Water from the atmosphere with using minimal power?

xcodemustdie

Ars Scholae Palatinae
779
This is for use in the Ukrainian trenches, so water use is very restricted (no showers for > a month), limited washing of everything like body, clothes, dishes etc.
Thus has a special set of constraints.
Can't use much electricity.
Can't use solar (i.e it's visible from the air)
Can't take up too much area

Is there any tech that possibly fits the bill.
 

demultiplexer

Ars Praefectus
4,944
Subscriptor
You're very physics-limited, as you need to condense water which requires energy, 2.2MJ/kg aka 0.6kWh/kg to be exact. To be clear: that energy needs to be extracted, i.e. it's excess energy in the water that's around in the atmosphere, but it's still a lot of energy to move around.

Your best bet to do that is a heat pump, which in common size-optimized (read: mobile dehumidifier) form factors has a COP of 2.5-3.0, so you still need about 1kWh per ~6L of water. This is the best you can do without a physically large installation that isn't visible from above. Anything else involves using e.g. clear night skies to condense water (but that requires a LOT of area and specific weather).

Obviously another big option is (solar) distillation of wastewater. That can be done almost fully without technology, and on sunny days the distillation rate is pretty useful, a few square meters of solar collector can yield gallons of water per day.
 

Megalodon

Ars Legatus Legionis
36,631
Subscriptor
This is for use in the Ukrainian trenches, so water use is very restricted (no showers for > a month), limited washing of everything like body, clothes, dishes etc.
Thus has a special set of constraints.
Can't use much electricity.
Can't use solar (i.e it's visible from the air)
Can't take up too much area

Is there any tech that possibly fits the bill.
I think there's some other implicit requirement here: it can't have a large heat signature (too detectable) and it can't use consumables that would be as bad or nearly as bad as just carrying the water (no combustion generators).

So, no, I don't think there's anything that fits these requirements. You can do passive condensation collectors but they need a large surface area to radiate heat. You can do powered heat pumps that retain the condensation, but you're going to be burning fuel.
 

FranzJoseph

Ars Centurion
2,145
Subscriptor
Not much of sea fog biomes over in Ukraine, so fog collectors won't really work (and wouldn't even fit the concealability requirements, unless doubled use with camo and drone netting). And that's basically the most advanced form of a passive humidity collector, while still requiring a lot of real estate space to be anywhere meaningful.

I guess the question is more for the trenches in Zaporizhzhia and SE Ukraine, which is more of a dry steppe ("Cold semi-arid") according to the Köppen? With mostly sand dunes on the spits as I remember from before the invasion?

Basically, you are left with what long‑distance hikers use – local sources with appropriate filtration, or hygienic wipes in the case of not enough local water sources at the position. And delivered barrels, but that's better for drinking use.

Wipes my be the easiest, as the volume and weight of them is the lowest, and nobody really cares about their CO2 and microplastic footprint when fighting to save your very country and peoples.

Pretty sure the frontline soldiers know the very best what works for them by now, as Ukrainians have so far shown so much ingenuity that it shames all the Western fighting forces.

So if you are trying to do this as a helpful project, I'd at first ask them what their needs are (I am assuming here you are not part of the fighting forces there, but of course I could be wrong, in which case have my apologies and Slava Ukraini!)
 

xcodemustdie

Ars Scholae Palatinae
779
So if you are trying to do this as a helpful project, I'd at first ask them what their needs are (I am assuming here you are not part of the fighting forces there, but of course I could be wrong, in which case have my apologies and Slava Ukraini!)
Yeah I am currently in the trenches, we have enough drinking water, it's the water for other things that's an issue. Like washing your hands after a shit.
I was listening to SGU podcast 2 weeks ago they were talking about a new method.
Even if we could get 10 liters a day that would be brilliant
 
  • Like
Reactions: FranzJoseph

demultiplexer

Ars Praefectus
4,944
Subscriptor
Yeah I am currently in the trenches, we have enough drinking water, it's the water for other things that's an issue. Like washing your hands after a shit.
I was listening to SGU podcast 2 weeks ago they were talking about a new method.
Even if we could get 10 liters a day that would be brilliant
OK, give us some more info. What are your constraints on available power, physical size budget, etc.? What do we have to work with.
 

xcodemustdie

Ars Scholae Palatinae
779
Size can be pretty big, we can dig a dead end trench and stick it in there.
Power, I'm not sure about (noone speaks English which is a pita) we have diesel generator, which we run for a couple of hours a day.

Alternatively I was thinking they use machine learning nowadays to detect birds from their songs, or what a single person is saying in a crowded room.
What would be amazing is some device using AI, using 3 or so microphones that detects the sound of a drone (Vs the sound of a mosquito, wasp etc cause they do sound similar) and preferably it's able to say which direction it is in.
At the moment this is all done by people listening, so it's not fool proof.
This device would save a lot of lives and I don't think would be that difficult to make.
 

Nvoid82

Ars Scholae Palatinae
720
Subscriptor

Nvoid82

Ars Scholae Palatinae
720
Subscriptor
Very interesting, I googled a bit more, there's a few patents about the idea, but afaiks there's no actual purchasable device made thus I assume there's some thing that doesn't work in reality like the theory suggests

The issue as far as I understand it is that the water is dirty. Generally full of soot and uncombusted hydrocarbons, which makes it not very useful for re-injecting into an engine. However, if you just need grey water for rinsing things, or can treat it afterwards, it might be worth pursuing.
 

demultiplexer

Ars Praefectus
4,944
Subscriptor
If you’re running a diesel generator, you can use a cyclonic seperator to get water from the cooled exhaust gasses.

https://saemobilus.sae.org/papers/extraction-liquid-water-exhaust-a-diesel-engine-2015-01-2806

Roughly speaking the lower temperature you can get the exhaust gasses the (around 8% vs 35% theoretical at 49 vs 45 C). In this paper, the collection vessel was under vacuum, though I’m not sure how necessary that is.
Don't do that, unburnt Diesel has detergent additives in it and exhaust gases have both water and oil soluble nasties that will all accumulate in condensed exhaust. The result isn't just dirty, even as gray water it's toxic to your skin and releases volatiles that are pretty nasty too.

It would be a really good source of water though, you get about 2L of water for each liter of diesel burned.

Have you tried to collect rainwater? What about digging or drilling for groundwater?
A groundwater rig seems like a good place to start.
 

MilleniX

Ars Tribunus Angusticlavius
7,786
Subscriptor++
Don't do that, unburnt Diesel has detergent additives in it and exhaust gases have both water and oil soluble nasties that will all accumulate in condensed exhaust. The result isn't just dirty, even as gray water it's toxic to your skin and releases volatiles that are pretty nasty too.

It would be a really good source of water though, you get about 2L of water for each liter of diesel burned.
Given the fairly consistent, predictable contaminants, it seems like it should be possible to develop a cost-optimized filter medium that can make the condensed water suitable for at least some purposes. That feels to me on the scale of something that chemical or materials engineering grad students could brainstorm and prototype various possibilities over the course of a summer internship.
 

demultiplexer

Ars Praefectus
4,944
Subscriptor
Given the fairly consistent, predictable contaminants, it seems like it should be possible to develop a cost-optimized filter medium that can make the condensed water suitable for at least some purposes. That feels to me on the scale of something that chemical or materials engineering grad students could brainstorm and prototype various possibilities over the course of a summer internship.
Eh, sure, you can eventually filter anything, but the nasty thing about exhaust water is that it's all atomized and partially soluble. You don't get that with e.g. raw water - raw water maybe has a small amount of salts dissolved in it, but all the other stuff is comparatively large and generally not soluble. So you can run it through a simple(ish) filter and get potable if not very healthy water out of it.

Exhaust water requires waaaaay more complicated filtering. And not just a grad student to design it, but also to safely operate it and know when to swap out your media.
 

demultiplexer

Ars Praefectus
4,944
Subscriptor
With complicated stuff like this, purification is a multi-step process where you need to take samples in between to check if solute levels are below some predetermined level to be able to continue. It's not something a single process can take care of. This is real chemical engineering, stuff people study for and need a bunch of additional experience to be able to tackle.

Long story short: idk 🤷‍♂️
 
  • Like
Reactions: continuum

dzid

Ars Centurion
3,230
Subscriptor
Size can be pretty big, we can dig a dead end trench and stick it in there.
Power, I'm not sure about (noone speaks English which is a pita) we have diesel generator, which we run for a couple of hours a day.

Alternatively I was thinking they use machine learning nowadays to detect birds from their songs, or what a single person is saying in a crowded room.
What would be amazing is some device using AI, using 3 or so microphones that detects the sound of a drone (Vs the sound of a mosquito, wasp etc cause they do sound similar) and preferably it's able to say which direction it is in.
At the moment this is all done by people listening, so it's not fool proof.
This device would save a lot of lives and I don't think would be that difficult to make.
This is a cool idea. A couple of years ago I found one particular project that I thought was pretty neat in terms of the sample code available which is the whisper.cpp project on GitHub. It's principle use is translation/transcription from file or real-time, but some of the code was pretty straightforward and perhaps adaptable, for example, identifying the predicted reliability of the translation by color-coding the text in the real-time code. I just looked, it's still up.

There are some really good use cases for this stuff, and yours is one.
 
  • Like
Reactions: xcodemustdie

xcodemustdie

Ars Scholae Palatinae
779
This is a cool idea. A couple of years ago I found one particular project that I thought was pretty neat in terms of the sample code available which is the whisper.cpp project on GitHub. It's principle use is translation/transcription from file or real-time, but some of the code was pretty straightforward and perhaps adaptable, for example, identifying the predicted reliability of the translation by color-coding the text in the real-time code. I just looked, it's still up.

There are some really good use cases for this stuff, and yours is one.
Thanks I only have access to a bad phone at the moment, but will have a deeper look later. I have mentioned my idea to a couple of relative ppl with the military so why knows.