Print

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

[jez9999]

I've been trying to figure out how water pressure works in DF based on this wiki article, so I did a little experiment; here's a snapshot of it (the number in the top-right is the z-level):


I'm tapping a river to the north (UPDATE: this river flows into the ocean instead of flowing from one edge of the map to the other, which means it counts as DAMMED, which makes all the difference here).  As you can see, level -1 is in the process of flooding.  I don't understand why this is.  It looks like the river's water is at level -1 (if you use 'k' to look around at level 0, the '7' tiles show as an open space, apparently indicating that the water is really one level below).  The u-bend "diagram A" in the wiki article says that even with pressure, the water should only fill up to one level below the source river level, which would be level -2.  So why is it coming up to level -1?

[Sir Crashalot]

The bottom of the river is at the bottom of level -1, the top of the river is at the top of level -1 so yes, it will flood that level if you allow it to, the pressure will push it up despite your U bend. This is why I dont open up the top of a well until the cistern is full and I have closed the floodgate to prevent more comming in and pushing it up when I channel the wells out. However, you can have the cistern at level -1 and the well at level 0 without worrying about that as the water level will not rise above the rivers level.

[jez9999]

The bottom of the river is at the bottom of level -1, the top of the river is at the top of level -1 so yes, it will flood that level if you allow it to, the pressure will push it up despite your U bend. This is why I dont open up the top of a well until the cistern is full and I have closed the floodgate to prevent more comming in and pushing it up when I channel the wells out. However, you can have the cistern at level -1 and the well at level 0 without worrying about that as the water level will not rise above the rivers level.

But my point was that "Diagram A" on the wiki page I linked to shows the water on the right of the u-bend only filling up to one level below the level of the undammed river.  Is the wiki article wrong?  It seems that pressure is causing it to fill up to the same level as the undammed river.

[Sir Crashalot]

I think it might be because even though the river is flowing off the map there is still the constant pressure of water flowing in regardless of the dam. It didn't make much sense to me tbh, perhaps there is a variable missing like the top of the ubend also flowing off or something.

[slothen]

But my point was that "Diagram A" on the wiki page I linked to shows the water on the right of the u-bend only filling up to one level below the level of the undammed river.  Is the wiki article wrong?  It seems that pressure is causing it to fill up to the same level as the undammed river.

I would treat any river as diagram B.

[BoredVirulence]

But my point was that "Diagram A" on the wiki page I linked to shows the water on the right of the u-bend only filling up to one level below the level of the undammed river.  Is the wiki article wrong?  It seems that pressure is causing it to fill up to the same level as the undammed river.

I would treat any river as diagram B.

I agree. Honestly I don't know why it should make any difference. The river is going to be filled to a constant level because of the flow of water. And water in a U bend (in real life) fills to equal levels on each side, so it should fill up to the source of the river.

My bet is that whoever did that science mined from the river at a diagonal.

Code: [Select]

#~~~###
#~~~##
####~#~#
####~#~#
####~~~#
########
Rather than

Code: [Select]

#~~~###
#~~~#~~
###~#~#~
###~#~#~
###~~~#~
########


[vanatteveldt]

I always place a diagonal at the top level I want to fill, that is a full proof way of avoiding floods. I would also advise always placing a floodgate/block in the inlet before you connect to the water source. This functions both as an emergency stop and can be useful if you need to work on the plumbing later...

[jez9999]

But my point was that "Diagram A" on the wiki page I linked to shows the water on the right of the u-bend only filling up to one level below the level of the undammed river.  Is the wiki article wrong?  It seems that pressure is causing it to fill up to the same level as the undammed river.

I would treat any river as diagram B.

I agree. Honestly I don't know why it should make any difference. The river is going to be filled to a constant level because of the flow of water. And water in a U bend (in real life) fills to equal levels on each side, so it should fill up to the source of the river.

My bet is that whoever did that science mined from the river at a diagonal.

Code: [Select]

#~~~###
#~~~##
####~#~#
####~#~#
####~~~#
########
Rather than

Code: [Select]

#~~~###
#~~~#~~
###~#~#~
###~#~#~
###~~~#~
########


Erm, how would that work?  Water can't travel diagionally on the z axis.

[jez9999]

I always place a diagonal at the top level I want to fill, that is a full proof way of avoiding floods. I would also advise always placing a floodgate/block in the inlet before you connect to the water source. This functions both as an emergency stop and can be useful if you need to work on the plumbing later...

Yep I definitely agree on the "diagonal on the top level you want to fill" rule.  Only trouble is, diagonals also slow the speed of the water down quite a bit so I guess for a significantly quick fill you'll want several channels that combine, all of which have dialogals on them.

Any way if this wiki article is wrong it should probably be changed.

Another question: does water's falling actually create pressure?  From my experiments, it seems to.  So even if you eliminate the pressure on level -1 with a diagonal, if the water falls down to level -2 and then level -3, it will rise up through any holes to level -2 again presumably because of the pressure generated by its falling down from -1 to -2.  This may only apply if the water is taken from some flowing water, I'm not sure.

[Ashery]

I always place a diagonal at the top level I want to fill, that is a full proof way of avoiding floods. I would also advise always placing a floodgate/block in the inlet before you connect to the water source. This functions both as an emergency stop and can be useful if you need to work on the plumbing later...

Yep I definitely agree on the "diagonal on the top level you want to fill" rule.  Only trouble is, diagonals also slow the speed of the water down quite a bit so I guess for a significantly quick fill you'll want several channels that combine, all of which have dialogals on them.

Any way if this wiki article is wrong it should probably be changed.

Another question: does water's falling actually create pressure?  From my experiments, it seems to.  So even if you eliminate the pressure on level -1 with a diagonal, if the water falls down to level -2 and then level -3, it will rise up through any holes to level -2 again presumably because of the pressure generated by its falling down from -1 to -2.  This may only apply if the water is taken from some flowing water, I'm not sure.

Alternatively, if you're not vulnerable to building destroyers, you can combine floodgates and a diagonal bend to create a togglable pressure kill. Not something I'd likely do, though, as I'm absolutely paranoid about the reliability of my fluid systems.

[spook54321]

I'll have to check when I get home, but I'm fairly certain I have something that follows diagram A and sources from a river.  Stopped 1 z-level lower than I anticipated and didn't rise any further.

[slothen]

Another question: does water's falling actually create pressure?  From my experiments, it seems to.  So even if you eliminate the pressure on level -1 with a diagonal, if the water falls down to level -2 and then level -3, it will rise up through any holes to level -2 again presumably because of the pressure generated by its falling down from -1 to -2.  This may only apply if the water is taken from some flowing water, I'm not sure.

This isn't really 'falling'. You're merely filling a u-shaped container from the top.  It falls when the container is empty, but the z bend on -1 ensures that the water will fill up to that level always.  When you talk about 'falling' water, one thinks of water falling through space from a waterfall.  Under normal circumstances the source water and where the water collects after falling are separated (at least on the downward trip).  Therefore there are no pressure tricks to ensure the bottom part doesn't flood, unless the water flow rates happen to be just right.

[Spitfire]

Contrary to everyone in this thread, my opinion was that Diagram A is correct. I noticed that effect back in 31.25. (No diagonal inflow, no outflow in the reservoir, come on guys, neither I nor the folks at the wiki are stupid.)
So Toady must've changed liquid physics (I haven't heard of that), or your setup is truly odd and worthy of a forum post.

To pressure in falling water: Pressure is only transmitted through 7/7 water. At least it was in 31.25. I don't know what to believe anymore...

[spook54321]

One quick test proves diagram A is correct.  One moment while I upload the pictures.

[jez9999]

Another question: does water's falling actually create pressure?  From my experiments, it seems to.  So even if you eliminate the pressure on level -1 with a diagonal, if the water falls down to level -2 and then level -3, it will rise up through any holes to level -2 again presumably because of the pressure generated by its falling down from -1 to -2.  This may only apply if the water is taken from some flowing water, I'm not sure.

This isn't really 'falling'. You're merely filling a u-shaped container from the top.  It falls when the container is empty, but the z bend on -1 ensures that the water will fill up to that level always.

But if you have a diagonal on -1 (BEFORE the water falls to the next level), you've removed pressure, right?  So, from then on, the water shouldn't be able to go "up" any levels, should it?  I thought only pressure could make water go "up".