Page 1 of 1
Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 5:04 am
by NateDangerous
So, I was thinking the other day that it'd be interesting to try running a chemical plant that takes two liquid inputs, but feed it using only one pipe. I came up with a fairly simple concept that works well, and isn't very circuit-heavy (it actually uses no combinators).
I made a short video explaining and demoing it.
Video:
https://youtu.be/fUqqjbooAEE
Here's a screenshot with brief explanation of how it works. Read it from the bottom up.
Disclaimer: This is not meant to be practical. This is mad science.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 7:32 am
by daniel34
NateDangerous wrote:Disclaimer: This is not meant to be practical. This is mad science.
AntiElitz uses something similar in his sub-2h world record speedruns, but without using the circuit network. Although his goal there is to vanish some of the pipe contents while yours is to transport both to the chemical plant and then 'clean' the pipe.
He connects the light and heavy oil outputs of a refinery together and passes them through a steam engine, which consumes the (in his case unused) light/heavy oil. He saves time this way as he doesn't need to crack it or set up tanks to store it, making the piping very simple and fast. Later in the speedrun he adds a chemical plant directly to the steam engine that creates solid fuel from the light oil.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 7:39 am
by Neotix
Instead of using steam engines to drain pipes, you could use combinators (SR Latches) to pump exactly amount that is needed. In that way shared pipes will be drained by chemical plant. Also it can be sequence that pump oil -> water -> oil -> ....
I didn't try it but concept of shares pipes is interesting.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 8:14 am
by XKnight
Concept is very interesting, and can be even used in regular factories to increase space effectiveness and reduce pipe mess.
Although, I have some concerns about using steams for this purpose.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 8:20 am
by NateDangerous
XKnight wrote:Concept is very interesting, and can be even used in regular factories to increase space effectiveness and reduce pipe mess.
Although, I have some concerns about using steams for this purpose.
I like that design. That was kind of what I was initially trying for, but never quite got to. Gonna have to try that and see where I went wrong. It looks like my first attempt was over-complicating the issue.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 8:44 am
by Neotix
XKnight could you add diagram and link that solution into TDT thread (
viewtopic.php?f=18&t=5553)?
I think that this will be helpful for many people.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 9:27 am
by XKnight
Neotix wrote:XKnight could you add diagram and link that solution into TDT thread (
viewtopic.php?f=18&t=5553)?
I think that this will be helpful for many people.
Perhaps my design is very complex because it is a bug workaround...
Correct build should look like
this.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 9:37 am
by NateDangerous
XKnight wrote:Neotix wrote:XKnight could you add diagram and link that solution into TDT thread (
viewtopic.php?f=18&t=5553)?
I think that this will be helpful for many people.
Perhaps my design is very complex because it is a bug workaround...
Correct build should look like
this.
That's very interesting! That would explain some of the odd behavior I was seeing in my first attempt on this.
The other odd thing I was seeing is that both pipes and storage tanks would show ALT information stating they still contained liquid that they did not. Like a circuit would report a storage tank had zero petroleum gas in it, yet the petro gas icon still displayed on it.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 9:49 am
by XKnight
NateDangerous wrote:
The other odd thing I was seeing is that both pipes and storage tanks would show ALT information stating they still contained liquid that they did not. Like a circuit would report a storage tank had zero petroleum gas in it, yet the petro gas icon still displayed on it.
Actually, this is well known thing. Information in status window may display "0.0 water" but in fact there is 0.01 water inside (or another number which is less than 0.05). So tank is not empty, but amount of water inside is 0.0
Also, if you connect small pump to this tank you will get this water from it and truly empty tank.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 1:44 pm
by XKnight
Unfortunately,
this bug won't be fixed in the nearest time, so...
This build is similar to four-stroke engine, where first stroke - input water, second - push fluid next, third - input petroleum, fourth - push fluid next.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 6:24 pm
by DaveMcW
Three liquids, one pipe.
- refinery-filter.jpg (244.25 KiB) Viewed 4777 times
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 8:30 pm
by golfmiketango
Wow.
So... do I understand correctly that putting multiple liquids into a pipe doesn't create some kind of slurry, but the liquids remain separate and exit in the order they are input (at least, once any bugs are fixed)? What if we replace the pipes with tanks, same behavior? I always wondered how that AntiElitz trick was possible.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Sun May 29, 2016 9:54 pm
by XKnight
golfmiketango wrote:Wow.
So... do I understand correctly that putting multiple liquids into a pipe doesn't create some kind of slurry, but the liquids remain separate and exit in the order they are input (at least, once any bugs are fixed)? What if we replace the pipes with tanks, same behavior? I always wondered how that AntiElitz trick was possible.
Yes and no simultaneously. I don't know how liquid physics will look after bugfix, but if everything remains the same it won't be possible to use several liquid in the same pipe (at least without additional combinator-pump stuff). The reason of this is limited push amount (only 1.0 unit of liquid could be pushed) and limited pressure difference (neighbourhood pipe should have 2.1 times higher pressure to start pushing process).
So, pipe will stuck if water amount in one pipe segment is greater than 1.0, this segment will be unpushable and it will stop the entire flow.
From the other side, if pressure in every segment is less than 1.0 it also will stuck, because there is no pressure difference.
But, the pipe might work if you create this pressure difference by draining liquid to assemblers. Although it will be very unstable.
Re: Two Liquids, One Pipe (proof of concept)
Posted: Tue May 31, 2016 3:26 pm
by Qon
DaveMcW wrote:Three liquids, one pipe.
That's so cool. Didn't know you could pump out liquids from the plant input. But the liquid sorting will lock up if you get non-heavy in the underground pipe connected to the heavy solid fuel plant and some heavy at the light/petrolium inputs. Tested. Will happen at high pressure at least when the liquid doesn't all go into the plants as input but slips past. And for this to be usable for me it has to work in high speeds and with a sorting that takes small enough space for it to make sense to compress refineries for it. Little point if the refinery space saving is less than space sorting is used. You might have beacons at refineries though which makes space saved at refineries worth more, but there are limits still. And I would need input that is fast enough to keep up with it one pipe too, and the solution by XKnight...
XKnight wrote:Concept is very interesting, and can be even used in regular factories to increase space effectiveness and reduce pipe mess.
Although, I have some concerns about using steams for this purpose.
...doesn't seem to be fast enough either if you are doing it for beaconed refineries.
But fantastic stuff both of you still.