Talk:Viscosity

From Wikipedia, the free encyclopedia

WikiProject Physics
Portal:Physics
 Physics Portal
This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.
B This article has been rated as b-Class on the assessment scale.
High This article is on a subject of high importance within physics.

This article has been rated but has no comments. If appropriate, please review the article and leave comments here to identify the strengths and weaknesses of the article and what work it will need.

This is the talk page for discussing improvements to the Viscosity article.
This is not a forum for general discussion about the article's subject.
Article policies

Contents

[edit] ASTM Cps

  • What is ASTM Cps measured used for viscosity???

[edit] Formula Explanation

  • I'd like to see explained the formulas of dynamic viscosity η (1 Pa·s = 1 N·s/m2 = 1 kg/m·s) and kinematic viscosity ν (m2/s). I understand what viscosity is but I don't understand much about the formulas. I'd like to understand what they represent -in common language- and how viscosity can be measured or calculated.

==Bulk Viscosity Can anyone explain what bulk viscosity is? I ought to know--I work in compressible flow and shocks but a simple explanation eludes me.

Could bulk viscosity refer to the dissipation that occurs in a fluid in pure dilatation? Diogenes 20:21, 21 Aug 2004 (UTC)

Dilatational/bulk viscosity is the "kappa" term in the "complete" general expression of Newtonian flow. The rest of the Newtonian equation comes fully from momentum balance; the kappa term is (I am not certain on this point) an after-the-fact correction. It accounts for the resistance, due to molecular interaction, of the fluid particles to expand or contract. In texts I've seen, it is an order of magnitude smaller than the viscosity, difficult to measure, and generally ignored, even in compressible flow. Alchemy3083 07:25, 10 Nov 2004 (UTC)

I understand that the answer to some of the questions above is not simple but has been carefully put together, in the simplest possible terms, by S. Vogel in his "Life in Moving Fluids; the Physical Biology of Flow", 2nd. edition (1994). Princeton University Press.

[edit] Mistatement concerning kinematic viscosity

The statement

"In many situations, we are concerned with the ratio of inertial to viscous forces, the latter characterised by the fluid density ρ. This ratio is characterised by the kinematic viscosity:"

would be correct if modified as follows:

In many situations, we are concerned with the ratio of

                                VISCOUS TO INERTIAL

forces, the latter characterised by the fluid density ρ. This ratio is characterised by the kinematic viscosity:

If no one comes up with a contrary argument in a few days, i'll edit the article elzorro

OK.  Done.   elzorro

[edit] Misstatement concerning inertial forces

This is incorrect. The kinematic viscosity can be thought of as the "constant of dissapative momentum" because it gives the equation τ=ν(ρv(x))/dy. But dividing viscosity by density does not tell you anything about inertial forces. Kinematic viscosity is not a viscous/inertial relationship.

The ratio of inertial to viscous forces is given by the Reynolds Number, which is determined through the constituitive momentum balance equation, such as the Navier Stokes Equations, of which Newton's law of viscosity forms a major part. Alchemy3083 07:42, 10 Nov 2004 (UTC)

[edit] Bulk viscousity

If you add pressure to a viscoelastic sample then it takes time for the volume to relax. In the old days they used to model this with a variant of the Maxwell model (See Maxwell_material). This model contains a term named the bulk viscousity. I don't think that bulk viscosity is relevant to this article.

[edit] Can solids have a viscosity?

The article should stress that viscosity is an equilibrium property.

It is correct that you can perform rheological experiments on a viscoelastic liquid, but in these experiments the ratio between strain rate and stress is not be constant in time. However if you wait long enough the ratio between strain rate and stress will approach a number. This number is called viscosity.

You cannot define the viscosity of a non equilibrium system. The problem is that viscousity is only defined in the long time limit, and in the long time limit the system is no longer at equilibrium.

Of course you can always define the viscousity as a parameter of a model. This may allow you to speak of the viscousity of a non-equilibrium system. However there is a difference between defining a quantity within a model and defining it as the result of a measurement.

Similarly you can describe linear perturbations of a viscoelastic system by allowing the viscousity to be a function of time. Perhaps some day in the future wikipedia will have an article on linear response, but until this happens please stress that viscousity is an equilibrium property.

Are you referring to rheids? (Sorry, I don't know much about them - I just found that page by random, which is why I got here.) I was wondering if there is a category that would fit for them in this context. Common Man 19:40, 22 May 2005 (UTC)

[edit] Viscosity and temperature

The article says:

Viscosity tends to fall with temperature.

Is this correct? I thought molasses was slower in January than in July. -- Dominus 10:08, 26 Sep 2004 (UTC)

I clarified this point by changing the above sentence to:
Viscosity tends to fall as temperature increases.
H Padleckas 18:59, 1 Oct 2004 (UTC)
What is the viscosity of molasses in January? (say -10 C) How about adding a table showing the viscosity of molasses vs temperature. --69.5.156.155 21:54, 24 Nov 2004 (UTC)

I would say viscosity goes down when temperature goes up is just as obvious as thinking that viscosity, in reality, is just a reversed speed measurement.--Cacumer 20:21, 18 February 2006 (UTC)

[edit] Viscosity of Liquids

I'm not sure that the value given for glycerol is correct the CRC handbook list 934 milliPascals second Where did the value on this page come from?

It is 954 mp at 25C and 1490 mp at 20C (according to handbook of chemistry and physics 69th edition)

It is not only the viscosity of glycerol, the viscosities fot methanol, ethanol are also wrong if one compares them with those in the CRC book. Whoever publishes this type of data should put a reference...


The value for pitch disagrees with that given at ref. http://xtronics.com/reference/viscosity.htm. Not sure which is correct. —The preceding unsigned comment was added by 128.250.204.118 (talk) 08:09, 21 March 2007 (UTC).

[edit] the flowing glass urban myth

Is in fact not an urban myth. One can clearly observe the effects on middle-age vitrals of cathedrals...

--CyrilleDunant 15:48, 18 May 2005 (UTC)

What one can observe on middle-age vitrals is indeed at the origin of the myth. However, the correct explanation is not flowing glass, but a consequence of the Crown glass process used to build the vitrals. Jerome.Abela 10:30, 14 August 2006 (UTC)


I agree with Cyrille, why is that called an urban myth? I've not studied dinamics of fluids, but I've heard of that. I'll try to ask someone expert in that issues, in my university there is a glass-engineering course, if I find usefull info I'll post it here. Afonso Silva 22:49, 19 May 2005 (UTC)


--

hmm... i have not yet seen any proof of that, meaning a real "measurement"; on the other hand, glass is a so-called "undercooled fluid" and one would expect it to be viscous - nontheless it break into shards which is rather a property of a crystal than a fluid. usually, viscosity in "glasses" (anything that is physically similar to glass) depends on temperature, at room temperature this would be somewhere arounf 10^13Pas. this means that if glass flows, it flows so slow that even after a few thousand years it would only result in a change of thickness comparable to the diameter of an atom (grab the formula and calculate for a plate of glass 1m x 1m and 1cm thick). the main reason that glass flow has not been observed yet is most likely that glass is also elastic to small stresses, and i would consider glass under its own weigth to be a small stress for window-sized pieces. the reason that church windows are not homogeniously thick is more likely to be found in the method of making glass back then: use a plate of metal, dip the edge into the melt and pull it out to get a sheet of glass. this will never result in a homogenious glass plate. also, soldering such a window includes heating the glass which may lower its viscosity by many orders of magnitude until it cools, especially if its a low melting point glass --Sparks25 14:46, 26 July 2005 (UTC)

    • Even the earth's mantle, with a viscosity 10^21 Pa s is often considered a viscous fluid, so why wouldn't we consider glas one? That it cannot be seen flowing in one's lifetime, doesn't have any physical meaning, we are not the measure of everything. Thijs!

The technical answer is that glass is a fluid, but I think some of the wording in the article needs to be changed so that the casual reader does not come away with the impression that glass behaves like a fluid on any scale that is observable on an every day basis. Carl 18:42, 27 February 2007 (UTC)

On 2nd reading I think the article gets the point across, just in need of some minor edits. Carl 19:23, 27 February 2007 (UTC)

[edit] Significance

--LouisBodo 23:00, 20 July 2005 (UTC)

I would suggest the inclusion of a section just before Contents. My suggestion is the following. Significance Viscosity is an inherent property of a fluid. Knowledge of its magnitude is indispensable for calculation of rates of flow in ducts, channels, etc, calculation of energy requirements of machines conceived for moving fluids, such as pumps, mixers, etc and in lubrication. In hydrodynamic lubrication at relatively low loads and high speeds (journal bearings) the load is supported (surfaces kept apart) by viscous forces within the lubricant which depend on their viscosity. Pressure drop in ducts is a function of the Reynold's number, which depends critically on the viscosity of the flowing liquid.

            Further I would like to stress the importance of the effect of temperature on the magnitude of viscosity, saying.

[edit] Temperature dependence of viscosity

Because viscosity of a fluid is critically dependent on temperature it is important that its value is quoted for a fixed temperature and it is measured in thermostatically controlled environment (equipment and fluid kept constant within, say ±0.1°C).

For petroleum oils between a limited range of temperatures the change of viscosity with temperature is near logarithmic. ASTM (American Society for Testing and Materials) has produced a chart somewhat better than an ordinary logarithmic chart, with a near linear temperature function, on which the slope of the viscosity-temperature lines of petroleum lubricants can be suitably compared. The smaller the slope of the line representing the viscosity change of the oil, the better it is, as its viscosity decreases less due to temperature rise caused by internal friction.


[edit] Eddy Viscosity

Would someone be so kind as to expand this section, and maybe add some references?

[edit] Spelling error on figure

Sadly, the figure describing Newton's theory has a spelling error, twice reading "boundry" where it should say "boundary". Since it is such a beautiful figure, it would be great if the original contributor could correct these small errors, which is a far easier job (if the original file still exists) than re-doing the whole thing. Is it too much to hope that (s)he may one day check back here? And is this a deficiency in the Wikipedia concept (that the original file is not available for easy editing)? 144.213.253.14 06:12, 3 October 2005 (UTC)

I will try to fix this problem now. H Padleckas 16:16, 3 October 2005 (UTC)
IAMANIDIOT, thanks H Padleckas. --Duk 17:07, 3 October 2005 (UTC)
You're welcome. I have fixed the diagram and uploaded the corrected file to Image:Laminar_shear.png. The corrected file usually does not show up immediately, but only after a few days; so let's check back in a couple of days.
H Padleckas 17:17, 3 October 2005 (UTC)
I still don't see these changes I made to Image:Laminar_shear.png, but I will wait further. H Padleckas 00:51, 7 October 2005 (UTC)
Your update went through fine, I saw it right away. You probably need to purge you browser cache or do a forced reload of the page. Also try http://en.wikipedia.org/w/index.php?title=Viscosity&action=purge. This purges the wikipedia squids. --Duk 02:42, 7 October 2005 (UTC)
Today I checked this picture and I see the corrections finally came through. H Padleckas 00:21, 11 October 2005 (UTC)

[edit] Misleading figure

The figure showing the parallel plate setup is misleading, as it seems to indicate that the shear stress changes along the y axis. --Slashme 05:44, 8 December 2005 (UTC)

I agree, particularly about the second picture (Laminar_shear_flow.PNG). It looks like a graph, but it's actually a diagram. I think adding some color to the fluid and/or the plates, and maybe using short arrows that look like vectors rather than axes would help. I can't fix the pictures myself, but I hope someone can. —Mister K 18:43, 22 September 2006 (UTC)

[edit] unitconversion.com

Do we really need four links to a single website or is it a link spam? abakharev 23:25, 10 October 2005 (UTC)

[edit] flowing

I must say, to me, all this discussion is blinded looking to the wrong side.

Viscosity seems to be an attempt to measure how much something can flow.

Anything can flow, flowing doesn't depend on vicosity, it's the other way around.

Flowing is movement of one thing from A to B. There are flowing of cells, molecules, atoms. As long as it's not zero Kelvin, it is flowing, at some level.

Now, if it's hard to measure pitch flowing, and if glass do flow, how can one tell about flowing metals?

The only way will be when some easier way to measure that pops around.

--Cacumer 17:54, 18 February 2006 (UTC)

[edit] viscosity

This is also looking to the wrong side. Since viscosity comes from flowing, we should first understand flowing and then we can understand viscosity.

I believe viscosity is overlooked. There's something in there, and it does not apply just to gas or just to liquid. If it applies to solid, I think it would also not stop there. Who knows how many states of matter are possible, in fact? We just know there are 4 today, but we didn't know that yesterday. Keep in mind that plasma is a state more energetic than gas, and it's hard to understand how it works.

If it is a measure of flowing, then it can work to measure any kind of movement.

Basically I think I would call flowing as molecule speed reversed.

--Cacumer 17:54, 18 February 2006 (UTC)


Why don't you find some sources which discuss that? And while we're at it, I'm putting a weasel words tag on the "solid viscosity" section. It shouldn't be too hard to dig up some sources for those views. Tenebrous 03:40, 16 April 2006 (UTC)

[edit] Error in equations

The equation that separates velocity gradient into a trace and a traceless term is incorrect. You probably want to change the left hand side into \partial_i v_j + \partial_j v_i and all the factors 1/3 into 2/3. According to Fluid Mechanics by Landau and Lifshitz, the equation of stress tensor should read \sigma_{ij}=-p \delta_{ij} + \eta \left(\partial_i v_j + \partial_j v_i - \frac{2}{3}\partial_k v_k \delta_{ij}\right) + \zeta \partial_k v_k \delta_{ij}, where ζ is called the cofficient of bulk viscosity (second viscosity in the book), and the combination of the second and third terms is called viscous stress tensor.


The update of the equations still contain errors. Note that the stress tensor is defined with a minus sign with respect to the momentum flux density tensor in Fluid Mechanics by Landau and Lifshitz.--Huaiyu Duan 19:41, 11 May 2006 (UTC)


Please show colour diagrams!

[edit] invisicd fluid

"an idealized fluid which has no resistance to shear stress is known as an ideal fluid (Symon 1971)." In the field of aerospace, we often use the word inviscid to refer to or model an ideal fluid (according to Symon). This word is not mentioned here. 21:40, 29 May 2006

So be bold and add it. --Slashme 05:52, 30 May 2006 (UTC)

[edit] Unclear paragraph inc typo (?)

This paragraph needs to be expanded with proper equations and more consistent variable names. (Not t for distance!)

The relationship between the shear stress and the velocity gradient can also be obtained by considering two plates closely spaced apart at a distance t. Assuming that the plates are very large, with a large area A, such that edge effects are neglected and that the lower plate is fixed, let a force F be applied to the upper plate. Incidentally, if this force causes the plate to move, the substance is concluded to be a fluid. The velocity of the moving plate and the top , the applied force is proportional to the area and velocity of the plate and inversely proportional to the distance between the plates. Combining these three relations results in the equation F = μ(AU/t). Where mu is the proportionality factor called the absolute viscosity (with units Pa-s or slugs/s-ft). The equation can be expressed in terms of shear stress; ρ = F/A = μ(U/t). U/t is the rate of angular deformation and can be written as an angular velocity, du/dy. Hence, through this method, the relation between the shear stress and the velocity gradient can be obtained.

This sentence in particular needs immediate editing:

The velocity of the moving plate and the top , the applied force is proportional to the area and velocity of the plate and inversely proportional to the distance between the plates.

I don't know what it intended to say so I can't do it myself.

Until this is fixed I'll add an "in need of expert attention" template to the top.

--cfp 10:16, 18 July 2006 (UTC)

I've given the paragraph a once-over. It had some bad mistakes. It looks as if some of the copy comes from a description of cup-and-bob viscometers, because they refer to angular velocity, not shear velocity. Let me know if you find it acceptable now. --Slashme 11:51, 19 July 2006 (UTC)
Great, thanks. (I've removed the "expert attention needed" template.) --cfp 12:29, 19 July 2006 (UTC)

I still think this article needs help. For one, it uses pitch as an example of a solid that flows when the experiment was done to prove that it is in fact a liquid, not a solid. Also, glass does NOT flow. Dragon of the Pants 21:46, 21 July 2006 (UTC)

K well feel free to sort it / to put the expert attention box back (^_^). I'm no physicist so I'm not really the best person to do it. --cfp 01:02, 22 July 2006 (UTC)

[edit] Teaching the concept of viscosity to undergraduates

The textbook I'm using as I've seen with other textbooks introduces the concept of viscosity by drawing analogies to solids and shear stress. I'm teaching this concept for the first time and I wanted to ask if there were clever ways of presenting the concept of viscosity to students in their first course in Fluid Mechanics? Any interesting figures or mpegs or anecdotes or examples that students can relate to would be much appreciated. Thanks.

[edit] Quark viscosity

would this be a good place to put a discussion of the quark viscosity (as opposed to behaving like a gas) from the Scientific American article in July or August of this year, and also under quarks?

I don't see why not as long as it's a small separate section and fully cited. --cfp 16:15, 14 August 2006 (UTC)

[edit] the viscosity of glass

since the article quotes the viscosity of glass then specifically says it needs a quote, i had a look arond the net and found:

http://www.a-m.de/englisch/lexikon/viskositaetglas.htm has a chart of glass viscosity at different temperatures and compositions - however there are a wide range of values given and none of them are close to the 10^40 quoted, although extendign the graphs you could get to about 10^30 by 0c.

http://web.umr.edu/~brow/PDF_viscosity.pdf has some detailed maths concerning viscosity of glass and mentions that the expected relaxation time for cathedral windows is about 10^32 years


http://hypertextbook.com/physics/matter/viscosity/ quotes the viscosity of glass at room temperature at 10^18-10^21


http://www.jstor.org/view/09501207/ap000455/00a00100/4?frame=noframe&userID=8266023c@uq.edu.au/01cc99333c00501ef9900&dpi=3&config=jstor shows glass having a viscosity of 10^16 at 450c agreeing with the first website - neiter of these sources extend their graphs to 0c and i have been unable to find any that do.

http://www.cmog.org/index.asp?pageId=745 (also mentions that the viscosity of lead might be around 10^11 - and glass 10^20)

This question about the myth of glass flow very decisively addressed in a paper in the Journal of Chemical Education back in 1962 (vol 39, p. 84-85), by David Dingledy.

just a few notes on the viscosity of glass - i wasn't sure how to incorperate these, as it almost seems like a diversion from actually talking about viscosity itself

i hope someone with a little more time and understanding than i can sort this mess out . . .

2c worth, Andrew

130.102.0.178 06:53, 28 September 2006 (UTC)

Can someone with more knowledge and a proper library explain the assumption made in the equations for viscosity that probably can not be made with glass? Carl 19:35, 27 February 2007 (UTC)

[edit] molecular origins

The viscosity of a system is determined by how molecules constituting the system interact.

I'm not sure "system" is the right word here. I was gonna replace it with "fluid," but then I wondered about fluids with components that weren't whole molecules, and confused myself. this section needs some work, I think. Ojcit 18:47, 2 October 2006 (UTC)

[edit] Merging Stokes (unit) and Poise

The information on the pages Stokes and Poise is very limited. Both pages are stubs. Much of the info there is repeated here. I propose to merge both pages with the page on Viscosity and have them redirect here instead. Mausy5043 07:48, 9 February 2007 (UTC)

Agree. Also include Pascal second for merger and delete category units of viscosity. -Myth (Talk) 14:50, 9 February 2007 (UTC)
Support! GGenov 12:58, 16 February 2007 (UTC)

'Tis done Mausy5043 17:06, 17 February 2007 (UTC)

[edit] Use of eta or mu for dynamic viscosity

The use of mu for dynamic viscosity is very common for many fields which do not use eta. Even as an engineer I was thrown off by the eta. This is a field-specific problem, and while IUPAC does govern chemistry, it does not govern engineering fields. As reference you can check most thermodynamic books, which will use μ.

[edit] Please look over for accuracy.

I was thinking about placing this near the top so there is a quick and concise explanation of what the units of viscosity mean. This is probably one of the top reason why someone would look at this page so I hope this additional text would help them find that information quickly. This is without having to read to far into the article.

Here is my 2nd try at it, I like this one better. I typed this one up after looking up viscosity in Physics 4th Edition 2nd printing, by Serway.

When looking at a value for viscosity the number that one most often sees is the coefficients of viscosity, simply put this is the ratio between the pressure exerted on the surface of a fluid, in the lateral or horizontal direction, to the change in velocity of the fluid as you move down in the fluid (this is what is referred to as a speed gradient). For example water has a viscosity of 1.0 x 10-3 Pa∙s and motor oil has a viscosity of 250 x 10-3 Pa∙s. pg 440 Physics For Scientist & Engineers 4th Edition 2nd Printing, Raymond A Serway, Saunders College Publishing 1996

Here is the first try, I include it here for compression and to make sure I got my facts right.

Viscosity for a fluid is expressed in two ways, dynamically or kinematical, dynamic is concerted about the pressure it takes to move the fluid, and kinematics is how fast an area of fluid moves. In Dynamics_(mechanics) Dynamic units what is being measured is pressure times time. If the fluid in question is placed between two plates the number tells you how much pressure in the lateral direction it takes to make the top plate move an equal distance as the thickness of the fluid in some amount of time. The SI units are Pa s or Pascal seconds. In Kinematic units what is being measured is area over time. This is to say, how quickly a given area of fluid is moving. The Si units are m2/s or meters squared per second.

Pleas look over to make sure that I not saying some thing wrong before I post to the article. Thanks Carl 19:51, 1 March 2007 (UTC)

Retrieved from "http://en.wikipedia.org../../../v/i/s/Talk%7EViscosity_46b7.html"