cables does not consume any nitrogen. As long as there is some in the filling station, the cables will be filled and this takes zero time. Also, I would have preferred to make ice using RotaryCraft machines, but I couldn't find any method that didn't amount to "go to an arctic biome and use the fluid crystallizer. RotaryCraft and temperatures - a neverending story.

Now back to the actual reactor. I made the components for the solenoid, but didn't get to it today. Instead, I significantly extended the rooms below the toroid and built the remaining hydrogen preheaters and the basics for the power distribution. See below the inside of a hydrogen preheater and the whole room finished. Or rather, almost finished, since the power infrastructure is not yet in place.




Also, right under the solenoid, there will be my power hub. The auroral battery is expensive in diamonds, but more surprising was the insane amount of glowstone used up - IIRC it was about 2000. Anyway, while its capabilties are right on par with those of a fusion reactor - even that gargantuan machine should take a few hours filling it - visually it's singularly unimpressive. So I built a pseudo-multiblock around it. This is how such a powerful device should look - collaboration of Chisel, ProjectRed and EnderIO:



The following picture shows where the power is transferred from my 5GRF capacitor bank filled by my Big Reactors reactors. Four tier 5 Magnetostatics generate shaft power which is transformed into electricity by four Induction Generators and sent into the battery. The battery has so far refused to let any energy into itself in spite of everything else being fully functional. I hope I can squash this bug in short order. Meanwhile



This is an impression of the other side while I was building. The modular sockets are used to output redstone directly into the engines, since they only react correctly to that and wires won't connect. The second shot shows where to look if I want to check the power throughput on the "shaft power" side.




If you've read this far, one thing should have been brought home in no uncertain terms. If you want to build a fusion reactor, building the actual reactor components is the least part of the whole project, except in resources. As far as I'm concerned, it's not even the most interesting since the reactor components have a fixed structure and with the exception of the preheaters, even a fixed layout. The supporting infrastructure is thus the more challenging part, and it's also where you can get creative. Oh, and one last comment: I mentioned in the previous post that you need to run 25000+ lodestone through a compactor to create the magnets. The exact number you might want to use is 26112, which results in 204 permanent magnets of 4096 Tesla, of which 84 are used for the auxiliary magnets and the remaining 120 put through the compactor one more time to get the 60 permanent magnets 16384 Tesla you'll need for the central magnet. Using 26624 Lodestones at the start you'll end up with some leftover, but have less manual work around the end.

So much for today. I really hope I can find out what's causing that battery to misbehave since I won't be able to move on until this is either fixed, or I decide I'll power it all with Magnetostatic engines directly after all.
 
Last edited:

Kirameki

New Member
Jul 29, 2019
310
0
0
I'm really impressed at the aesthetics of your design, love the Auroral battery "multiblock" you created. Out of curiosity, what kind of block are you using for all of the walls/floors?
 
  • Like
Reactions: Pyure

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
I'm really impressed at the aesthetics of your design, love the Auroral battery "multiblock" you created. Out of curiosity, what kind of block are you using for all of the walls/floors?
Thank you! The walls and floors are all variations of Chisel marble. That's why I went out mining more of it. Main walls: Marble block. Floor in the capacitor bank room: arranged marble tiles. Floors everywhere else: marble blocks (note the plural, as to distinguish it from the wall variant). Ceilings, archway frames and a few other selected elements: classic marble panels. I had to use the panels in some of the wall lining because neither Carpenter's Blocks nor EnderIO facades inherit the connected-textures capability of the original blocks. I would like to use some of the elaborate variants for adornment here and there but unfortunately, almost all of those more elaborate variants aren't covered by my texture pack. The results don't look very good.
 

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
It spins! An important stepping stone towards a functional fusion reactor is completed.

But first things first. I debugged the power input into the Auroral battery. The problem was that all inputs must be either active and outputting the same torque/current, or disconnected - and I had two inactive engines that were connected but unpowere because this was to be a test run.

After that, I went mining in the Nether again - because I had decided to make more batteries, albeit one class smaller than the big one. The reason? The auroral battery has only one output, and if I connect several subsystems to it the power level of the whole system changes if I add a new component. Resistors can help with that, but the calculations can get rather involved. Imagine, for instance, that I have 7 different outputs. Depending on how many subsystems are in the system, I might have to adjust my gearboxes every time I add something. A battery has the advantage of taking power in at whichever voltage and current comes its way and outputting it at a fixed voltage, and with a resistor I can adjust the current. That way I have control over both parameters. I suspect I don't really need to do this - I'm still new enough to ElectriCraft that I mix up the relation between voltage/current and torque/speed in my head - but this also has the advantage that I have backup power in case of having to break something temporarily, and a convenient switch location as well. Thus, the batteries are expensive, but they fill several important roles. For now I made one Graphene Battery for the Solenoid power, and one for the preheaters' heat rays. Both subsystems need 8 MW power and the batteries output 16. I can tell them to output less with resistors.

Then I started working on the solenoid. I've described crafting the components in an earlier post, this is about building it. I'll post several pictures to illustrate how this is built. It is actually not hard at all, if several conditions are fulfilled: you know how many of each element you need and have built them, you have a vague recollection of Reika's tutorial video, and you're using the blueprint highlighter. I didn't need to consult Reika's video again while I was building it. For the solenoid, you need: 1 Solenoid Magnet, 17 Solenoid Hub, 40 Hysteresis Rod, 20 Central Permanent Magnet, 28 Auxiliary Magnet, 40 Ferromagnetic Base and 56 Magnetic Linkage. You start with placing the Solenoid Magnet block right on top of the blueprint highlighter. The 17 blocks of the solenoid hub are placed around it (8) and above that platform (9). The hysteresis rods are placed - intuitively - on the same height as the solenoid magnet block and follow this pattern:



You have 40 hysteresis rod blocks, and there isn't any pattern that suggests itself except the one shown. Then you have 20 Central Permanent Magnet blocks, which intuitively go 5 each at the end of each hysteresis rod pointing in a cardinal direction. You may recall the structure is 3 high, and from there you can conclude that the 40 Ferromagnetic base blocks go above and below the central magnet blocks. Considering the blueprint layout, yet again a pattern suggests itself for the 28 auxiliary magnets:



And you have 56 magnetic Linkage blocks, so again there isn't any pattern that suggests itself rather than above and below the auxiliary magnet blocks. Taken that way, building the solenoid isn't difficult at all. Only crafting the magnets has taken a few hours of Compactor time. After that, I also finished the remaining three plasma injectors, so that the central components of the fusion reactor are now built and ready to be powered and connected. I just hope I didn't make a mistake when having to replace and reorient one toroid magnet I accidentally broke. I can only say be careful where you point your pickaxe. The defining block space of a toroid magnet is transparent, and if you break one neighbouring a plasma injector you have to tear down one side of the injector to place it again. Anyway, here's the finished main structure.



The next step was to power the main systems. I was going to build the power systems for the preheater and the solenoid before getting to the many Van De Graafs - which I'll be doing last. Less probability of getting shocked that way. I also built the start of my control chamber and made the quantum link to my base as well as another wireless access point, so that I can now access my ME network from everywhere on the fusion reactor construction site. There isn't much else in that room yet and I'll show it later.

Now for an inconvenient and unexpectedly frustrating delay. I was testing the Electricraft resistors when I noticed that I didn't have any brown dye (or cocoa beans). Brown codes for a rather important number: the 1. I hadn't been out exploring for a while so I didn't mind, and anyway exploring the overworld - as opposed to the Nether and the underground - is fun, so I went out to find myself some dye trees (there is no jungle biome in the area I have explored so no cocoa beans). As the RNG would have it, it didn't take long to find 14 of the 16 colors. Ironically given that the world is covered in forest and grass, the missing colors were - you knew this was coming, right? - green and.....brown. An hour and a used-up ruby sickle later I was about to give up and cheat when I found a single brown dye tree, which dropped one cocoa bean and several tree dyes and saplings. Good. That was the inconvenient part. The frustrating part - that was when I discovered that coloring the resistor was only possible with vanilla dyes. In this case, you need cocoa beans but tree dyes or other oredicted dyes won't work. I suspect that this is a bug.

With that out of the way at last, I could start building the power supply system for the solenoid. I'd need a resistor, an induction motor, an 1:8 bedrock gearbox and some bevel gears and bedrock shafts. No problem, I already have all that, but several of the components can't be placed vertically, so how will I make this symmetric? What I came up with is this:



Except for two of the cable blocks, all components are part of a symmetric 3x3 structure centered on the solenoid magnet above and the auroral battery below. The unfinished corridors are to be maintenance corridors which will contain power lines and fuel and plasma pipes. I opened up the ceiling for a close-up of the power supply system:



Note the color coding on the resistor "520 amps". 520 amps go into the system are converted into 4160 Nm torque. The gearbox takes that up and converts into slightly more than the necessary 32678 torque, thereby wasting small amount of power because I can't code the resistor for 512. Meanwhile, the 16384 V output by the graphene battery get converted to 2048 rad/s speed by the induction motor and down to 256 rad/s by the gearbox. Exactly what I need.

I should say that it is satisfying to see the solenoid spinning after that much work. That finishes today's work. The next update will probably have to wait for a few days since I do have a life outside of Minecraft....
 
Last edited:

Pyure

Not Totally Useless
Aug 14, 2013
8,349
7,216
383
Waterloo, Ontario
@Ieldra, this is fascinating stuff. I see you ran into the same gripes I have with electricraft: precision-amps, and cocoa beans. Dear god, cocoa beans are annoying to manage for that mod.
 

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
@Ieldra, this is fascinating stuff. I see you ran into the same gripes I have with electricraft: precision-amps, and cocoa beans. Dear god, cocoa beans are annoying to manage for that mod.
I consider that a bug. I can't imagine that Reika intended that the dyes of his own Dye Trees mod don't work in ElectriCraft. Also, Mimicry to the rescue: one mimichite crystal makes three cocoa beans from one, and you can snowball that effect. At least now I have something to do with my hundreds of mimichite ore blocks. Until the bug is fixed, that is. Who would've thought I'd ever use Mimicry for something as mundane as cocoa beans (Though the stuff is worth it in RL ;)).

Also, about precision amps: I could've gotten the precise amount of power I needed by combining the output of two smaller batteries, but I didn't consider it worth the expense.
 

Pyure

Not Totally Useless
Aug 14, 2013
8,349
7,216
383
Waterloo, Ontario
Also, about precision amps: I could've gotten the precise amount of power I needed by combining the output of two smaller batteries, but I didn't consider it worth the expense.
I went to the trouble once, and decided it took up too much space and too much of my patience. With fission reactors running, the loss is minor.
 

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
Ignition....ACHIEVED!





....immediately followed by this:



...and this:



I'll tell the full story later, I just couldn't wait to post these shots. No idea what went wrong. Maybe 16KW isn't enough for the toroid Van De Graaff's, as opposed to what the recently published build claimed. Also, I found the Van De Graaff's don't work in the rain. I just hope covering up those parts of the toroid will be enough. Also, you see I don't have the power generation up. This was to be a test run for the preheaters and the fuel supply. That part worked. Also, you may not see it in the screenshots, but the toroid was powered.
 

Kirameki

New Member
Jul 29, 2019
310
0
0
Yeah I'm suspicious of the 16KW claim, myself. I have Steam engines powering mine but the colors are so faint it makes me nervous what'll happen when I finally start it up.
>Dat spiral pattern
Looks pretty cool, except for the whole "this isn't supposed to happen" part.:D
 
  • Like
Reactions: Padfoote

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
Yeah I'm suspicious of the 16KW claim, myself. I have Steam engines powering mine but the colors are so faint it makes me nervous what'll happen when I finally start it up.
>Dat spiral pattern
Looks pretty cool, except for the whole "this isn't supposed to happen" part.:D
My batteries power them, and there are reserves. That part should be no problem. I also have a suspicion that power is not the problem. I'll test that tomorrow. As far as I can tell, as long as the color isn't black there should be no problem.
 

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
Today I built a lot and prepared for another ignition test, which will have to wait for another day. Regarding the containment breach of yesterday, I changed the direction of the plasma injection, because the solenoid drew the plasma in a different direction than it was injected. In his video, Reika said the direction doesn't matter as long as it's the same in all injectors but I'm not quite ready to believe that. Now everything travels counterclockwise from the start - the toroid charge, the solenoid and the plasma. Anyway should that fail, I have big power reserves. Which brings me to the power systems I built: I already had two Graphene Batteries to power the preheaters and the solenoid, now I added four Procyon batteries, one for the two subsystems in each quarter of the reactor: the pipe containment and the toroid containment Van De Graaff's. I also added a rudimentary control system using wireless redstone and some redstone logic.

This is one of the four subsystems powering a pipe containment and a toroid Van De Graaf generator each. The switch for the battery is below it. You can see a wireless redstone receiver above the relay, which is of course to switch it.


The subsystem over the preheater room looks a little different, but functionally it's exactly the same as the others. You can see I have added ways to access everything. Corridors mostly because of the way I build, but also walkways where necessary. These corridors are, btw, the "hot" area. You don't want to be here while the toroid Van De Graaff's are working because you'll get shocked. I wonder if there is a way to prevent that. Anyway, you're safe in the corridors below.


Here you see the power system for a preheater. Power comes from a battery in the roof of the corridor to the main battery.


Next, I've started on a rudimentary control system. The idea is to have a relatively failsafe power-up sequence. You see all but one subsystem start disabled. That's because the systems on the right are dependent on the systems on their left running, and you won't be able to start the injection before all other subsystems are running. The most critical part is the pipe containment power, that's why it's the leftmost component. If any of the other systems fail, that's inconvenient but not necessarily (!) immediately dangerous. If the pipe containment fails, that can turn your reactor into a lava lake. Later, I'll try to find a way to make this temperature-dependent, so that you won't be able to power down the systems unless the reactor components have cooled down. At the moment, you can only power-up the reactor in sequence, but you can switch the leftmost lever while it's running, resulting in disaster.



Here's how things look behind the control display. I've used colored redstone channels in EnderIO insulated redstone conduits, and Engineer's Toolbox modular sockets for the redstone logic. Again I would like to draw attention to this completely awesome mod: Note how I can have an AND gate in four directions, which lets me control the lamp, the wireless transmitter (above, not quite visible) and two pipe signals in a rather compact way.


Also, I built more of the aboveground components, the start of the actual power generation systems: neutron absorbers, steam boilers, water supply and steam lines. Later I'll replace some of the neutron absorbers with irradiation chambers - I want to have a reactor that makes its own fuel - but I'm not sure of the final layout yet. Here's how things look at the moment. Water supply is by highly upgraded XU transfer nodes, one in each quarter. Who would've thought I would ever teach my ME network how to make iron pickaxes? The water is meant to stay as decoration. Once the reactor is functional, I'll add some more stuff to make the site look better.


That was it for today. Tomorrow I'll run another ignition test, and if that works, I'll start on the turbines. Also, I should mention that I've let my Big Reactors run all the time while I was building for a few days now, and they have managed, with 40 KRF/t output, the heroic task of filling.....3% of the Auroral Battery. The fusion reactor, once it's running, will fill that battery in 2-3 hours.

BTW, I do know that those arcane lamps are a bit jarring from a stylistic viewpoint, but I know of no other light sources that light things up with equal power, and I hate having to place light sources every 7 blocks or so.
 
Last edited:

Padfoote

Brick Thrower
Team Member
Forum Moderator
Dec 11, 2013
5,155
5,914
563
If you need to access the areas where the VDGs are running, chain mail allows you to walk by without taking damage. I'm not sure if it causes them to ignore you, or if you simply take no damage.

Overall I like the way you're doing this. That control panel is quite cool, and a very good idea for a reactor. I will probably be stealing that idea when I start messing with reactors.
 
  • Like
Reactions: pizzawolf14

Ieldra

Popular Member
Apr 25, 2014
1,811
734
129
Chainmail, eh? Not my style, but I'll keep it in mind. Hehe....the anachronism of having an armor stand with chainmail in a reactor control room. That's worse than the arcane lamp. And thanks, btw.
 
  • Like
Reactions: Padfoote