Practical Volcano Magma Pump

[MustardWarrior]

So I've been testing out magma pumps for volcanoes for the purposes of making a petroleum boiler.   So it turns out that if you can pump magma, then building a petroleum boiler is pretty simple, you just run the magma through some radiant pipe in a hot petroleum pool and then connect that to the steel door heat regulator and build your counterflow heat exchanger like any other build.

I understood the cross mechanic how pumping magma works in principle but it seemed really finicky in particular you run into problems when the liquid that is being used to activate the pump hits the magma as often happens if the pump experiences a power outage or it's pipe gets blocked you often end up with overheating problems.

So basically what I ended up doing was putting the pump in a small room only exposed to one magma tile, then putting some metal tiles on the side and a pool of petroleum next to it so that it cools the room, and also I put in a diamond tempshift plate too.   Basically if the pump stops for whatever reason and you end up with sour gas, the sour gas stays cool enough that nothing overheats and just for the sake of insulation I pump it out, since it's really easy to pump such a small room out.

This was all done on a test map and it works, if I deactivate the pump it won't overheat.  I know there are other designs that involve dropping magma and using a robo miner, I prefer using this method for a few reasons.

1.   This design is more flexible, the magma dropper is a very calibrated design.  I tend to avoid pre planned designs, petroleum boilers are one of the most satisfying things to build in ONI and I feel like using an existing design is less satisfying than building a new one by intuition, when you get a different design to work, that is very satisfying so every time I play I use a different design.

2.   This is more modular. namely because you can pipe around the liquid magma to wherever you want.  Or you can make the hot petroleum pool into a column that connects to multiple boilers and basically combine the heat of all of your volcanos, even the metal ones and use all of your heat and build multiple boilers.

3.  You get a bit of power from piping it into a steam turbine and a bit more igneous rock from this method.

 

Anyway in the screenshot, the magma gets pumped, then sent to the vacuum storage tanks and if those are full it gets sent back around to the magma pit to avoid it breaking the pipes.  From the storage tanks it goes through the hot petroleum box and then into the steam turbine box and then it gets vented and turns solid.  Obviously this involves using the 1kg/s trick that prevents it from hardening in the pipes.   I find that 100g/s works well for the valve that regulates how much petroleum is dropped on the magma pump.  No space materials, just steel, obsidian, aluminum is fine.  The pump inside of the petroleum is just to start the process and the thermo regulator is just for the steam turbine and the petroleum box that cools the magma pump.

[Craigjw]

Good.  However...

People should really stop dripping oil into their refinery conversion pools.

Better:

 

The vent blocks when 1000kg of oil is present in the conversion chamber, this negates ALL oil from ever escaping into the counter flow.  You need 3 tiles in height for this.

The drop in 1 tile into the counter flow helps prevent the conversion chamber heat from transmitting into the counter flow, this would ideally be 2 tiles high, but I ran out of space.

Best:

The input has a temp sensor on it, when the petrol is above 450, it switches flow to the flow valve, restricting it's flow to 3kg/s.  Prevents overheat and broken pipes at the head of the oil counter flow.

On the subject of conversion rates, I run this on a refinery, which turns out to be really impractical for steal production, as I'm currently only able to make about 1 to 2 steel bars per cycle running at a full 10kg/s of input oil.

On another note, your number of posts is mustard.

[biopon]

7 minutes ago, Craigjw said:

The vent blocks when 10kg of oil is present in the conversion chamber, this negates ALL oil from ever escaping into the counter flow.  You need 3 tiles in height for this.

Did you mean 1000kg of oil? I'm not sure how it would stop at 10kg.

10 minutes ago, Craigjw said:

The drop in 1 tile into the counter flow helps prevent the conversion chamber heat from transmitting into the counter flow, this would ideally be 2 tiles high, but I ran out of space.

1 tile drop should be fine, you have vacuum here:

 

[Craigjw]

You are correct, it's 1000kg.  Confusing pipe contents to tile capacity.  It caps at a full tile, so it's an intrinsically better design. 

The materials view for liquids would suggest that it isn't a vacuum.  However, when it's not active and no input heat, this is a vacuum.

When I built an earlier version in sandbox, a 2 tile drop does not show yellow.

[biopon]

1 minute ago, Craigjw said:

it isn't a vacuum.

That's surprising, just how long is that horizontal flow?

[Craigjw]

That might be the reason, it's 12 tiles long before the drop.  What's the max length of petrol creep on a surface?

[biopon]

12 tiles isn't too wide. And I was wrong, it's not a permanent vacuum with just a 1 tile drop. It alternates between this:

And this:

Cool tips, both of them, much appreciated.

[Craigjw]

Steel production was always a problem, as it still overheated the head of the counter flow oil pipe, the drop helped alleviate this.

The metal & heat sinks at the head of the counter flow is probably not needed, I left it from an early build.  I was trying to collect cooled petrol that was below 395 and send this to the metal tiles to temper the head of the counter flow oil pipe as steel production still caused overheat problems, however this always resulted in all petrol collecting in this tank, leaving none for the refining process.  The flow valve to the conversion chamber solved this completely.

My last tip is to loop the petrol and mix it with fresh hot petrol from where ever, as this will average out the temperature, which also helps preventing overheat of the oil pipe.  The valve at the bottom is just to switch off the system either manually, or if there is no oil being input, there's an element sensor at the start of the counter flow.

You can probably get away with less tanks, however, 2 tanks is minimum, one to mix and one to store used petrol, as it doesn't work so well if the flow stops.

We are getting off track here to be fair.

Not wanting to destroy the OP's OP post count, can you include a SS of your piping and automation please.

I guess that you decided to use a thermo regulator, as there was quite a lot of liquid piping and gas pipe avoids pipe overcrowding, allowing the bonus buff to pipe reproduction be applied.

[biopon]

Oil business is tricky with high temp deltas. If you're happy with the valve it's all good, but two options:

You could use the steel coolant to directly heat an intermediary, like a 3x6 block with water (-> steam) in it, and leech heat from this with a door connecting to your conversion chamber. I use magma rocks like this and open the door above 404 degrees, it works really well.

Alternatively, you could have some petrol in a reservoir on a constant loop just to measure its temperature; if it's below what you want it to be, use automation to enable the refinery so a dupe comes and makes a batch of steel and warms it up a bit. Adding ~90mDTU to 400kg petrol is a lot but if you average it into a full reservoir it's only about 10 degrees.  

[Craigjw]

The heating loop effectively does this, if it's below 405, use it to fill the refinery, which becomes enabled by default when it fills, otherwise, mix it back into the supply tank along with the refinery liquid.  Because it's too slow at cooling the petrol from the refinery, the heating chamber supply tank remains full, allowing averaging of the petrol temperatures.

[Lancar]

In all the large amounts of petroleum boilers on this forum, I've always noted that the ledges on the oil/petroleum heat exchanger after the conversion chamber are always some sort of insulator.
Why is that? I would think metal tiles with radiant pipes inside would achieve the same and wouldn't the heat equalize around the middle of the construction regardless? Not to mention that with more surface-to-liquid area, less space would be needed to achieve equilibrium. Especially if you throw in some tempshift plates.

Am I missing something fundamental here?

[Craigjw]

It's because you don't want the heat to travel along or between the ledges.  Insulators can be used instead of airflow tiles, but airflow tiles don't conduct any heat at all.  Nakomoro did state in a different post that it is feasible to place tempshift plates along the ledges so long as they have a 2 space gap between each plate.

If you have temperature creep along or between the ledges, it averages out the temperature more and the temperature differential is less, making cooling the petrol and heating the oil more difficult.

[nakomaru]

Yes - any heat transfer between the same medium is an inefficiency, so metal tiles and tempshifts usually work against you.

[Lancar]

I still don't get it.

Isn't the goal of the heat exchanger to make the incoming oil and outgoing petrol the same temperature? Why not achieve this faster with a shorter & more conductive exchanger?

[Angpaur]

44 minutes ago, Lancar said:

I still don't get it.

Isn't the goal of the heat exchanger to make the incoming oil and outgoing petrol the same temperature? 

Yes. And if you use metal tiles instead of some insulator how you want to achieve that goal?

[nakomaru]

1 hour ago, Lancar said:

I still don't get it.

Isn't the goal of the heat exchanger to make the incoming oil and outgoing petrol the same temperature? Why not achieve this faster with a shorter & more conductive exchanger?

A simple exchanger with high conductivity everywhere would take e.g. 0C and 100C water inputs and make both 50C outputs.

With counterflow you can do much better by only letting the hot hot stuff touch the hot cold stuff, etc. Taking the temp at a few locations of transfer might look like this.

• 100C hot stuff with 99C cold stuff.
• 60C hot stuff with 59C cold stuff.
• 1C hot stuff with 0C cold stuff.

When you add metal tiles and tempshifts, generally you are adding these types of heat transfers to the system:

• 65C cold stuff with 70C cold stuff
• 45C hot stuff with 50C hot stuff.

These only reduce the efficiency. You want as much of the top stuff and as little of the bottom stuff.

[gaucho_tche]

On 10/27/2019 at 4:50 AM, Craigjw said:

Good.  However...

People should really stop dripping oil into their refinery conversion pools.

Better:

 

The vent blocks when 1000kg of oil is present in the conversion chamber, this negates ALL oil from ever escaping into the counter flow.  You need 3 tiles in height for this.

The drop in 1 tile into the counter flow helps prevent the conversion chamber heat from transmitting into the counter flow, this would ideally be 2 tiles high, but I ran out of space.

Best:

The input has a temp sensor on it, when the petrol is above 450, it switches flow to the flow valve, restricting it's flow to 3kg/s.  Prevents overheat and broken pipes at the head of the oil counter flow.

On the subject of conversion rates, I run this on a refinery, which turns out to be really impractical for steal production, as I'm currently only able to make about 1 to 2 steel bars per cycle running at a full 10kg/s of input oil.

On another note, your number of posts is mustard.

Craig can you post pictures of your full design and/or a save file? I'm interested in a petroleum boiler

[Lancar]

23 hours ago, nakomaru said:

A simple exchanger with high conductivity everywhere would take e.g. 0C and 100C water inputs and make both 50C outputs.

With counterflow you can do much better by only letting the hot hot stuff touch the hot cold stuff, etc. Taking the temp at a few locations of transfer might look like this.

• 100C hot stuff with 99C cold stuff.
• 60C hot stuff with 59C cold stuff.
• 1C hot stuff with 0C cold stuff.

When you add metal tiles and tempshifts, generally you are adding these types of heat transfers to the system:

• 65C cold stuff with 70C cold stuff
• 45C hot stuff with 50C hot stuff.

These only reduce the efficiency. You want as much of the top stuff and as little of the bottom stuff.

Ah, i think i get it now!
The reason is because both the cooling and heating mediums are the same mass and similar heat capacity, and therefore one cannot get an advantage over the other. Hence, a cooling block of metal works great for cooling oxygen with water, as water has a much higher heat capacity & mass than the O2, and thus will stabilize the metal blocks at a target temperature and save cooling power by buffering the heat/cold, whereas petroleum and oil will not. When you just want to equalize them, any interference will actually make the system bigger, not smaller. There's no buffer needed.