Print

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

[PatrikLundell]

I've been trying to induce flow in light aquifers to power water wheels, but every attempt so far has failed (while it has worked with only a couple of attempts with heavy ones in the past). However, I see a gear has changed state, indicating a water wheel has actually generated a pulse of power at some time, and I think I've seen a flicker occasionally that might have been a turning of a wheel, but it's never happened while I've stared at it.

Could it be that power is generated only when the aquifer would have release some water, which happens too rarely to be useful?

[Quarque]

This question has been on my to-do list of things to try, so your post made me carry out the planned experiment. It succeeded. Here is how I did it.

Symbols used in schematics below:
#: wall
_: floor
.: open space
=: water wheel
-: axle
F: fortification
^: ramp upward
v: ramp downward

Let's call the lowest aquifer level z=0. At z=-3 I dug a channel up to the map edge and carved fortifications into the edge.

Code: [Select]

#FFF#
#___#
##_##
##_##
##_##
##_##
##_##
##_##
##^##

At z=-2 I dug a walkway along the channel and pre-installed a water wheel. Also dug ramps up and down:

Code: [Select]

###.###
##_._##
##_=_##
##-=-##
##_=_##
#__._##
#_#v###
#_#_###
#_#^###

Finally, I dug out a massive water collection area one level below the light aquifer (z=-1). Water will slowly drip down into any squares you dig out here, but because it happens in the entire area the total rate of water collection is quite high:

Code: [Select]

#################################
###############v#################
###############_#################
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#_______________________________#
#################################

The collection area fills up to water of mostly depth 1 to 2. The channel has a steady water depth of 4, until it drains away at the edge. The water wheel displays 100 power 24/7.

[PatrikLundell]

An interesting design, showing that you can use light aquifers for moderate water production (making e.g. obsidian farms reasonable). It doesn't answer my question, though, which was about induced flow, i.e. getting aquifer tiles to generate power without any actual movement of water (my rationalization for using them has been that they're a less FPS draining way of getting power than using actual flowing water).

[knutor]

The aqua layer reabsorbs its leaks wiki says, so..
but not fast enough, I imagine, my steps are always slippery.

Maybe try...

checkerboarding aqua layered rm, will induce a wonderful flow.
pulse that leak with a water repeater(wr)
link pressureplate(pp)in wr to a floodgate
feed looped pipe, 2z deep
fun part.. put screwpump and wheel in the loop
run SP til wheel starts
stop SP, wheel keeps spinning, run its power back to SP
lotsa leftover juice
if doesnt work, add drain off map to edge, then seal it back up with FG, for perpetual flow

I never did this, but maybe it just might work. Good Luck

[PatrikLundell]

knutor, it sounds like you've just reinvented the dwarven water reactor (which does work). However, I'm not after any real world flow, but the property of the tile called flow. With heavy aquifers you can use screw pumps to set this property and have water wheels turn forever without any actual water movement (which is what you get with the map edge block trick, with the disadvantages of requiring more work and appearing below the aquifer rather than on top of it [as I want the power to run a dwarf washer]).

[knutor]

Thats more advanced than I am. I'm in need of a new playbook. Above the aqua layer that is special achievement.

[hanni79]

I used my light Aquifer slightly differently, I also made an collector area, but used the resulting water to fill up my dwarven reactor, which in turn was used to pump up water and magma from the caverns.

Three 1*30 Tiles long tunnels were enough to fill it in a somewhat reasonable time, but would have been much too slow to ever fill up a reasonably sized Obsidian cat basin.

I know, Dwarven Reactor is kinda cheaty, but I just love building them  To me, they are simply "early Alpha Dwarven Magic" ^^

[Quarque]

It is not surprising that the DF physics allow a working perpetuum mobile.

On the contrary, it is absolutely incredible how nature manages to prevent ANY sneaky perpetuum mobile design from working, while at the same time making sure that no elementary particle has a definite speed or position.

[Leonidas]

Has it been demonstrated that the water accumulation in an empty tile is proportional to the number of orthogonally adjacent aquifer tiles? Meaning, does a checkerboard pattern definitely produce the most water?

[muldrake]

Has it been demonstrated that the water accumulation in an empty tile is proportional to the number of orthogonally adjacent aquifer tiles? Meaning, does a checkerboard pattern definitely produce the most water?

I'm not sure how you'd measure flow but it would seem the test would be simply to start by digging a checkerboard, draining off into whatever reservoir/water wheel/whatever you're using to measure, then measure the flow, then dig out more and more of it until/if the flow changes.  If it decreases, the answer is probably yes.

It seems intuitive, but I've never tested it.  Also, so long as it isn't actually worse than other patterns, it seems a lot easier/faster to do.

[PatrikLundell]

To test it you'd need to make a save with things prepared, dig out the various patterns, and check the amount of water collected in your measuring reservoir after a standard amount of time, probably done multiple times per pattern.

My current assumption is that a checkerboard would produce the largest amount of water, but there's also a risk that the limited diagonal water flow rate may stall large patterns such that you'd end up with 7/7 water in remote parts because it can't flow away fast enough. That's pure conjecture, though, but if that's an issue it might be better to provide some straight paths for the produced water to flow away, with checker boards feeding these channels.

[Leonidas]

Since it seems that nobody has done this science yet, here's my experimental design.

In Setup A, there's a single light aquifer tile surrounded on seven sides with contructed walls. On the eighth side, which is orthogonal, there's a channeled-out tile blocked by a lever-controlled hatch.

In Setup B, there's a single non-aquifer channeled-out tile, surrounded on seven sides with aquifer tiles. The eighth tile, a diagonal, is for building access.

Code: [Select]

A = Aquifer
0 = Constructed Wall
H = Lever-controlled hatch over a channeled-out tile

Setup A:
0000
0AH
0000

Setup B:
AA
AHA
AAA

In both setups, the down staircase leads to idential columns of channeled-out tiles that serve as measuring tubes. Both hatches are linked to the same lever. There are lever-controlled hatches at the bottom of both measuring tubes that'll drain out any excess water before the experiment. If the water produced in an open tile is truly related to the number of aquifer tiles it's touching, then the difference in the setups should appear pretty quickly. My guess is that they'll be the same.

Does this design seem right?

[PatrikLundell]

That should answer that question, although I'd run it about 3 times to be sure.

The other question asked is whether a single aquifer tile produces the same amount of water per time unit or whether each face produces that amount. That could be tested with Setup A compared to a copy where the orthogonal "O" tiles are replaced by "H" columns (and surrounded by walls to not get water in from other sources).

My expectation from the base experiment is that B produces much more water than A. In part that's based on experiences from heavy aquifers, where things get considerably easier once the first wall has been built. My guess for the second proposed test is that each face produces water independently of the other faces.