How to get large amounts of power in rocketry upgrade?

[Carnis]

On 23.9.2018 at 8:41 PM, Carnis said:

@onlineous

I meant this build:

1 heat exchanger oil vs 360-420kg chunks of magma

2 conveyor loader full of 700-1100 igneous

3 using staged heat recapture to cool The 1100 igneous to 400C, then conveying this downstream from 380 pool 2, to eventually 110C.

[The Flying Fox]

8 hours ago, wachunga said:

Another thing worth mentioning is that the more drops or levels or stages the better. Along a horizontal run, each tile of petroleum touches others and will transfer heat the wrong way. From a certain point of view, the "ideal" exchanger is very tall and just one tile wide. Each petroleum tile is isolated and there is no counter productive heat flow between tiles. Of course that's not practicable to build. So my rule of thumb is to make the exchanger as tall and as with as many levels as possible. Then widen it to achieve the necessary efficiency. A 1kg/s exchanger can be much smaller than a 10 kg/s one.

I like to make each level dupe accessible to ease construction. This does mean the exchanger is less space efficient than strictly necessary, but I think it's a worthwhile trade off. I also make the drops 2 tiles wide in an effort to prevent "sticking" that sometimes happens. Droplets effectively don't exist while a slug of petroleum sticking to the wall allows heat transfer the same way as a horizontal level does, ruining the efficiency of the drop. This may no longer be a problem, but it was in the past so it's become a habit of mine. My builds end up something like:

 

That works reasonably well up to 4kg/s (preheating to within ~25C). More than that and you should really widen it some as you start needing to draw too much heat out of your source. Also note that the efficiency of the exchanger seems to increase when you play at a higher speed. At least for me, 3x speed gets the preheat temperature about 5C higher than at 1x speed.

This is exactly pretty much what I found out and came up with in my own designs.  Long flat pools of oil and petroleum, over-time, average out their temperatures across them and narrow the thermal gradient, thus reducing the effectiveness of the thermal exchanger.  This also applies to gas ones, although not as much, but it is still noticeable.  So, vertical stages for gas thermal exchangers are better then horizontal stages.

 

The major reason for this has more to do with how ONI moves heat around.  It highly prefers moving heat upward and cold downward while thermal transfer side to side is more average.  So, having the cold source on the top and heat source on the bottom just ends up averaging out the temperatures into a much smaller gradient.  But, putting the cold source on the bottom and the heat source at the top more favors a thermal gradient because ONI doesn't want to move the heat in that direction.

 

I did notice in past builds having petroleum sticking to the walls as well, but that seemed to only happen in vacuums and when the stages were only 1 tile tall. (Debug builds, basically)  If I filled the gaps with a tiny bit of chlorine, it would prevent the petroleum from sticking to the walls.  That being said, I haven't had that issue with 2 tile tall Dupe-serviceable builds.  I know the 'stickyness' that's seen in my build is intentional though for make-shift liquid air-locks.  I wanted to make absolutely sure that my vacuum would be maintained in my heat exchanger even if there was any accidents. 

 

2 hours ago, onlineous said:

I can't completely figure out how you heat the petroleum reservoir at the top to the right temperature without overcooking it.

Chlorine temperature sensor:

This is the basic principle of it.  A tiny amount of chlorine passes through a single insulated non-abyssalite gas pipe through the thing your trying to measure.  Due to the pipe's nature and chlorine having basically nothing for conductivity, it only rises to about 60% of the temperature of the heat source.  The first gram takes the most heat away while each gram added takes less and less then the last.  Since it's chlorine, it takes a very tiny amount of heat away from the heat source.  In the below example, the chlorine passing the pipe temperature sensor is now within it's use-able range.  The chlorine is then cooled back down by a stable cold source and loops again.

Here is that setup in my current machine:

 

The only major difference is that I pulse the packets of chlorine once every 5 seconds to reduce the amount of heat taken. (Not that it's that much, but why not?  I don't need the reading to be updated every second anyway, it could even be less.) So, with this setup, I can keep the upper petroleum tank around ~500C easily although I could set it lower and the oil refinery would still work just fine.  When the temperature gets too low, it tells automation that controls the doors just below the gold volcano to drop more gold into dropper setup.  A small amount of liquid gold drops onto the horizontal door and hydro-sensor and cools into a solid chunk.  This then starts a filter gate timer that's connected to a memory toggle which gives the gold chunk more time to cool before the door is opened and the cooled gold chunk drops into the hot petroleum tank.  A sweeper then corner grabs the chunk out of the tank and sends the spent chunk away.

[wachunga]

1 hour ago, The Flying Fox said:

This is exactly pretty much what I found out and came up with in my own designs.

Convergent evolution is quite eerie at times. This and some other of your complex builds are crazy similar to stuff I do. But with enough testing and iterations, it's not crazy at all. Like my suggestion to neotuck which he expanded into his SG refinement thread is based in part off the gas heat movement you are describing. Pretty neat.

1 hour ago, The Flying Fox said:

If I filled the gaps with a tiny bit of chlorine, it would prevent the petroleum from sticking to the walls.

Very interesting, I never did do enough testing to figure out what causes it to happen or not happen.

 

And for anyone who wants to tinker with the build I showed, I forgot to add the save to my earlier post.

Crude to Petro.sav

 

[The Flying Fox]

10 minutes ago, wachunga said:

Convergent evolution is quite eerie at times. This and some other of your complex builds are crazy similar to stuff I do. But with enough testing and iterations, it's not crazy at all. Like my suggestion to neotuck which he expanded into his SG refinement thread is based in part off the gas heat movement you are describing. Pretty neat.

Very interesting, I never did do enough testing to figure out what causes it to happen or not happen.

 

And for anyone who wants to tinker with the build I showed, I forgot to add the save to my earlier post.

Crude to Petro.sav

 

There is a method to the madness, as they say!   I also noted your lack of ladders on, at least, the bottom rungs.  Stupid heat stealing ladders!  People may not realize it, but abyssalite ladders can slowly leach an enormous amount of heat out of systems like these over-time. 

[StarSquid]

On 9/29/2018 at 6:01 AM, Carnis said:

If they eat cooked foods they generate almost no coal.

But If you Feed them dirt (sages), then the conversion rate is 100%, they generate More coal.

Problem is, dirt is better For glossy dreckos & other organics are really expensive to mass produce.

Why would you give them cooked foods instead of raw food?

Also, of course, rock is renewable if you have a volcano, should it really come to that.

[Icywolf]

@StarSquid more calories compared to raw food. Also Pickled Meal does not rot as fast