40kg/s Regiolith Melter

[Angpaur]

If Saturnus's values of temperature after melting comes from in-game testing then there is something strange. 

If melting 200kg@400C of regolith in 3000kg@1500C of magma gives final temperature of 1481.78C then it lacks 3.75C, because in theory it should be 1485.53C.

It can be game real simulation thing again or fault of testing method. I will do some testing as well later today. It is interesting case  

[Grimgaw]

@Angpaur Differences less that 5deg could be attributed to phase change hysteresis in ONI.

@Saturnus OK, sorry you're right in that heat needed to maintain the temperature of whole system is constant for set amount of regolith. What you're not seeing is that the mass of system increases with each addition. If you remove that extra mass and use it's heat to preheat the incoming regolith you'll need to add less heat next time. And so on.

[Saturnus]

47 minutes ago, Angpaur said:

If Saturnus's values of temperature after melting comes from in-game testing then there is something strange. 

If melting 200kg@400C of regolith in 3000kg@1500C of magma gives final temperature of 1481.78C then it lacks 3.75C, because in theory it should be 1485.53C.

It can be game real simulation thing again or fault of testing method. I will do some testing as well later today. It is interesting case  

Please do. I may have seen a downward trend now but only if we have 10000kg magma per tile and add 200kg regolith. Then it seems to be 290.7kDTU per kg added needed.

There's several obvious problems with that. You can't really have 10000kg magma per tile under normal circumstances while at the same time having to have a method of adding more heat to the system as that would normally need to be done via adding more magma which is difficult in an overpressurized magma system.

The other one is that it's very very far from the predicted values.

[Saturnus]

I found a giant problem.

Up until now I've melted regolith spawned in.

However, if I have a one tile wide chamber of magma and dump in dug out regolith the regolith doubles in amount. 

For example here I've dumped in 20kg of dug out regolith into 600kg magma. But the result is 640kg of magma.

This obviously completely skews the results. So my current assumption is that if anyone have gotten such a system to work in any way in the past, it's due to this doubling of mass effect. So they've, probably unknowingly, used the liquid multiplier exploit.

 

[JohnFrancis]

2 hours ago, Saturnus said:

Aren't you talking to the wrong guy here?

oops, sorry about that, it was early and I had not had breakfast yet.

At the people arguing that this is possible, please build a prototype or a proof of concept design and share the save file. Without that this is all just numbers that may or may not work in game.

[Grimgaw]

10 minutes ago, JohnFrancis said:

At the people arguing that this is possible, please build a prototype or a proof of concept design and share the save file.

I think I'm done beating a dead horse here. @Saturnus I'm baffled you're not getting this. It's the same reason your old pH2O boiler had massive cooling effect. (That 6SHC of pH20 vs 4SHC of clean water).

[JohnFrancis]

@Grimgaw So regolith melting is possible I did not think that was the debate. I though the debate was on if it can be made self sustaining where the change from regolith to magma gives you a material with more thermal capacity so effectively you created free heat. That can then be used to help melt more regolith.

[caffeinated21]

26 minutes ago, Saturnus said:

I found a giant problem.

Up until now I've melted regolith spawned in.

However, if I have a one tile wide chamber of magma and dump in dug out regolith the regolith doubles in amount. 

For example here I've dumped in 20kg of dug out regolith into 600kg magma. But the result is 640kg of magma.

This obviously completely skews the results. So my current assumption is that if anyone have gotten such a system to work in any way in the past, it's due to this doubling of mass effect. So they've, probably unknowingly, used the liquid multiplier exploit.

 

Saturus is correct; I think we are seeing a doubling coming off the rails. That explains why i'm seeing a build up of rock on the cooling plate. Explains why I was seeing thermal runaway of the cooling plate, and had to add an intermittent thermal dump to earlier in the heat exchangers. I guess I"m generating 80kg/s of rock now...

Regarding the discussion of input heat required: this design preheats the Regolith to nearly melt point (1403C from the previous post). A bigger heat exchanger might be able to get you a little bit hotter. The last few degrees to melt point are taken from existing magma (like from a volcano), or another heat source (metal refinery, glass forge, etc.) When I ran this thing for tens of cycles it averaged to using <150kg / cycle of an external magma source. 

On a related note, the heat generated by this thing is insane. Output rock (after preheating regolith) is 1200C. If my math is right I could run 140 steam turbines with this: 1100K * 80 kg/s * 1 DTU/g / 75K* / 2 kg/s / 4.179 DTU/g =~ 140. Good for a cool 112000watts. 

Other fun facts: you could support 343 stone hatches with this thing. 

[Angpaur]

So I tested it a bit now and I found interesting thing happening. Please take a look at below screenshots:

One tick before melting:

Game tick after melting:

Before melting the regolith there is 1810kg of magma at 1479.6C

After regolith melts temperature drops to 1472.6

A sudden drop of 7C. This is why it doesn't work as in my theorethical calculations.

39 minutes ago, JohnFrancis said:

I though the debate was on if it can be made self sustaining

It is not possible to make it self sustaining. We established it at the beginning of the whole discussion.

But what me and Grimgaw is saying that each melting should decrease a bit energy needed to heat up magma back to 1500C.

But there is something strange happening that when regolith melts magma temperature drops by 7C and this prevents this from working as expected.

[hackcasual]

They way I've explained this before is heat chicanery due to SHC adds (or removes) heat, but not temperature.

To address the original design, I've found a small (maybe 3 stage) counterflow heat exchanger helps quite a bit with efficiency. I'm not really sure if there's an ideal design, since the heat exchange of conveyed items inside solid blocks doesn't seem to make much sense to me, and you're exchanging heat between things with a 5x difference in heat capacity.

It's also worth pre-heating the input regolith as much as possible. Even using an aquatuner, you'll get a net win, since the heat you put in, gets multiplied by 5x, and can then get extracted by a steam generator. 

[Saturnus]

12 hours ago, Angpaur said:

But what me and Grimgaw is saying that each melting should decrease a bit energy needed to heat up magma back to 1500C.

Should be noted that I don't disagree that it should but observation of how it actually works in the game does not support that assumption.

[Yunru]

13 hours ago, Saturnus said:

Should be noted that I don't disagree that it should but observation of how it actually works in the game does not support that assumption.

But... people have built it?

[Saturnus]

20 minutes ago, Yunru said:

But... people have built it?

See above.

On 16/04/2019 at 6:11 PM, Saturnus said:

So they've, probably unknowingly, used the liquid multiplier exploit.

 

[caffeinated21]

Just chiming in here that the melted regiolith doesn’t get mixed in with the heat source regiolith. As soon as it melts, it drops down to the solidification plate. Why heat something up farther than you have to? 

[Saturnus]

2 minutes ago, caffeinated21 said:

Just chiming in here that the melted regiolith doesn’t get mixed in with the heat source regiolith. As soon as it melts, it drops down to the solidification plate. Why heat something up farther than you have to? 

Absolutely. However, we needed to dump it in for testing so we could read the heat exchange from observations. Had we not done it, we probably also wouldn't have noticed that the regolith doubles in mass when melted, and thus throwing a spanner in the works.

[caffeinated21]

I’d like to think i’d have eventually wondered where those tons of extra rock were coming from...

[MustardWarrior]

On 4/16/2019 at 9:00 AM, JohnFrancis said:

I have not played around with regolith melting but I have played with petroleum boiling and I can tell you the numbers the game gives you and the reality don't match when you start involving state change. If you believe it's possible then try building it, but as has been said before this game is a harsh mistress who laughs when you try and use math on her.

For an example of one quirk try making a heat exchange stairs that drops one tile for every three across. Make a left facing one and a right facing one. Both of them will output at very different temps.

I think the math works, specific heat capacity calculations are pretty reliable in my experimentation although state changes can screw things up on the margins, I've looked into this and even wrote c++ programs that have done accurate thermal exchange rate calculations for very simple tile to tile heat exchange examples.  

Here is a design that I threw together haphazardly that could likely be tuned to working.  An issue you run into is cooling the thermium auto sweeper.  you'd need to make a cooling plate and run the magma room in a vacuum and then build something like a counterflow heat exchanger.  Since I just threw it together it has way too many temp shift plates you can space them out just make sure there is at least one overlapping tile in the 3x3 area.

 

 

Could you do this?  Yes crude oil has like 8.5 times as much heat capacity as regoltih and you have people have said that you get ~ 20000 kgs of regolith per cycle on average coming in from space.  So 33.3kg/s should be something you can do with volcano especially considering the igneous rock flowing down the counterflow heat exchanger will has 5x the heat capacity of the regolith going in. 

The issue is that why would you do this? You probably get enough igneous from the volcano to provide all of your late game coal needs (just the kiln).  I calculated it out and if you converted 20,000 kgs of regolith per cycle into igneous, fed it to your stone hatches and then burned all of that coal you would get around 400kg of carbon dioxide per cycle which from about 20 slicksters under ideal conditions would net you 200kg of petroleum. per cycle which would net you 200*(750/1750) = ~86kg of polluted water per cycle.  Or about 127 grams of oxygen per second.  Even when it's all up and running, I don't see how this is going to not take at least one duplicant's worth of labor to keep all of the hatches and slicksters and everything fed I suppose maybe a ton of unmaintained stone hatches in a small room or something.  This seems like a complete waste of time to me.

[caffeinated21]

Dude- I did do this. This is a thread of the design that already does this. 

Is it pointless? Isn't everything in this game pointless? Why? Because you can? Essentially infinite power? Infinite meat? Shaking your fist at the comet gods? 

(grossly oversized) heat exchange pic here:

Save file attached if anyone wants to poke around. You could add another two conveyers for the doubled mass of the rock, but I don't really like relying on that bug...

Power Sandbox.sav

[MustardWarrior]

Yeah I mean it is very satisfying getting a self powered petroleum boiler up and running I imagine the feeling would be similar with this.  I guess it's a decent monument to OCD.

[JohnFrancis]

 

1 hour ago, MustardWarrior said:

I think the math works, specific heat capacity calculations are pretty reliable in my experimentation although state changes can screw things up on the margins.

That was pretty much my two cents on the issue, I never got into regolith melting. I approach the game from a straight vanilla view point. What builds and designs can get me the most benefit in survival play. By the time I'm able to farm regolith all major issues are solved so dumping it into shove voles for mass meat production was a simple and beneficial solution. It's the same reason I never got into sour gas boiling or farming sleet wheat other options are more time/cost effective.

But I can appreciate a monstrous perversion of the laws of dupe physics.

[Dead Squirrel]

Ok, I did a test. I get that there are some weird side things going on (I didn't run into the duplication bug). But there is a net positive energy in heat (DTU) when you melt regolith. Importantly, I think many tests are not measuring the temperature and volume of every tile in a liquid lake, and that definitely throws off calculations. 

 

I hope this doesn't stir the pot ...

[caffeinated21]

200kg * 0.8 change in specific heat * 1399.9 = 223984 kDTU. Remaining 1640 feels like a rounding error?

[Dead Squirrel]

2 minutes ago, caffeinated21 said:

200kg * 0.8 change in specific heat * 1399.9 = 223984 kDTU. Remaining 1640 feels like a rounding error?

 

Yeah, maybe. Also, the regolith does have to heat up a little to melt - about 9°C to 19°C. Don't phase changes happen just passed the expectation? So I guessed the difference was the energy lost raising the last bit of temperature, or else some quirk in how ONI calculates exact phase change values.

 

(For reference, 1409.9°C regolith has 56 396 kDTU, and should create 225 584 kDTU when swapped for magma. That's a lot closer to the 225 624 I calculated.)

[Arkaon]

When i was reading this post i remember when mom told me, "dont play video games that would make you dumb", yeah right

[Neotuck]

6 minutes ago, Arkaon said:

When i was reading this post i remember when mom told me, "dont play video games that would make you dumb", yeah right

and vaccines make you autistic...