View Issue Details

IDProjectCategoryView StatusLast Update
0001635Dwarf FortressDwarf Mode -- Flowspublic2015-01-10 12:53
ReporterMalicus Assigned Touser6 
PrioritynormalSeverityminorReproducibilityhave not tried
Status confirmedResolutionopen 
Product Version0.31.03 
Summary0001635: Water pumped through a u-bend near an ocean does not rise to or above ocean level
DescriptionWhile trying to test something else regarding u-bends, I ran into a problem. Water that was pumped seemed to just VANISH, instead of being "teleported" to its destination due to water pressure. I initially thought that this was because of a floor hatch I had installed, but further testing (with another pump elsewhere and a deeper u-bend) revealed that this happens without the hatch. The water seems to refuse to go up to the highest level of the ocean due to pressure (which is where it would stop if the water was not being pumped), but the water that SHOULD be pressurized to that level instead vanishes.
Steps To Reproduce1) Download the save in which I discovered this at http://dffd.wimbli.com/file.php?id=2268
2) Follow the directions outlined on the file description on DFFD to see that the pumped water will vanish in both locations where I installed pumps to test this
Additional InformationI ought to go to bed, so this may be unintelligible. It should be easy enough to figure out what I'm talking about from the save, if nothing else.
TagsNo tags attached.

Relationships

related to 0001177 new pressurized water starts flowing, then stops 

Activities

Quietust

2010-04-29 05:48

reporter   ~0005655

I seem to recall an issue similar to this back in 40d (possibly the exact same issue), where water pumped through a U-bend near an ocean would mysteriously vanish as if an aquifer was present.

Logical2u

2010-04-29 06:08

manager   ~0005656

Hmm, well, if you can find that topic, Quietust, it would be appreciated!

That being said, it looks like the F2 pump is working almost correctly, if maybe slightly unintuitively. The pumps are applying water under pressure at Z-level 99. Water under pressure only goes up to the z-level below that which it started. As expected, there's a 7/7 block of water at Z-level 98 on the opposite side of the U-bend.

However F3 is a little weirder - pressure is being applied at Z-level 99, but the water in the u-bend is only going to z-level 97. It doesn't look like there are any pressure reseting arrangements in the design, either - only a shift from indoor-to-outdoor tiles at one point. It also doesn't look like there's an aquifer that's absorbing the water, either.

cephalo

2010-04-29 06:13

reporter   ~0005657

This happened to me in 40d also while I was trying to build some dining room fountains in a coastal fort. I'm sure this happens so you can drain water into the ocean without causing a flood. There might be a similar mechanism for underground caverns now.

DoctorZuber

2010-04-29 12:51

reporter   ~0005691

it it u-bending to exactly one level below the ocean level? If so I don't understand what the problem is here exactly. I guess I'll have to grab this save and try to figure out what you're talking about.

Malicus

2010-04-29 13:11

reporter   ~0005698

Oh, there are no aquifers in the equation anywhere here, since they're modded out.

@Logical2u:

If the water at F2 were working correctly, then either the water in the u-bend would go higher or the pump area would fill up with water. Neither of those is happening, as the pumped water goes to oblivion when it touches the water that's already there. ...unless it's doing something completely different for you.

cephalo

2010-04-29 21:49

reporter   ~0005731

Here is the fountain I had an issue with in 40d. There are aquifers in other biomes on the map, but not here, just the nearby ocean. I had to redesign the u bend to be one level higher than the ocean, hence the bottom of the pipes are kinda awkward looking. Equal to the ocean level lost all pressure, I could pump water endlessly.

Snarlingtool
http://mkv25.net/dfma/poi-19662-themainhall

Logical2u

2010-04-30 07:16

manager   ~0005742

Huh, yea. According to the wiki, the water should be going to 99 at F2 and not 98. It's acting like the pumps are a river source instead of a pump source...

I think you forgot to mod out Conglomerate as an aquifer stone, since there's an aquifer to the east!

Malicus

2010-04-30 14:34

reporter   ~0005775

There ought not be an aquifer there, as I just checked, and the aquifer tag had been removed from conglomerate already.

The main issue I'm trying to point out here, though, is that the water at the level of the pumps is vanishing, rather than teleporting somewhere through pressure. Granted, the water source IS infinite, which makes this less of an issue, but it still shouldn't happen. (Unless something completely different is happening on your computer.)

Logical2u

2010-04-30 14:40

manager   ~0005777

Last edited: 2010-04-30 15:48

I'm just trying not to think too much about teleporting water - it's bad physics!

Yes, I agree that's what going on, although the exact circumstances seem to differ slightly between F2 and F3's pump setups.

And I did check, and Conglomerate is missing its aquifer tag... but I've definitely got an aquifer. Hopefully that's not important.

Edit: I'm assuming you removed all the aquifer tags before you generated the world?

DoctorZuber

2010-04-30 17:47

reporter   ~0005791

I must say this is pretty odd looking. If there's an aquifer there though this is working as expected since it can just keep endlessly dumping the water off the edge of the map through the aquifer negating the normal pressure of the pump.

If there's not an aquifer there, well that would be a bug...

Logical2u

2010-04-30 17:52

manager   ~0005792

No, there's no aquifer in the pump setup, just a random one off to the side.

Malicus

2010-05-01 00:50

reporter   ~0005806

Last edited: 2010-05-01 00:52

I figured out why there is an aquifer there. I forgot to remove the [AQUIFER] tag from puddingstone, because I didn't think to look at inorganic_stone_mineral.txt (as I didn't think to look at the non-layers), and guess what was just behind that seemingly aquiferous conglomerate?

ptb_ptb

2015-01-10 09:20

reporter   ~0031846

Testing in DF 0.40.24 with aquifers removed. Water does seem to behave strangely near the ocean. Essentially, water seems to disappear from the 'up' side of the U-bend in the z-level that is directly below the beach floor (but not instantly). If you have a natural flow (no pressure) this results in a water level one lower than the ocean on the 'up' side. If you use a screw pump to keep forcing water through with pressure then it is enough to keep the water on the 'up' side level with the ocean, and the odd bit of water leaks through. There is still clearly a lot of water going missing.

I have no idea why the 'up' side leaks water but the 'down' side doesn't.

Possibly this is something to do with the handling of ocean waves?

Save file: http://dffd.wimbli.com/file.php?id=10412

Add Note

Note

Issue History

Date Modified Username Field Change
2010-04-29 03:57 Malicus New Issue
2010-04-29 04:14 Logical2u Relationship added related to 0001177
2010-04-29 05:48 Quietust Note Added: 0005655
2010-04-29 06:08 Logical2u Note Added: 0005656
2010-04-29 06:13 cephalo Note Added: 0005657
2010-04-29 12:51 DoctorZuber Note Added: 0005691
2010-04-29 13:11 Malicus Note Added: 0005698
2010-04-29 21:49 cephalo Note Added: 0005731
2010-04-30 07:16 Logical2u Note Added: 0005742
2010-04-30 14:34 Malicus Note Added: 0005775
2010-04-30 14:40 Logical2u Note Added: 0005777
2010-04-30 15:48 Logical2u Note Edited: 0005777
2010-04-30 17:47 DoctorZuber Note Added: 0005791
2010-04-30 17:52 Logical2u Note Added: 0005792
2010-05-01 00:50 Malicus Note Added: 0005806
2010-05-01 00:52 Malicus Note Edited: 0005806
2015-01-10 09:20 ptb_ptb Note Added: 0031846
2015-01-10 12:53 user6 Assigned To => user6
2015-01-10 12:53 user6 Status new => confirmed