Circular Kovarex Enrichment Setup

[MBas]

Both stack inserters are set to read content (hold) only.

kovarex.png (432.04 KiB) Viewed 14819 times

[thereaverofdarkness]

Very clever, I think this is the best I've seen yet.

[MBas]

Anyway, i think that "reducing" U235 in centrifuge is not very good goal. Even with this setups, i prefer to have 80 pieces of U235 in. So the process is smooth and dont wait for recharging.

[thereaverofdarkness]

Even with this setups, i prefer to have 80 pieces of U235 in.

Me, too. I found it didn't take all that long to saturate the process for my setup, once I went to do something else to occupy myself for a while:

[Guu]

Enrichment scheme that works by itself, without combinators and any settings

[thereaverofdarkness]

There is a flaw in your design: occasionally an inserter will be busy scooping U-238 which may cause it to miss some U-235. Ultimately it's probably not a huge problem once the thing is properly buffered but my setup above adds inserters filtered to handle U-238 so that the stack inserters can focus on U-235. That way they run it perfectly every time.

[Guu]

There is a flaw in your design: occasionally an inserter will be busy scooping U-238 which may cause it to miss some U-235. Ultimately it's probably not a huge problem once the thing is properly buffered but my setup above adds inserters filtered to handle U-238 so that the stack inserters can focus on U-235. That way they run it perfectly every time.

I showed this scheme not because it works fine or it is better than others. This scheme is very very simple, there is practically nothing superfluous in it, only two inserters without any adjustments, a belt and a filter inserters for picking up the finished product

[MBas]

There is a flaw in your design: occasionally an inserter will be busy scooping U-238 which may cause it to miss some U-235. Ultimately it's probably not a huge problem once the thing is properly buffered but my setup above adds inserters filtered to handle U-238 so that the stack inserters can focus on U-235. That way they run it perfectly every time.

I showed this scheme not because it works fine or it is better than others. This scheme is very very simple, there is practically nothing superfluous in it, only two inserters without any adjustments, a belt and a filter inserters for picking up the finished product

I also belive that any solution which "works" and dont steal from base 40 pieces is good enough, there is definetly much more task then having 72 or 85 or 40 pices inside exactly... The solution of mine is trying to use fact that inserters are much more faster when it move things "from instance to instance" instead of waiting for any belt stacking and can be used even for one simple kovarex.

Anyway i never get why people use more than one koverex process. Well, maybe two of them for really huge bases and maybe three or four if you need massive atomic bomb production. But still... Why you use a beacons to support something which is not filled with prod modules? It just drain electricity with no benefit against just multiplying your pattern.

[Guu]

But still... Why you use a beacons to support something which is not filled with prod modules? It just drain electricity with no benefit against just multiplying your pattern.

It was a test circuit and I accelerated it to shorten the time of checks

[Killcreek2]

I use a simple method that is combinator & wire free, scaleable, & works for both belts & bots.

It relies on "stack size override": 4x stack filter inserters limited to size 10 each, +1x filter inserter limited to size 1, results in 41 U235 removed simultaneously. The 40 are fed straight back in for the next cycle, and the 1 goes off into storage.

Simple Kovarex

Kovarex Enrichment.png (642.87 KiB) Viewed 14669 times

It runs perfectly well [unless the u235 output belt backs up completely...]
Downtime between enrichment cycles is less than 2 sec, if you set the normal stack inserter [greens] to size 10 also [this eliminates the brief pause they would usually do when picking up only a partial handful].
Note the wired belts: This is only to allow the "waste" U238 to be easily re-used [via the red belt], by pausing the main U238 input feed unless the pickup tile is empty.

Even in early game you can still use this method with 5x stack filter inserters set to size 8, or even 8x set to size 5, with some minor adjustment to the feedback loops.

Obligatory Blueprint

Code: Select all

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

[Kryptos]

Why not try something like this:
(see attached image)

No circuits, never clogs, never runs dry (will be back at full power eventually if you have enough U235 for one cycle, but will also prioritize powering up over production), and is easily expandable by extending the middle bit.

Works by keeping things on their side of the belt. The top inserter at the end will pull any U238 off the top half of the belt, allowing it to cycle to the bottom half. As long as the supply inserter and the cycle inserter have enough belt lag that there is a continuous gap (this design works with just 2 centrifuges, never tried one, might need an extra belt loop), you will never have the system clog. Inserters between centrifuges are entirely optional, and quite likely counterproductive if you expect this to ever wind down to a single cycle's worth of U235, but they shave a few seconds off the lag period between cycles. Which is interesting, considering I haven't taken the time to upgrade them to stack inserters (or to add S3s all around). I went a bit overboard on centrifuges this map. Note that the U238 inserters, both the cycle and the supply, have to feed the same side of the belt.

Edit: Oh, there's a second page...Well, this is very similar to Guu's design above, but it avoids the possible issue of U235 slipping by the feed inserter and a centrifuge shutting down. I specifically avoided that in this design.

Also, there may be some superfluous power poles.

Also note that the inserters have to be set up the way they are to prevent clogs. If the U238 cycle inserter is on the U238 side of the belt, it won't prioritize the ones on the U235 side, allowing it to clog. The U235 inserter just needs to be in the same square to avoid clogging.

[nr2117]

Why not try something like this:
(see attached image)

No circuits, never clogs, never runs dry (will be back at full power eventually if you have enough U235 for one cycle, but will also prioritize powering up over production), and is easily expandable by extending the middle bit.

Works by keeping things on their side of the belt. The top inserter at the end will pull any U238 off the top half of the belt, allowing it to cycle to the bottom half. As long as the supply inserter and the cycle inserter have enough belt lag that there is a continuous gap (this design works with just 2 centrifuges, never tried one, might need an extra belt loop), you will never have the system clog. Inserters between centrifuges are entirely optional, and quite likely counterproductive if you expect this to ever wind down to a single cycle's worth of U235, but they shave a few seconds off the lag period between cycles. Which is interesting, considering I haven't taken the time to upgrade them to stack inserters (or to add S3s all around). I went a bit overboard on centrifuges this map. Note that the U238 inserters, both the cycle and the supply, have to feed the same side of the belt.

Edit: Oh, there's a second page...Well, this is very similar to Guu's design above, but it avoids the possible issue of U235 slipping by the feed inserter and a centrifuge shutting down. I specifically avoided that in this design.

Also, there may be some superfluous power poles.

Also note that the inserters have to be set up the way they are to prevent clogs. If the U238 cycle inserter is on the U238 side of the belt, it won't prioritize the ones on the U235 side, allowing it to clog. The U235 inserter just needs to be in the same square to avoid clogging.

I'm interested in this one the most of the ones I've seen so far, please may I have the blueprint?

[impetus maximus]

here is a simple layout i use.
BLUEPRINT IS FOR INSERTER CAPACITY BONUS 5+
you can limit the wired stack inserter (along with ≥ enable cond), and filter stack inserter to 2, 4, or 8 if you don't have stack bonus 5 (10 stack) researched and it should work.

Kovarex.png (157.13 KiB) Viewed 14185 times

feed the top belt with U238 (wire keeps room for the 2 excess from process)
put 40 U235 in centrifuge.
when done 40 U235, and 2 U238 gets recycled.
1 U235 gets placed on lower belt.
they can be tiled together if you need high output.
it will keep running as long as you feed it U238, and take the 1 U235.

inserter stack 10pc blue print string

Code: Select all

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

[Kryptos]

Why not try something like this:
(see attached image)

No circuits, never clogs, never runs dry (will be back at full power eventually if you have enough U235 for one cycle, but will also prioritize powering up over production), and is easily expandable by extending the middle bit.

Works by keeping things on their side of the belt. The top inserter at the end will pull any U238 off the top half of the belt, allowing it to cycle to the bottom half. As long as the supply inserter and the cycle inserter have enough belt lag that there is a continuous gap (this design works with just 2 centrifuges, never tried one, might need an extra belt loop), you will never have the system clog. Inserters between centrifuges are entirely optional, and quite likely counterproductive if you expect this to ever wind down to a single cycle's worth of U235, but they shave a few seconds off the lag period between cycles. Which is interesting, considering I haven't taken the time to upgrade them to stack inserters (or to add S3s all around). I went a bit overboard on centrifuges this map. Note that the U238 inserters, both the cycle and the supply, have to feed the same side of the belt.

Edit: Oh, there's a second page...Well, this is very similar to Guu's design above, but it avoids the possible issue of U235 slipping by the feed inserter and a centrifuge shutting down. I specifically avoided that in this design.

Also, there may be some superfluous power poles.

Also note that the inserters have to be set up the way they are to prevent clogs. If the U238 cycle inserter is on the U238 side of the belt, it won't prioritize the ones on the U235 side, allowing it to clog. The U235 inserter just needs to be in the same square to avoid clogging.

I'm interested in this one the most of the ones I've seen so far, please may I have the blueprint?

Hey, sorry, I haven't really been on the forums in a bit. Here you go!

Code: Select all

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

[impetus maximus]

here is a simple 1 wire method i use now. credit goes to DaveMcW.
i improved the cycle speed a bit by using chests instead of an express belt. (steel pictured, wood in blueprint)

Kovarex.1.wire.png (26.63 KiB) Viewed 13980 times

just feed the near side of the belt with U238, and leave the far side blank for the 41st U235 to exit.
put the initial 40 U235 in the centrifuge.
when the 2 U238 exit into the chest to be reused, it triggers the standard inserter (set to stack size 1) which spits out a single U235 onto the belt.
the remaining 40 U235 get recycled as also. it tiles well.

blueprint string

Code: Select all

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

[Kryptos]

Sorry to double(-ish) post, but I just want to add that my design does *NOT* work with blue belts. They move too fast, and so the middle part that shares U-235 and U-238 must be yellow or red.

[impetus maximus]

Sorry to double(-ish) post, but I just want to add that my design does *NOT* work with blue belts. They move too fast, and so the middle part that shares U-235 and U-238 must be yellow or red.

you can edit your post.

[hummel124]

Hi,
I designed a loop for Kovarex enrichment which outputs U-235 at a rate of 2-3 per second when fully equipped.
The upper left requestor chest requests 200x U-238
The middle requestor chest requests 100x U-235 which I use as a buffer.
When the U-235 chest contains more than 120x U-235 these are taken out of the loop by the long handed inserter to the passive provider chest. This is the only circuit condition I use just to prevent the loop from being underequipped by the long handed guy

I put all the stuff at one side of each belt to enable the centrifuges output to be placed on the other side. The top left inserter prevents the outer belt from being completely filled by U-238.

You can easily replace the chests by belts if you don't like the robots or even skip the beacons or use less centrifuges if you need less output

Factorio_Kovarex_Enrichment_2-3ps.png (639.52 KiB) Viewed 13847 times

I hope you like it
Hummel

[Junit151]

Hi, I'm quite new here, so please move this thread elsewhere if it is in the wrong place.^^

I just wanted to share a Kovarex enrichment setup that is working quite well for me.

Old thread but just wanted to say thanks for the design! It's working well for me, too well in fact. It was eating up all of my U-238!
I had to add some super basic logic that turns it off when it has a big enough surplus of U-235.

It just disables the inserters that feed into the centrifuges when U-235 >= 2500 in each of the two output chests, here's a pic:

Notice the severe shortage of U-238 on the input belt (bottom center)

5 years later the design is still solid

Cheers, Laur

[spiral_power]

I seem to have created it 4 years ago, very simple one.
I use the vehicle trick.
By the way, why was I using the blue modules at the time...was red unavailable?

DpCUhuiUYAEz9tT.png (1.14 MiB) Viewed 4843 times