Liquid O2/H with inbuilt petro boiler

[Majestix]

I have not found a magma volcano on my map yet; and since I was about to start transitioning to LOX anyway, I wondered whether I could use the same aquatuners that would make the liquid gases to also boil petroleum for me. I have not seen this done, so before I ask my dupes to build it, I wanted to ask you knowledgeable folks to critique it/warn me of impending problems. Or maybe you will just tell me that I am wasting my time, since I will get more energy out of the cooling system by just slapping two steam turbines on top of the aquatuners - I did not do the math on this.

I just threw this together in a very scrappy way in sandbox, but the build is extremely simple:
 

Spoiler

Liquid overlay:

Spoiler

Automation:

Spoiler

The left thermosensor in the aquatuner room opens the door when the temperature goes above 404C, and opens the valve for fresh oil at the same time. It keeps the valve open for another 5 seconds after the door closes, so that some more heat is extracted from the outgoing petroleum; otherwise the incoming oil that has to wait just outside the boiling room gets too hot.

To prime the system initially, one can just fill up the boiler room with crude oil, about 2 tiles high. While new LOX/H is produced, the system seems to process fairly close to 10kg/s of crude oil, but of course there will be a fair bit of downtime whenever no new liquid gases are needed. I did not stress test it over 100s of cycles, but it seemed to have no problems starting up after a lull.

The heat exchanger can be quite inefficient at times. Sometimes the temperature of the outgoing petro is 120C, which is great, but sometimes it's above 200C. In the present form I am not sure how to make it more efficient without also making it more fragile, given the very uneven flow that is being processed. I guess the next step will be to try turning it into a waterfall exchanger and have temperature measurements at various intervals, where the oil gets the opportunity to bypass sections of the counterflowing petroleum if it gets too hot, so as to make it adaptable to different regimes of the aquatuners.

Any comments welcome, I am quite new to almost all aspects of this: gas liquification, petro boiling, and sandbox experiments!

Edit: this has got to be vastly more efficient than deleting the heat with steam turbines, even with a suboptimal heat exchange! I am powering the whole thing with petroleum generators, and once it has reached steady state, it produces about 3800kg more petroleum per cycle than it consumes.

[KittenIsAGeek]

I'm all for building something in a way that works for you.   I'm especially in favor of moving the heat from one place to where it needs to be in order to do work. In your case, you're moving the heat out of hydrogen in order to boil petroleum.

As for your outgoing temperature oscillations.. I think that has something to do with the stair-step design of your counter-flow heat exchanger.  There are a number of posts showing a more vertical design gives a better result because of the 'back-flow' of thermal heat along horizontal exchangers.  

In this thread, a lot of the conversation is based around heat exchanger design.  It might be useful and may give you more consistent results.

 

[Majestix]

I am frankly shocked how well this is working, given how simple it is:
 

Spoiler

Liquid overlay:

Spoiler

The pipe thermo-sensor inside the waterfall checks whether the crude oil is already too warm, and if so, then it redirects it via an insulated route. The whole thing is extremely robust, has no problems starting and stopping, and almost always delivers the oil at a fairly consistent 380-390C. With more intermediate temperature checks along the waterfall it could be dialled in extremely precisely, but it makes the waterfall look less pretty, so I probably won't be bothered. I will now build this in survival. Yay!

Edit: Sorry, I have just realised that the insulated route that bypasses sections of the waterfall will never have the opportunity to joint the main pipe. The simplest solution is to give it its own vent. That vent does not need any automation, it can just always be open.