Friday Facts #167 - Reactors Operational

[Proxy]

one Thing i just Notcied...
the Reactors won't need Cooling as they are Shown now.
as they basically seem to use Nuclear Fuel, to make heat.
it is a RThG (Radioisotope Thermal Generator).
No Mechancial parts, no Coolant.
and it will make Heat for a long time (Depening on the Fuel).
a RTG (Radioisotope Thermoelectric generator) would use the Heat Difference bweteen the Fuel and Outside to make Energy, basically a battery that holds VERY VERY Long.
(i know it is WIP, but that is what they are now xD)

because a Real Reactor would take 2 Water In/Out.
1 Water Cycle to cool the Rods (Fuel). and 1 to make Water into Steam using the Produces Heat, which the nagain Spins a Turbine/Steam Engine to make Energy.
or just Breeder Reactors with are used to make Plutonium for Weapons.
or Thorium, for Eco Friendly. Weaponless and Long Living Reactors without Melt-down Danger.

Final note:
Reactors don't Explode. all they could do at most is Melt-down and all the Rods would fuse to a Radioactive ball os Heat that Sinks down into the ground, Making the Surroundings Inhabitable for quite some time.

[MeduSalem]

one Thing i just Notcied...
the Reactors won't need Cooling as they are Shown now.
as they basically seem to use Nuclear Fuel, to make heat.
[...]

They do need cooling.

As I understand the Boiler-Heat-Exchangers basically use the heat that comes from the reactor to vaporize water into steam, thus probably "reducing" accumulated heat of the Reactor in turn.

So I bet if you don't place enough Boiler-Heat-Exchangers around the reactors and provide enough water to be vaporized then the reactors probably face heat build-up and once they reach critical heat levels a melt down follows. If they explode though is another question.

I'd say if anything blows up at all then it should be the Boiler-Heat-Exchangers if the Steam builds up in the pipes too much for some reason. So you'll have a nice little Hydrogen explosion or something. That in turn might damage the reactors nearby, and a melt down is probably also imminent due to how without an intact Heat-Exchanger you'll face critical heat levels in the reactor.

So basically there you have it... Chernobyl in Factorio.

[AcolyteOfRocket]

If we must have nuclear reactors I would like them to be able to be set up safely, if I pay attention to their needs.

So if they need a secure water supply then I want to be able to use water stored in a tank as a backup for example. Having hazards with no way to mitigate them would be just trolling imho.

If they do blow up then the result should be lots of pollution - similar to a forest fire - man the turrets, here they come !

[DonRobo]

Every time you post screenshots of new art it looks amazing. Just one thing has been sticking out to me for a while now. The stones under the train track just look weird to me somehow. It kind of sticks out next to all the other really amazing art in my opinion.

I'm wondering if I'm the only one who doesn't like them.

[psychomuffin]

one Thing i just Notcied...
the Reactors won't need Cooling as they are Shown now.
as they basically seem to use Nuclear Fuel, to make heat.
[...]

then the reactors probably face heat build-up and once they reach critical heat levels a melt down follows.

I thought the idea with nuclear reactors is if they get too hot, the rods retract, thus lowering the atomic reaction, and heating less water. So rods are positioned based on how much water is needed. Furthermore, I thought the intended design is they use power to stay held up, so in case of power outage, they automatically retract, instead of staying in a position that could result in meltdown.

...Or am I totally wrong?

[Sigma1]

one Thing i just Notcied...
the Reactors won't need Cooling as they are Shown now.
as they basically seem to use Nuclear Fuel, to make heat.
[...]

then the reactors probably face heat build-up and once they reach critical heat levels a melt down follows.

I thought the idea with nuclear reactors is if they get too hot, the rods retract, thus lowering the atomic reaction, and heating less water. So rods are positioned based on how much water is needed. Furthermore, I thought the intended design is they use power to stay held up, so in case of power outage, they automatically retract, instead of staying in a position that could result in meltdown.

...Or am I totally wrong?

That is correct as far as I know. Maybe in the game the rods could be circuit network controlled to add a bit more challenge.

[MeduSalem]

I thought the idea with nuclear reactors is if they get too hot, the rods retract, thus lowering the atomic reaction, and heating less water. So rods are positioned based on how much water is needed. Furthermore, I thought the intended design is they use power to stay held up, so in case of power outage, they automatically retract, instead of staying in a position that could result in meltdown.

...Or am I totally wrong?

That is correct as far as I know. Maybe in the game the rods could be circuit network controlled to add a bit more challenge.

I'd say yes... they probably have control rods... and I'd also expect them to be controlled by circuit network.

But I also expect the reactor to work even without controlling it by circuit network, just that you'll probably have to provide enough coolant in the heat-exchangers since the reactor is basically running at full power 24/7 in that case. That would be the "non-smart" solution for the people who are not that much into the circuit network. It would be inefficient of course and probably waste a lot of fuel so the additional challenge of the circuit network will be worth it in the end.

Let's wait and see how it really turns out in the end.

[huhn]

The most important thing in this preview, is that after several experiments, we decided not to render anymore in normal-resolution from Blender. Instead we are going to render just in high-resolution and re-scale the output bitmaps to normal-resolution using a bilinear interpolation.

does bilinear reality results in a better picture quality than bicubic 150 or bicubic 50 in linear light when downscaling?

[rolfl]

The most important thing in this preview, is that after several experiments, we decided not to render anymore in normal-resolution from Blender. Instead we are going to render just in high-resolution and re-scale the output bitmaps to normal-resolution using a bilinear interpolation.

does bilinear reality results in a better picture quality than bicubic 150 or bicubic 50 in linear light when downscaling?

For what it's worth, when down-scaling, I have had much better success with Sinc-Lanczos when down-sizing highly-detailed images. The results are significantly sharper. (when u-scaling, it's not the best, for sure).

I do this often when creating much smaller images (thumbnails) through to 1080 versions of photos that have. Consider the two downscales:

Cubic:

fff-165-train-3000.cubic.800.jpg (262.28 KiB) Viewed 10219 times

Sinc Lanczos:

fff-165-train-3000.sinclanczos.800.jpg (269.31 KiB) Viewed 10219 times

Regardless, the differences are subtle, and subjective, but the right tools for upsizing and for downsizing are different, and have to be assessed properly. Just saying that for my past work that Sinc-Lanczos has worked out better for me.


Oh, the original image is the 3000pixel-resolution image from the fun facts: https://us2.factorio.com/assets/img/blo ... n-3000.jpg and the resizing/scaling was done with Gimp 2.0

[Proxy]

i didn't really mean Active Reactors need no Coolant.
but if you just had a Radioactive Material cased inside a Box, Connected via Thermal Conducting Wires to the Outside and Coated in something Heat Resistant...
you would have a RTG without the Electric and Generator part.
it would Provide a ~Constant amount of Heat for as Long as the Material in it hasen't Decayed enough.
only Downside. it Leaks Radiation to the Outside.

also, some Reactor Types really don't need an Extra Fluid as Coolant. like a LFTR (Liquid Fluoride Thorium Reactor). it is a Molten-Salt-Fueled Reactor where the Liquid Fuel is the Coolant itself.
they are by themselfs Pretty Safe too.

(warning, i might be saying something wrong, but i just really love this Idea of Nuclear Fission (or more) in Factorio)

[Sigma1]

IMO now that they are introducing some level of nuclear physics maybe they should add some fuel requirement to the portable fusion reactor or name it something else like alien reactor because IRL fusion reactors do need fuel.

[Gertibrumm]

I thought the idea with nuclear reactors is if they get too hot, the rods retract, thus lowering the atomic reaction, and heating less water. So rods are positioned based on how much water is needed. Furthermore, I thought the intended design is they use power to stay held up, so in case of power outage, they automatically retract, instead of staying in a position that could result in meltdown.

...Or am I totally wrong?

That is correct as far as I know. Maybe in the game the rods could be circuit network controlled to add a bit more challenge.

I'd say yes... they probably have control rods... and I'd also expect them to be controlled by circuit network.

But I also expect the reactor to work even without controlling it by circuit network, just that you'll probably have to provide enough coolant in the heat-exchangers since the reactor is basically running at full power 24/7 in that case. That would be the "non-smart" solution for the people who are not that much into the circuit network. It would be inefficient of course and probably waste a lot of fuel so the additional challenge of the circuit network will be worth it in the end.

Let's wait and see how it really turns out in the end.

As I understand it, what happend in Fukushima was a immediate reactor shutdown just before the quake and later the tsunami.
Reactor shutdown means:
- fuel rods retract (they still get awefully hot)
- diesel engines provide water cooling to all fuel rods
- no more power to the powerlines due to turbine shutdown (not in all cases)
Later the tsunami killes off the diesel engines or at least some parts of the cooling system

In factorio the tsunami would be replaced by a rolling wave of biters.
Killing off your backup colling systems, your power lines, and active cooling together with pumps and turbine.
And than the biters start chewing on your precious reactor core. Thats where your creativity is asked!

As far as circuit challenge goes, it should only be used by players who want to double their efficiency/power-output at the cost of higher risks when designing the plant. Reactor cores should not fully stabilize without external help from cooling systems.

[GuiltyBystander]

I know you don't know what ratios you'll end up using eventually, but please make them better than the current ones. I know most people think 1:14:10 is the best, but I find that 1:13:10 is better on many counts. Please pick numbers so people won't get stuck with wrong/bad ideas about how math works.

[huhn]

The most important thing in this preview, is that after several experiments, we decided not to render anymore in normal-resolution from Blender. Instead we are going to render just in high-resolution and re-scale the output bitmaps to normal-resolution using a bilinear interpolation.

does bilinear reality results in a better picture quality than bicubic 150 or bicubic 50 in linear light when downscaling?

For what it's worth, when down-scaling, I have had much better success with Sinc-Lanczos when down-sizing highly-detailed images. The results are significantly sharper. (when u-scaling, it's not the best, for sure).

I do this often when creating much smaller images (thumbnails) through to 1080 versions of photos that have. Consider the two downscales:

Cubic:

fff-165-train-3000.cubic.800.jpg

Sinc Lanczos:

fff-165-train-3000.sinclanczos.800.jpg

Regardless, the differences are subtle, and subjective, but the right tools for upsizing and for downsizing are different, and have to be assessed properly. Just saying that for my past work that Sinc-Lanczos has worked out better for me.


Oh, the original image is the 3000pixel-resolution image from the fun facts: https://us2.factorio.com/assets/img/blo ... n-3000.jpg and the resizing/scaling was done with Gimp 2.0

these are 1/4 resolution screens with different scalers:

cat rom mull LL/bicubic 50 LL: https://abload.de/img/bicubic50llzvuco.png
bicubic 150 (my clear winner for this image): https://abload.de/img/bicubic150e3ui3.png
bilinear (clear loser for this image): https://abload.de/img/bilinearv8umm.png
lanczos3: https://abload.de/img/lanczos36yur5.png

"tool" used madVR.

of cause for a proper test a game sprite is needed and the effect of the different scaler should be way bigger for that source.

and i would really like to scale the game with something different when it is zoomed out i'm really interested in the difference.

[rolfl]

......

these are 1/4 resolution screens with different scalers:

cat rom mull LL/bicubic 50 LL: https://abload.de/img/bicubic50llzvuco.png
bicubic 150 (my clear winner for this image): https://abload.de/img/bicubic150e3ui3.png
bilinear (clear loser for this image): https://abload.de/img/bilinearv8umm.png
lanczos3: https://abload.de/img/lanczos36yur5.png

"tool" used madVR.

of cause for a proper test a game sprite is needed and the effect of the different scaler should be way bigger for that source.

and i would really like to scale the game with something different when it is zoomed out i'm really interested in the difference.

I would agree with your ranking of those images. Of that set, the bicubic150 is best, and bilinear is worst. Heh. All told, the point is that choosing the right scaling algorithm makes a significant difference. I should probably do another round of evaluation for my use on my photos.....

[blackdove]

I thought the idea with nuclear reactors is if they get too hot, the rods retract, thus lowering the atomic reaction, and heating less water. So rods are positioned based on how much water is needed. Furthermore, I thought the intended design is they use power to stay held up, so in case of power outage, they automatically retract, instead of staying in a position that could result in meltdown.

...Or am I totally wrong?

That is correct as far as I know. Maybe in the game the rods could be circuit network controlled to add a bit more challenge.

I'd say yes... they probably have control rods... and I'd also expect them to be controlled by circuit network.

But I also expect the reactor to work even without controlling it by circuit network, just that you'll probably have to provide enough coolant in the heat-exchangers since the reactor is basically running at full power 24/7 in that case. That would be the "non-smart" solution for the people who are not that much into the circuit network. It would be inefficient of course and probably waste a lot of fuel so the additional challenge of the circuit network will be worth it in the end.

Let's wait and see how it really turns out in the end.

As I understand it, what happend in Fukushima was a immediate reactor shutdown just before the quake and later the tsunami.
Reactor shutdown means:
- fuel rods retract (they still get awefully hot)
- diesel engines provide water cooling to all fuel rods
- no more power to the powerlines due to turbine shutdown (not in all cases)
Later the tsunami killes off the diesel engines or at least some parts of the cooling system

In factorio the tsunami would be replaced by a rolling wave of biters.
Killing off your backup colling systems, your power lines, and active cooling together with pumps and turbine.
And than the biters start chewing on your precious reactor core. Thats where your creativity is asked!

As far as circuit challenge goes, it should only be used by players who want to double their efficiency/power-output at the cost of higher risks when designing the plant. Reactor cores should not fully stabilize without external help from cooling systems.

Pretty much the key thing to remember is that nuclear reactors produce large amounts of heat after the control rods have been inserted. The control rods stop the Uranium reaction but there is still a large amount of byproducts undergoing decay that the control rods can't stop. This is why fuel rods have to be stored in pools with active cooling for years after they're removed form the reactor since the ongoing decay of the byproducts produces enough heat to do things like burn the outside coating off and ignite the fuel causes rather severe problems. hence one of the things done at Fukushima was to get cooling to those pools as well as the reactor.

And while it's correct that nuclear plants can't go "nuclear" steam explosions, hydrogen explosion, fuel cladding fires, and other things can go wrong and explosions are a decent substitution for reality unless they plan on detailing more information in the game. Though if they wanted too they could also do something like some newer reactor designs that don't need active cooling and thus can't melt down but those mostly require much more effort when it comes to making the fuel than just make a fuel rod. Balance out the lack of danger with the need to spend more resources making the fuel for their safe reactor with less power vs a plant that can produce more power but can really screw things up if they break

[space-exe]

i didn't knew that

[loganb]

It'd be awesome if your model of nuclear reactors modeled the Iodine Pit: https://en.wikipedia.org/wiki/Iodine_pit

It would make controlling the reactor a challenge!

[matjojo]

but who's ranium are we mining?


sorry I just had to

[inteljoe]

What about running reactors like in the Minecraft Mod (http://wiki.industrial-craft.net/index. ... ar_Reactor)

More specifically where they talk about the "Mark Level" the reactor becomes. From what I remember, a "Mark I" is a reactor that produces energy and requires 0 input from the player making it stable as long as fuel exists. While a "Mark V" would require a significant input (which can be automated) from the player but also has the highest amount of energy output.

Overall the idea would be that a "Mark I" would require a very basic resource investment but overall would be easy to maintain/automate. While a "Mark V" would give the most amount of energy but require more a complex use of automation and more specialized resources.

I really like how the chambers are set up in the mod above. Where if you place one reactor chamber next to another it extends the size of the reactor. I'd imagine in Factorio some clever scripting could do the same? Basically it allows for one itemID for any size reactor you want depending on what your needs are.

Also, I know mods have larger capacitors available but I wonder if there are plans to implement "large" style capacitor's in the vanilla game.