Hacker News Reader

Comment by ChuckMcM | Hacker News Reader

As a person who has run through this particular loop once with local politics (Sunnyvale's northern border is the SF Bay which has former Morton Salt ponds in it) what I ran into was two things, one that there wasn't a cost effective power solution for just desalination, and two, the players don't care about water to individuals (drinking water and what not) because there is absolutely no risk whatsoever in the bay area of not having "enough" of it. The political block that kills it is a combination of local pop saying "we don't want nuclear/cheap power in our back yard" and the folks farming the desert in the central valley saying "If you can't supply farms, why would you do it?"

My hope is that one day, small scale nuclear power (sometimes called "modular nuclear power") will make this a reality.

View on HN

FWIW there is a really fun "disruptive" desalination option that could work in places like Monterey Bay which is just dropping a really really long pipe down into the ocean that has the desalination membrane at the bottom of it. Top of the pipe is above the surface. Water pressure at depth drives water through the membrane and up the pipe. At the surface you just pump fresh water out of the pipe like an ordinary fresh water well.

Maintenance is basically pulling the pipe up every couple of years and replacing the membrane which will become fouled over time, but generally it would be deep enough that it would not get barnacles and other mollusks on it.

On paper this sounds like a really obvious great idea

Have we ever built a working prototype/plant that does this?

There was a startup pursuing a test setup using an oil derrick of Santa Barbara if I recall. These guys[1] were proposing a different system (basically providing fresh water to an underwater habitat) however the paper goes over the basics pretty effectively.

[1] https://www.hindawi.com/journals/ijce/2021/8898472/

It does sound clever at first glance. It caught me at first, too.

The problem is, your fresh water is now several hundred feet below the surface. To raise it up to where you are, you need pumps to push up the same exact pressure differential that you got by putting the membrane so deep. And if you need to do that anyway, just use surface level water and pumps for ease of maintenance and access.

Have you seen a "wave" pump?

The pressure at the bottom of the ocean will need to push thru the column of water as well as the membrane - there's not enough energy to do that (otherwise, there'd be spontaneous water sprouts in the ocean!).

Not the full column.

You pump out the top x feet of water from the pipe. The equalizing levels of inside and outside the pipe is what drives the upward flow through the membrane.

Pumping x feet of water creates x feet of pressure difference. How is that better than doing reverse osmosis at the surface?

Perhaps the advantage is that you can lift in multiple stages, so each pump doesn't bear the full column pressure. Also the pumps aren't exposed to salt.

Why would you need to lift in different stages? If you take the top 3 feet out of the pipe, the water pressure will push three more feet into the pipe.

Am I missing something here?

The membranes require hundreds of feet of head worth of pressure, not 3 ft.

The pressure at the membrane will only ever be the difference. It has nothing to do with the depth the membrane is actually under the water.

That makes sense, but then, why put it on a pipe at all?

Near as I can tell, it offers no actual advantage to do so, and creates several serious disadvantages.

In theory, wouldn’t you be able to just keep pumping from the surface of the pipe (increasing x) until x provided a sufficient differential pass through the membrane? It would then stay at x even while you continue pumping. The pump would merely need to lift the water from x below sea level.

Pumping from the surface, like with a vacuum? The maximum height of a drinking straw is about 10 meters. It's generally easier to lift water by pushing from the bottom than pulling from the top.

10 meters at sea level. A lot more when you have an ocean’s weight of water pushing on the other end.

That's irrelevant, because the pressure at the water surface is ~1 atm regardless of its depth below sea level.

Air pressure increases slightly as you descend a well, but not enough to change the outcome.

The site pressure at a bottom of a well is irrelevant. You do realize air pressure is not what matters, but pressure as a whole? That includes water pressure. The bottom of the straw is not in a well; it’s more akin to a cassion, so you need to factor in the water pressure as well.

What you’re saying would imply that the Titan submersible would be at the same pressure whether under an ocean of water, or in a well the same depth. No. The former is 375atm, the latter is a bit more than 1atm.

You only need to lift the top layer of water a little bit. The rest of the water column would come up after it. Of course, you cannot lift the top of the siphon more than 10m above sea level.

In order for the reverse osmosis membrane to function, you need a rigid pipe at atmospheric pressure, without water in it. The pressure difference inside/outside the pipe is what powers the desalination process.

While operating, it's basically a deep open well with a bit of water at the bottom. You need to pump that water out to keep the process going.

I mean the “pump” probably isn’t a vacuum pump. Maybe it’s a water screw. Maybe it’s a rope and bucket. The effect is the same, no? The water must be lifted X.

Yes, the effect _is_ the same. Regardless of the mechanism, you can only lift water from above with a sufficiently short hose.

It is incontrovertible that this system would have to push the water up from below.

Maybe the thought is that if you drop a 4000m-deep-to-surface pipe into Monterey Canyon, there will be a pressure differential at the base of the pipe (since sea water is denser than fresh water).

I don't think the math works out all that spectacularly though. The pressure difference would be a about 1.1 MPa, and desalination of seawater takes ~6MPa, so by running a pipe down to the bottom of Monterey Canyon you could... save almost 20% of the energy costs.

Which makes sense. You can't build a perpetual motion machine by sticking a pipe with a membrane at the bottom and a turbine at the top into the ocean. Even if it's a very long pipe.

Seawater has a higher density, about 2,5% according to Wikipedia. So 40m of water column fresh water vs seawater would be equal to 1m water column fresh water. 400m would result in a 1 bar pressure difference.

Reverse osmosis systems (like the one speculated with a membrane) has operating pressures between 40 and 82 bar when using seawater. The ocean is pretty deep but having a 16 kilometer pipe straight down seems like it has its own engineering problems.

Have you heard of oil?

If we can run those for hundreds of miles through water, pumping thick sludge that is crude oil....

> there wasn't a cost effective power solution for just desalination

There absolutely is.

All you need is a bunch of mirrors to direct the sun at a chamber filled with water. The water will evaporate, the expansion into steam creates significant pressure so you can run it through a turbine, and then once the steam cools down and liquifies again, it won't have any salt in it.

It's not even remotely expensive. Mirrors are cheap. The turbines/etc are basically the same as a coal or gas power plant and mirrors are a hell of a lot cheaper than mining coal or gas. Also if you're boiling a usable quantity of water you'll be generating a lot of power, which is a handy byproduct (more valuable than the water to be honest).

"Not in our back yard" is the real issue. The chamber is generally at the top of a tower, and even painted black all that sunlight focused onto it is very bright. People generally don't want a really bright object on top of a tower anywhere near them. I guess you could build a wall around it or something, but you'll still get complaints from aircraft/etc.

Driving into the sunset or sunrise is annoying. Imagine if that was all day every day instead of just for 20 minutes or so.

Having said that, if you're in a remote area to counter the NIMBY issue, and clean water is hard to find, then the system I've described is very effective and they are being used for cost competitive farming of cheap crops.

Anyone who actually wants to desalinate seawater can do it. It's just that almost nobody wants to right now.

This is really interesting; I of course know of salt retrieval via evaporation but I don't have a strong understanding of steam capture for pressure.

How big a tower are we talking about for the steam volume necessary to turn a reasonably sized turbine?

Big.

https://en.wikipedia.org/wiki/Crescent_Dunes_Solar_Energy_Pr...

isn't molten salt quite a bit different?

And how much land do you need for the mirrors, to evoporate the water at a reasonable tate?

How are you handling the byproducts of the concentrated solar distillation process (ie remnants not evaporated in the focal point chamber, salt and anything else you’re not filtering out prior to steam production)? Does this solution assume a closed loop heat exchange system, where the primary loop feeds a steam generator (which is fed raw saltwater as feedstock)?

Brine is the one word term to replace that sentence defining desalination byproduct...

Why use so many words when there is literally a very old term well established in use?

I didn’t think of the word while writing my comment. I should’ve! Thanks for correcting me.

I don't quite understand the rest for the turbine? Once it's steam can't you just cool it and then you have a distilled water source? I.e. steam through coils becomes water again.

It's more efficient to use mechanical filtration than to use distillation.

A survey paper from 2019 on CSP Desalination techniques. https://drive.google.com/file/d/1kuDYo__EdM2-ew0SJ0RuZ_MdhOZ...

They have their own issues and they require quite a bit of real estate. Not something that was easily obtained in Sunnyvale but I'm sure that there are places where they would work.

Evaporation cannot create pressure needed to run turbine, while boiling will, but boiling will create highly corrosive steam with salt.

  > People generally don't want a really bright object on top of a tower anywhere near them.

There is such a tower 50 km south of my house. We see it brightly even at this distance.

There is a stretch of road near the tower that is uncomfortable to drive on during the day, the tower is so bright. I honestly can imagine that many a child has hurt themselves staring at it, it is extremely eye catching and there's really nothing else to look at in that desert.

I'm all for green power (I have an electric car that I feed from roof-mounted solar) but these towers should be very far from population centers.

Desalination seems ideal for cheap but unreliable energy like solar or wind. You just need a water reservoir to act as "battery".

https://www.moderndescartes.com/essays/factobattery/

Using the above methodology, I ran numbers for a desalination factobattery and it's really bad, like two orders of magnitude more expensive than just buying a Li battery to buffer the unreliable energy and running desal 24/7.

This right here. Even low grade continual energy supply is better for desal.

By "battery" here, I just mean a dam that holds a few days worth of fresh water.

I think you misunderstood the comment. I agree there has to be a buffer somewhere. You're arguing that we can get a free buffer by piping the water behind a reservoir, but the cost is that the very capital-expensive desalination plant will be idle half of the time. I'm proposing that the more economical option is to buy a lithium ion battery to buffer the renewables and run the desal plant 24/7. (You can still pipe the water behind a reservoir if you want)

I see. It's better to have a half as big plant running 24/7 with a battery.

Thanks for educating me!

Even with battery degradation?

See: collapsing European industry, finally now even in Germany, due to thinking "demand-side response" was a good solution.

The problem, of course, is utilization.

Amortized capital costs for desalination runs about 40%,[0] so if you "charge" for 12 hours and "discharge" for 12 hours then your facility doubles in size, so your desalination cost increase by (at least) 40%. This doesn't even account for added maintenance btw.

If you charge for 6 hours (ie the solar peak) and discharge for 18 hours, you need a facility 4x as large, and your desalination costs increase by at least 160%.

Desalination isn't 'free' energy storage.

[0] https://www.advisian.com/en/global-perspectives/the-cost-of-...

Desalination plants do REALLY badly when run infrequently. All the salt water, brine, etc. creates absolute havoc when it sits for random period of time, and maintenance costs are already a huge part of opex for desal plants.

Also, the concentrated brine is a real problem in any area that has regular sealife. Maybe not a big problem in the salt ponds already, but it will kill off anything normal.

Yeah, it isn't going to hurt the salt ponds. What is more, we did the math and found you could harvest more salt faster for less money when starting with a higher salinity brine which would be a win for Morton.

The smaller the reactor, the worse neutron efficiency gets, good old cubes and squares. Good practices also require as many hands/eyes for normal operation a GW class reactor as a 50MW one.

At the time we were thinking NuScale type, see https://www.energy.gov/ne/advanced-small-modular-reactors-sm... for efforts by DOE to make small modular reactors feasible in the US.

>Advanced SMRs offer many advantages, such as relatively small physical footprints, reduced capital investment, ability to be sited in locations not possible for larger nuclear plants, and provisions for incremental power additions. SMRs also offer distinct safeguards, security and nonproliferation advantages.

I studied in a tangent area and either I'm dumb or DOE had a lapse of reason that day.

Having five reactors per site is going to be worse than two of equal installed power in almost every respect. USA is the fourth largest country in the world, how is reduced footprint even a consideration?

There is probably a literal ton (if you printed it out on 20# paper stock ;-)) of stuff written about how small modular reactors are a much better answer to deploying nuclear power than big LWRs. In the desal pitch the ability to scale them as needs scaled "quickly" as well as the ability to decomission them when they were done by picking the entire reactor complex up and transporting it off site were the primary selling points.

If you're still interested you can start at the DOE site and follow those links, NuScale (https://www.nuscalepower.com/en) has a bunch of stuff to, they will send you white papers). Anyway, I doubt they are going to go anywhere before the first round of the climate apocalypse gives people the "You can either put in a nuclear power plant or you can all die, which do you want to do?" choice.

The downside of more waster per GWH and/or greater enrichment requirement far outweighs that.

A power plant is only ever "done" when it finds itself on a far more valuable land than when it was built or when your politicians decide to go back to fossil fuels, something that should pretty much never happen.

That said, if it does need to happen, traditional 500MW PWR sites have been restored to green field condition.

SMRs are expected to by 1.5-2x as expensive as traditional nukes, it's borderline fraud at this point.

Is there somewhere I could read to become very knowledgeable on all things desalinization?

I know I’m already at that point. Just experiencing the southwest heat and lack of water for my life has been enough. It strikes me that the US could absolutely resolve so many of its problems with known tech today, but refuses. And definitely it takes some shattering moment to wake people up. We recently had the hugest wildfires known and I don’t quite see much difference locally. But these things take time.

Have you ever done a masters degree? If not, this is how it works (and this works for ANY subject, not just desalination)

1) Get a library card for university (this will more reliably get you papers than sci-hub). Sometimes you can just pay them for one, sometimes you have to register for a course (1 unit courses in the US can be inexpensive and even if they are in basket weaving as a "student" you get to get a library card :-)

2) With the assistance of the reference librarian find the journals that cover the topic you're trying to master and go through the papers. There will be some "highly cited" papers that everyone refers to, some "medium levels of citations" and then typically you'll see papers that take the research in one of a few directions. You can choose to either stay in "survey" mode at this point or dive down one of those directions.

3) Read the papers and learn from the results, you may find subjects that you didn't learn as much as you needed to in order to read the paper (math, chemistry, physics, etc) for each of those, you find text books or things like the Chaum's notes for the subject to bring you up to the level you need in order to understand the paper.

If you keep at this for a while you will start finding that when you read a paper you both understand what they are doing and may instantly develop an opinion on whether or not you think it is "good" and will be able to explain why it is or isn't good.

At that point you are now the "master" of that topic. You don't have the degree but you didn't have to write a thesis either so there is that :-). If the university you chose in step 1 has professors doing stuff in the area you could probably co-write papers with them.

Yep have done! Was hoping it was a little more accessible though. Luckily I do have some library access.

Sorry to disappoint, if it was easier we would perhaps have a wider base of expertise on which to rely upon. On the plus side, the more areas you develop the faster the new ones are to develop as the base subject (math, physics, chemistry, biology, Etc.) overlap a bunch of different disciplines. I've been working through some tomes on Organic Chemistry for a plastic waste processing idea/project and that is a pretty slow slog. Beyond its use in semiconductors I've not been a huge chemistry fan.

Good practice should really follow the joke about airliners: the ideal flight deck crew consists of pilot and dog. The pilot is there to look after the dog, the dog is there to make sure the pilot doesn't touch anything.

Even the RBMK reactors, which had lecturers in the 1960s pointing out safety flaws, needed human intervention (and quite a lot of materials fraud) to actually go boom. TMI similarly needed some humans to fix it until it was broken (also why there is a genuine good-faith lobbying argument to mandate less maintenance sometimes, especially as higher-quality equipment wears slower).

You'd still need security, and frankly security for the security, though.

Reply on news.ycombinator.com