Is this a crazy idea?

[james247]

Originally Posted by MrPix

The cost of the RO filter system for saltwater plus the solar system, combined, is well into four figures. I think I can run a condensate cooler on 60 watts (and even then, it might be too cold and freeze - it just needs to get below the dew point).



I looked at peltier elements because I could use PWM to operate them only to the temperature needed. However, they're quite inefficient and draw over four times the current per BTU that phase change systems do.



I'm definitely open to suggestions.



My developing notion is to develop a control circuit, which can monitor temperature and humidity to calculate the dew point. Then it would cool the element to maybe 5 degrees below that point. In a humid environment with free air flow, a vertical metal plate 100x160mm in a 80% RH environment at 80F would extract around 0.2GPH for around 60 watts. It would likely cost under $25 to produce. It would be extremely easy to use and maintain.



I'll do some experimenting and report back.

I operate major buildings in the Middle East. The humidity here is very high and condensate is huge by comparison to most areas of the world. We recycle the condensate for irrigation on the particular property.
Consider this condensate is generated from the air passing over the fins of the cooling element of the ac unit. This is where pathogens concentrate. I would never drink this water without nano filtration and chl injection. This would rather defeat the cost Benefit analysis of use of condensate.
It would be equally efficient to use salt water and reverse osmosis I feel.
There is some break through water purification technology coming from Japan you might research.

[Lodesman]

Originally Posted by MrPix

Production: 280ml/hr (1.77G/day)

?? - Curious about the math - 280ml/h over a 12 hour period yields 3.36L or about 0.74 Imp gal / 0.89 US Gal. What am I missing?

Originally Posted by MrPix

I am personally happy with 1.5 gallons - 6.8 liters.

You're in the US, but converting litres from Imperial gallons - is that intentional?
BTW - I applaud your efforts to think outside the box. Look forward to seeing where your research leads.

[ahoymon]

Take a look at Seawater Pro https://seawaterpro.com/ They have a modular water maker that runs on a a Honda 2000 and costs under $1500. I am installing one now and have seen it in operation on others boats.

[MrPix]

Originally Posted by hellosailor

I know that without an 80# pack, I can go through 6-8 pints, consuming well more than 64 ounces, in one day of moderate work in a hot humid climate.

Well yeah. The figures were worked out with soldiers in the peak of physical fitness. We're all overweight, old, and have the stamina of an asthmatic ant.

But really this device isn't for you, and I'm not here to make y'all happy. I have, in 3 days of tinkering, make a machine that with no input but 12V, 5A can keep one or two people *alive*. Not comfortably, but alive. That power draw sounds huge, but it's one 80W solar panel.

I have had a couple of ideas on how to increase the efficiency of this. If the evaporator side was painted matte black and faced the sun, and the condenser side was painted white and well insulated, it would increase efficiency a lot. It would also increase the chimney effect by a factor os maybe 100% on a clear day, or maybe half that on a cloudy day.

Once, someone said, "hey, what if we got a membrane and used osmosis to only pass water through it?" and everyone was all "but the power!" or "those high pressures you'd need!" or "but there'd be tiny tears in the membrane!" and that one guy who would go "wow, gee, I ain't drinkin' that swill, it'll still have like 1% of the salt in it! TOXINS!"

So far, I've done this with a peltier element. They're cheap, reliable, and super inefficient. My next option is evaporative cooling. It could use sea water as both the source liquid and the evaporative liquid to cool the recovery chamber. This would eliminate 90% of the current draw. Or, it could be used to supplement a peltier cooler to allow it to cool a much large surface. I don't feel like doing the math on that, so I'll instead experiment and see what actually happens.

[MrPix]

Originally Posted by Lodesman

?? - Curious about the math - 280ml/h over a 12 hour period yields 3.36L or about 0.74 Imp gal / 0.89 US Gal. What am I missing?

You're in the US, but converting litres from Imperial gallons - is that intentional?
BTW - I applaud your efforts to think outside the box. Look forward to seeing where your research leads.

I'm sorry!

I live in the US but am of British descent. To me a gallon is 4.564 liters, or 3.9 liters, depending on whichever number pops into my head first. I do mix them up. I prefer working in metric, being raised post-1973. It's not deliberate, I promise.

I appreciate your interest. Some people are really down on it. Especially those that don't read ahead and follow the evolution of the (really very old) idea.

I did order a solar still, just for comparison. The main inefficiency of it is that the outside skin that is the condensation layer warms in the sun too. This slows down the process significantly. If the outside were bathed in a mist of evaporation, the process would speed up considerably. So my focus right now is on a solar-based design with a sun-heated evap side and a swamp cooler based cooler side. If humidity is too high, a peltier element could give a cooling boost.

One of my big discoveries is about cooling below the dew point.

I ran the peltier element with a PWM signal so I could turn it up or down with 256 steps of variability. I measures the temp and RH to obtain the dew point, and the temperature of the air passing the condensation surface and the temperature of the condensation surface itself. PWM is interesting as an arduino can monitor the temperatures and only run the peltier element the amount needed to obtain just the minimum amount of cooling required. This could reduce current draw by anything from 5 to 95%... Experimenting will tell us how much.

The cooler's surface needs to be quite a bit cooler than the air to cool the AIR to below the dew point. If the cooler is 2C cooler, the air right by the cooler dries but a thermal layer forms and the flow becomes laminar. This prevents mixing, because the moisture becomes exhausted in that boundary layer. With another 5C of cooling, the air cools more, and then heat and humidity causes turbulence within the air close to the cooler surface which breaks down the laminar flow. This allows more mixing of the air, and more humid air to come close to the condenser where it is robbed of heat.

However, once that state has been reached, quite large extra increases in cooling to not dramatically increase the condensation rate.

So, right now, this magical test article is a cat siphon-style water dispenser with an ultrasonic fogger in it ($12 + $8.99) running up a black painted PVC pipe that is 100cm tall, around a u bend, then down a 100cm copper pipe that is covered in sponge. It has a flattened area with a peltier plate 10cm from the top.

You first wet the sponge with seawater, and place the capture bottle under the outlet. You fill the input water bottle with seawater and put it in the siphonic waterer (to maintain a consistent water level). You turn on the peltier element. After about 2 minutes, it cools enough and the PVC side warms enough in the sun that the air starts to rise up the warm side them fall down the cold side. Once this flow happens, you turn on the fogger. The mist is drawn up the pipe and evaporates. Salt precipitated back down into the source water. Then water vapor goes over the top, hits the cool side. This causes immediate slight condensation that drops down the walls. This combined with the evaporative cooling draws further heat out of them, causing additional condensation down the pipe.

I have some thinking to do on separating the swamp cooler run-off from the output water. That water needs to go through a sawyer filer just to be doubly sure there's no bacteria, fungus, slime mold or other contaminants. However, their exposure, if any, to the water would be so brief that toxins would not be an issue. The pipe interior is easy to access and clean. Inverting the apparatus and pouring in a 1% bleach solution for 20 seconds will do it.

I'm waiting for some new supplies to arrive, and will do some more experimenting next weekend.

When I have a fairly evolved design, I will publish an open source build plan, list of materials and construction guide, plus any code used by the controller.

I anticipate the finished result will cost well under $100 and provide a cheap yet minimally functional water maker that weighs in at around 2.5 kg (see, I've gone full metric now!)

[john61ct]

Well you're obviously knowledgeable motivated and having fun.

I originally thought just for yourself, not worth the trouble.

But if this ends up as a potential product on Kickstarter,

especially as an open-hardware project,

I'll definitely support it even if it doesn't fit my needs.

Best of luck!