40d Talk:Water wheel

Perpetual motion machine

I've not used waterwheels yet, so I'm unsure if this is the case, but couldn't you, theoretically, set up a perpetual motion machine using a waterwheel and a screw pump?

The article contradicts itself, it says on the first line "in" a flow, but the next line refers to a flow underneath, which is correct? Matryx 17:34, 31 October 2007 (EDT)

A perpetual motion device is easy enough to setup once you have an understanding of screwpumps and power. I currently have a water wheel placed between two underground resevoirs that runs a mill and pumps water from the lower tank to the higher one. Its very energy efficient aswell. Three axles, a gear, the mill and the pump only draw 40 power leaving me 60 for other devices.

I'll try get up screenshots of it or maybe a tidier one later --Lucid 19:58, 31 October 2007 (EDT)

edit comments

Someone removed my edit, but the water wheel actually only requires one square of water underneath it, not three. - Sludge Man

We could really use better pictures. The tileset in these screenshots is terrible, (lets use the default one) and they are very unclear. I do not understand how to build a working waterwheel after looking at this page.

Perpetual motion again

I removed the unclear example of a perpetual motion machine with a forum link to much clearer designs. We still need clear pictures and elaboration of the method of construction. I'll get around to it once I understand it myself, if no one beats me to it. --Turgid Bolk 15:48, 5 November 2007 (EST)

Sorry about the tilesets but I didn't think anyone would mind my custom one considering the only noticibly difference is the pump which looks like 2 barrels.

Although my screenshots were specifically for a perpetual motion machine which is why they lacked indepth wheel and pump construction. I was hoping to create a video or some sort of tutorial to add in the construction section which dealt with creating waterwheels for someone who has absolutely no idea of any of the mechanics.

Also how reliable are the new designs? I checked out that link and I previously toyed with designs similar to those and found them to be excellent power generators but not true perpetual motion machines, they all lost power intermittedly for varying lengths of time. --Lucid 18:30, 5 November 2007 (EST)

I made a shoddy vid of my own design which also makes a waterfall. here's the link if you want it on the main page. [1] --Mattex 14:11, 9 July 2009 (UTC)

Water circulates counter clockwise in bi-level circuit. A pump lifts the water to the top of the channel, where it then falls back to the bottom, turning a wheel as it circulates. The wheel powers the pump by a shaft in the middle.

I made a potentially erroneous illustration in an attempt to understand perpetual motion. Is this correct? I did not attach it to the article fearing that I might confuse somebody. --SnowCat 04:16, 30 August 2009 (UTC)

I must say that Reactor design is very efficient. I set one up expecting it to have issues with evaporation and while I've only left it running a few years so far, I see no evidence that it's allowing any water to evaporate at all. I do strongly advise that you build the hatch first since you may have some flooding issues if it activates before you get your water chamber filled properly. Doctorzuber 01:53, 19 March 2010 (UTC)

Perpendicular to water flow

Strange as it may sound, water wheels do 'not' need to be parallel to the water flow direction to work. VengefulDonut 16:31, 23 November 2007 (EST)

This is as in real life - you can build a less efficient water wheel by putting the "buckets" at an angle - and indeed might put the "buckets" on a 360 degree pivot for some purposesGarrieIrons 07:21, 8 January 2008 (EST)

Because "flow" has no direction in DF. You either "have flow" or you "do not have flow".--Albedo 17:07, 20 March 2010 (UTC)

Water Flow Needed?

Actually there is no need for water flow at all. A water wheel seems to generate power even if it is built to a water body with no current (no "entry" or "exit" points). So basically the water wheel just needs to touch water and that's it. This is what I've noticed when confining a canal with floodgates at both ends. --Flaa 07:47, 22 April 2008 (EDT)

This seems to have changed in a recent version. In previous versions you just had to connect a canal to a brook and the whole canal would count as 'flowing' even when nothing was moving. In my latest game this hasn't worked. So it's either changed, or the fact that I tried a 4-tile wide canal has stopped it working. --Juckto 18:22, 10 May 2008 (EDT)

Powering a Water Wheel

It seems that you can connect a Water Wheel to a Windmill, or other power source and cause it to spin - is there a purpose to doing this? can you move water along channels this way? What are the benefits? --SeiferTim 12:47, 12 March 2008 (EDT)

I think that it's because water wheels use power to run. Hence, they produce 100 power by themselves, but end up using 10 to spin, making the net power output only 90. Therefore, most likely the windmills provide that 10 power to the waterwheel. I don't think it does anything, but it does look cool.

--Umiman 05:34, 27 May 2008 (EDT)

Verify Carpenter

Does the water wheel require carpentry to build?--Richards 04:11, 21 April 2008 (EDT)

Yep--Dorten 04:15, 21 April 2008 (EDT)

Thanks Dorten.--Richards 04:26, 21 April 2008 (EDT)

Minimum water level

What is the minimum water level and flow required to power a waterwheel? --Sphexx 16:48, 27 May 2008 (EDT)

1/7 water Hight. Hoborobo 04:59, 8 June 2008 (EDT)-

That doesn't sound correct to me... Juckto 20:33, 14 June 2008 (EDT)

Definition of "flow"?

The article states that waterwheels need flow under them, but how exactly does the game calculate this?

I was thinking that it means any water changing depth below it, but that would rule out building on a river/ocean full of 7/7 tiles. Will a waterwheel work on top of water full to the brim but supposedly "flowing" (i.e. ocean, full river)? G-Flex 8 June 2008 (EDT)

I just redirected a river to run through my fortress and then back out again. As my channel was filling up it provided NO power. However, once the water filled the spaces at the edge of the map (the OUT point of the original river), my waterwheels suddenly gained power. It seems that flow takes place either a) when the depth of the water is shifting around or b) when the tile is defined as a river. Rivers seem to be defined as any body of water that is connected to both the in and out points on a river. Lymojo 22:04, 6 May 2009 (UTC)Lymojo

I believe that right now a waterwheel will provide power as long as there is at least 4/7 water depth below it, or at least that is what has been the case in my experience, as I don't think a dead end channel would really provide any water flow yet a waterwheel will still provide power if place over one. --Elvang 04:15, 8 June 2008 (EDT)

Are you sure, though? Juckto (above) seems to believe the state of affairs is a bit more complicated. I guess I can just test it out for myself at some point. G-Flex 04:28, 8 June 2008 (EDT)

I'm sure about the minimum being 4/7, never had a waterwheel that would work below that level. Maybe it counts the water as flowing if the water depth changes during x ticks? If thats the case then a closed system would work unless you manage to get all the tiles at the same depth. I doubt the game uses a system much more complicated than that, as it would cause incredible lag checking each tile for depth, direction of flow, speed of flow, and path of least resistance. Not to mention factoring in water pressure and evaporation. Additionally, waterwheels seem to work no matter what their orientation is to the water. --Elvang 04:37, 8 June 2008 (EDT)

Hrm, yeah, it seems like you need change of depth for it to count as "flow". Putting in the ocean just don't work.G-Flex 04:23, 16 June 2008 (EDT)

Change of depth not needed. Putting it in a full river works. --Savok 22:47, 20 July 2008 (EDT)

I believe to have found the key to it: I placed three waterwheels in the bed of a drained underground river; the first one with its outer tiles on the riverbed, the second one hanging in a channel of three tiles, the third one with its outer tiles on top of two constructed walls, which I constructed in a channel in the riverbed, i.e. on rough stone-block floor.

So I built them on top of this:

 1.     2.     3.  (  4. )
rrr    rrr    rrr  ( fff )
rrr    r_r    rfr  ( f_f )    r=river-tile
r_r    r_r    r_r  ( f_f )    f=rough floor (top of constructed wall)
rrr    r_r    rfr  ( f_f )    _=open space (channeled underneath)
rrr    rrr    rrr  ( fff )

When I let the water in again, only the second one was powered, even when I interrupted the flow again. (As I the water flew out, it still turned at water level 1. That seems to be the minimum level indeed.)^[Verify] So the generation of power seemed to depend on the underground you build them in, not on anything like flow. Then, to verify my theory that you need river-tiles around waterwheels to make them work, I built a fourth one upstream, which was surrounded by floor-tiles. This one, however, streched all across the river, and so divided it into a lower part (with the three waterwheels in it) and an upper part. In the lower part, no waterwheel worked anymore, as I built one of the second type in the upper part of the river, it worked again!

Therefore, I suppose waterwheels must be surrounded (at least partially?)^[Verify] by river-tiles, which additionally must not be disconnected from the source of the river. --Doub 12:25, 4 September 2008 (EDT)

Construction Key?

The construction area needs a key. I can't figure out what the O is supposed to be, and I only use the ASCII version of DF.--Rusty Mcloon 11:19, 18 July 2008 (EDT)

It is a vertical axle. You should know, since you play the pseudo-ASCII version. --Savok 22:47, 20 July 2008 (EDT)

Waterwheels & waterfalls? & magma?...

would water wheels be powered by water falling down a Z axis onto a waterwheel?

can a waterwheel be used in magma or would it burn?--Althalus 10:05, 9 August 2008 (EDT)

Water wheels need to be placed in a channel with water to be powered. AFAIK the channel doesn't have to be any longer than the water wheel, doesn't need water coming in nor going out. The game isn't clever enough yet... (shhh!)

I think the water has to be depth:3.

Water wheels have to be made of wood. Any machine that has magma flow through it, if it isn't made of magma-safe material, will burn/melt. You might get a bit of power for a short time but the water wheel would burn (assuming temperature is turned on which it is by default).

To sign your name properly use the signature button on the toolbar or type "~" four times: ~~~~ GarrieIrons 21:26, 8 August 2008 (EDT)

Waterwheels may not be magma-safe, but the actual structure of the wheel is built one z-level above the magma, rather than in it. Thus it is entirely plausible that a waterwheel could survive being used with magma. Whether it does actually survive such circumstances, and if so whether it starts turning, could do with being tested. --Raumkraut 23:09, 8 August 2008 (EDT)

Well i guess i will have to do my own experiments on magma wheels, i will report my findings. --Althalus 10:05, 9 August 2008 (EDT)

Sounds something like Cog's wooden magma pumps. --Schwern 20:22, 18 December 2008 (EST)

Not really - of course, waterwheels suspended above magma will not self-destruct, since they are not IN magma, but it is not known whether magma powers waterwheels, or if that is only hardcoded to be water. --Savok 09:42, 19 December 2008 (EST)

I tried powering some waterwheels above a stream of flowing magma, powered by a pump. it did not produce any power. *sigh* there goes my plan for magma based perpetuum mobilae... Roderik 00:54, 11 March 2010 (UTC)

Yeah, in the year plus since the above posts, that's been confirmed. Will edit.--Albedo 04:15, 11 March 2010 (UTC)

Waterwheel on rivers

Yesterday I build a waterwheel above a river, it did not work. I build a second wheel in a moat channeled from the river it did work. Either I made some stupid error, or there is a bug at the current ver. that makes wheel not working on rivers themselves... --Catpaw 06:37, 29 August 2008 (EDT)

A brook or a river? HeWhoIsPale 09:24, 29 August 2008 (EDT)

Oh it was a brook, since my dwarfes always waded through it. okay I understand it now... Is kinda quirky in the simulation... you turn a brook into a river by digging its surface... uh. common logic anyone? :-) --Catpaw 09:50, 29 August 2008 (EDT)

Brooks have a special layer of floor on top of them that allows fishing and similar activities, but discounts it being considered as running water, channeling can remove this layer. Apparently this has been the best fit way of allowing a brook to be traversable while still being a body of water. HeWhoIsPale 10:51, 2 September 2008 (EDT)

Water and Flow Answers

After a few(a lot of) tests, I have deduced the following:

• A water wheel needs to have any one of its three tiles above water
  • The water must be at least 4/7 depth
    • The exception:water falling down a z-level

  Placing a waterwheel with one tile stuck into the side of a water fall with make it active, and can be effectively done by placing several next to each other under the waterfall

  Note: if the water flows over all three or the two opposite tiles, it will only be active for a second here and there, but it's great to make tons of mist

  • It can not be submerged, a submerged water wheel will not work
• The water must be flowing, stationary water will not power the wheel
  • A screw pump can make water flow (See the section of the article on Perpetual Motion)
  • Many rivers do not flow, you may have to channel a side river to make it flow out of the actual river and then into some other body of water

  P.S. it helps to stagger the channels downwards so that the flow continues

• Lava will make a water wheel turn, (if it flows of course) if you turn off temperature (the water wheel will simply go up in flames otherwise)

This information was gather in version 40d Feel free to add anything here to the article or correct anything that might be wrong

Very simple design for a water wheel
+~W~+
+~W-*
+~W~+
(W) = wheel tile with flowing water under it
(-) = horizontal axle
(*) = gear assembly
(~) = river/brook tiles
(+) = ground/wall/whatever else you may have there

--Wizjany 23:14, 25 October 2008 (EDT)

I'm not entirely sure about submerged water wheels. I had a water wheel placed in a (channeled) brook, connected to a screw pump. I built a wall/box around both the screw pump and the water wheel to create a reservoir which was drawn from by another screw pump one z-level up. Perhaps the waterwheel only ceases to work after the water reaches a certain depth? --RomeoFalling 23:31, 25 October 2008 (EDT)

Not totally sure what you mean, but if the water wheel is totally submerged in "flowing" water, it shouldn't be active because the water is not actually flowing through the tiles, but "teleporting" to where it would end up, according to Toady in an interview. (See the picture) --Wizjany 19:15, 10 July 2009 (UTC)

natural flow from a river or brook behaves very differently, so I can envision a case where it might be possible to get a fully submerged waterwheel to turn. Artificial flow however has a different system and does stop flowing when it fills up to 7/7. Having said that, I think I'll go run a test for the submerged waterwheel using natural flow. Doctorzuber 18:54, 20 March 2010 (UTC)

Safe water wheel power

Imagine that I had a totally walled off above-ground area. Now I redirect a river to flow under my walls with an above-ground area between my walls. Will this provide the necessary flow for a water wheel? Gairabad 18:13, 15 November 2008 (EST)

I imagine if it would work without the walls in place, it would work with them in place. But swimming creatures (crocodiles, etc.) might be able to swim up the channel.--Maximus 19:27, 15 November 2008 (EST)

A well-placed grate or two would deal with them. Gairabad 23:55, 17 November 2008 (EST)

Any flow ought to be able to power a water wheel. I've had three ganged water wheels running off the power of one aquifer flowing into another. (much lava work was needed to breach the aquifer in various places.) --Corona688 14:04, 19 December 2008 (EST)

Underwater Waterwheels

Will a waterwheel continue to provide power if it is submerged, so long as there is running water in the appropriate tiles? Say, suppose you drain the ocean, and set up water-wheels along the ocean floor to provide power to an underwater fortress, so it can auto-drain itself if you abandon and reclaim. Lastofthelight

Not sure on your whole scenario, but i recently flooded 3 waterwheels and despite what other ppl say they still worked powering just the pumps that flooded the room when being in 7/7 water. This was underground, right at a river, with a ceiling above, if that makes any difference. --Birthright 16:58, 12 July 2009 (UTC)

Yes, that makes a difference. The example you describe involves the oddities of naturally flowing water. Here is an example of a submerged waterwheel that turns. a brook is naturally flowing source of water. the pumps are putting a layer above this that behaves like naturally flowing water. Note this is a two wide channel with two pumps because apparently the center tile of a waterwheel is solid and cannot be submerged. The waterwheel is connected back to the wheel with gears to submerge itself. This is a very special case however, since it is using the quirky properties of naturally flowing water to get two layers of flowing water on top of each other. Doctorzuber 20:21, 20 March 2010 (UTC)

    ≈   Natural flowing water as from a River or Brook
    ▒   Solid Ground
    %%  Screw Pump
    ||| Water Wheel

             %%|||≈≈▒
    ≈≈≈Brook≈≈≈≈≈≈≈≈▒
    ▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒

Adjacent water tiles

I'm not going to get caught up into this edit skirmish, but having three adjacent water tiles is a requirement for neither the construction nor the functioning of a water wheel. VengefulDonut 02:34, 22 June 2009 (UTC)

Agreed. Moving on to the real puzzles? --Höhlenschreck 13:52, 22 June 2009 (UTC)

Water Flow Complexities

My current thinking seems to be that sourced water (river, brook, spring) behaves very differently than unsourced water. With sourced water it seems to have flow for free, even in pointless dead end channels on the far end of the map, a waterwheel will still turn as long as it is channeled directly from sourced water. In fact in my experimentation I found that a floodgate thrown down on this channel will NOT shutoff the waterwheel. Somehow even though water won't pass the floodgate, flow will. Note, I've read that some natural rivers in the game don't have flow.

Unsourced water, taken from a lake, or otherwise seperated from the river is a lot less forgiving about water flow. here it seems the only mechanic for deciding if the water is flowing or not is to watch for a difference in the water depth. This leaves me wondering if this will become a problem for stagnant water in my underground drinking area. Doctorzuber 18:41, 5 July 2009 (UTC)

I opened up a water flow testbed on a plain forest map with a river. Lucked out and scored a flowing river so ran a bunch of fun tests with water. Discovered a quirky bug. As has been mentioned before, a connection to a flowing river or brook will power a waterwheel, even if it's a dead end. Testing this, I found that distance seemed to have no effect on this as long as the 4/7 rule is satisfied. I also discovered a quirky bug with floodgates. Apparently, while floodgates block water from moving, they don't actually block the flow from a flowing river or brook. Even after slamming the floodgate down, the waterwheel still churned away happily in a completely isolated dead end channel blocked from the river by a floodgate.

Playing later, I found another quirky bug with floodgates. I was piping water down a few levels and figured water pressure would happily push it back up the 3 z-levels to fill my water reservoir and stared at it for a bit while it failed to do so. I ended up fiddling with my flood gate controls opening and closing it a few times and it managed to reset the water pressure and fill suddenly, but until I did, I had somehow managed to bug the water pressure to zero with an OPEN floodgate. More testing is needed to make sense of this bug. Doctorzuber 03:46, 7 July 2009 (UTC)

If you disable SHOW_FLOW_AMOUNTS in init.txt, you can very easily see whether or not water is flowing, as it will alternate between ~ and ≈. With the setting turned off, I dug a long channel next to a river and then dug out the last wall to fill it in, and the water mostly stopped flowing once the channel filled, but after a minute or so, the entire channel spontaneously started flowing at full force. Presumably, there's a process that happens every so often where the game traces all paths from water sources and marks those tiles as flowing, and they keep flowing forever (and providing water wheel power if they're at least 4/7). Presumably, this check also marks all "isolated" water as non-flowing (such as those formed from ponds being filled via pumps or being expanded via channeling). --Quietust 13:59, 8 September 2009 (UTC)

I've recently came back to dwarf fortress, and I'm still rather obsessed with water flow... I'm firmly of the belief that rivers and brooks have some sort of "magic flow" that behaves differently than any artificial flow you can try to create from stagnant water. I've checked the various perpetual motion generators and am still not really satisfied. If a power source stops, even briefly, it makes a mill worthless, since it will cancel jobs every time it stops. Sooo.... I am back to running tests again, and looking to develop a true perpetual motion device. This reactor is a new one to me, I'll have to check that out, although looking at it I strongly expect it's going to halt periodically as well. Here's some theories I'm working on...

• Water from a natural source (river/brook/ocean/aquifier) flows naturally. I'm calling this magic flow.
• Water pumped directly from a flowing source, remains flowing water. The pressure changes, the flow seems to remain.
• floodgates stop water from moving, but keep the flowing state. (Bug?)

These are all still in the theory stage, so by all means, if you disprove them, let me know, but no flaming.Doctorzuber 16:59, 17 March 2010 (UTC)

Mobius Wheels

New PEG design; [2] [3]

I honestly have no idea why this works, but it does. The design is simply an enclosed 4 *2 space, two pumps on lvl1, water wheels directly in front of the pumps on lvl2, and a gear assembly linking the wheels to the pumps. One manual pumper switches it on (fill the chamber), and it will never ever stop. It does NOT work with single wheels. --Salk 16:40, 8 September 2009 (UTC)

• I've found that water pumped directly out of a river and into an enclosed reservoir appears to flow indefinitely (turn off SHOW_FLOW_AMOUNTS in init.txt and you can see whether water is flowing or not), so your setup might work even without linking the pumps to the water wheels. --Quietust 16:45, 8 September 2009 (UTC)

Non-Flowing Rivers or Brooks?

Can anyone verify if these actually exist or is this simply a myth floating around from someone who didn't understand how to channel through a brook or something? Doctorzuber 13:32, 20 March 2010 (UTC)

To my knowledge, there is only one way to get a brook or river which does not flow, and that is to embark in a region which contains two or more brooks/rivers flowing away in opposite directions - one will have flow, while the other will not. --Quietust 16:24, 20 March 2010 (UTC)

Never seen a double, no idea. But if you dam a brook/river, does the upstream still have "flow"? Not sure - it's always dangerous to try to apply "logic" to DF. It's clearly not the waterchannel itself, but the beginning tiles at the map's edge - but do they in turn rely on the draining tiles at the far end?? --Albedo 17:05, 20 March 2010 (UTC)

By my testing on natural flow from a river or brook I would expect it would continue to flow in that case, but this is strictly because it's a river or brook which behaves differently than other methods of generating flow. Doctorzuber 19:24, 20 March 2010 (UTC)

Expectations, assumptions and guesses are all based on "logic" - which in DF is as arbitrary as any code. If you haven't tested it, maybe it will shed light on something. <shrugs> --Albedo 19:31, 20 March 2010 (UTC)

true, but I have tested natural flow in other dead end situations and natural flow remains. I should test it with river damming at some point. I'll get around to it, eventually. I'm trying to look back over this material and verify it against 0.31 to see if anything has changed. I don't expect it has, but it's good to check. Doctorzuber 18:45, 8 April 2010 (UTC)

Hypothetical 'best' power plant design

After investigating that submerged waterwheels work fine, I came to the conclusion that the absolute best possible use of space in regards to a powerplant is stacking waterwheels directly on top of each other in a room full of flowing water, as outlined here: http://www.bay12games.com/forum/index.php?topic=51302.msg1100425#msg1100425. This design can be easily altered to be wider north/south, longer east/west, or really as tall up/down as you like while giving practically perfect space conservation for a power plant. Dunno if it's worth adding to the main page - but thought it was at least worth adding here. Other thoughts? --Retro 21:42, 21 March 2010 (UTC)

I saw your design on this and looking at it I don't think it will work although I am curious about it. However right now I am trying to focus on testing to see if there are any changes in fluid mechanics for 0.31 right now. I'd like to get these pages pulled forward as soon as I can. Doctorzuber 18:45, 8 April 2010 (UTC)