A simple germ killer that processes a continuous full pipe of input.

[Tonyroid]

Just now, Yunru said:

Ah, fair point. Packet's only have a fraction of the content of the tile. As the save will probably reveal.

I see you have 130,000 kg per tile of ph20 in your output. Are you sure there isn't some kind of averaging thing that's making the germs dissappear?

[Yunru]

46 minutes ago, Tonyroid said:

I see you have 130,000 kg per tile of ph20 in your output. Are you sure there isn't some kind of averaging thing that's making the germs dissappear?

No idea, but feel free to stress test the design. I'm going to sleep.

Edit: A thought that's keeping me up is if you extend the time the water spends in the reservoir (via door disabling), you could cut down on the number of reservoirs needed. 

[Yunru]

So doing some quick napkin maths on some assumptions I'm not even sure are right:



A toilet produces 110k germs on a 10kg packet (these numbers are purposefully inflated slightly because they're both easier to work with and because it gives a margin for error).



This enters a tank filled with 4990kg of germ free water in the first tank. Anything leaving this tank will thus carry 220 germs(?).



This enters the second tank. Due to rounding, the 0.44 germs that would leave this tank become 0.



 



So for toilets, no more than two purification tanks are needed. Are there any situations that would result in more germs than that?



 



More on the fly maths, effectively you want the number of germs to be less than 0.5. Each tank divides the number of germs by 500.



So if X is the number of germs and Y is the number of tanks;



0.5=X/(500^Y)



(500^Y)/2=X



...



Not the formula I was going for, but it's neater than solving for Y. From this you can work out the threshold of each number of tanks. 250 germs/packet for 1 tank, 125,000 germs/packet for 2 tanks, 62,500,000 germs/packet for 3 tanks, etc.