Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Fenwah]

EDIT: Now with repeatability!

I recently did some dwarven science and I have a method for bringing magma up from the magma sea without using a (large) pump stack

Basically, you use the fact that cave ins displace liquids above them. If you carve out a long column of rock (refered to as the piston) and drop it into the magma, every unit of magma underneath will move to the top of the column. If you prepare a catchment area at the top, then it wont fall back down the shaft and you have yourself a nice pool of magma.

Reasons you would build one of these instead of a traditional pump stack:

• No fps drop
• Very little iron is needed. I moved magma up over 150 levels on a map with no iron ore or sand. The only magma proof pumps needed are to fill the cistern at the bottom and empty the reservois at the top
• Much less fiddly designating and managing production of components. Just designate it in one go and let the miners do their thing
• Very dwarfy

Heres an mspaint drawing of one before it was triggered

Spoiler (click to show/hide)

and one after

Spoiler (click to show/hide)

Here is a map of my final repeating version.

EDIT: As someone has pointed out below, semi molten rock will eat the piston, so you can't drop it directly into the magma sea, and you'll have to use a cistern. To reinterate: DO NOT DO THIS DIRECTLY INTO THE MAGMA SEA. IT WILL NOT WORK

The hatches / retracting bridges are there to catch the magma without supporting the piston. Also make sure they are magma proof or you will have a problem (I made that mistake the first time round with some granite hatches). It's also important that you make sure they line up at the right level (or a bit below would aslo work).

I tested this originally using a tank filled by a volcano and transported a 14x10x5 volume of 7/7 magma (thats 4900 units) up around 20 z-levels.

Things to remember

• Scout your location. Your pump can only be as high as its shortest z-column so you need to find a way down to the magma sea with as much natural rock as possible. Send military or feline explorers to map out the caverns for you and pick a good spot. Dig into the magma sea in a few spots to map its shape as well. Or use reveal.exe.
• If your piston has bits missing in the middle because of caverns, anchor the bottom section, and drop the top parts. The gaps will fill and you can then clean up the top of the piston
• Mine, don't channel. Channelling out the shaft takes a ridiculously long time and is dangerous for your miners. Just mine out the shaft and drop a floor from above when you are done to break through all the floors. You can mass designate the shaft by holding down the mouse and going up and down in long z-level columns. Designating 150 z-levels takes around 5 minutes, and digging takes less than a year with 3 miners

Making a repeatable pump

The method above will give you a big hit of magma, suitable for setting up your metal industry, pits for trash/goblins, and maybe a nice trap or two if you recycle it. If you want tons of magma to throw around everywhere you need a repeating pump, which takes a little more work than the one shot method.

The basic principle behind this is to cast obsidian on top of the piston. In order to gain magma, you need to minimise the amout of magma present when you cast. You'll need some bridges above the piston which can be filled with water, then dropped on the piston. Make sure the shape and size of the falling water matches the piston exactly (don't worry, it wont spread as it falls). You will also need a set of pumps taking the magma up to at least the level above the orignal height of the piston (if your piston drops 3 levels each cycle, you need to go up at least 4 levels), and a way of returning this magma to the top of the piston. The procedure goes like this: remove magma untill there is around 2/7 magma left on top of the piston (it will work with 1/7, but you run the very real risk of tiles evaporating at the last second and ruining the casting, which is very hard to fix). Then dump water. Now return magma until there is a 2/7 layer on top if the piston and cast again. Repeat as needed until the piston is rebuilt, then pump out all the delicous magma that is sitting on the sides of the piston. Now all you have to do is mine out the cistern at the bottom, refill it, and your pump is reset.

More things to remember:

• Make sure the whole thing is subteranean if your map freezes. My map was frozen half the year and it was a massive pain in the ass, as the obsidian caster will ice up.
• Don't try and cast with a 2/7 layer on top of a 7/7. The water doesnt all fall together, and as the obsidian sinks it throws magma everywhere and the obsidian will not form a flat surface, ruining your piston
• Everything about this device's performance can be worked out in advance: how much magma will come up, how much is needed to rebuild, etc. Run through the numbers before you start building to make sure that there will be enough magma to rebuild. A thin piston will be able to rebuild less layers that a fat one, as a greater proportion of the magma will end up on the sides
• And for Armok's sake, get familiar with how cave-ins work before attempting this

Anyway thats my contribution to dwarven science. Hopefully it will be used to bring magma to dwarves in need everywhere.

[Psieye]

Outstanding piece of dwarven research, commendations!

[Grimlocke]

I did not know this, excellent! Its like a giant one-use piston pump. This will be particularly useful for making an early start with the magma forges.

[afoninv]

Awesome. There is no other word for it. Elegant, useful, epic way of doing things.

You are a model dwarf, I attest. I will tell my grandchildren that I dwelled on the same forum with you.

[Psieye]

I did not know this, excellent! Its like a giant one-use piston pump. This will be particularly useful for making an early start with the magma forges.

Personally I get my magma forges set up within 1st season of embark (including anti-magma-life measures), foregoing building/digging out anything else beyond the bare essentials. But the beauty of this method is that you don't have to worry about pesky magma-life annoying you later. Whatever magma men or imps were down there, they're dead and you just get the magma filtered free of hostiles.

[Unshaven]

I have used collapses for swift irrigation of an area.. when I am busting an aquifer. but to do the same with the movement of magma is a wonderful idea.. though I can't help but think there may be fair amount of trail and error when it comes out getting the right amount up the shaft.. just saying..

[Jetsquirrel]

Thats... thats GENIUS!!!!

You get a Dwarven SCIENCE! Stamp of Approval!

[Sphalerite]

Fascinating.  I had no idea cave-ins would displace liquids upwards.

Now we need to work out how to make it automatically repeat, and we have a piston pump.

[Captain Hammer]

I hate to rain on the parade but doesn't anything that touches semi-molten rock get absorbed never to be seen again? Your entire pillar would disappear into the magma sea

[afoninv]

I hate to rain on the parade but doesn't anything that touches semi-molten rock get absorbed never to be seen again? Your entire pillar would disappear into the magma sea

Hm. But BEFORE disappearing it should still displace magma upwards, no?

Any willing testers? =)

[Fenwah]

Thanks for the kind words

@ Captain Hammer:
I haven't heard of that before but it shouldn't be hard to test. If that is the case then it would just be a matter of tranfering the magma from the sea to a constructed cistern using a few pumps. This would be annoying but 5 pumps and a big hole is easier than 150 pumps. I'll have to test out the effect of SMR on cave in rock i guess.

EDIT: A quick wiki check showed that you are quite right, and this will not work going directly into the magma sea. So it looks like cisterns are the way to go.

@ Unshaven:
There wasn't much trial and error involved. Everything in the system is predictable and with a bit of thought it works out. I got my hatches to line up properly with the top of the pillar on the first try. If it wasn't a flat surface af the bottom or there was caverns in between then it would be much harder but certainly predictable. That said, its always a good idea to have a save backup before you pull the lever.

[Z1000000m]

So its only use is to get magma?
i though you can make a cavein that will pierce hfs..

anyway, nice work

[denito]

Oh, hey, I could use this same technique to transport ocean water up to the top of a mountain!  (By digging a channel from the ocean to a cistern under the mountain.)  I have a volcano on a mountain/cliff next to the sea and I wanted to try dropping some water into the middle to make an obsidian column.  This'll be much faster than building all those pumps.

Although it sounds like my obsidian is going to disappear when it hits semi-molten rock at the bottom of the tube -?  At least it should reveal what's down there.

[goge]

Great idea!

Can't you use a reservoir 1 z level below and to the side of the piston catch tank to drain the magma into before the piston melts out?  I drew a picture

Spoiler (click to show/hide)

now no pumps needed again

[Patchy]

I've thought about something like this before, but I didnt try it cause I thought the piston would insta melt into the semi molten rock. Since yanno cave-ins are like nearly instant and thus the magma would start immediately falling back into the sea. Anyways if this method does work, awesome and very dwarfy idea you had ;P