[raiguard][2.0.32] Liquid voiding on automatic recipe change on chained units

[BBBBSammich]

out.mp4

(1.48 MiB) Downloaded 24 times

This is a continuation of the following issue.

In a direct insertion processing unit recycling setup, an unexpected amount of sulfuric acid was used, after which the above issue was opened. After the gained information an appropriate change was made to the design in which the fluid system was kept half full. Liquid was however still disappearing at about the same rate. After further investigation I have discovered that not all liquid is returned specifically from chained plants as in the above video, with method as follows.

• The fluid system is filled and is shut off from the source so that no more fluid can be introduced into the system.
• A note is taken of the amount of fluid in the tank as, 10201 units.
• The first plant’s recipe is removed, and its fluid is moved into the pipe system to where the tank reads, 10749 units, which is a positive difference of 548 units.
• The first plant’s recipe is then reset, and the fluid returns into the plant to where it once again reads 10201 on the tank, signifying that no fluid is lost.
• This is repeated to reinforce the point.
• The second plant’s recipe is then removed and it’s fluid contents moves into the system to where the tank reads 10359 units, a change of 158 units.
• The second plant’s recipe is then reset, and the fluid returns into the plant where the tank now reads 9818 units, a negative change of 541 units, signifying a voiding of fluid.
• This is once again repeated to force the point that fluid is being voided.

Testing setup blueprint:

Code: Select all

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

[Muche]

The issue seems to be related to the behaviour of EMP's thru-pipeline when it's not directly connected to an outside pipe.
When the recipe is cleared, only the thru-pipeline's fluid is flushed to the storage tank's pipeline. Recipe input's fluid (400 units) ignores the neigbour's thru-pipeline.
Adding an external pipe solves it:

127305-OneExternalPipe.jpg (121.72 KiB) Viewed 1568 times

Both thru-pipeline's and recipe input's fluid get flushed into storage tank's pipeline, then thru-pipeline is disconnected (and the fluid amount in that one external pipe rises proportionally, it does not try to take all the fluid being flushed out).

This "thru-pipeline not connected to an external pipe doesn't count as pipeline" also affects when fluid mixing warning during a manual pipe placement is issued, see viewtopic.php?p=635413.

[BBBBSammich]

Is there an update on this? The fluid mixing topic was solved a different way.

[raiguard]

Thanks for the report, this has been fixed for 2.0.44.

[tinker9]

I'm guessing this was 2.0.44 as well

[raiguard]

Indeed.