Print

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

[dragdeler]

Hi,

My goal is to have an artifical canal with tons of waterwheels that is still connected to the river in order to bear all forms of aquatic life. The problem is maintaining water flow and avoiding water teleportation. I had bad experiences with designs where water falls down 1z level; even with pumps and obstacles in order to revigorate flow, the system remains unpredictable and treacherous, every "improvement" seems to ruin flow in a former sweetspot.

So I was thinking: I have the water fall into a cistern that is blocked by my pumps. The pumps bring pressure to their Z level (-2 (0 being the surface and -1 the riverbed)). That, in conjunction with some obstacles but no pumps, should deliver a constant flow without too much fps drain. So logically my fish can only come from downstream, the problem however, is that it will remain dry unless I pump waste water up there and if I connect them, well then I'm back to my old waterpressure but conquering the canal upstream in this case... At least that's what I'm guessing.

So the question is: Will I be able to create enough drainage, using one OR two Z levels for the drain, in order for this not to happen? Or is there a way of allowing fish acces without a direct connection between the canal and the river? Something like a waterfall where a part of the fish could to fall into the canal while all the the water is safely drained...? The alternative would be to not dry the river and have both it and the canal, which would certainly have less preferable effect on my fps than some advanced drain.

This is the design I'm going to try, it is worth noting that the canal will be wider than the number of pumps connected to the cistern, but will have as many draining pumps as it is wide.

Spoiler (click to show/hide)

[PatrikLundell]

Flow is tricky. I don't know enough to give advice beyond not relying on logic.

Vermin fish does not move as creatures do. Instead they flick in and out of existence (virtual vermin?), and I think there is something in DF that tags 16*16 blocks of the map as vermin fish supporting (or not). This means that artificial bodies of water within such tiles will get vermin fish as well (I've had them flick in and out of existence in a dwarven foot bath in one fortress, although I don't think I't tried to fish there).

[dragdeler]

Now that you say it... I was kind of counting with vermin fish anyway. But my OCD wanted those big beautiful ASCII signs, even tough I'll never actually get around to air drowning them. Saying it out loud, I realize it might be a hinch futile.

[PatrikLundell]

"Real" fish would probably move about without those restriction, but as far as I've seen rivers have "real" critters on embarks, but none replace them as they move out (or are killed, in the case of giant sponges). Getting real ocean creatures is supposedly hard as well, as there seems to be a big preference for airborne or land critters. I've heard of projects where the land access has been closed off to improve the chances for ocean critter appearance.
I suppose embarking in an exclusively ocean biome (i.e. on the shore with "embark anywhere" would presumably remove the land critters, but you'd still get birds and "amphibian" critters such as capybaras. You'd also be reduced to underground farming, as oceans support no above ground crops (or trees, so wood would have to be imported, as drift wood is a useless decoration).

[Mostali]

Hi.  First let me apologize for not directly answering your question(s).  I don't entirely understand your description or your picture.  However, I do have a pretty good grasp of flow, so I will make a recommendation based on your objective.  I'm going to assume you've read the wiki on pressure, flow, pumps, etc and so I'll try not to just repeat the info from there.

I'd recommend forgetting about pumps.  Pumps cause pressure, and pressure kills flow.  You're better off letting gravity and obstacles do all the work.  Don't just use a few obstacles, use a lot.  Most designs seem to like a few blocks across a channel - for a channel n tiles wide, yielding n-1 flow points - but I've found that to be too few to really get water flowing.  Instead, dig a channel four tiles wide with the center two alternating blocks and you can make as many flow points as you want, depending on space.  With one side pressured and the other not, you might be surprised how fast water fills an area with ten flow points.  The unpressured side can be open and should be able to support all the water wheels you want.

The part I didn't catch is what you plan to do for an exit, and that's the most important part.  If you're not providing an exit at all, then flow will be very difficult if not impossible.  If you can get your water to a map edge then flow will just happen.  If you can drain it into an aquifer or a cavern lake then it should work fine as well.  If you're trying to pump it back up into the river or loop it back into the canal again, then it can be tricky.  I've made it work and also failed to make it work - I think it's a matter of tweaking volume control.  I've never tried a 'portable hole', but I'd bet that would be a way to get flow in your canal as well.

As for fish, I have no clue.  I don't recall ever getting vermin fish in my artificial canals, but I also didn't really pay attention.  The consensus of a quick search seems to be what Patrik said - either the area will support them or it won't.

Hope that helps, good luck!

[dragdeler]

It's a normal river so 0 is the surface, -1 is the riverbed and waterwheels, and -2 is where the water flows (at least the biggest portion of it that generates power). Gravity has done me nothing but harm, so I was counting on blocking off the source with the pumps, that way there is pressure but not infinite amounts of it.

For my drain I'd prefer to pump the water back up into the river, but fortifications at the same spot one level below is fine too, what would you advise? I was thinking about three pumps in, 5 tiles wide and 7 pumps out.

The canal itself will be empty of pumps, they're just not worth the hassle so I wanted to try a combination from the wiki checkboard depressurizer and this:

Where do you plan to put the water after it's been used for your water wheels?

One useful note: building destroyers may destroy vertical grates.  Instead, use floor grates with the water coming from below:

This

Code: [Select]

Note:  This must be either 1z below the level of your brook OR your brook must be completely dammed for this to work.

W = water
X = wall
# = floor grate

(vertical view)

Z+0
XXXXXX
W#XWWW (water source this way -->)
XXXXXX

Z-1
XXXXXX
XWWWXX
XXXXXX

(horizontal view)

W#XWWW (water source this way-->).
XWWWXX
XXXXXX


is completly useless

BTW: The river start and end don't touch.

[dragdeler]

Most designs seem to like a few blocks across a channel - for a channel n tiles wide, yielding n-1 flow points - but I've found that to be too few to really get water flowing.

That sticks with my experience so far, but I've only implemented half of the measures I want to try out. (It's not like I rolled back a few RL days allready hehe...) One odd thing tough, instead of a diagonal row I'm using a 3 tile wide checkerboard, so it's even worse than n-1 atm (if I got that right?!). But I've been more foreseeing this time; the save and the design can go many ways from here on...

Instead, dig a channel four tiles wide with the center two alternating blocks and you can make as many flow points as you want, depending on space.  With one side pressured and the other not, you might be surprised how fast water fills an area with ten flow points.  The unpressured side can be open and should be able to support all the water wheels you want.

Um so it is some thing like this you're suggesting? Which is closer to what has proven efficient, A or B? And which is better: up (blue) or down (red) // along the stream (blue) or perpendicular (red)?

Spoiler (click to show/hide)

[Mostali]

So I'm still unclear how or why you want pumps, but that part isn't actually important.  Let's just assume you have pressure to your satisfaction.

The exit:  Digging an exit to the map edge and carving a fortification is clearly optimal.  There are two types of flow, let's call them natural and artificial.  Natural flow is when water is connected to a map edge - such as an exit I've described or any river edge.  Any water on the same z-level as that map edge will have natural flow.  Natural flow is also consistently directional so water wheels don't falter when there's natural flow.  Artificial flow is harder to create and control - it comes from when water literally spills over from one tile to another.  Without natural flow, a water wheel sitting over three 7 deep tiles won't move or generate power.  You have to create a local imbalance in the water levels so that water spills from one side of the water wheel to the other.  Pumps at the end of your canal might achieve this, but you'll have to maintain some perfect balance of water levels gradually rising from the end to the beginning.  This is in fact the theory behind water reactor designs.

So again, my recommendation is to use natural flow with diagram B.  (And I still think you can forget the pumps).  It won't matter how you arrange water wheels then, everything will be powered.  If you insist on artificial flow, then you'll just need to experiment and constantly tweak the design.  You will likely need more like diagram A to keep water from rising to 7 and never falling.

[dragdeler]

this shit is hard to communicate. But I think I understood what you meant what the first time, it just seems so incredible. My intuition would have been, that the flow coming out of diagram B would be perpendicular to the pressurized source, so I was allready imagining how I would have to control those branches. But if it just flows toward fortifications that is way simpler. I just didn't think pulling could be as effective as pushing.

I'll share my final design in any case, promised. But now it would just be confusing to upload screenies where I still have to explain all the elements that aren't finished yet.


Oh and all fish seem to do fine in any case, tough as usual it's impossible to judge a system before running it.

[dragdeler]

Well I've done it... In retrospect I probably would have spared myself a lot of sorrows if I went for your design but I didn't know if I would be able to make it as compact and afraid to make mistakes and start over even more times, so I grinded trough it, knowing I was kind of close.

https://imgur.com/a/qkNEa

Generates ~4k power with the absolute lowest ever spotted, being 1278 net. Occasionaly fish can be seen in any corner tough they seem to allways come from the edges.

Thanks for your time guys.