Updated: The compact supercharged 10Kg/s distiller

[Saturnus]

In continuation of the discussion in this thread: 

Ok, so made a new very compact completely stable 10Kg/s boiler based on the supercharger pre-heater I developed earlier (see pic).

It's completely self-regulating. Has ultra-fast start up if turned off for whatever reason. Very compact as it's 10 tiles high and 29 tiles long, so in fact more compact than any of the 5Kg/s+ boilers in the linked thread. Output water temperature is a manageable 58C. It's serviceable through naphtha airlocks. And it's easy to build in a survival game as there no place you need to trap specific air pockets or anything like that. Just turn on the air pump to make it a vacuum before starting it up, and it's good to go.

Hope you like it, and can use it.

EDIT: Measured power consumption over 10 cycles is 2760W(avg). Consisting of 3.333 pumps, 0.8333 aquatuner, and 1 tepidizer.

NOTE: It's perfectly possible to build a more power efficient by going larger. Just remember that the sweet spot for the supercharger is 40C to 42C. Outside that feed temperature range the output water temperature drops. That's why the system as posted is self-regulating. It will automatically fall back into the optimum temperature range as long as the input water is between 24C and 30C 

I've optimized a lot and found new solutions to some of the heat deletion bugs. So much so that we might have found a heat creation bug instead.

Anyway, this thread is to show case the result. It's a working 9400g/s water distiller that uses about 2650W on average, so not the most power efficient solution but with just one aquatuner and one tepidizer as the active components the throughput is astounding. The power consumption includes the feed pump though, and the consumption is so high because the throughput is high so there's a lot of pumps that need to run constantly.

The boiler section that uses crude oil as base to avoid the all too familiar heat bug along with the wire bridges and the powered airlock to keep temperatures to a safe level is the new part. Compared to previous versions the entire distiller is also mirror imaged in order to force condensed water over the entire condenser floor. Flip it around and you get thermal run away with steam pressure building up into the 1000s of kg per tile.

The clean water output is germ free no matter how dirty the input water is. I tested it with 100 million germs per tile which is about 100 times as many as you'll find in game. The output water temperature is 35C with an input water temperature of 27C.

Anyway, hope you can use some of the ideas. And it's certainly possible to build this in a survival game should you want to.

EDIT: I misread power consumption. It was total system consumption, not the distiller part before. Sorry.

 

 

 

 

Supercharged 10Kgs Distiller.sav

[Sevio]

Nicely done on developing it into a complete working build. It's quite large but we already knew that the bigger the heat exchanger, the better this stuff works.

[Coolthulhu]

What's the heat creation bug?

[Saturnus]

39 minutes ago, Coolthulhu said:

What's the heat creation bug?

We don't know really. But if you calculate the amount of J an aquatuner can put out then it should only be possible to get about 6000g/s with the boiler input temp.  So something strange is going on unless they silently changed the aquatuner stats in the code without make note of it anywhere.

[Saturnus]

The big challenge with a build that has this high throughput is that since you're running so close to the maximum of pumps, you have to be really careful making sure there's no choke points in the design. If you make the condenser or heat exchanger channel too long the channel itself will limit the throughput. Too short and you will not hit high enough temperatures to feed the boiler efficiently. The condenser design is also tricky, too high pressure and you get thermal runaway and the steam never condenses but just builds up higher and higher pressure, too low pressure and the steam will not effectively heat the feed channel.

I've made some changes to the design and think I have it stable to 9900 now. Effectively the maximum efficiently given that it's not stopping the feed because of there being anything wrong in the design but simply down to occasional lag spikes in the program. It even runs at lower power consumption but I'll test it a bit more before posting. It's also huge, well very long but not that tall.

[Cilya]

18 hours ago, Saturnus said:

If you make the condenser or heat exchanger channel too long the channel itself will limit the throughput.

I didn't have this problem. Sure, there is a limit... but the limit is so high, I never run into it, even though I made it much bigger.

18 hours ago, Saturnus said:

The condenser design is also tricky, too high pressure and you get thermal runaway and the steam never condenses but just builds up higher and higher pressure, too low pressure and the steam will not effectively heat the feed channel.

I'm not sure about that. The steam will turn into water and the temperature will be exchanged when mixing into the channel. On the contrary, you want to condensate the steam as fast as possible to improve the heat exchanger efficiency, i.e. have the highest possible gradient there. Or did I miss something ?

[Reaniel]

21 minutes ago, Cilya said:

I'm not sure about that. The steam will turn into water and the temperature will be exchanged when mixing into the channel. On the contrary, you want to condensate the steam as fast as possible to improve the heat exchanger efficiency, i.e. have the highest possible gradient there. Or did I miss something ?

I agree... wouldn't the upper level exchanger be more effective when the water condenses sooner rather than later?  I mean having a thin layer of 99C water there should, theoretically, be better than having a thick 101C steam, isn't it?

[Saturnus]

24 minutes ago, Cilya said:

Or did I miss something ?

Yes. When pressure in the condenser reaches a certain point the condensed water starts to re-evaporate into steam, which increases the pressure and temperature even higher, and you have thermal runaway. Too low pressure and you can't reach the roughly 101C feed temperature required to run at 10kg/s constantly.

   

[Saturnus]

8 minutes ago, Reaniel said:

I agree... wouldn't the upper level exchanger be more effective when the water condenses sooner rather than later?  I mean having a thin layer of 99C water there should, theoretically, be better than having a thick 101C steam, isn't it?

No. And no. The water level needs to be at least 100kg and at around 102C or there's not enough energy to feed 10kg/s at 101C.

[Saturnus]

Updated!

[WanderingKid]

I'd thought the purpose of most of the boilers discussed was to steam PH2O into H2O without sand.  Is the purpose of this simply to clean PH2O of food poisoning?  .. Also didn't realize you can stick a liquid switch on top of a bridge, that'll be handy!

[Saturnus]

14 minutes ago, WanderingKid said:

I'd thought the purpose of most of the boilers discussed was to steam PH2O into H2O without sand.  Is the purpose of this simply to clean PH2O of food poisoning?  .. Also didn't realize you can stick a liquid switch on top of a bridge, that'll be handy!

This does that. It's a distiller. It steams PH2O into H2O,without using sand, and in the process removes all germs.

Or are you just looking at the supercharger pre-heater?

[WanderingKid]

@Saturnus I have no understanding of the mechanics of how this works then.  The Tepidizer can't get to boiling, afaik, on the right side I thought that was polluted water output you were getting, not water, and I'd have expected a gas pump to collect the steam, not a water pump.  I may have to build this simply to understand it.  Well, thanks for the design!

EDIT: I brainlocked.  The supercharger is to pre-heat the water FOR the distiller!  AH!  Thanks!

[Saturnus]

Just download the save file and have a look around. It'll explain most of the features

Correct, a tepidizer can't boil water under normal circumstances, and if tweaked to, it's throughput is horrendously low. Only a few grams per second.

Try putting a tepidizer in water. The normal maximum temperature you can reach is about 90C. I reach 108C with the supercharger. The challenge with high throughput distillers have always been to get the polluted water temperature high enough so the aquatuner could boil it faster. Previously, the maximum temperature you could reach with a heat exchanger was just under 100C. With 108C feed water the aquatuner doesn't even have to run constantly to boil 10kg per second. It's a big breakthrough in distiller designs.

You should not collect steam with a pump. That's a very ineffective method, and you just risk phase change destroying your pipes. Instead you should have a condenser which is the statues to the left.

EDIT: When I say ineffective I mean it. Remember it will take minimum 20 air pumps running non-stop to keep up with the steam generation.. A air pump maximum is 500g/s. This generates 10000g/s steam.

[NanoD]

Well I have to say you have created a ridiculous machine there. 

I just run all the polluted water and just watched it mesmerized.

[Reaniel]

I've been playing around a bit with this setup, and I can now see how it's really dependent on there being sufficient steam pressure for it to work best, since the bridges are mainly conducting heat from the steam (gas) tiles to the door tiles.

[Saturnus]

Made a couple of additional tweaks. Mainly fail safe measures but also added a small zig-zag in the the heat exchanged to destroy as much heat as possible. It's a weird thing when the thing is running so effectively you have to actively find ways to destroy heat whereas before we were trying to eliminate those bugs.

Anyway, throughput still unchanged (after all, it is the maximum possible) but the small charges means the supercharger output temperature is now a staggering 112.2C. Which in turn means power consumption is reduced even more. Still running the 3.333 pumps and a tepidizer which can't be changed. But the aquatuner now only runs 2/3rds of the time so total average is 2560W(avg).

Build is also slighty more compact now. Overall dimensions are unchanged but it takes up less tiles. In fact it takes up considerably fewer tiles now than any of my previous builds.

Supercharged 10Kgs Distiller.sav

[Whispershade]

How do you handle dirt debris. Do you have issues with entombed aquatuner?

[Saturnus]

Just now, Whispershade said:

How do you handle dirt debris. Do you have issues with entombed aquatuner?

There's a double naphtha airlock in the top right corner with stairs leading directly to the boiler basin. With the oil upgrade came atmo-suits so there's no chance of scalding either. Just set check point at the entrance.

[Whispershade]

But does the dirt ever entomb or just collectable fragments?

I saw the naptha access way. I thought, "I guess membrane airlocks are a thing now."

I haven't built a distiller before and I am very big on automous action so I was thinking of ways it could run without intervention. I settled on using the dirt for mushbars and letting them rot next to pufts.

But if they entomb I will have to configure the polluted water pit in such a way as to break the dirt blocks with water pressure while still being dupe accessable for gathering. I was happy I could enter with atmo suits without scalding. As healing dupes is the micro I am trying to avoid.

[Saturnus]

I've never experienced it entombing itself ever. Theoretically it could but it would take 3333 cycles, give or take a few hundred. But you will want to clean it once in a while as the dirt build up affect efficiency. In this set up that just means it'll use a little more power over time as you're not endanger of it not being able to run a full capacity.

[manu_x32]

19 hours ago, Saturnus said:

There's a double naphtha airlock in the top right corner with stairs leading directly to the boiler basin. With the oil upgrade came atmo-suits so there's no chance of scalding either. Just set check point at the entrance.

How do you build a naphtha airlock in a real game?

Actually, I guess that it's gonna be so hot in there that you could just built plastic tiles where you want the naphtha doors, and they will melt very quickly. 

Good job with the new super boiler!

[Saturnus]

Just now, manu_x32 said:

How do you build a naphtha airlock in a real game?

First you melt some plastic which is dead easy as plastic melts at 76.9C. Then you have naptha. Naptha in the game is a special liquid that flows vertically (upwards) instead of horizontally. So put a valve on the floor where you want the naptha airlock and pump in the naptha. Once the first tile is full it'll fill the tile above. A tile of naptha holds 740kg, so you need about 750kg per airlock.

Alternatively, you can do it manually with the liquid bottler and a bottle emptier.

[eloy2030]

I´ve made a couple of boilers befor, max I reached was 6000 g/s so congrats!

BTW, how do u handle different inputs, I mean, can this handle changes in the PH2O input?

[Saturnus]

3 minutes ago, eloy2030 said:

I´ve made a couple of boilers befor, max I reached was 6000 g/s so congrats!

BTW, how do u handle different inputs, I mean, can this handle changes in the PH2O input?

Yeah. No problem. Obviously you can't feed it more than 10kg/s as that the maximum you can run through a liquid pipe. It's self-regulating, and even power outs isn't a problem.