Help me check my liquid oxygen heat transfer math

[Supraluminal]

I recently went to a lot of trouble to start liquefying oxygen, thinking I could use it as a coolant for my base. I was going to run liquid pipes through my floors and use automation to run LOX through them as needed.

On my first trial, I found that the LOX warmed up really quickly and the (granite) pipes started breaking within just a few tiles of where they branched off from the main abyssalite supply line. What's going on here? I ran the math before starting out on this project and I thought I'd pretty much determined that LOX shouldn't absorb heat that fast.

LOX shows a specifric heat of 1(J/g)K and a thermal conductivity of 2(W/m)K. So, for a 10kg packet of LOX, it should take 10kJ to heat it 1K. There's no mass term in thermal conductivity, so assuming a 250K degree difference between the 10kg packet of LOX and the rest of the cell, it should still max out at 500W of heat transfer, right? And that gets us 10kJ / 500W = 20 seconds to heat a packet by 1 degree.

Instead, from what I can tell it's heating at more like 2.5K per second while it's running through granite pipes, about 50 times faster than I expected. Did I mess up my math that badly or is there something else I'm missing? Is thermal conductivity giving W per g of material or something? Do pipe contents not have a 1m/tile surface area?

[Kasuha]

Each pipe segment has 5 kg (of e.g. granite) for heat transfer purposes. When you send liquid oxygen at -190 C to pipe at 30 C, the temperature difference is 220 C. Now, I don't know exactly how the heat conductivity values are applied in the game (since there's distinct lack of "distance" measure between things inside the pipe) but the heat exchange between the pipe contents and the liquid oxygen is quite rapid at such temperature difference. And for each 1 C the pipe segment cools down, the 10 kg packet of oxygen heats up by 0.4 C. Plus there is of course also rapid heat echange between the pipe and the air around it, which is trying to keep the pipe at the ambient temperature, so the temperature difference between the pipe and its contents doesn't change much before the liquid oxygen starts evaporating.

Don't use liquid oxygen for cooling. Use safe media that don't change state when getting close to the room temperature. Hydrogen and polluted water are probably best choices as they have the greatest heat capacity in their respective categories, some people however prefer crude oil as it can be cooled to lower temperatures while staying liquid.

[Rotintin]

I always use liquid o2 to cool my base by running it through abyssalite pipes and dripping it.

 You can use temp shift plates to disperse it or it can be used in this build: 

 

You'll notice Kasusha helped with the math there too. 

[Supraluminal]

Thanks. I totally get that one packet of LOX wouldn't be able to absorb all that much heat from the environment before evaporating. Like any coolant system, I was going to push a large volume over time. I just thought, based on the math, that I could do reasonably long stretches of uninsulated pipe with it.

I know hydrogen is a safe and generally decent choice for a coolant, but I figured that the higher thermal conductivity of LOX combined with higher total heat capacity per 10kg packet vs.1kg of hydrogen would make it even better... as long as I could keep it from boiling in the pipes.

Now that I see how quickly that happens, I guess I can either abandon the LOX plan and use hydrogen, or experiment with very short heat exchanger runs for LOX. Which seems awkward, so I dunno.

I guess I can keep the liquifier setup on hand purely as a storage device for surplus oxygen. Is there anything else fun or interesting to do with liquid oxygen?

@Rotintin, I suppose you're right that it can be just dripped out into the base to cool things off since it'll harmlessly evaporate. It bothers my sense of tidiness though! And it wouldn't be suitable for all applications, obviously including cooling areas where you don't want gaseous oxygen floating around.

[Rotintin]

@Supraluminal You might be able to make a radiator by dripping it on to water or even crude oil in a tank, then pumping out the cold oxygen that evaporates. I've never made that, it could be a cool build. Also, mixing the excess cold oxygen oxygen with oxygen from electros is a great way to cool it.

[Yoma_Nosme]

@Supraluminal

You could try to cool your base in segments. Open new segment when first one reached temperature so the transfer of heat won't break pipes. And so on.

Maybe automation valves with temp sensors?

I don't think liquid oxygen is very good at cooling .thermal conductivity of 1 and heat capacity of 2. Try liquid co2 (or liquid Nat gas)or just oil or petroleum 

Edit. Actually Nat gas and liquid co2 are even worse. Sorry

[Supraluminal]

30 minutes ago, Yoma_Nosme said:

@Supraluminal

You could try to cool your base in segments. Open new segment when first one reached temperature so the transfer of heat won't break pipes. And so on.

Maybe automation valves with temp sensors?

I don't think liquid oxygen is very good at cooling .thermal conductivity of 1 and heat capacity of 2. Try liquid co2 (or liquid Nat gas)or just oil or petroleum 

Edit. Actually Nat gas and liquid co2 are even worse. Sorry

You got the conductivity and heat capacity of LOX backwards, it's conductivity 2 and heat capacity 1. That's pretty comparable to the other liquids with low freeze points. (Although I don't actually know LNG's specs, it's not on the wiki and I haven't made any yet.)

Oil and petroleum are fine, but I preferred LOX for various reasons including infrastructure requirements and the extremely low freeze point (which affects what else you can do with it, and plays into thermal conductivity).

As far as the system design goes, I was doing it in segments controlled by valves and sensors. It just turned out that my segments were too long - the pipes were breaking after only 5 or 6 tiles.

[Supraluminal]

Small update, I tried cutting back the heat exchange section of my test run to just 6 granite pipe tiles and it still didn't work, but it was behaving oddly. By hovering over the LOX moving through the pipes I could verify that the packets were arriving adequately cold and that they didn't appear to be heating past the evaporation point. Occasionally, though, a packet would disappear and break the piece of pipe it was in, despite being just as cold as the packets before and after it in line (which would make it through the whole thing just fine).

It looks as though some packets were randomly evaporating, maybe due to an incorrect spike of heat transfer. I think I've seen mention of that before on here? It was frustrating, anyway. I had hoped I'd be able to salvage my project this way but I don't think it's going to happen.