Liquid Reservoir Output bug... I think?

[Crimsontide]

I have a funny issue with my liquid reservoir.  It occasionally outputs small packets (200g) despite having room to put out more and having the liquid to do so.

Ok so I know it looks a little crazy, the reservoir I'm talking about and the packet are highlighted.  Its a chlorine de-germing loop followed by a cooling loop.  At the end of the cooling loop (just off screen to the left) there is a valve set to 9900g/s.  The liquid reservoir works fine with the valve set at 10000g/s (a full packet), and will spit out a consistent 10,000g packets.  But set the valve to 9900g/s and it will occasionally (every 4th packet) spit out a 200g packet (the other 3 being 9900g as expected 10000g).

It might be hard to see from the screenshot, but the reservoir in question is not on a door.

[JRup]

I've had something similar happen. Even with gas reservoirs.

I've made my mind to think of them as a very long stretch of pipe that fits up to 5T of liquid (or 150kg of gas) that happens to average liquid temperatures of the same type but does not do anything else for contents...

With that in mind, whatever comes and whichever volume does so must go out just the same. So maybe a small packet got pumped, then the same will come out eventually. It is virtually an unnoticeable thing if pipes back up with a single element.

[nakomaru]

A basic mechanic of pipes is that any amount of liquid in a pipe overlapping a green output will entirely block output.* (except for bridge and valve outputs) Easy to fix with bridge merging / buffering.

I can't see your whole system, but something like this is happening.

1. Sometimes your bottom or side piping will have an incomplete packet in it, such as 200g.
2. The top piping tops it off by adding 9800g to it.
3. 200g is now in your top reservoir output pipe, which blocks output from the top reservoir.*
4. The flow continues forward with e.g the bottom reservoir empty, allowing full flow from the top reservoir.
5. No self stacking piping exists so 200g continues freely.

15 hours ago, Crimsontide said:

despite having room to put out more

* Basically this is the source of the confusion. It doesn't have room.

Visualized:

A bridge buffer solution.

A bridge buffer is the same thing as a packet stacker without the automation. The second bridge input pipe acts as a 10kg special reservoir without the partial exit limitation. There is a break in piping between the inputs and outputs.

By the way, only a 6x3 room, 3 full reservoirs of unmixed liquid, no doors and no automation is necessary for decontamination.

[Crimsontide]

Thank-you, appreciate the gifs, I understand now.  Had a feeling it wasn't a bug, rather a misunderstanding of it.

As for the decontamination.  My understanding is that the liquid has to be in chlorine for 3 cycles right?  As soon as germy water hits the reservoir, the germs are mixed and that reservoir will immediately start emitting germs as well.  That system you presented would only work as long as the input/output was low enough for mixing to occur for a long time.

[nakomaru]

3 hours ago, Crimsontide said:

Thank-you, appreciate the gifs, I understand now.  Had a feeling it wasn't a bug, rather a misunderstanding of it.

As for the decontamination.  My understanding is that the liquid has to be in chlorine for 3 cycles right?  As soon as germy water hits the reservoir, the germs are mixed and that reservoir will immediately start emitting germs as well.  That system you presented would only work as long as the input/output was low enough for mixing to occur for a long time.

You're welcome.

The system presented will take any same element input at up to 10kg/s and immediately output up to 10kg/s.

This is because it always maintains 15T of that same liquid to mix with, and it dilutes the germ concentration by a factor of 125,000,000 ((5000kg/10kg)^3), which equals 0 for any natural concentration of germs. It's poop homeopathy.

[Crimsontide]

Ahh I see, that is clever.  Though would that work for a continual 10kg/s input/output?  I guess it would be time to break out the sand box...

[Yunru]

1 minute ago, Crimsontide said:

Though would that work for a continual 10kg/s input/output?

The way the feedback loop is constructed, feedback is prioritised over output, and input is prioritised over feedback. 

Thus whenever an amount enters the system, an equal amount leaves. 

[nakomaru]

28 minutes ago, Crimsontide said:

Though would that work for a continual 10kg/s input/output?

Here's the steady-state numbers for 10kg/s with the maximum natural concentration of germs for polluted water input.

[Crimsontide]

17 minutes ago, Yunru said:

The way the feedback loop is constructed, feedback is prioritised over output, and input is prioritised over feedback. 

Thus whenever an amount enters the system, an equal amount leaves. 

That I understand.  And for most systems this is fine.

Mine is a bit different.  The cooling loop, and water loop are the same, and spam the entire map.  So I need to feed a continual 10kg/s through this.  Obviously you still need to keep the reservoirs topped up (to keep your dilution factor high), so you still prioritize the feedback over feed out, but once the reservoirs are full, then the system needs to handle a full 10kg/s continually.

My worry would be that even if a small amount got in, despite the high dilution factor, would start to accumulate.

edit: and mid-reply you already post an answer

[wachunga]

@nakomaru

First I would like to applaud your dogged determination in fighting the good fight in regards to this subject.

Second I would like to nominate "Poop Homeopath" as a custom title if you ever tire of your current one.

Cheers.

[Crimsontide]

On a side note, what software are you using for video capture to make the gifs?

[nakomaru]

ShareX