Question on fluid throughput

[vanatteveldt]

From an old post by DaveMcW (viewtopic.php?f=5&t=6066) I learned that for medium/long distances pipes have max throughput of 90 fluid/s. This has never been a big problem for me, but my plastic production was stalling because of PG shortage. I have 2x12 plastic plants each working at speed 5.875, so each requires 17.625 PG per second, meaning a 90/s pipe can supply only 5 of them. So, I figured I should run at least 4 pipes from my refinery/cracking (which is about 10 underground pipes away, so comfortably in the 14 - 224 pipe range for 90/s; I guess I would really need 5, but four seems to be enough to saturate 4 blue belts)

This is my current setup that seems to have solved the problem:



But I feel like I don't really know what I'm doing, so maybe this is massive overkill, and maybe this is just plain stupid design.

Practical questions:

1) Is the 90/s value for 14-224 pipes still correct?
2) Is it OK to connect pipes? For example, in the bottom right of the screenshot I reconnect all 4 pipes to branch to the solid fuel processing (which is activated only if PG is too high) and then branch them off again:



Does that lower/alter total throughput? Does that depend on whether fuel processing is switched off, i.e. whether PG is diverted north?
3) Does connecting pipes like above balance the fluids? I.e., if PG is diverted north, does it come equally from all 4 pipes, or only/mainly from the top one?
4) Suppose I reconnect two of the pipes just before they go into the plastics rack. Does that give more flow in the connected part? If half of it is consumed within 14 pipes, is the max for the remainder of the rack still 90/s?

(and is there a "flow meter" mod somewhere? They should really allow wires to connect to pipes like they do to belts...)

[DaveMcW]

90/s is correct for 224 pipes. You get more with 223 pipes and less, up to 120/s at 14 pipes.

Combining pipes usually works, except when it doesn't. You can usually fix it by building the pipes in a different order.

[vanatteveldt]

A right, so 14-224 pipes is 120/s. That explains why I get 4 blue belts of plastic

So the answer is that there are no answers. For a game like factorio where everything else is OCD-able that is really a shame...

[Blurb]

(1):
It is still correct for exactly 224 pipes, not for 14-224 which would indicate an interval.
You need to understand that the throughput of a pipe isn't some hardcoded value based on pipe length.
There's a few other posts (such as viewtopic.php?f=18&t=33685&, viewtopic.php?p=125742#p125742 and viewtopic.php?f=18&t=33324&) going into more detail on how fluid is transferred through pipes, but people tend to ignore them in favour of the simple post you linked to.

(2):
It should have no negative effect on the throughput, aside from what adding an extra piece of pipe does.

(3):
It should balance the fluids to some degree, but how effective the setup will be is up for debate.

(4):
I don't follow.
Could you provide a more detailed explanation of what you have in mind, and/or a picture?

Regarding a flow meter, you can simply build your own using two liquid storage tanks.
Of course, you'd need to perform a series of experiments to get anything useful out of the fill levels - or start digging into the provided material.

On a side note, your setup consumes a lot less PG than you think it is - because you've completely overlooked that your output belts are a massive bottleneck.
Knowing that you produce 160 plastic per second (4 blue belts), you'd only need 240 PG per second to supply the production (assuming no production modules).
Using 2 T3 production modules, you'd only need 200 PG per second - which could probably be supplied by two separate pipes.
This means the average consumption is only 8.33 PG per second, only 47% of what you've calculated.