Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
I find that hard to imagine given that the required logic is incredibly simple - two inserters connected by a wire:
https://steamcommunity.com/sharedfiles/filedetails/?id=2284910140
U238 filter inserter is set to read hand contents, U235 filter is set to activate when U238>0, and has stack size set to 1.
@op just put the two uraniums on two different lanes or two different belts and loop them around. The 100% fool proof way of doing it is having 1 belt each and only looping on the inside lane.
This is not my way of doing it but it is a very simple way to understand and it can't break. I could post a picture tomorrow.
https://steamcommunity.com/sharedfiles/filedetails/?id=2287221769
This one actually does work.
It is not susceptible to problems originating from blockage on the U235 output or on the U238 return cycle. It can cope with production modules just fine. And it uses the circuit network to manage inserter stack sizes to ensure an exact 40 U235 are fed back in, and not a single more, regardless of researched available max stack size. It also automatically bootstraps.
It almost tiles; there's an issue with one of the underground exits not aligning correctly.
(Still have to fix that.)
Unfortunately this does use quite a few combinators and wires. However; you won't have to touch any of it after plotting down the blueprints. There's no need to attempt to reset things, because it doesn't foul up to begin with. (Well sure; if you start messing with the chests manually, you could probably break it, but that goes without saying...)
https://steamcommunity.com/sharedfiles/filedetails/?id=2293094306
Probably you'll find it usefull :D
(Can be made without bots, belts only, if needed)
-It has no circuits and should be easy to understand how it works.
-1 splitters have filter set, one has only output priority and 1 has only input priority.
-Works with or without modules/beacons.
-You connect the two different uranium inputs (see the arrow of the loader) and it will start and keep going forever without further intervention.
-Once all of the centrifuges have started you can disconnect the u235 input if you want, but this is not necessary.
https://steamcommunity.com/sharedfiles/filedetails/?id=2294124555
Edit: noticed that the beacons are misaligned for maximum efficiency. Obviously beacons should always be offset 1 tile. This does not change how the system works though.
You've forgotten to list one very obvious downside:
It is over-eager and takes in the maximum (of iirc 120) U235 per centrifuge, rather than limit it to the minimally required 40 units. It means it will take a long, long time for the production cycle to fully stabilize and give decent output.
This is worsened by the fact that the U235 flow in your setup is arranged with a priority output splitter which forcibly reintroduces all produced U235 back into the centrifuges until all of them are fully saturated. In other words: this particular over-eager setup won't produce any U235 at all until it's fully saturated.
You can of course force it to only fill 40, but only by complicating things with circuits. The point here was to show a simple design without circuits that is easy to understand and can't fail.
Also the internal buffer is <3 nukes. If you want a quick first nuke for some unexplained reason you can manually take out the internal buffer and build your nuke with it.
I really don't see this buffer time as a problem. The only reason I use circuits in my own base is because I can, not because I need it. There are buffer times for many things in the game. For example when you start shipping things with trains do you send your first one early to get the first resources in quickly or do you wait for it to fill up? My guess is that in that case you don't care about the initial buffer time and let the first train fill up before sending it.
The difference is that while most buffers will fill relatively fast, U235 won't.
Each centrifuge requires 40 U235 to start up the Kovarex process and stocks up to 120 internally.
Each centrifuge only produces 1 U235 per run and the ratio for getting U235 from raw ore is incredibly low. If you stack 3 or so centrifuges, as per your design, you'd be waiting a long time for that buffer to fully fill and the system to spit back out the first U235 for use.
You only need 3 combinators, actually. That and a few chests. I've built a further refined version of my earlier Kovarex setup; the one that I linked to earlier in this thread, which still uses 5 or 6.
The trick behind it is very simple and mainly relies on the chests forming a discrete state machine. Combinators set up a circuit controlled stack inserter size, where that signal can double for the purpose of controlling the enabled state. The most complicated thing there is really the idea that you combine signals of U235 held in chests and inserter hands for a total count of 'what's left in the system.'
If you don't care about automatic bootstrapping, you even only need a single arithmetic combinator and the entire thing becomes resilient against power loss. (The bootstrapping needs a latch, so it can sadly fail on power loss.)
I guess I'll work on a new set of blueprints to publish later.
For reference a single red circuit train wagon takes 17000 copper plate, 8000 iron plate and 11400 plastic to fill up assuming you have fully moduled everything. For a single train wagon. That's what I would call a big buffer. 7 minutes of work for a centrifuge is no big deal in comparison.
2: Here's what I have in my actual base: https://i.imgur.com/HyYRRia.png
It uses 4 combinators which can be shared across all centrifuges. And it is only because I have productivity modules. If you don't have productivity modules you need 0 combinators (though still circuits). But there are several things that can go wrong with this design if you don't do things exactly as I intend it, and because of the combinators it's not easy to understand, so that's why I didn't post it.
Op said in his post that he tried several other blueprints and all of them failed. I also watched streamer imkibitz struggle with blueprints that failed because he didn't have the modules it expected, or he primed it wrong, or he thought it didn't work even though it did and broke it when "fixing it", or didn't have enough inserter capacity research, or the creator simply didn't test it enough. That's the problem, more complicated isn't always better. Saving a bit of buffer time once over the entire life span of the base may not be worth struggling for 2 hours just to get a design you don't understand up and running correctly.
3. I downloaded your blueprint from the other thread (it has 5 combinators, not 3 like you suggest here?) and it can fail if you get low power, not 0 power but low power. That's a general issue with combinators though and hard to safeguard against. I tested your design while getting low power at two different times and it failed in two different ways. You can most likely start it again, but imagine op trying to "fix" it not knowing how it works.
The new one isn't published yet.
It should be more resilient against outages as the core loop only uses a single combinator to compute max allowed stack size. If that returns 0 due to low power, the inserters will halt as > 0 stack size is also the enabled signal.
The actual state to ensure only 40 units of U235 re-enter the centrifuge is carried by the chests (as well as reading inserter hands in hold mode) so as soon as power is restored everything snaps back where it should be.
The new one should only have outage issues with the bootstrapper circuit, which requires a latch to close once the 40 units of U235 demand is met.