Slashdot Mirror
Heat Engines Shrunk By Seven Orders of Magnitude
KentuckyFC writes "The vast majority of motors that power our planes, trains, and automobiles are heat engines. They rely on the rapid expansion of gas as it heats up to generate movement. But attempts to shrink them by any significant amount have mostly ended in failure. Today, the smallest heat engines have a volume of some 10^7 cubic micrometers. Now group of Dutch engineers has built a heat engine that is seven orders of magnitude smaller than this. The engine consists of a piezoelectric bar that expands and contracts in the normal piezoelectric way. However it also heats up and cools at the same time causing a thermal expansion and contraction, which lags the piezoelectric displacement. By carefully choosing the frequency of the driving AC current, the Dutch team found a resonant effect in which the thermal expansion and contraction amplifies the mechanical motion, making it a true heat engine. Operating the thermodynamic cycle in reverse turns the device into a heat pump or refrigerator. The total volume of the device is just 0.5 cubic micrometres."
Yeah a Beowulf cluster of them
It's not like a heat pump turns heat into nothing. One side of a heat pump gets cold, the other side gets hot. At half a micron across, it's hard to see how such a device could help evacuate heat from a CPU.
Somehow "heat engine" directly translates into "internal combustion engine" for me. But this piece uses electricity, exactly how useful is that? This is bound to be less efficient than to use the electricity to just power an ordinary electric motor. I suppose scaling a motor down to that size might be kinda difficult, though, if that was the point, why emphasize that it is a heat engine?
How many beer cans fit in a 0.5 micrometers refrigerator?
Depends. Are we talking micro- or macrobrews?
You know how big a millimeter is, right? A micrometer is one thousandth the length of a millimeter.
A cubic micrometer is the volume occupied by a cube one micrometer on each side.
10^7 cubic micrometers would fill a cube about one-fifth of a millimeter on a side. Smaller than a pinhead.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
It is so small that it produces a very minimal amount of horsepower, which is not useful for any actual way.
Unless of course you have several billion of them on a gram sized object. If you can't see the value in jet powered ants you should turn in your nerd card.
Seven puppies were harmed during the making of this post.
Fun fact - Wolfram Alpha can serve as your 'self-checkout line' for things like this.
http://www.wolframalpha.com/input/?i=1+cubic+micrometer
Here's a bit of scale - a cubic micrometer is about the same size as a calibration bead for microscopy. A red blood cell is about 8 micrometers across. http://learn.genetics.utah.edu/content/begin/cells/scale/ Or, there's this video showing the "powers of ten" (also its title...): http://www.youtube.com/watch?v=A2cmlhfdxuY
Also, chemists work at these dimensions, too! (So do biologists. And others.) :*P Don't snub the other disciplines!!! Or I'll weep. And not gently, nor to a guitar.
Be careful of your thoughts; they could become words at any minute...
Assuming we had micro engines, we can take full advantege of many things that are better smaller than bigger.
For example, a small device that turns heat into power could power an IMPLANTABLE MEDICAL DEVICE using the bodies own heating/cooling systems? No more changing the battery for the pacemaker every
Then there are small flying devices. I am sure the military would love a flying camera the size of a real fly that uses the solar heat of the sun to power it.
Then there are phones and musical devices. Want one that uses half of its' own waste heat to recharge itself, perhaps doubling battery life?
excitingthingstodo.blogspot.com
Already done: see peltier device. They are already made to the correct size and probably better efficiency.
http://www.peltier-info.com/
http://en.wikipedia.org/wiki/Thermoelectric_cooling
about one-fifth of a millimeter on a side
That's about the thickness of a sheet of paper. (Round here, and probably in a lot of the world, the thickness and density of paper is specified, for instance "160 g/m^2, 200 micrometres".)
If the heat were produced externally it would be a sort of Stirling engine. So I guess one this size would be Sterling sliver.
Free Martian Whores!
And I despair of the lack of English education, specifically reading comprehension.
This isn't internal combustion, which is what your argument is based on. It uses the fact that solids expand and contract when heated and cooled, including some piezo materials.
Please read the summary *again*.
It's almost as pathetic as the idiots who assume heat engine == combustion engine.
0.00000000000000000852167911 beer kegs
If the fridge interior happens to be shaped optimally so that no space is wasted and the entire 0.5 micrometer fridge is filled with keg, then.. exactly 8.52167911 * 10^-18 beer kegs (if each keg is 15.5 gallons). [Incase someone wants to out-pedant me: Yeah, I understand you can't optimally shape a 0.5 micrometer fridge for a keg, when the size of 1 unit of keg > 0.5 micrometer fridge.]
Citation: http://www.google.com/search?hl=en&q=(0.5+micrometers%5E3)%2F(1+keg)&aq=f&aql=&aqi=&oq=
New webcomic updated on Sundays: HERE
It's the basic SCALING LAW that Galilleo figured out like 600 years ago.
As you make things smaller, their volume, which is their abilitry to burn fuel, goes down as the CUBE of its linear dimension.
But its surface area, which is how it loses heat, only goes down as the square.
That'd be Newton's law of cooling, no more than 300 years old.
Slashdot: news for Apple. Stuff that Apple.
Yes, someone explain how many of them would fit into the library of congress.
A metric assload.
Out of modpoints but really liked a post? 1BDkF6TtmmeZ3yqXbz9yhdYVqRYnwFoXDj
That's no help - most Slashdotters are American. What's that in imperial assloads?
++ Say to Elrond "Hello.".
Elrond says "No.". Elrond gives you some lunch.
It always amuses me when people try to raise performance as a point against a first generation lab prototype vs. a tenth generation refined technology in production. The question is not whether these piezoelectric heat engines/pumps are more efficient than peltiers now, but rather can they be more efficient than peltiers in the future after further development, or is there a foreseeable upper limit to the technology that makes such an application unlikely even with development?
There *is* a need for heat reduction at very small scales, especially in mobile devices or even the implant devices of the future. Of course heat has to go somewhere, the only issue is that the destination of the heat be better able to deal with it than the source.
I support the Slashcott and will not be reading or commenting from 2/10/14 to 2/17/14. Beta is steaming pile of dog shit
What has happened to Slashdot? Who do you have to be a Guru in every subject to read Slashdot?
Looks like gone are the days when all you needed to good discussions on Slashdot was genuine curiosity and decent , not necessarily perfect, grasp of English language. And no, being a know-all, done-all master of the universe was not required either.
While I can perfectly understand saying "You are making a mistake" or "That's not what the article says", I have never really understood calling someone pathetic for not knowing something.
The range of topics covered here is very wide and I don't know abc of several things discussed here. Does that make me stupid and pathetic?
If you've got a thing that small, it's time to give up on it...
The range of topics covered here is very wide and I don't know abc of several things discussed here. Does that make me stupid and pathetic?
The key point is that you recognize that you don't know everything about the topic at hand. The post that sunking2 was responding to was essentially a spew of vitriol against the researchers, claiming that it's impossible to make such a small engine with any sort of efficiency, and that they're stupid and ignorant for even trying. According to that post's replies, the writer is completely wrong and doesn't know some basic facts about the subject they're yelling about.
So, no, you're not at all stupid and pathetic for not knowing everything about everything, and I'm in the same boat with you (I've learned a fair amount from this story's discussion), but neither of us is telling everyone (including the Dutch engineers in question) that they're stupid and don't know what they're talking about.
WRONG!
Library of Congresses are a perfectly cromulent unit of volume. Just because the necessary measurements to derive the value are not easily google-able doesn't invalidate that fact.
In the past, when deriving the conversion from Library of Congresses to BTU's, we've used the assumption that we're talking about the books that make up the Library of Congress, not the building itself. This is because, back in the mists of time, Library of Congresses were originally used as a measure of information in the collection of the Library of Congress.
Anyhow, as a back-of-the-envelope estimate, 29 million books at 1" x 10" x 8" gives us a value of ~50,000 cubic yards. That gives us a value of ((10^7) (cubic micrometers)) / (50 000 (cubic yards)) = 2.61590124 × 10-16 Library of Congresses.
Screw this "metric system" with it's plethora of different units for different quantities. I strongly endorse that everybody normalize on Library of Congresses for units of any quantity. Just imagine how it would simplify your life!
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
I read the attached paper on arxiv, and from what I could tell, they passed a DC current through the thing, which caused the small engine beam to expand, causing it to heat up and move the mass. The piezoelectric effect causes the resistance in the small engine beam to change, which causes the beam to cool down and move the mass back with help from the larger spring beam. Rinse, repeat. Effectively a thermoelectric buzzer. The buzzing of this particular device was measured to be about 1.255 MHz at a DC current of 1.045 mA.
Unlike what the Technology Review article says, the paper shows no application of an AC current to get the thing vibrating. In fact, the measured voltage is alternating because the resistance is alternating. The current remains the same. There is no complicated application of a DC current and an AC current. There's just an applied DC current.
Am I understanding the paper correctly?
Towards the Singularity.