Print

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

[ancistrus]

Problem: in a current fortress there is no good place to build a pumpstack since it would run into some obstacle (bedrooms, workshops, dining room.....) no matter where I would build it.
So question is, if I created one pumpstack that pumps magma to a 20 tiles long pipe that ends in another pumpstack, would those be able to bring magma to the surface with the same efficiency as a single pumpstack?

[lue]

It should be possible, just mind the power sources. See this wiki page for some details on magma, if you need it. In particular this paragraph:

Without screw pumps to impart pressure, magma flows rather slowly (though no more slowly than unpressurized water). A pipe to bring magma across the full map can take as much as a year to fill. This, combined with the fact that it will evaporate, can make filling a reservoir difficult and tedious. As a rule of thumb, the area coming out of a 1-wide-pipe shouldn't be more than three squares wide and 20 squares long, or else it will evaporate as fast as you fill it.

So, I'd recommend not going too far between pumpstacks.

[jcochran]

It will work fine as long as you can keep the pump stack full, you can have a channel as long as you want between two pump stacks. The issue of slow flow of magma is for un-pressurized magma. The major limitation will be how fast magma reaches your bottom pump. That is the real limit on how fast the pump stack will pump magma. You might want to have several pumps at the bottom level widely separated (so they don't 'steal' magma from each other pumping into a common manifold. Then on the Z level about that, have a conventional pump stack sucking magma from that manifold

[Merendel]

You might want to have several pumps at the bottom level widely separated (so they don't 'steal' magma from each other pumping into a common manifold. Then on the Z level about that, have a conventional pump stack sucking magma from that manifold

This if you want significant throughput.  I tend to do either 3 or 5 seperated by 3 tiles (so I can put a 3 long axle between them for power) with the center pump being the botom pump of the main stack.  The auxiliery pumps feed into a short tunnle leading to the output of the primary one so it can all be picked up by the next pump in the stack.  This is normaly plenty unless I need more magma than a single layer of the magma sea can supply in one go.  A 5 pump manafold can actualy hoover out most of a layer in a surprisingly short amount of time if you have somewhere for all of it to go.

[nanomage]

If two pump stacks are going to be separated by long 1-tile wide pipe, then this pipe will be the (horribly inefficient) performance bottleneck of the entire project.
To transport magma long distances with speeds comparable with those achieved in pump stacks (so the the phase of horizontal transfer wouldn't hinder them) you need a horizontal row of pumps where each next pump's intake is either on the same tile or directly below the previous pump's output. Essentially, it's a horizontal (or diagonal) pump stack. Also, one of the good things you can do for your PC and your dwarves is not build any casing unless necessary to protect the stack from FB's. Stacks more magma through midair fine, just watch out that there are no dwarves around when you switch it off.

[Quietust]

If two pump stacks are going to be separated by long 1-tile wide pipe, then this pipe will be the (horribly inefficient) performance bottleneck of the entire project.
To transport magma long distances with speeds comparable with those achieved in pump stacks (so the the phase of horizontal transfer wouldn't hinder them) you need a horizontal row of pumps where each next pump's intake is either on the same tile or directly below the previous pump's output. Essentially, it's a horizontal (or diagonal) pump stack.

This is not true - as long as the beginning pump's intake is kept primed, it can rapidly fill a horizontal channel of practically any length. This is the reason behind having multiple intake pumps from the magma source itself feeding the base of the pump stack.

[jcochran]

If two pump stacks are going to be separated by long 1-tile wide pipe, then this pipe will be the (horribly inefficient) performance bottleneck of the entire project.

Not true at all. Once the tunnel is full of magma, magma 'teleports' from the output of one pump to the input of the other. Doesn't need to flow at all. Also the build order of the pumps has minimal effect (if you build the stack from the top down you'll see output from the top pump on the first tick of the stack being powered. If you build from bottom up, you'll see output N ticks after the stack is powered with N being the number of pumps). But once fluid is being pumped, the throughput is entirely dependent upon how fast you can make magma available at the very bottom of the pump stack. Now one thing you'll notice about pump stacks where the demand of fluid is interment is that once the stack is full, when you require fluid (just opened a floodgate or drawbridge), you'll gets extremely rapid flow for a short while and then the flow will reduce to a smaller, but still steady rate. That initial rapid flow is the pump stack emptying itself (the output tile of each pump in the stack being drained by the pump above). After that happens, the remaining flow is how fast you're getting magma to the bottom pump. So a very tall stack has a built in reservoir and can handle intermittent burst demand. But the volume of magma in the tunnel that's connecting the two pump stacks that the OC asked about, doesn't contribute to the burst capability of the pump stacks. That tunnel is "dead" as far as burst capacity. But it has no effect on the overall throughput that the stacks are capable of providing.