ReactorCraft Fission Reactor Designs

[Demosthenex]

I just confirmed in creative that I can pull waste from a chest and distribute it using item ducts and it places one item per slot in the short term waste containers. I don't see it stacking.

That said, I'm not sure why my survival waste container blew up. The arctic ambient is -20C, and they are submerged near bedrock. I may pad my containers with cobble so they can't be more than half full.

Much to my surprise and dismay, my little 6 core reactor blew up in survival when I logged into the server.

I must say everything that works great in single player has consistently bombed (pun intended) serverside. I think it happens on chunk load or unload during initialization because I was standing in that room when I logged out. This was the first time I tried to let it run across sessions. My PBR has never had a problem, but I never run it except in supervised short bursts, never always on.

Yes, I know I run MCPC so I can't really ask for support but i"m climbing the walls because now not only is waste overreacting but a tiny 300MW 6 core cool reactor doesn't work either. Pardon my rant.

 

[Reika]

Much to my surprise and dismay, my little 6 core reactor blew up in survival when I logged into the server.

I must say everything that works great in single player has consistently bombed (pun intended) serverside. I think it happens on chunk load or unload during initialization because I was standing in that room when I logged out. This was the first time I tried to let it run across sessions. My PBR has never had a problem, but I never run it except in supervised short bursts, never always on.

Yes, I know I run MCPC so I can't really ask for support but i"m climbing the walls because now not only is waste overreacting but a tiny 300MW 6 core cool reactor doesn't work either. Pardon my rant.

Are the chunks getting unloaded? That is a guaranteed cause of meltdown.

 

[MajPayne21]

Much to my surprise and dismay, my little 6 core reactor blew up in survival when I logged into the server.

I must say everything that works great in single player has consistently bombed (pun intended) serverside. I think it happens on chunk load or unload during initialization because I was standing in that room when I logged out. This was the first time I tried to let it run across sessions. My PBR has never had a problem, but I never run it except in supervised short bursts, never always on.

Yes, I know I run MCPC so I can't really ask for support but i"m climbing the walls because now not only is waste overreacting but a tiny 300MW 6 core cool reactor doesn't work either. Pardon my rant.

As did my design that copied your 6-core. I hadn't even had it running for 24 hours. I, too, am using MCPC, because it's pretty much required to keep our server in good shape. I think my meltdown occurred on server restart. I found my 3x3 water pools empty, so I think on server restart, some lag or block update issues might make the 3x3 pool of water restore itself too slowly, running dry. Then, the predictable boiler explosion happens. Are 4x4 water pools more reliable at preventing this?

I also thought that in single player, your 6 core design was completely safe even without water, so I'm surprised that an interruption of the water supply caused a major meltdown (I'm also surprised that my single fuel cycle caused a meltdown).

Oh well, guess I'm moving all my stuff now.

I wouldn't worry about PBRs; I have run them rock-solid continuously, and as far as I'm aware, they only melt into lava. I don't think they produce meltdowns or radiation damage events.

@Reika, is v24 safe for HP turbines?[DOUBLEPOST=1405872721][/DOUBLEPOST]

Are the chunks getting unloaded? That is a guaranteed cause of meltdown.

My reactor was 100% inside the chunkloaded area. The only time the chunks would have been unloaded in my case was on server restart.

 

[Demosthenex]

Are the chunks getting unloaded? That is a guaranteed cause of meltdown.

Well crap. Our server ChickenChunk chunk loaders are allowed to run for 4 hours after a player logs out. You think that the chunk unloading would cause it?

I built the reactor entirely within one chunk including redundant water. The steam lines should be able to cache output until the turbine chunk loads.

Can you explain why unloading causes instant melt?

If that's the cause, I can try using an admin anchor. If it works, I'll use an alternate chunk loader that runs nonstop instead of obeying the 4 hour limit.

 

[Demosthenex]

As did my design that copied your 6-core. I hadn't even had it running for 24 hours. I, too, am using MCPC, because it's pretty much required to keep our server in good shape. I think my meltdown occurred on server restart. I found my 3x3 water pools empty, so I think on server restart, some lag or block update issues might make the 3x3 pool of water restore itself too slowly, running dry. Then, the predictable boiler explosion happens. Are 4x4 water pools more reliable at preventing this?

I also thought that in single player, your 6 core design was completely safe even without water, so I'm surprised that an interruption of the water supply caused a major meltdown (I'm also surprised that my single fuel cycle caused a meltdown).

Oh well, guess I'm moving all my stuff now.

I wouldn't worry about PBRs; I have run them rock-solid continuously, and as far as I'm aware, they only melt into lava. I don't think they produce meltdowns or radiation damage events.

@Reika, is v24 safe for HP turbines?[DOUBLEPOST=1405872721][/DOUBLEPOST]

My reactor was 100% inside the chunkloaded area. The only time the chunks would have been unloaded in my case was on server restart.

We also have a daily restart... so that's another concern regarding unloading chunks.

Regarding moving, I'd suggest creative admin mode to clear the area and use Opis to delete the radiation entities. When something has clearly been demonstrated in single player and well documented and it's a server integration issue, that's fair.

What are you using for a water supply? The RoC pumps eat water source blocks, so a 5x5 pool is most reliable per pump. I'm using 4 liquid transfer nodes on a 5x5 pool and they don't remove source blocks.

As to the single player tests, remember the cool vs hot reactor test is done without piping and all parts top/bottom exposed to air. When you put in a functional reactor with steam pipes and water inputs, you have insulated all blocks from air. That means they must have active cooling. The reason I test air exposure is to demonstrate that the heat gradient across the parts is spread out. If it only has air cooling, I have to diffuse the heat over a larger surface area. When active water cooling is applied, it means that I expect less sensitivity to water throughput than a reactor with high heat hot spots. It will likely be somewhere less efficient, but I assert it should be safer and believe most of the heat is still captured despite more blocks. It does not mean it could function without water.

It would be interesting to see if I can create a heat pipe design where if water fails heat would be drawn outside to boilers exposed to air... Like a passive safeguard built on heat conductivity.

 

[MajPayne21]

We also have a daily restart... so that's another concern regarding unloading chunks.

Regarding moving, I'd suggest creative admin mode to clear the area and use Opis to delete the radiation entities. When something has clearly been demonstrated in single player and well documented and it's a server integration issue, that's fair.

What are you using for a water supply? The RoC pumps eat water source blocks, so a 5x5 pool is most reliable per pump. I'm using 4 liquid transfer nodes on a 5x5 pool and they don't remove source blocks.

As to the single player tests, remember the cool vs hot reactor test is done without piping and all parts top/bottom exposed to air. When you put in a functional reactor with steam pipes and water inputs, you have insulated all blocks from air. That means they must have active cooling. The reason I test air exposure is to demonstrate that the heat gradient across the parts is spread out. If it only has air cooling, I have to diffuse the heat over a larger surface area. When active water cooling is applied, it means that I expect less sensitivity to water throughput than a reactor with high heat hot spots. It will likely be somewhere less efficient, but I assert it should be safer and believe most of the heat is still captured despite more blocks. It does not mean it could function without water.

It would be interesting to see if I can create a heat pipe design where if water fails heat would be drawn outside to boilers exposed to air... Like a passive safeguard built on heat conductivity.

Thanks for the tip. Killing the radiation entities worked just fine. I think the effort I would have had to go to in order to move my entire base might have caused me to stop playing until the next world reset. I'll try the transfer nodes. I'm pretty sure running out of water is what caused this. I think I'll attempt to modify your design to include control rods and a central CPU. Hopefully they would be able to react in time and SCRAM if the fuel core temperature skyrockets.

 

[Demosthenex]

Thanks for the tip. Killing the radiation entities worked just fine. I think the effort I would have had to go to in order to move my entire base might have caused me to stop playing until the next world reset. I'll try the transfer nodes. I'm pretty sure running out of water is what caused this. I think I'll attempt to modify your design to include control rods and a central CPU. Hopefully they would be able to react in time and SCRAM if the fuel core temperature skyrockets.

You absolutely must have a robust water supply. I'd be interested to see how you update that tiny reactor to use control rods.

OTOH, if server restarts kill things it won't help.[DOUBLEPOST=1405877282][/DOUBLEPOST]

Thanks for the tip. Killing the radiation entities worked just fine. I think the effort I would have had to go to in order to move my entire base might have caused me to stop playing until the next world reset. I'll try the transfer nodes. I'm pretty sure running out of water is what caused this. I think I'll attempt to modify your design to include control rods and a central CPU. Hopefully they would be able to react in time and SCRAM if the fuel core temperature skyrockets.

I build mine in an arctic base far from my main... I figured if I legitimately borked the area, I'd abandon it.

 

[Reika]

Well crap. Our server ChickenChunk chunk loaders are allowed to run for 4 hours after a player logs out. You think that the chunk unloading would cause it?

I built the reactor entirely within one chunk including redundant water. The steam lines should be able to cache output until the turbine chunk loads.

Can you explain why unloading causes instant melt?

If that's the cause, I can try using an admin anchor. If it works, I'll use an alternate chunk loader that runs nonstop instead of obeying the 4 hour limit.

Neutrons pile up because entities do not simulate in unloaded chunks.

 

[Demosthenex]

Neutrons pile up because entities do not simulate in unloaded chunks.

Ok. So do you think it's the 4 hour chunk loader or the server restart?

I can use a nonstop chunk loader, but I can't change the restart.

 

[MajPayne21]

You absolutely must have a robust water supply. I'd be interested to see how you update that tiny reactor to use control rods.

OTOH, if server restarts kill things it won't help.[DOUBLEPOST=1405877282][/DOUBLEPOST]

I build mine in an arctic base far from my main... I figured if I legitimately borked the area, I'd abandon it.

Well, I assign you partial blame for this () since you SAID it was safe! I've put everything back together and deleted the radiation entities.

I'm going to give this a try: (R = Control Rod, C = Core, B = Boiler, N = Neutron Reflector. Please ignore my poor choice of color palate; I'm colorblind and working quickly)

It fits in a 9x9 area and has rods intercepting the core interactions. The cores are still exposed to neutron reflectors on the outsides, so it's not 100% safe, but it may prevent a runaway by severely retarding the reaction rate. Obviously, some of the neutron reflectors aren't doing anything here, but I figured I'd just border the whole reactor with them to keep it simple. In essence, this is just a scaled-up version of your 6-core that capitalizes on the fact that distance between cores doesn't affect neutron interaction. I'm just hoping that basically doubling the number of boilers doesn't significantly gimp efficiency by causing a lot of the boilers to be under 100C too often. I may replace some non-critical boilers with steel bocks or something. I'll test all this.

This results in adding a LOT more boilers. Based on your testing, do you think I'd be better off using steel blocks in place of boilers that are not directly touching the cores? Can the central control go 1 block above the control rods, or will I need to replace one of the center boilers with a central control?

Are you running v23C from Monster 1.1.2? I can't get the HP turbine to work without crashing me. I may try 5 or 6 stages instead of the full 7 and see if that helps. I don't think I can get my fellow players to upgrade all Reika's mods without some significant handholding.

 

[Reika]

Ok. So do you think it's the 4 hour chunk loader or the server restart?

I can use a nonstop chunk loader, but I can't change the restart.

...?

 

[MajPayne21]

...?

He's asking if the chunk loader shutting off (and thus unloading the chunks without shutting off the server) is the proximate cause of the meltdown, or if, during server restart, the order in which items load might cause the meltdown.

If the reactor is fully within one chunk and surrounded by neutron reflectors, no stray neutrons should escape to collect on the chunk boundaries, correct? Thus, loading/unloading the chunks shouldn't be an issue unless the water supply is somehow affected. I'm inclined to believe the server restart is what caused my issue.

 

[Demosthenex]

Well, I assign you partial blame for this () since you SAID it was safe! I've put everything back together and deleted the radiation entities.

*sob* It's all my fault!

I'm going to give this a try: (R = Control Rod, C = Core, B = Boiler, N = Neutron Reflector. Please ignore my poor choice of color palate; I'm colorblind and working quickly)

It fits in a 9x9 area and has rods intercepting the core interactions. The cores are still exposed to neutron reflectors on the outsides, so it's not 100% safe, but it may prevent a runaway by severely retarding the reaction rate. Obviously, some of the neutron reflectors aren't doing anything here, but I figured I'd just border the whole reactor with them to keep it simple. In essence, this is just a scaled-up version of your 6-core that capitalizes on the fact that distance between cores doesn't affect neutron interaction. I'm just hoping that basically doubling the number of boilers doesn't significantly gimp efficiency by causing a lot of the boilers to be under 100C too often. I may replace some non-critical boilers with steel bocks or something. I'll test all this.

Add more two rods between the set of cores in the middle. The rods should limit as many core to core interactions as possible. The reflectors are only 25% reflective.

This results in adding a LOT more boilers. Based on your testing, do you think I'd be better off using steel blocks in place of boilers that are not directly touching the cores? Can the central control go 1 block above the control rods, or will I need to replace one of the center boilers with a central control?

Steel blocks halt 90% of neutrons. Just use boilers. The temperature exchange isn't lossy, just based on the relative temperatures of the two blocks. I think the central control needs to be touching a reactor block, but not in the reactor itself. Let me know if you can place it outside the reflector, that's be great.

Are you running v23C from Monster 1.1.2? I can't get the HP turbine to work without crashing me. I may try 5 or 6 stages instead of the full 7 and see if that helps. I don't think I can get my fellow players to upgrade all Reika's mods without some significant handholding.

I'm on the same version. I had some issues where my HP turbine wouldn't assemble at all without setting the server sync rate to 1, but I'm not confident that's really a fix. I never crashed while assembling one though, it just refused to connect together.

 

[Demosthenex]

...?

Exactly what MajPayne21 said.

Do you think the meltdown was caused by the chunk loader unloading those chunks while the server was still running?

Or do you think it is a server shutdown and startup that causes the melt?

I can try another kind of chunk loader, maybe the MFR one powered off the reactor. I can't change that there is a daily shutdown of the server.

 

[MajPayne21]

*sob* It's all my fault!



Add more two rods between the set of cores in the middle. The rods should limit as many core to core interactions as possible. The reflectors are only 25% reflective.



Steel blocks halt 90% of neutrons. Just use boilers. The temperature exchange isn't lossy, just based on the relative temperatures of the two blocks. I think the central control needs to be touching a reactor block, but not in the reactor itself. Let me know if you can place it outside the reflector, that's be great.



I'm on the same version. I had some issues where my HP turbine wouldn't assemble at all without setting the server sync rate to 1, but I'm not confident that's really a fix. I never crashed while assembling one though, it just refused to connect together.

I'm guessing you meant 'more THAN two rods', but I don't understand why. Shouldn't one rod placed in the neutron path between two cores stop all neutron interaction?

I can get the HP turbine multiblock to form no problem, but when it runs, it will occasionally rapidly 'flip' its orientation and render and then reassemble itself. During this time, my client crashes but the server is not affected.

You can observe the same phonomenon with the small turbines if the steam is routed poorly, but since the HP turbine doesn't deal with free steam, it mystifies me.

I suspect this behavior is fixed in 23d or 24, because the changelogs say 'fixed HP multiblock' and 'made HP turbine more tolerant of nearby blocks.

Sent from my SCH-I535 using Tapatalk

 

[Demosthenex]

I'm guessing you meant 'more THAN two rods', but I don't understand why. Shouldn't one rod placed in the neutron path between two cores stop all neutron interaction?

Nevermind, my derp. I transposed your cores and rods mentally. It's fine.

As to flipping, i see much less of it with sync = 20 and careful destruction of all steam blocks.

 

[MajPayne21]

Nevermind, my derp. I transposed your cores and rods mentally. It's fine.

As to flipping, i see much less of it with sync = 20 and careful destruction of all steam blocks.

Where (and how) do you extract Low-p water from the HP turbine?

 

[1SDAN]

So not to intrude but I think the reactor partially unloading thing seems a bit annoying. Couldn't the Reactor act as a partial chunkloader so that it loads all of the reactor chunks if one is loaded? Or is that a design choice to force the use of Chunkloaders?

 

[Demosthenex]

Exactly what MajPayne21 said.

Do you think the meltdown was caused by the chunk loader unloading those chunks while the server was still running?

Or do you think it is a server shutdown and startup that causes the melt?

I can try another kind of chunk loader, maybe the MFR one powered off the reactor. I can't change that there is a daily shutdown of the server.

So I ran a (potentially destructive) test... I replaced the 4 hour ChickenChunks loader with a Railcraft Admin Loader and then performed a server restart. Nothing went wrong.

That leads me to believe it was the chunk loader time limit forcing an unload. I'll leave the reactor running for a few days on the server to test that.

I also believe this is the problem with my waste storage, that unloaded chunks caused neutron buildup like Reika mentioned. If I have no reactor problems, I will reattach my reactor to the waste pond storage instead of the trash can and test that.

That also means as a compromise, if I have other players on my server which are using these kinds of reactor I will offer them an admin anchor exclusively for the reactor room, but that can't void the waste in return.[DOUBLEPOST=1405888376][/DOUBLEPOST]

So not to intrude but I think the reactor partially unloading thing seems a bit annoying. Couldn't the Reactor act as a partial chunkloader so that it loads all of the reactor chunks if one is loaded? Or is that a design choice to force the use of Chunkloaders?

So I don't think any of Reika's machines include a loader. Even the boring machine requires a separate chunk loader to keep it running, unlike a quarry.

I purposefully designed my reactor to fit in one chunk to keep crossing the chunk from causing an issue. I think this was caused by the chunk loader timeout our server enforces to keep the load down.

 

[MajPayne21]

So not to intrude but I think the reactor partially unloading thing seems a bit annoying. Couldn't the Reactor act as a partial chunkloader so that it loads all of the reactor chunks if one is loaded? Or is that a design choice to force the use of Chunkloaders?

Unfortunately, it's just the nature of minecraft to have to be careful with chunkloading. Even in the days of Railcraft boilers being the best available power source, you had to ensure your water and fuel supply were chunkloaded if the boiler was chunkloaded, lest you starve it and cool it down or cause an explosion.

If you're to the point where you're building a ReC reactor, chunkloaders should be effectively zero-cost for you, so just use F9 to make sure your reactor is centered in a single chunk and place a chunkloader in the same chunk (I set mine to radius 3 for a 25-chunk square), and you should be good to go.

Demosthenex - I still don't know where to extract low pressure water from the HP turbine at the end of the cycle