Bill, I've never used (or even thought much about) heat pipes. Are the contaminating gases bad only because they cause more scattering of the 'working' gas, or is there something else going on too?
Do you always need a wick? If you arange things such that the cold end is above the hot end then gravity could do the 'liquid return'.
Do you design it for a specific temperature? Or could I go from, say, room temperature to 100 C, with a single unit?
Hey maybe even a liquid nitrogen heat pipe.
Sorry for all the questions, don't fell like you have to answer them all. Do you know a good heat pipe reference? (There seems to be a lot of architectural stuff on the web.)
I made a low power switcher around 350kHz last year, and was amazed at the voltage drops along tracks: significant fractions of a volt in a few mm. Apart from the obvious tip (use fat traces to minimise L and dc R,), the skin effect was cutting in. I sorted it out on the prototype board by tinning those tracks. So 2oz copper, or tracks you can tin, is a really good idea for your high current, high freq switcher tracks (ie between switcher IC and smoothing capacitors) and for the power levels you're talking about, could significantly reduce heat dissipation.
contaminating gases are just dead space, and as you get more and more the pressure inside builds up and the boiling point of the working fluid goes up
there maybe a limit to how small you can make the pipe, but gravity return heat pipes are used to extract heat from the ground and keep permafrost buildings etc.
it will work in a range, below some temperature the liquid will no longer boil above some temperature all the liquid is vapour or the pressure is so high it can no longer condense
you should be able use pretty much any fluid it is just a matter of finding some that has a reasonable pressure at the working temperatures
I'm sure google has some, might also look at refrigeration diagrams pv,pt,ph
I think that's it, but that "scattering" of the working gas really makes a difference - when I was distlling from one cold trap to the next as a graduate student, and all I had was condensable gas, you could hear it whistling as it went through the pipe-work and the liquid nitrogen around the receiving vessel would boil furiously. Any kind of vacuum leak and that didn't happen.
I think the wick also increases the surface area doing the evaporation at the hot end.
They don't work too well if the working fluid freezes at the cold end of the pipe, and you might need excessive pressures of water vapour to get water to condense if the cold end was much warmer than 100C (373.15K).
For the right source and sink temperatures.
Never seen one.
The wikipedia article
formatting link
is pretty good. It mentions stuff that I'd never heard of.
The only time I personally ordered a heat pipe was back in 1985, and I never got the chance to try it. I knew that we were eventually going to need something like it, and when we did need it, the engineer who'd taken over nursing the machine did try it, but it didn't make a great deal of difference, and I never got to hear what it actually did, or didn't do. It was intended to work close to room temperature and it might well have contained too much non-condensable gas to work well with only 15 torr of water vapour pressure.
That not the whole story. The condensable vapour has to diffuse through the non-condensable gas, which gets pumped down to the condensing end of the heat pipe, so you've got a gas concentration gradient in the pipe as well as the extra pressure.
Thanks Bill, A liquid nitrogen heat pipe may be the perfect solution to a problem. A foot long (0.3 m) piece of copper, with daimeter x, is a "comprimise" as a heat conductor. I wonder what the time constant of a heat pipe is? At low pressure it could fly, the molecules traveling with out scattering. (at least those that start in the right direction.)
I read a post from some Ham (radio amateur) who had a transmitter that was cooled by a steam heat pipe that was open to the air. (or maybe it was second-hand info, I don't recall.) I think they call this "vapor cooling." What the poster claimed was that when the Xmitter tubes (vacuum tubes, or valves) reached the boiling point, the boiling of the water would cool the tube envelopes (or at least hold them to 212 F), and the steam would rise through a pipe (like a chimney) to the roof where he had an arrangement that, for lack of a better term, was a sort of a "cooling tower;" the steam rose, got to this box with slats, where it would condense and the water would run back down the "chimney" and get reused.
I'm not an "admitted communist". Are you posing as krw or being stupid on your own account?
And while Libertarianism may be - in part - about individual liberty, it is primarily about the elimination of the state. Since human evolution sems to have been concentrated on evolving the skills which allow us to cooperate in larger social groups than the rest of the great apes, this is to reject what makes us human. Significantly, the wkipedia article
formatting link
cites Herbert "nature red in tooth and claw" Spencer as one of the thinkers who formed Libertarian political philosophy.
In so far as it justifies tax evasion, it was one of the justifications claimed by the founding tax evaders.
No, I want more effective and efficient cooperation. Humans can cooperate better than any other vertebrate, and rejecting that skill seems to me to be profoundly stupid. And - of course - better cooperators can make mincemeat of groups who can't coordinate effective action.
No. Not even close. *Some* Libertarians maintain anarchist leanings, but there are also Minarchists - seeking the minimal state. And there are other stripes.
If anything, it rejects the assumption that social bonds can be expressed formally or through government.
It cites Spencer as one writer referred to by another.
Anarchism *assumes* that the social nature of man makes the State unnecessary. Spencer was a Positivist, and really doesn't have any place at the table with modern Libertarians.
never mind that anarchists are the most .. anachronistic of the lot.
Not all of Spencer is useless; Social Statics is a good start.
Then you want markets as much as possible and Ricardian rents as little as possible. Modern government does not arrange for those.
As soon as you show *any* interest in efficiency, you're headed directly into the arms of Libertarianism - almost full stop.
There certainly are cases of clearly delineated public goods - and even public distribution of private goods - which are more efficient than market solutions, but it frequently takes the better part of a century to achieve that. US famr policy probably mostly produces better outcomes than pure market, for example ( there is overproducton that is "cheaper" than monetary insurance against shortfall ) but you're talking 80 plus years of goofing around with it, and it's not very clear that all negative externalities are accounted for.
But Libertarians don't reject cooperation. Not even Rand.
Coordination isn't that hard. Design is. Seriously, if we really were interested in coordination, we would have a nation of logisticists. We don't.
Only if you share the right-wing delusion that free markets automatically generate the optimal distribution of resources. As Keynes pointed out, they don't. Participants in the market don't - in fact - have perfect knowledge of the state of the market, and human participating in markets don't behave rationally, though some of their "irrationalities" - like spending money on punishing cheats - work to make the market more nearly optimal.
In the real world, the free market has to be supervised and policed, to protect against fraud, monopoly power and "irrational exuberance" amongst other problems. Libertarians don't get to first base when it comes to seting up that kind of supervision.
Actually, it comes a lot closer than any Libertian solution would.
Only if you believe in the perfection of tyhe unsupervised free market, which is the economic equialent of the perpetual motion machine.
Any more than they are in less regulated markets.
Just the prerequisites for truly effective cooperation.
It does *for truly free markets*. Truly free markets are pretty scarce. Part of that is people who are identified as capitalists who very much are not - they seek rents based on government power.
They don't have to. Markets are tiered and it works pretty well when people specialize.
So what? For hu-mans, "irrationality" is one thing they do.
Yes. To an extent - you need at the very least criminal and contract law.
... a Libertarian stalwart, along with force, and ....
monopoly power
Doesn't actually exist in nature. I know, I was as surprised as you...
That one I might grant you, although diversity is a better guard against that sort of thing than monolithic systems that fail all at once.
See also David Friedman, who gets well beyond first base. I will stipulate to this bein in fairly early days - so there's some validity to what you are saying. Over time, though...
No, it simply creates rents.
So you don't understand it, then? Again, this is a pure information theory thing, and nobody who understands it disagrees with it.
Markets get there faster when they're kept clean and free. Doesn't happen often... but we're tlaking oodles of bandwidth for markets, and almost none for Commanding Heights mechanisms.
Whatever you mean by "effective". I expect that means "to a social standard", which isn't a very good standard. People are irrational, you know? :)
No, we do not, sir. We have in essence lawyers and lawyer-wannabees. People who think that they can solve flow problems by the use of declarative logic ( for lack of a better analog ). By rule-making. Feh.
It would probably fail in a year or so. Or a month. Making reliable heat pipes turns out to be difficult.
And a heat pipe just moves heat. You've still got to get rid of it. A copper heat spreader on a heat sink might do a better job, and have an MTBF of a million years.
If everyone were sane, and had a healthy sense of self-respect and self- determination, there would be no need for "the state." Admittedly, there would be a "need" for someone to coordinate road-building, and sewage- schlepping, but other than that, not so much.
But I guess we've got a ways to go, more's the pity.
Public health care (vaccinations, quaranteening people with infectious diseases and the like), defence, a tax system to collect the money to pay for all of this, a police force ...
In order for the market to be perfectly optimal, the agents involved need to know everything affecting the market as soon as it happens. No amount of specialisation is going to make that possible.
Well, it does destroy the mathemtical proof that markets will deliver an optimal distribution of resources.
In your dreams.
Some things very like it do exist - Microsoft come to mind.
Diversity it perfectly useless. In fact the obvious sign that a bubble is close to bursting is the way that your neighbours start teling you that you could make money out of it. Once that level of "diversity" has been reached, your are into the pyramid selling phase of the game and the burst is incipient.
Everybody understands the proof, and everybody with any sense understands that the proof applies to a situation that can't exist in the real world.
,
People have an "irrational" enthusiasm for punishing free-loaders, but this isn't their only irrationality. Imperfect knowledge in real markets is effectively a delay around the control loop, and without tolerably sophisticated control mechanisms feedback loops with delay are prone to oscillation. In the market this is called boom and bust.
Keynes wasn't a lawyer, and his scheme didn't involve merely declarative logic.
As noted, a heat pipe doesn't perform miracles, it just moves heat. If it's literally a pipe, it will have thermal contact problems on both ends, and you may well wind up with a copper spreader on one end (to get the heat from the semiconductors into the pipe) and some radiator structure on the other end, with fins and fans. Net benefit of adding the heat pipe may be zero or less.
Heat pipes are a favorite of armchair engineers. Faced with real semicondictors, real dissipations, and real economics, they rarely make sense.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.