Water plumbing heat experiment

[Sevio]

5 minutes ago, Saturnus said:

@Sevio I wasn't sure but I did seem to remember that vents don't become overpressurized in polluted water (due to the lower mass per tile than regular water), so I didn't propose that exact thing before I had a chance to test it.

I've used that feature before so I know it's possible but now it's actually turning out to be quite handy both for avoiding the heat destruction bug and avoiding the crossed flows!

[Saturnus]

Btw. Just to satisfy my OCD. Could you finish the sculptures. Cheers

Here's a granite The Thinker on a vacuum background.

 

TheThinker.yaml

[Cilya]

11 hours ago, Sevio said:

I have to note my doubts here about creating heat via the dropping water exploit.

I've just done the experiment. The exploit only works for cooling, not heating. I've droped 1kg of water at 380°K in 3975kg of water at 260°K. The temperature change is 0.2K on one tile only.

It's more than the theory would predict. But when you swap the two temperature, the result is spectacular. When I drop 1kg of water at 260°K in the same pool but at 380°K the temperature change to around 355°K, near one fifth of the temperature difference.

[Sevio]

And now I can present the third and probably last iteration of my boiler, which I think is now worthy of this save's name, The First Law. @Saturnusyou'll be happy to see that no other boiler rivals the aesthetic refinement of this one. "I think, therefore I boil."

Savegame attached in case you'd like to see for yourself how it's running.

Overlays in the spoiler:

Spoiler

The Tepidizer has now been moved into a separate basement so that it can preheat the incoming polluted water before it gets to the heat exchanger. To avoid any potential heat destruction from dropping water, as well as crossing the (water) streams, this design pushes polluted water into its respective rooms from below. I increased the size of the heat exchanger, made it all out of sculptures and made the build more horizontal to keep the height down. It is now 21 wide by 17 tall, a bit wider than the previous design.

I tested this for a bunch of cycles in Power mode to see what it's capable of and I was quite astonished! I think this design pushes the heat exchanger to the limit as the polluted water reaches just over 97 C before being pumped to the aquatuner. This has allowed me to increase throughput in Power mode to 5000 g/s! It's been running stable for quite a few cycles now. Steam pressure is immensely high at the left end of the boiler (over 60 kg, seen over 100 kg at times) but at the far right the steam pressure is alternating between vacuum and 100-300 g so it's condensing the steam at a stable rate. This high throughput rate does come at the cost of a significantly higher water temperature though.

A recap and update of the features and stats:

• Input temperature: 26.9 C - vaporization temperature: 122.3 C -  thermal gradient: ~ 95.4 K
• Two modes available: Cold and Power
• Throughput: 2250 g/s polluted water in Cold Mode, 5000 g/s in Power Mode
• Power usage: 1.40 kW in Cold Mode, 2.08 kW in Power Mode. (Power usage tracked with a large battery bank and averaged over 1 cycle, includes remote polluted water input pump)
• Water output temperature: 5-7 C in Cold Mode, ~ 53.2 C in Power Mode (depends somewhat on environment temperature around the cooling tank)
• Maximum Aquatuner effectiveness through use of a polluted water cooling loop
• Power mode also doubles as Efficiency Mode due to Tepidizer preheating being more efficient than relying mostly on the Aquatuner.
• Polluted oxygen siphon
• Aquatuner feedback loop that can save on pumping power when running in Power mode

The First Law.sav

[Saturnus]

@Sevio Very cool... or should I say hot. I might have a few small improvements I'll play around with, even not involving putting a geyser in the middle of it.

First given the higher than input temperature of the cooler, it might be possible to just dump the incoming water directly in the cooler reducing the need for the tepidizer to run as hard. And then given the 5000g/s throughput I might be temped to run the aquatuner and input water feed on a single pump output split in two and then use a feedback loop on the aquatuner.

Great work!

[Sevio]

12 hours ago, Saturnus said:

First given the higher than input temperature of the cooler, it might be possible to just dump the incoming water directly in the cooler reducing the need for the tepidizer to run as hard. And then given the 5000g/s throughput I might be temped to run the aquatuner and input water feed on a single pump output split in two and then use a feedback loop on the aquatuner.

This might be a bit awkward as that pump is dedicated to feeding the Aquatuner with cooling medium, but I do recognize that with the high output temperature at maximum throughput there is still more heat that can be reclaimed to reduce power usage. The feedback loop is something I thought of doing as a further improvement ever since you mentioned it, but I'm not sure either of those two improvements can be done without sacrificing the ability to cool the output water to 2 C.

But maybe I haven't been able to wrap my head around the concept just yet. If your improvements work out I'd be very interested to see them!

[The Plum Gate]

17 hours ago, Cilya said:

I've just done the experiment. The exploit only works for cooling, not heating. I've droped 1kg of water at 380°K in 3975kg of water at 260°K. The temperature change is 0.2K on one tile only.

It's more than the theory would predict. But when you swap the two temperature, the result is spectacular. When I drop 1kg of water at 260°K in the same pool but at 380°K the temperature change to around 355°K, near one fifth of the temperature difference.

I have noticed similar effects - it seems to be a sort of cooling bug and people trying to make boilers should take note of this: Dropping water directly into your boiler basin without some degree of preheating is probably having an adverse effect on the heating performance of some boiler designs.

My own experience with this involved melting Ice into my water basin - this drastically lowered the temperature of the basin - after doing this twice- the water was very near freezing temperatures and the mass of ice was very small compared to the mass of water.

[Saturnus]

@Sevio I've added a feedback loop on the aquatuner and it works flawlessly so now the pump is only using half the power. I've also shortened the feed line to the boiler but that's just for my own comfort as I feel it reacts a bit faster.

The valve is at the red arrow and set to 5000g/s. The bridge is at the yellow arrow. Both can just be plugged in to your original design by deconstrucing only one single piece of pipe.

Output water temperature is now 51.2C, ie. down 2C, for some reason which I'm not entirely sure about.

Edit: Oh, and btw, you write there's a P-O2 siphon but it's actually not hooked up, and the atmoswitch for it is not set correctly either. It should be on if above 1600g 

[Sevio]

@Saturnus Clever change with the bridge at the yellow arrow! One point of attention with your change in plumbing for the boiler feed line: The liquid bridge is thermally connecting the boiler with the outside environment, so it has to be abyssalite to avoid heat losses.

I've been testing Cold mode today and while the tepidizer basin has worked very well in making Power mode more effective, in cold mode its job is to add extra heat so the cooling tank doesn't freeze over. But with the large amount of polluted water in its basement and the distance from the clean water dripping point, the temperature correction goes through pretty slowly so I've had to increase the thermo switch threshold from -2 C to 0 C.

I think the feedback loop would work out very well for running in power mode, but in cold mode I notice the polluted water in the Aquatuner output pipes is -7.7 C, so this would definitely kill the ability to run it near a threshold of 0 C. It would have to be increased to about 14 C to be able to use the feedback loop, making the cooling tank less useful as a heat sink for the base.

[Saturnus]

10 minutes ago, Sevio said:

@Saturnus Clever change with the bridge at the yellow arrow! One point of attention with your change in plumbing for the boiler feed line: The liquid bridge is thermally connecting the boiler with the outside environment, so it has to be abyssalite to avoid heat losses.

Thanks. I was aware of that. I use the colour mod so I can directly see what everything is made of so it's easy for me to notice any mistakes.

I must admit, I have not tested cold mode, nor do I really think it necessary. You really only save a small amount of power for a large drop in efficiency. For me it's maximum throughput per Watt that counts.

[Sevio]

8 minutes ago, Saturnus said:

I must admit, I have not tested cold mode, nor do I really think it necessary. You really only save a small amount of power for a large drop in efficiency. For me it's maximum throughput per Watt that counts.

Fair enough, for me personally I think when I have the power generation to run this, I probably don't need the full throughput so I would like to use it in cold mode to provide a heat sink for the base to make the aquatuner's work doubly useful.

Anyhow, I've implemented your feedback loop with one small change that should make it work for both cases: the valve has been moved to the power transformer area so the feedback mix can be adjusted by dupes as needed.

[Jackblac]

Why are you building the tiles on the outside of this as normal tiles? Dont you want to keep the heat/cold inside the system to preserve it?

[Saturnus]

15 minutes ago, Jackblac said:

Why are you building the tiles on the outside of this as normal tiles? Dont you want to keep the heat/cold inside the system to preserve it?

They're made of abyssalite (except those in the bottom cooler bulge). It really doesn't matter if abyssalite tiles are insulated or not, the heat conductivity is almost non-existent. And regular tiles have higher decor values, and take half the material to build than insulated tiles.

The only place where making somthing of abyssalite in insulated version makes sense is if you're building something where you're at the verge of phase change happening in the pipes, and you really really don't want anything to cool down or heat up by even a fraction of a degree. For example in feedback looped aquatuner liquid oxygen freezers where the intake is a couple of degrees lower than boiling temperature and the output is a couple of degrees higher than freezing temperature.

[Cilya]

1 hour ago, Jackblac said:

Why are you building the tiles on the outside of this as normal tiles? Dont you want to keep the heat/cold inside the system to preserve it?

Actually, I think the only one to use insulated tiles or pipe are those who want to show which parts needs to be insulated and built with abysallite and which parts needs to be conductive and use other materials. It's then much easier to spot a heat exchanger. I have to admit I use them also out of debug mode just for the immersive feeling.... but twice the cost in abyssalite is a bit tough for that.

[Sevio]

I updated the save game in the presentation post with the feedback loop and also updated the new stats for power consumption and output temperature. Power mode got a bit more expensive than before (1.7 to 2.08 kW) but that's explained by the increased throughput. The feedback loop helped save a decent bit of power though, I was able to lower the valve to 3000 g/s while running at maximum throughput in Power mode.

[Saturnus]

@Sevio I've managed to successfully compact the build a little so it's now 15 tiles tall, down from 17. Same 21 tile length. It looks a bit cleaner, and all control valves and switches are in the bottom left corner.

Currently I'm working on maybe getting some fail safe features built in so if the boiler is swarmed it shuts down feed etc but it's tricky to get the same performance.

 

[Sevio]

@Saturnus Nicely done compacting it! What are the performance problems you ran into when trying to make it fail safe?

[Saturnus]

Making the boiler wider to have a overflow valve makes the boiler a lot less efficient. And getting an inferred pressure sensor to operate a liquid filter by pass on the polluted water feed seems to delicate to get working properly so I might just drop that. I'll continue to reworking it as time permits.

[Sevio]

Well, in a survival game I'd probably just try to lower the throughput valve slightly to provide some extra margin of safety and equip it with large battery banks. That would give me enough time to shut down the valves if it runs out of power just after dupes go to sleep.

But I tend to run smaller colonies so I don't think I will need its maximum throughput.

[Saturnus]

@Sevio I did some minor clean ups and some minor tweaks. For one I remove the boiler temperature sensor and replaced it with the feed valve so there's less water, and a very short pipe from the feed pump. I also made a single power connection point and wrapped the power connection all the way around instead. Having the heavi-watt cable in the outside channels instead of through the heat exchanger also seems to improve performance slightly. I think it can go over 5000g/s safely now.

(see updated version below)

 

 

 

[Sevio]

@Saturnus Yeah in general less heat capacity in the boiler seems to improve it, other than the sculpture mass needed to help conduct heat. Is there any particular reason for the airlock doors in the cooling tank though? I thought granite tiles were shown to have better heat conduction.

The new basin for the aquatuner looks like very tempting, since I was not happy about the increase in heat capacity that having the liquid vent 2 tiles under the aquatuner caused. And the thermo switch I've been using in there doesn't seem necessary once the boiler has stabilized.

Seeing how you have a lot of heavy-watt wire in there now, I take it you have an aversion to using regular wire for large throughput power provider circuits that don't have consumers on the network.

[Saturnus]

16 minutes ago, Sevio said:

@Saturnus Yeah in general less heat capacity in the boiler seems to improve it, other than the sculpture mass needed to help conduct heat. Is there any particular reason for the airlock doors in the cooling tank though? I thought granite tiles were shown to have better heat conduction.

There's really almost too much cooling now due to surface area having almost doubled. At 5000g/s throughput output water from the aquatuner is 5C, and clean water output is 32C. But I also just like the look of it, and wanted to be sure the tiles could take the water pressure. I'm sure you could replace almost all with granite tiles if you really wanted to.

16 minutes ago, Sevio said:

Seeing how you have a lot of heavy-watt wire in there now, I take it you have an aversion to using regular wire for large throughput power provider circuits that don't have consumers on the network.

Yes. Definitely feel that's a bug that should be fixed asap.

[Saturnus]

Right so I played around with it a bit more.

It can run stable up to about 5450g/s throughput. Can't quite reach 5500g/s. The steam pressure is immense at this throughput. Almost 300kg of steam per tile. I've settled on a feedback loop at 2500g/s which is pretty safe at high throughput. 2222g/s is pushing realm of possibility at a return water temperature of -18.6C to -20.2C which is just a hair above freezing.

The usual screen shot, and the modded savefile. Credits to @Sevio for the original design.

I'll go back to my oxygen distiller design now.

First_Law_Take_Two.sav

[FR4NK]

I've been trying to build this in game and my PH2O in the aquatuner loop is getting too cold before the boiler gets hot enough to start producing steam...I've tried to pre heat the PH2O up to 90C before adding it to the system and it still gets too cold before I get steam. 

Next I might try putting a 2nd tepidizer in the cold PH2O section to keep the everything from freezing until the system stabilizes. 

Also, what about adding some nat gas generators to the system to power it and also add to the PH2O? Basically it will input nat gas and output power, water, and CO2. Would be possible to skim the CO2 to PH2O which would then convert back to clean H2O. I still need to run the math to see if that would work or else I can just freeze the CO2 to soild in a hydrogen bubbler to get rid of it.

[Saturnus]

It'd be immensely difficult to build something this compact in a play-through. It can be done but it'd take extra-ordinary levels of patience. It'd be especially hard to make the small vacuums that fills the heavi-watts wire runs-throughs, and all the pump spaces. And getting the water locks to form correctly. What I think can be gained is the general principle and then you can slowly work your way into something more compacts. But realistically it doesn't need to be this compact, or high performing.

To answer your questions. If you look at my modded design it has a much larger heat exchanger between the output clean water and tepidizer looped water. Note the tiles are granite to get maximum possible heat transfer. This is key to getting stable performance. If you use a feedback loop on the aquatuner, you might want to essentially bypass it by setting the control valve to maximum in order to get high temperatures running.

If you load my modded file, start by upping the feedback valve (the one on the left) to say 3000g/s and dial back the input valve (the one on the right) to 5000g/s. Note that the system is very slow responding although all measures has been taken to make it as fast as possible, so if there is a failure it takes several cycles before you can start building up the throughput to optimum performance again.