Fluid rate drop with even math

[SLAINE]

hi, i started my second playthrou recently.

one pure oil runs OC with 600 oil / min
pump supports a mk2 pipe
segments show 600/min

I split this into 2 300/min pipes (thus one side will not get more...)

300 runs 8 (plastic/rubber) refineries -> consuming 240/min
1 refinery runs fuel consuming 60/min

no, output is blocked/crowded all at 100%

the other 300 oil went into
5 fuel refineries (all 100%)
consuming the 300 oil

the fuel is pumped in a buffer and feeds 8 fuel-gen with only 198 fuel/min consumption.

thus there should no shortage at all.

nevertheless, once in a while one fuel-gen stops and only perform at ~96%

Is the simulation "hanging" or what am I missing?

Should I reduce the clock to only use 196 fuel/min to have a bigger buffer?

[Fenix]

So many possibilities, as to what is causing this.
Try turning off a couple generators and let the buffer fill to about half, then turn them back on.
Also maybe put valves in to prevent backflow.

[JBlaze]

If I recall correctly, fluid pipes suffer from a floating point value error. You should not expect to get the full 600 M3 per minute from a mk2 pipe. Instead plan for at most 580 M3 per minute.

[SLAINE]

Thanks for the response. I will use more "save-buffer" thus running the fuel-gen a little lower. and despite I have 4GW left at the moment, I build a 2,4GW for one hour power storage.
That should prevent any harm from mini blackouts of one or two generators.

[SkootyWang]

Fluids suck so hard in this game. Forge ahead and fight it if you want. I coped by just building everything with water/liquid at ground level or below the extraction point. Then I went and downloaded the "fluids are gases" mod and life has never been more like a flowing river made of gas that just works.

[Casurin]

Originally posted by SLAINE:

Thanks for the response. I will use more "save-buffer" thus running the fuel-gen a little lower. and despite I have 4GW left at the moment, I build a 2,4GW for one hour power storage.
That should prevent any harm from mini blackouts of one or two generators.

be aware that the fluid-buffers do NOT work like most people would expect.
Buffers and gravity can be used together to form a reliable network, but if not understood they can be the source of most trouble.

[Ryzilynt]

Having experienced all manner of fluid nightmare known to man. I can only say good luck.

I have ultimately been able to make everything work, one way or another, in the end.

[>< V ><]

Originally posted by Casurin:

Originally posted by SLAINE:

Thanks for the response. I will use more "save-buffer" thus running the fuel-gen a little lower. and despite I have 4GW left at the moment, I build a 2,4GW for one hour power storage.
That should prevent any harm from mini blackouts of one or two generators.

be aware that the fluid-buffers do NOT work like most people would expect.
Buffers and gravity can be used together to form a reliable network, but if not understood they can be the source of most trouble.

I would also look here first.

I understand people say or believe that mkII pipes cannot carry 600 m3/min, but they can. I have no issues with mkII piping and have all of them at 600 m3/min with my factories at 100%.

Yes, there can be problem solving at times, but this is in a sense, a problem solving game.

I've heard, but never tested because I see no need for fluid buffers, that fluid buffers will shut down output in order to fill. I would remove your buffers as they are overall, not needed.

I have no idea what the actual programmed physics are in this game, but I get a sense that water does not have friction. That once you get sloshing in your system, that there seems to be no friction to slow that action and the sloshing can continue indefinitely because of no friction. This is my hypothesis and my solution is to shutdown the last factory in the line and let the rest fill. Once they are filled I slowly ramp up the last factory to 100% and this has always solved what I hypothesize as frictionless sloshing.

[Fenix]

Originally posted by Casurin:

be aware that the fluid-buffers do NOT work like most people would expect.
Buffers and gravity can be used together to form a reliable network, but if not understood they can be the source of most trouble.

This is why you use Valves on both side, Creating an input and output, plus having the buffer on a raise platform is usually a good idea, too

[Wurum]

I had a heck of a time with balancing fluids last play through. So, this play through, I'm packaging excess fuel and sending it to the AWESOME Sink.

[dazkaz]

Originally posted by Fenix:

Originally posted by Casurin:

be aware that the fluid-buffers do NOT work like most people would expect.
Buffers and gravity can be used together to form a reliable network, but if not understood they can be the source of most trouble.

This is why you use Valves on both side, Creating an input and output, plus having the buffer on a raise platform is usually a good idea, too

This is what I have done in my game and it works great.
Pump fluid to a buffer, that is placed higher than the manifold pipes, connected to the machines.
I place a one way valve just before the buffer, on its feed pipe from the pump, to stop back flow.
The buffer works like a gravity fed, water system header tank.
You can also accomplish the same, over longer distances, by packaging the fluid at its source, then transporting and emptying it to the buffer (header tank).

[PsYcHo_O]

Originally posted by >< V ><:

Originally posted by Casurin:

be aware that the fluid-buffers do NOT work like most people would expect.
Buffers and gravity can be used together to form a reliable network, but if not understood they can be the source of most trouble.

I would also look here first.

I understand people say or believe that mkII pipes cannot carry 600 m3/min, but they can. I have no issues with mkII piping and have all of them at 600 m3/min with my factories at 100%.
...

Maybe it depends on the fluid or the consuming production machine (or RNGesus), but fully overclocking a nuclear reactor (consumes 600 m3/min of water) was not possible for me because the water supply would always drop off. I tried every possible combination of valves, pumps, buffers and elevation but the result was always the same. Slowly but surely (took at least a whole hour) the water level would sink until the reactor shut down.
I had 8 reactors at that time and it is probably worth mentioning that the shut down times were completely random and not synced with the run times in any way. Some reactors even remained stable over a whole day.

I could only fix it by reducing the consumption to ~595 m3/min (still providing 600 m3/min water) and building an additional reactor to pick up the slack.

[Casurin]

Originally posted by PsYcHo_O:

Maybe it depends on the fluid or the consuming production machine (or RNGesus), but fully overclocking a nuclear reactor (consumes 600 m3/min of water) was not possible for me because the water supply would always drop off. I tried every possible combination of valves, pumps, buffers and elevation but the result was always the same. Slowly but surely (took at least a whole hour) the water level would sink until the reactor shut down.
I had 8 reactors at that time and it is probably worth mentioning that the shut down times were completely random and not synced with the run times in any way. Some reactors even remained stable over a whole day.

I could only fix it by reducing the consumption to ~595 m3/min (still providing 600 m3/min water) and building an additional reactor to pick up the slack.

Sadly in vanilla there are no "smart" extractors that can downclock to meet the demand.
I normally run them underclocked but still slightly above the supply-need. The machines are always supplied from above and normally also get a small gravity-block. For longer distances i have a tank by the extractor and a tank at the consumer (funny thing is they do NOT need to be inline to work - a single connection is enough). Having them in series basically requires you to use valves to prevent sloshing.