Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

For those curious what the 385% refers to

> the team’s demonstration device can achieve an overall efficiency of 385 percent in converting the energy of sunlight into the energy of water evaporation.

Honestly I still don't know what that means, or how efficiency can be over 100%.



Usually, numbers over 100% mean that you are putting in less energy than is needed for the process. In this case, that would mean putting in a little over a quarter of the needed energy. That does not imply free energy or anything. The rest of the energy has to come from the environment.

A heat pump is another common example with efficiency numbers in the same ballpark. With a heat pump, the heat is being moved from outside to inside (or vice versa for air conditioning and refrigerators). In that case, it requires, for example, 1kW of electricity input to move 3.85kW of heat.


Efficiency is still the wrong name. What you are describing is usually referred to a coefficient of performance, abbreviated to COP.

See https://en.wikipedia.org/wiki/Coefficient_of_performance


The Deja vu sensation of this conversation happening almost identically 2 days ago (everything from wondering how >100% efficiency works, someone explaining it's from the environment, and then someone else explaining efficiency is inappropriate and COP exists) is kinda wild.


Nothing to worry about. Its people alpha-testing the Copilot internet commenting plugin.


Can't wait for all our text input boxes on the web to be GPT-3-enabled, and then all comments on HN and Reddit and Twitter will be just people accepting the defaults, and it will end up just being GPT-3 talking to itself, and we can all go back to doing something productive.


Pretty sure FB has several instances of itself with all the users played by GPT-3 trained on the users’ previous activities so they can monte carlo various changes, like pre-A|B testing or estimating impact of various new ad types or congressional testimonies.


Reminds me of a post here where it was some neural network chat/excuse generator. Just two people back and forthing about why they couldn’t meet up.

I lost it when one of the excuses was, 3 weeks later “oh, I can’t meet for lunch on July 29th, I have to go to my mother’s funeral”


Amy way you can link to that conversation? I tried to find it on Google.


I think the convo is different each time. You’d have to search here for the chat itself. It was 2-3 months Ago


Relevant xkcd: https://xkcd.com/810/


I wonder if concentrating solar cells should have similar nomenclature.


The 100% level refers to a system where all energy goes to heating up water in order to evaporate it, and then letting it cool down to condense it. All the energy that was spent to heat up the water, is lost to the environment in the "cooling down" step.

If some of the heat is instead recovered during condensation to heat up the next batch of water, then you have >100% efficiency.


Does it have to be batches?

Can't it be like the reverse of a rocket engine where they use regenerative cooling from the fuel to cool the rocket nozzle, but just in reverse.

Or like they way my grandpa was doing moonshine -- the alcohol vapors pass through a serpentine in a water tank, condense, at the end you obtain alcohol, the water in the tank gets warmer -- instead, heat water coming from the water cooling tank that is preheated by the vapors of water that is condensing in the serpentine pipe.


Unless you ignore energy sources from the environment you cannot exceed 100% efficiency. And that would be incorrect, applying that same standard to photovoltaic panels would result in infinite efficiency. That doesn't make any sense. Edit: Also, when you take heat/energy from a previous step in the process, you also need to account for the energy put into that previous step. In the end that will again be <100%.

As someone else already pointed out, this would be called Coefficient of Performance. Efficiency is clearly defined and cannot exceed 100% without breaking laws of physics. Call it "3.85 times more efficient than before" or something along those lines and it won't sound like a free energy claim.


> applying that same standard to photovoltaic panels would result in infinite efficiency

No it wouldn't. The equivalent for panels would be like... you want to run some number of watts through a diode, and you're using solar panels to collect this power. The diode happens to give off waste light. By aiming this light at your panels, you can recapture most of it back into electricity, and reuse it 2.85 more times.


the full sentence is:

the [..] device can achieve an overall efficiency of 385 percent in converting the energy of sunlight into the energy of water evaporation.

it seems quite clear how the 3.85 ratio is obtained


Others already explained where the >100% efficiency comes from, but I want to point out that a good 1/4th of the article is repeatedly explaining how this works, over a couple of paragraphs.


Reading it in the comments was still more efficient than reading the article.


220% more efficient.


My understanding: As the water condenses onto the next surface layer in the stack, the solar heat is recycled. This is because the transition from gas to liquid releases heat.

Really clever stuff!

Edit: mixed up evaporate/condense


For that to work, each successive layer will need to remain cool enough for water to condense on it.


It’s due to changes in pressure. https://upload.wikimedia.org/wikipedia/commons/thumb/3/33/Ph...

At 5ATM water condenses at a higher temperature than it boils at 4ATM.


I think it means 3.85x more water is evaporated compared to just leaving the water outside in the sun.


The paper[0] goes into more detail about how the efficiency is calculated (scroll down a bit to see the actual formula).

[0] https://pubs.rsc.org/en/content/articlehtml/2020/ee/c9ee0412...


It quantifies the heat re-used between stages. In the supplement, they note 600% is the maximum. The derivation is at the bottom of p834 from the journal pdf.

It's basically vapor produced at the measured average temperature divided by energy input.


If 600% is the maximum, then what does this mean?

> Theoretically, with more desalination stages and further optimization, such systems could reach overall efficiency levels as high as 700 or 800 percent, Zhang says.


I think it tracks how well the materials they're using move heat between stages or lose it to the atmosphere. Their modelling (supplement figure 2, mentioned on page 4) depends on their specific construction. I wonder what it would do with gold as the plate material and aerogel everywhere else...


Exposing a deep body of water to sunlight is probably the reference.

edit: deep as in practically having unlimited heat capacity and the heat conductivity of water.


I think maybe it means "3.85 times as efficient as sunlight in evaporative filtering" ?


> Honestly I still don't know what that means, or how efficiency can be over 100%.

They explain it in their article[1]:

"the solar-to-vapor conversion efficiency, defined as the ratio of total vaporization enthalpy to total solar energy input, for most previous studies has been limited to below 100% as the vaporization enthalpy is lost to the ambient environment."

[1] https://pubs.rsc.org/en/content/articlelanding/2020/ee/c9ee0...


My guess is that they get 3.85 times as much water evaporated as the energy used to just normally evaporate water. They did that by also using the energy from condensation to be put in the process again.


Its disingenuous. We know what the theoretical best efficiency of desalination is, by thermodynamics. You can calculate it by assuming a completely ideal reverse osmosis setup. Compared with that, the efficiency would be less than 100%. This article takes 'efficiency' as compared to just evaporating the water and not reclaiming any energy on condensation.


I thought it would be something impressive like somehow using the obtained salt for further powering of the device but nope, another clickbait trip into bullshitland


I'm not even sure what 'the energy of water evaporation' means? Gravitational potential of the mass of water evaporated (and condensed at some height)?


I think they mean latent heat/enthalpy of evaporation.


I think of it as taking the combustible fuel energy required for a car's motor to drive it somewhere, divided by the energy required to turn the ignition key and press the pedal.


It's an example of a good old scientifically flavored clickbait.

The energy efficiency of anything cannot exceed 100% (until I missed something groundbreaking in physics).


How can it be "click bait" if neither the title nor the headline contains that number? You have to read at least the first three paragraphs of the article to stumble about that number. The focus of the article is on the inexpensive design, not on its efficiency


The HN title has been changed. It used to refer to the 385 percent claim. Probably GP thought this was from the title.


It very much can exceed 100% and it’s not a measure invented for click bait purposes.

Look up heat pumps. You can make things hotter by moving heat than you could by directly heating it by burning fuel.


As another comment pointed out, it’s usually called the Coefficient of Performance in this case.

Which is to say this is click bait, because saying it has a COP of 3.85 isn’t anywhere near as sexy, whilst being technically more accurate.


The ‘more than 100%’ is only in comparison to a less efficient process. A heat pump itself is operating below 100% efficiency - it’s just kind of “cheating” by using an external source of external energy as one of its inputs in addition to the electricity running it.


I guess water evaporates naturally, adding sunlight speed up the process.


And my perpetual motion machine uses naturally running hamsters.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: