Talk:Rayleigh–Bénard convection/Archive 1

http://www.physiology.rwth-aachen.de/user/jaeger/diplom/index_e.html [broken link] is a really good to include. I post it only in discussion, because I don't know anything about the legal stuff.

"Memory"
These systems do not exhibit a "memory", they exhibit behaviour that can loosely, in a very simple sense, by understood by analogy to memory. Memory is not supported by any of the refs on the page (I've checked them all and been banned as a robot from arXive for my trouble!). The term inertia is a more accurate analogy, bt this all still needs supporting by refs. --DrEightyEight (talk) 10:03, 1 February 2008 (UTC)


 * They exhibit hysteresis. I think the question is whether hysteresis can be characterized as being memory-like. I do not think inertia is related to hysteresis and I think the current text is not correct as a matter of fact. &mdash;Whig (talk) 10:10, 1 February 2008 (UTC)
 * No, the question is in this instance can it be characterised as being "memory"-like. The answer is, only in a very simplistic (and slightly "mystical") sense. Inertia seems to be a better description (not bothered by small perturbations, only by large ones), but either needs supporting by sources. "Inertia-like" is a much better simplification --RDOlivaw (talk) 10:31, 1 February 2008 (UTC)
 * With all due respect I think you are either trying to redefine hysteresis or deny that B&eacute;nard cells exhibit hysteresis. It is alright to flag a fact as needing a citation, but changing it to a contested and probably wrong fact without any source backing it seems to be a violation of RS & V. But let's see what the editors of Hysteresis say about inertia being the same thing or a better simplification. &mdash;Whig (talk) 11:35, 1 February 2008 (UTC)
 * In this case, inertia or momentum are better analogies (I think). Hysteresis isn't a synonym for memory, and can't be accurately described as "memory" either --RDOlivaw (talk) 11:45, 1 February 2008 (UTC)
 * Let's agree to punt to Hysteresis. This seems like it was a reasonable edit. &mdash;Whig (talk) 18:51, 1 February 2008 (UTC)
 * Interesting discussion - I remember working with people making Bubble memory in the early 80s ! Magnetic cells, not convection cells, but similar ? I think I have the odd wafer somewhere !
 * Hysteresis is the physical effect, memory is the application to Computing. See also magnetic core memory.
 * --195.137.93.171 (talk) 00:16, 13 September 2008 (UTC)

Needs simple explanation
I wonder if there is any simple model or intuitive description for why the convection cells are (in the absence of turbulence and other chaotic effects) regular hexagonal, as opposed to other regular tilings or tilings of nonregular polygons? If there is, I think it should be added to the article.

A possible simple model is to start with cylindrical cells. When they press against each other, they pack together with body-centered packing (as opposed to the unstable regular packing where their centers form a rectangular array) to reduce potential (configuration) energy, and from there they obviously must deform to hexagonal prisms (reference, diagram, imagine the deformation happening slowly: the straight lines of the hexagons form from the middle of each edge due to equal pressure from adjacent cylinders).

The only thing missing is an intuitive explanation for why convection should happen in cells of a certain size, as opposed to one large cell. Is it because as the fluid cools it tends to descend a certain distance from its origin?

Examples of almost-hexagonal Bénard cells that could be added to the article are the crack patterns sometimes seen when mud dries out and the pattern seen in fluidic thermal convection. David (talk) 13:37, 26 May 2008 (UTC)


 * I suspect mud may be primarily tension cracking, like old paint, with convection cells maybe just giving tiny lines of weakness - see Talk:Giant%27s_Causeway and Wikipedia_talk:WikiProject_Rocks_and_minerals. We could really do with a photo of a chip pan or something. I would suggest linking to Giant's Causeway, but geologists assume tension cracking, rather than convection cells ... Incidentally if you actually look at mud, it's as square as it is hexagonal !


 * Size of cell will affect convection speed - large cells will take too long to circulate: small ones will cause too much shear-drag within each cell between up & down streams. A tougher question is 'Why does the centre of the cell rise and the edge sink? Why don't inverted cells form ?' Maybe they do ? Or the containers in experiments have cooler sides ? By the way, the diagram is ambiguous - really needs to be 3D. It would also be more believable if it had an even number of circules - half-a-cell will not exist stably !
 * Maybe that is what is meant by 'Hysteresis' in the discussion above ? If you have a single-cell container, heating the walls or cooling them could reverse the toroidal flow, which would be maintained, even if the wall-temperature returns to the fluid temperature or slightly beyond ?
 * --195.137.93.171 (talk) 00:30, 13 September 2008 (UTC)

Move to singular
Is there a reason why this article had a plural title, "Bénard cells", until I changed it a moment ago? That clearly violates WP:MOS except when there's a special reason to use the plural (e.g. The Beatles, Cauchy–Riemann equations, etc.). Michael Hardy (talk) 20:51, 28 February 2009 (UTC)

Move of page content
Note - I have just moved around some content related to this topic. For an explanation, please see Talk:Rayleigh-Bénard convection. Thanks David Hollman (Talk) 21:45, 7 September 2010 (UTC)