Waterfall Petroleum Boiler

[mathmanican]

1 hour ago, Prince Mandor said:

Are there some solution for such situation?

A left descending step (45 degree angles) will eventually loose any mass above 400g on each step. Decrease the angle, and a tad more will remain but then temps equalize on the same level, working against you. You can add thermal buffers in each region to somewhat offset this, but upon restart, every setup will have some issues. If you have a large thermal buffer, then it takes longer for things to equalize. This is true for towers, stairs, and waterfalls. 

You can add extra thermal mass to a waterfall boiler (liquid vents closed via automation, horizontal bridges, etc.). However, you can also just restart the thing in increments (500g/s, then 10s later 1kg/s, then a bit later 2kg/s, etc).  You can also design a boiler plate at the top region that guarantees any incoming oil always becomes petro.  It takes almost no time for the system to reach equilibrium (as such, I don't recommend thermal buffering).  Just make sure your boiler can handle on/off situations (every boiler has to deal with it in some way), and it evens out extremely fast. I haven't played with them for hundreds of cycles, rather noticed the speed at which they reach equlibrium. If anyone with more experience wants to add comments, I'd love to hear opinions on the topic. 

Pros:

• Least material used of all boilers.
• Least space requirement
• Quickest equilibrium temp reached (start/stop possible with very little extra heat loss on boiler)
• Bead version eliminates ALL backflow conduction (please find another one that does this - huge efficiency boost)
• No need to dismantle an elaborate system if something goes wrong (you broke a pipe because it got too hot). Just build a ladder to the side for servicing. Done.
• Looks cool (waterfall) or cool and almost invisible (bead).

Cons:

• Build in vacuum?  Don't we build just about all of them in vacuum). 
• Requires knowing how to force a bead/waterfall pump. Guides are out there, or play around with ONI physics to discover them. 
• Start stop does not have large thermal buffer so temps don't maintain (but they also almost instantly reset to equilibrium). You can design in a thermal buffer if you want, but then material costs are increased for very little effort, and it takes long to reach equilibrium and change valve levels.  Probably better to just not. 
• You are limited to 10kg/s (so is every other piped version - if you want 90kg/s, then open flow is required - escher falls are amazing)

What have I missed?

 

[redxaxder]

Running tiles and bridges down the bead path and zigzagging the pipe would add a thermal buffer.

[mathmanican]

4 hours ago, redxaxder said:

Running tiles and bridges down the bead path and zigzagging the pipe would add a thermal buffer.

Absolutely you can, at a cost. First thing I did when playing with this was to add extra mass (liquid vents closed via automation, bridges, ladders, gas vents, and other stuff). However, this just causes the system to resist change (what a buffer does). One of the beautiful parts about this boiler is how rapidly it adjusts to new inputs levels.  You want to go up to 10kg/s. Fine.  Need to pull back or stop. Done. The system very rapidly adjusts. Just design the boiler at the hot plate to deal with the increase (put the thermal buffer at the hot plate). 

[wachunga]

29 minutes ago, mathmanican said:

One of the beautiful parts about this boiler is how rapidly it adjusts to new inputs levels.

Agreed. While there may be some case where high thermal mass in the actual exchanger is useful, I think the opposite is true generally speaking. Typically I run these at low flow rates for better efficiency and so that they can run for hundreds of cycles with any player intervention. If one did want to start and stop the flow, it wouldn't be particularly difficult to setup a bunch of shutoffs to do it automatically. A bit of trial and error would be needed to find the values, but it could be something like 500 g/s for 20 seconds then switch shutoffs to give 1 kg/s for 20 seconds then 2 kg/s and so on.

[mathmanican]

With a filter set at 5010g, you can use partial evaporation (liquid flaking) to meter and control your heat loss very easily. 

You have to prime it with hot enough crude to begin with (as you still have to pass some breaking phase up delta point), but once began, I can run 0C into the thing and loose only enough heat to bring petro up to boiling. I believe they used the child's SHC when computing the energy required, so this may actually use up a tad more energy, but you also don't have to deal with the instant 1.5C (multiplied by number of surfaces in contact) heat loss upon phase change. 

[Prince Mandor]

34 minutes ago, mathmanican said:

With a filter set at 5010g, you can use partial evaporation (liquid flaking) 

What happens to 10g of oil? Does it disappears?

[mathmanican]

46 minutes ago, Prince Mandor said:

What happens to 10g of oil?

Flaking only converts 5kg. But you need 5010g to enable it for liquids. With 10kg you would only convert half to petro and have a huge mess to deal with.

The 10g crude is lost. A reasonable loss for near perfect heat bleed.

Technically the 10g splits into several pieces that shoot out from the point of flaking in a direction that rotates in a 4cycle pattern. But since each bit is under 10g, the liquid disappears on the next tick.

[Prince Mandor]

36 minutes ago, mathmanican said:

The 10g crude is lost. A reasonable loss for near perfect heat bleed.

Ok

Another small point. Can you, please, put some useless ladder segments or disconnected pipes near long things? It is hard to measure without any reference to count tiles.

How long this waterfall? Is it optimal in length or just some arbitrary long waterfall exchanger?

[mathmanican]

13 minutes ago, Prince Mandor said:

How long this waterfall? Is it optimal in length or just some arbitrary long waterfall exchanger?

This one was just built arbitrarily long (I won't say it's optimal at all). The point was to show that you can combine this with the updated liquid flaking mechanic. The longer the waterfall, the cooler the petro will be as it leaves, and the less heat you loose at the top.  However, the benefits offer diminishing returns. If you have a thermium aquatuner, then it's trivial to get high enough temps and maintain them.  If you are working with magma or other things, then you may want a longer waterfall. 

Here is a picture with a marker every 5 tiles. 

Spoiler

 

[Prince Mandor]

1 hour ago, mathmanican said:

Here is a picture with a marker every 5 tiles. 

Thank you very much!

:))) I just imagined some base, where this waterfall also cross-used as multilevel airlock :)))

[blakemw]

I built one of these in survival: 

I look forward to more petroleum Boogers adorning bases. 

What I especially liked about the Waterfall besides all its technical merits, is that it tolerates busted pipes so well, the crude oil would simply fall down into the pit where it can be swept. No flow blockage, no fuss. A great design for survival.

The only thing I've found a little tricky with stopping and restarting is that sometimes the molten lead tile gets displaced. Fortunately the lead tile normally just undergoes binary division so it could experience a bunch of displacement events before being entirely removed. When it finally disappears I replace it by building a lead conductive wire (bridge) to melt into a new lead tile, so it's not a big deal, but it's an obstacle to a boiler that can be stopped and restarted completely freely.