Thanks for you answer, Dear Judocus, focusing on other relevant aspects of the problem. In the case of very fine particles surface charge (and magnetic proprieties) is effectively an important variable, and for low diameter pipes, superficial tension must be considered too. However in the original question, I was thinking mostly about Newtonian fluids and turbulent flows, particularly water-sand (with low concentration of solids) in Wastewater Engineering and in the Petroleum Engineering (hydrocarbons -sand).

What I mis in your list of parameters is: surface-charge and magnetic properties. In medicine virtually all intravascular particles exhibit surface charge. In order to exploit any knowledge on transport of two phase liquid - solids in biomedical applications, surface charge should be included in the models.

Further more: when studying flow using MRI-techniques in the biosciences requires the inclusion of magnetic properties as wel.

This is not a hypothetical necessity, as research on nano-particles with non-neutral properties is already under way.

Thanks for you answer, Dear Judocus, focusing on other relevant aspects of the problem. In the case of very fine particles surface charge (and magnetic proprieties) is effectively an important variable, and for low diameter pipes, superficial tension must be considered too. However in the original question, I was thinking mostly about Newtonian fluids and turbulent flows, particularly water-sand (with low concentration of solids) in Wastewater Engineering and in the Petroleum Engineering (hydrocarbons -sand).

Then I'll be out of the loop. It seems that this is more of a thing for an engineer from Ballast Nedam to answer. These guys earn their living applying exactly the knowledge you ask for.

I hope that you continue participating, Dear Judocus. Your contribution and the issue that you raise in your first answer are also very well come to this discussion.

Hi Antonio,

there are two fundamentally different computational approaches: (a) Lagrangian, i.e. you have billions of real particles and model them by a couple of (say 1 mio) superparticles for which you compute trajectories following frictional force by the flowing fluid, and gravity, corresponding to ordinary diff eqns. A second approach is (b) to consider concentrations of particles. This leaves you within the same problem class like the Reynolds eqns - partial differential eqns. Both cases can be done for rough size spectra, e.g. say three classes, raw, finer, fine. Hope you can enjoy this topic. It's filling more than one evening ...

To continue on the biomedical field, I can give an example, to the large effects of interactions between cells, to exemplify that particle size and cellular coating/surface charge/properties do matter.: There is a rather old blood test: the erythrocyte sedimentation rate [ESR or in dutch BSE]]. Blood typically concists of 45% [by volume] of red bloodcells. This percentage is called hematocrite. In certain disease states and during pregnancy, when a blood sample is left standing in glass column for one hour, the sedimentation of erythrocytes is much , much enhanced. Theoretically in a 20 cm column, one would ultimately expect a sediment thickness of 9 cm [=45%]. Normally, when erythrocytes do not adhere, this sediment is only a few mm.

Erythrocytes [red bloodcells] do not adhere in this test because of clotting, because an anticoagulant is added.

In case of even mild inflammation anywhere in the body, the sedimentation rate rises rapidly.

This is mostly due to rouleaux formations: cells are stacked up like a stack of coins and thus act like larger particles, with an increased sedimentation rate as result.

This test was originally developed as the first pregnancy test ever. As you now understand, it was quite often false positive.

In case of severe anemia due to a decreased number of erythrocytes, the ESR is also low.

The proces of rouleaux formation is promoted by a large number of signal peptides that are released from sites of inflammation and affect all of the circulating blood. The effects of these peptides can to some extent also be measured by their effects on the membrane contents and by the behaviour of erythrocytes i.e. when being counted using Coulter counters [but here depending on the exact composition of the fluids used to dilute them, as normally Coulter counters are only optimized for the counting of erythrocytes.

From the fluid dynamics point of view, this model [stagnant fluid phase and particles sedimenting under the influence of gravity] is nice to show the large effects of particle size. The specific weight of rouleaux is identical to that of individual cells. Typically rouleaux of 5 to 20 erythrocytes are seen under the microscope and the effect on ESR is quite spectacular, increasing that speed by a factor of 20 or more. So it can be a large effect.

Judocus, I fully agree with you, but this is possibly not Antonio's point. He looks at turbulent flows wherein a steady-state equilibrium (in German language we use the word "Fließgleichgewicht" coined once nicely by Ludwig v. Bertalanffy, 1953 or so) exists between resuspension by bottom shear and deposition by gravity. Under varying external conditions like mean-flow shear, this equilibrium is shifted slightly towards the one or other end, and the laws acting behind these processes are sufficiently complex. Sorf of coagulations are often mediated by organic material in the flow. I.e. also here we have size spectra with certain inherent dynamics.

I would think that your medical application is relatively easy to model/compute. One needs to look at equilibrium size spectra which shift from an initial spectrum (when you fill the glas column) towards a final size spectrum close to the bottom of the column. If you wish to learn more about the physics you are welcome.

I once was studying some semesters biophysics at Humboldt university in Berlin under Roland Glaser and Paul Hoffmann, later in Moscow at Lomonossov university under A. A. Alexeyev whose grout prepared the Soviet mission to the planet mars where life-supporting systems via algae were at the center of interest. They gave up due to $$$ constraints in 1987 - default. Today this would be a nice new challenge in particular if superpowers would put their coins together: We need a challenge for human kind to distract from the ugly war games  with billions of casualties - give the dog a bone ...

@ Helmut

My example was just to underscore that the effects encountered in biological systems can be quite large.

You are totally right that my example was not about turbulent flow.

In a lot of bloodvessels turbulent flow is normal. For example in coronary arteries both laminar and turbulent flow is commonly encountered and can even be visualised using echo-doppler techniques.

Rouleaux formation has [almost] nothing to do with coagulation. I mentioned coagulation in a sense that coagulation was ruled out as a contributing factor.

This [and blood viscosity in general] is a major researchfield in cardiology.

I am currently not working in this field, but I do know for sure both that fluid modeling in the vessels of a beating heart is quite difficult to model [I left quite a few details out for simplicity] [I have worked with some of the technology needed for such research] and that especially in the field of atherosclerosis research a lot of $$$ is stil to be found for such research.

Such sedimentation is considered of importance for platelet aggregation and clot formation and the interaction between vesselwall irregularities and myocardial infarction.

The perfusion of a single coronary artery for example also depends on the electrical conduction system of the left ventricle.. For example in case of a left bundle branch block [a conduction disorder] the anteroseptal wall of the left ventricle may be perfused less, due to a phase delay in muscle relaxation, compared to the timing of the closure of the aortic valve, which in itself will effect the timing of the moment when the flow changes from laminar to turbulent in certain coronary arteries. 

All of this is not just physics, where I am sure you can learn me a few things, but also [patho]physiology.

Why is this subject so important for doctor  [i.e. for control through medication]?

Just count the deaths due to myocardial infarction........ or congestive heart failure...

And to go one step further: vessel walls are less stif than rigid piping as in industry. In humans we do know that the viscosity inside atheroma's in the vessel wall [atheroma's line the inside of affected vessel walls] themselves play a role both in turbulence and in the distribution of flow-speed across the cross section of a vessel and the probability of ultimately ending up with thrombosis. The flow-speed distribution can routinely be visualized using echo-doppler techniques.

Better understanding of [turbulent] flow and various interactions as mentioned may be of vital importance for further improvement in this subfield in medicine.

@Judocus

Please take my "coagulation" not too serious, it's kinda model notion standing for "initially separate but then sticking together". Thanks to your correction I realized that this is not the right notion. Fortunately it somehow came across what I have had in mind.

I am always curious and love to solve problems. So if you know somebody who needs support I would love to try what I can offer. The interface between bio and phys is challenging because of the great importance for man, as you wrote, and because it is an intellectual challenge to cover both fields, more or less and sufficiently. But again, Antonio is on another steamer and we should give him a chance to follow his agenda. To contact me separately from this platform simply use [email protected]

Thanks Dear Helmut for your assertive answers.

Hans Albert Einstein, 1904-1973 (first son of Albert Einstein), Civil Engineer and Professor of Hydraulic Engineering in the University of Berkeley, studied phenomena of sedimentology and solid transport in rivers. According to the link http://pt.wikipedia.org/wiki/Hans_Albert_Einstein “One of the father's advices was for him giving up of studying the phenomena of solid transport in rivers and dedicate to quantum physics, “because quantum physics was less complicated than sedimentology of rivers”.”

I must confess that I know a little more of water and hydraulic systems than what I know about blood and human circulatory system (where I suppose can exist also a strong interaction between blood flow and size of the “channels”, as pointed by Judocus). Dense slurries and heavy oil crude may be also non-Newtonian and due to higher viscosities and/or low velocities the flow may be also laminar. But in normal circumstances, in many practical Hydraulic Engineering applications, flows are frequently turbulent (smooth-pipe regimen, transitional turbulent or fully-rough).

http://library.ucr.edu/view/wrca/collections/portraits/einstein.html

In Engineering Practice of urban drainage of sanitary systems and storm water systems, the criteria normally used (respecting the official regulations) for pipe flow (normally force mains of pumping systems or inverted siphons) is to impose a minimum mean velocity in order to avoid sedimentation or to allow the re-suspension and transportation of solid materials. In force mains (see Manuals of Practice, for example), the minimum velocity imposed is normally between 0.6 m/s to 0.75 m/s, but sometimes values of 0.9 to 1m/s, 1.1 to 1.2 m/s, or more, are recommended. Portuguese regulation establishes as minimum 0.7 m/s. For inverted siphons, ASCE Manuals of Practice recommend that velocities between 0.6 m/s to 0.9m/s must be reached at least one time for day (0.7m/s to 1m/s are the corresponding values in the Portuguese regulation).  However, obstructions are frequently reported in inverted siphons and a permanent attention is normally necessary. Richard May (2003) in “preventing sediment deposition in inverted siphons”, in J. of Hydraulic Engineering, for example, even for very low volumetric fractions of solids, reported higher mean velocities required and also a strong dependence of the slope of the upward branch.

In addition to your answer regarding Newtonian fluids and turbulent flows: In the field of what is called "Civiele techniek" [civil engineering is the English word] in the Netherlands a lot of experience has been accumulated regarding the transport of sand and clay mixtures over long distances in such 2-phase  systems. It is a very major business here and many dutch companies are involved in this field internationally. There control over sedimentation is quite important. In Holland when transporting mud, silt and dredgings, but also when raising beaches. It is very common to trans[port water/sand mixtures over distances of several miles, for example using floating pipes. Basically it is the same engineering problem as what you describe.

Yes, the fabulous Mike Abbott, a Scotsman if I am not in error, once initiated the use of digital computers in applied hydraulics, up to the waterways engineer and also the groundwater expert. Major institutions were/are Delft Hydraulics and Danish Hydraulic Institute - together with the US Army Corps of Engineers these are three world-leading commercial players. But regarding research and innovation at least in my view the stage is open for everyone who has good physics skills and, most importantly, NEW IDEAS for the navigation through the extremely complex world of fluid-mud motion, sedimentation of mineral-organic mixtures, re-suspension in wavy bottom terrains etc. etc. At present research is still done mainly within national contexts. It would pay much if we could establish an international institute dealing with these processes in (A) aquatic environments and (B) the atmospheric transports near the ground. The basic physics does not differ too much in both fields and the role of these processes for man is huge, namely in the harsh environments of developing countries. E.g., I remember a sand storm in Gobi desert which governed the daylight in Beijing/China. Today natural air-quality issues are there possibly dominating over anthropogenic issues (they've taken two-stroke motors from the streets and replaced them by e-bikes etc. - progressive and impressive at the same time). The UN system should join superpower and other forces ( sensu Kishore Mahbubani, 2014) to establish such an institute. IIASA in Vienna, established during the cold-war period, was a good start but possibly with a focus which was/is too general and too much focused at computing only. One should include experiments/observations. I am sure that the world's major deserts and the estuaries of the world's major rivers will play the roles of key study objects.

The US Army Corps has delivered so many pieces of public domain engineering software code and thereby set standards; including standards on "open" resource [knowledge] sharing.

On the other hand: the Dutch dredging companies apply more of that knowledge commercially and on a daily basis and most probably transport more mass per year using these techniques than the Danisch and the Army Corps combined, **worldwide**.

The UN should 'harbour' 'open' research projects to produce the software code for things like this and keep that code available in the public domain. In several aspects the IAEA does a similar thing for nucleair energy and other peacefull applications of radiation.

And re deserts: Now that solar energy has become so cheap for those living near the equator: desalination should also drop rapidly in price, so finally we can regain territory on the deserts.....

@Judocus: Yes, the world is full of great new opportunities. Maybe a new UN-based institution could be a mix of IIASA and IAEA. Full openness and systems-analysis orientation of IIASA with measurement- and practice orientation of IAEA.

@Antonio: What about periodic forcing, did you try already (e.g. numerical) experiments with periodic flow perturbations? You should look at the period in seconds, the characteristic sinking time of the typical particles, in seconds, at the amplitude-to-mean-flow ratio, and at the absolute value of the mean flow in m/s. Due to the non-linearity of the processes involved one might get (possibly) a better horizontal transport of solids. To drive the periodicity, one can take energy from the otherwise mean flow. Say, there is an upstream region where no sedimentation issues occur, and one constructs a facility downstreams which sucks energy from the mean flow and leaves far more downstream a periodic flow. One cannot exclude that this gives an improved functioning of the tube or channel under consideration. I am sure that this is an old idea and studied already somewhere.

Regarding this topic: I talked with my brother, who is an experienced drilling & well engineer for a -not-to-be-named- big oil company.

There they use a variety of complex compounds, collectively known as 'mud' when dtilling kilometer long bore holes. This 'mud' contains a lot of chemicals as wel as solids and must simultaneously perform several functions. For this type of application very detailed knowledge of sedimentation under conditions of turbulent flow is vital and errors extremily costly.

The 'mud' is used for cooling, removal of debris, obtaining 'counterpressure' to avoid a fountain of oil coming from the rig by means of carefully adjusting specific gravity. The choice of other additives are important for dealing with porosity of certain rock formations or for preparing for the placement of casings in the borehole.

One can expect a whole lot of practical knowledge in the research centres of oil-companies involved in drilling and especially with their subcontractors.

Very probably, there is more working knowledge there than in technical universities, because they have both the big money and the incentives and the entire globe to experiment with and they are doing just that for decades and with quite some succes!! 

If you can drill a 5 kilometer long hole in a rock fromation without loosing your 'string' you definitely have to understand the hydraulics of 'mud'.

Why re-invent the wheel if there is already so much expertise?

And half the knowledge so obtainable is 'universally' applicable. That is why I gave examples from medicine / dredging / oil-exploration [and I can give even more]......

Boundary spanning [looking in the research yards in other fields of research] is both fun, really challenging, often entertaining and most rewarding. And as the sole non-engineer in my family I do get plenty of such challenges  ;-)

@ Helmut, regarding your comments on periodic flow perturbations:

see my previous answer using the example of oil-exploration: there such models are already applied in large scale "experiments'' .

Whenever a new pipe segment must be added to a drill 'string' the rotary motion is halted, leaving the solids in the 'mud' to sink by the forces of gravity alone. But it should remain possible to restart the drilling because loosing a string due to sedimentation causing too much friction to rotate or withdraw the string simply is not an option [money-wise].

Thanks for all your answers, Dear Helmut and Dear Judocus. I never did investigate before in the field of solid transport in pipe flow, and I do not know also yet the state of art, particularly because this is a specialized area not very common in the text books of Hydraulics. It is well known that Netherlands (“Pays-Bas”) and Germany have some tradition in hydraulic constructions and hydraulic research. I am interested in steady state flows (as Helmut correctly pointed initially). The purpose will be to develop (if possible) a new computational approach for three-phase steady flows, with the presence of a gas phase too. However, in the moment, I am just trying to seek and analyze what are the possible approaches for computing scientifically (if possible) the solid transport in pipe flow.

@Sergei, yes, the typical critical Reynolds number for steady pipe flow is about O(10^3) where the exact value depends on the definition of the length scale use, radius or diameter etc. , and on the roughness of the walls... Surely blood has too high a viscosity to become stably turbulent. I did not compute the details, it's a back-of-the-envelope matter. All other aspects we did already discuss at some length. Wishing a nice Sunday, Helmut

Thanks for you answer, Dear Sergei. The transition between laminar to turbulent flow (critical regimen) is typically assumed for Reynolds Number of order 2000 (Re=VD/kinematic viscosity, where D is the diameter of the pipe and V is the mean velocity of the flow), but the region of the critical regimen is not very well-defined. The Kinematic viscosity of water is very low (at 20 degrees Celsius is about 1E-6 m2/s), which explains the reason for occurrence of turbulent regimen in the large majority of practical hydraulic applications (according to Moody Diagram, the critical regimen may occur for Re of 2000 to 4000, and for VD > 0.004, water flow is normally turbulent). Kinematic viscosity of blood is higher than kinematic viscosity water and depends also from shear rate (in normal conditions, at 37 Celsius degrees, is not far from 3E-6 m2/s to 4E-6 m2/s, but it decreases for higher shear rate?).

Thanks again for your answers, Dear Helmut. I have already some experimental and numerical research work published and an article submitted in steady two-phase flows (air-water). As I wrote already, the specific problem of transport solid in pipe flow is a specialized issue that is not frequently treated in general or applied hydraulics. In engineering practice of sewerage and storm water systems, the solid phase is frequently included in the liquid phase, for low concentrations of solids, with the adaptation of some parameters or flow variables (my PhD Thesis was in optimization of sanitary sewerage systems and storm water systems). 

The flushing technology in sewer systems that you mentioned in a previous answer has more than one hundred years and is still used for the cleaning of the pipes or eliminating obstructions. However, my purpose is not computing unsteady flows created for example by any flushing device producing a current with a wave type. In any way, thanks for the suggestion. I want to create conditions with the introduction of a gas phase in the wastewater for a continuous self-cleansing capacity without sedimentation (with a three-phase flow). Additionally (and also fundamental), the air injection in pressure pipes of sewer systems will allow the control of the sulfide generation, and will improve the hydraulic performance of the systems.

Thanks Dear Sergei for introducing new terms in the discussion. I am not sure if I understood your question. Nevertheless, I do not intend to diminish eddy viscosity. The idea subjacent of my (main) question is well expressed when I formulated the question.

Unless solid particles are fluidized by the action of speedy air, it is difficult to avoid sedimentation in pipes. Otherwise, the solid particles have to be solubilized in the fluid medium which moves them & this is not an easy option since dissolving a variety of solid materials simultaneously requires nearly a "magic" solvent. This is my simple note according to what I learnt in industrial chemistry

Thanks Sergei for your physical point of view.  With respect to open channel flow (like a river or artificial channel for example), the solid transport is presented normally in the generality of the current text books of hydraulics. The work of Shields and his diagram (1936), for example, are normally considered good references in this area, as several others in the last decades.

In pipe flow, observing many existing experiences, for avoiding sedimentation, when a high concentration of solid and heavy particles are present in the liquid (like sand in water, for example), it looks in fact necessary a mean velocity of flow relatively high, particularly for large pipe diameters (in result of the complex hydrodynamic systems of forces that act in the immersed particles). This may be in agreement somehow with what Dr. Nizar expressed about some aspects in Industrial chemistry.

With regard to blood flow, Max Hardeman has done a lot of research in the past decades. Back in 1983, as a med student, I helped him with some computercode for his whole blood viscosimeter, that could also assess shape changes in erythrocytes. [The early days of his LORD: Laser-assisted Optical Rotational Deformability meter]

Blood has very interesting properties here: It is non-Newtonian fluid: its viscosity drops with an increase in shear rate. Very probably erythrocyte deformability plays a role here. And if that is the cause, I expect vessel wall elasticity plays a role too in the smaller vessels. This is difficult to confirm experimentaly. At least, the whole blood viscosimeter set up at that time was completely unusable for that.

For a description of the measuring equipment, see:

http://extranet.rrmechatronics.com/products/lorrca/index.htm

Or see:

M.R. Hardeman, P.T. Goedhart and S. Shin, Methods in hemorheology, in:

Handbook Hemorheology and Hemodynamics,O.K. Baskurt, M.R. Hardeman, M.W. Rampling and H.J. Meiselman, eds, IOS Press, Amsterdam, 2007, pp. 242–266.

Or for the technical aspects, the PhD thesis of J Dobbe:

Title: Engineering developments in hemorheology

Faculty AMC-UvA

Year 2002

FULL BIBLIOGRAPHIC DETAILS:

http://hdl.handle.net/11245/1.197989

with regard to material transport even a pneumatic conveying technology can be used

Dear Lamba: Thanks for your contribution. In fact pneumatic injectors may be used in the upstream of the sanitary networks for pumping small flows of wastewater (they have low efficiency, but some advantages comparing with centrifugal pumps). Vacuum systems may also be used for very small wastewater flow rates in difficult topographies.

Dear Sergei: The introduction of an air flow rate in an upward pipe flowing wastewater increases the velocity of the mixture air-water; it increases the volumetric water flow rate (if the available hydraulic head is fixed) and the superficial velocity of the water. I suppose that this will improve the solid transport. It is well known that airlift pumps (vertical pipes partially immersed in a liquid with air introduction near the bottom) have capacity to carry sand and other coarse particles. For more details, you can see the paper below where I developed a numerical model for two-phase flow air-water and that I applied to air-lift pumps and to inverted siphons: 

Thanks for keeping the discussion alive, Dear Sergei. For great Reynold Numbers (due to high velocities of the mixture, for example), the flow will be fully turbulent (and the dissipation of energy will be due only to tensions tangential of origin turbulent, i.e., independent of Re). If I had a previous answer for my question, I would not make the question.  Qualitatively speaking, it may be possible that the average drag force in a pure turbulent regimen will increase more than the decrease of the average buoyance force when a gas phase is added to a liquid flowing upward (and perhaps this may explain the good capacity of the solid transport in air-lift pumps, for example), but I did never confirm this hypothesis neither theoretically nor experimentally.

In biological systems, gas bubbles in the bloodstream either do not occur or should be considered as dangerous if not life treatening events, It may induce infarctions both in the brain and bones [painfull], much like severe cases of decompression sickness in divers.

Is the use of Reynold numbers in models a valid [validated?] approach, when dealing with highly non-linear effects other than the transition from laminar to turbulent flow and back, such as in blood? Or do we just continue using it, lacking better theory?

The effects of deformability of erythrocytes do not scale in the same way as gas flow / fluid flow scales.

Reynolds number is a dimensionless parameter proportional to the relation between the inertial forces and the viscosity forces, and it is a very important parameter in hydraulics and fluid mechanics, particularly in pipe flow. Any characteristic velocity, and length characteristic, can be used; depending from the phenomena we are interested in study, only changes the absolute value (as induced by Helmut). Traditionally, the Vaschy-Buckingham Theorem (and Dimensional Analysis) has been an important tool in experimental research in hydraulics. Some works in this area seems to point in that direction, but the number of variables may remain considerably high in some cases.

Thanks for your answer.

Maybe I should have expressed myself clearer:

The Reynold number is used in a lot of modeling of compressible and incompressible flow phenomena, as a means to get the scale effect out of equotations, i.e. in windtunel experiments in aerodynamics.

What I wondered was, if you have more than one phenomenum that responds to scale [and both do so in a different manner], is it wise to keep on using the Reynolds number.

I refer to turbulent and laminar flow in blood vessels where typically 45% of the volume is taken up by erythrocytes, that are deformable and that show marked negative effects on viscosity in situations of high shear-rate.

@Sergei Esipov

May I add an other ingredient to the blood you described above:

Platelets are cell fragments, much smaller than erythrocytes and under condition of high shear-rate they can actively initiate the cascade of events known as 'clotting'. This results in the production of fibrin out of fibrinogen and the fibrin threads or fibers help catching and fixing erythrocytes.

Dear Judocus: The scale effect you mention is very common in reduced models in free surface in hydraulics (when gravity is the predominant forces, and forces of viscosity and surface tension are neglected). With respect to blood, I let the discussion for the experts.  Coming back to solid transport in pipe flow, one alternative may be to adapt some basic aspects of the successful Shields’ doctoral work, developed for horizontal open channels (1936), for inclined pipe flow. The following paper from 2009: “Threshold for sediment erosion in pipe flow”, for example, is available in the link below:

http://iusti.polytech.univ-mrs.fr/~aussillous/publiperso/IJMF2009.pdf

@Sergei Esipov  The phenomenum of formation of "dynamic aggregates" is named rouleaux formation in the literature on human blood [flow].

Human circulatory systems and blood may have in fact some connection with hydraulics systems and water. One Professor Retired of my Department advanced many years ago the possibility of researching on that field. The following paper looks interesting regarding to Shields work “The Legend of A. F. Shields”, by John Buffington, J. Hyd. Eng. 125 (4), 1999, 376-387:

“Summary: The well-known doctoral work of Shields is a tale that is frequently recounted by many authors and has spawned a large, continuing body of research over the last 60 years. Despite the success of Shields’ work … ”

http://cedb.asce.org/cgi/WWWdisplay.cgi?116314

Pp.3 Effect of flushing velocity and elapsed time on sediment transport in driplines

Pipeline flushing can be considered as a case of solid transport in liquids. There are four flow regimes for the solid transport in liquids:

• homogeneous suspension,
• heterogeneous suspension,
• moving bed, and
• stationary bed

http://www.k-state.edu/sdi/reports/2013/Puig13.pdf

Hi Krishnan, thanks for the very instructive paper on flushing and driplines/pipelines. I sullfy agree that there are the four regimes you mentioned.Actually they represent four different regimes of the same physical system, if you may you can call them different phases. A major point in the present discussion is in my view to describe the general phase transitions between the four regimes. This seams to be possible only if there is a sufficiently general system of eqns covering mean flow, turbulence, particle properties like form, stickiness, density. Real challenge ...

Thanks for the suitable reference (even if in micro irrigation systems), Dear Krishnan. According to the paper, velocities required for low diameter pipes seem really low (in horizontal pipes?). The well diffused Empirical Correlation from 1952 of Durand and Condolios is used. This correlation is exactly a very good and typical example of many empirical correlations and empirical procedures in steady flow that were developed in this area (a little far of the complex more scientific computational approach that was mentioned by Helmut). According to Durand procedure, velocities required are relatively high for large diameter pipes.

As another example of a similar approach to that used in Durand correlation, with a “modified Froude Number”, for taking into account the slope of the pipe, is the following ASCE Meeting Paper from 1972 “Critical deposit velocities for low-concentration sand-water mixtures”, By Robinson, M.P., and Graf, W.H.., available in the link below:

http://digital.lib.lehigh.edu/fritz/pdf/353_7.pdf

 António:

I had a quick reading about your subject and you may read my paper regarding to the FlowRider. The FlowRider is a surfactant which could help the gas bubbles to attached on the solids surface and transport the solids easily in the flow channel. The FlowRider is used for the proppant transport for the fracturing operation or the sand cleanout for the well intervention for the petroleum application. But it is potentially used for the slurry transport in a pipeline or your application. I have involved the solids transport studies  more than 20 years. I understand the complexity of the solids transport   process. Here I recommend you to read the papers:  SPE 171285 "Overview Solids Transport Study and Application in Oil-Gas Industry  Experimental Work". or you call tell me your email address and we can talk more about the subject.

Jeff

Dear Jeff Li: Thanks for your answer and for mention your paper in the field of well bore drilling. In fact, well drilling is frequently mentioned in scientific papers as an important field of application of multiphase flows and also of air-lift pumps.

One of the major impediments in pipeline transportation is the velocity of flow that depends on many factors. Some of the important parameters are:

• Carrier fluid properties like viscosity, density, etc.
• Properties of solid particles like sp. gr., size, shape, terminal settling velocity etc
• Properties of the slurry, solid concentration, viscosity, and particle size distribution, percentage of sub-sieve particles etc.
• Presence of fine particles increases the viscosity of the slurry and they offer increased resistance to settling behavior of larger particles.

The various design parameters that are required to be established for the design of the pipeline system can be classified under the following three categories:

Yes, you are correct, Dear Krishnan. In sanitary sewerage systems, for example, low velocities increase retention time in pipes (and sulfide generation in pipe flow) and sedimentation. Ventilation is fundamental in sanitary sewer systems in order to avoid sulfide generation (H2S), and the subsequent odors, toxicity and … corrosion. These systems are normally conceived by gravity and in open channel flow with pipes partially filled with air atmospheric for ventilation. Pipe flow is as much as possible avoided, and the national regulations and manuals of practice (like ASCE or EPA manuals, for example) establishes procedures for sulfide control (particularly in pipe flow) like injection of air or oxygen (in low points of the conduits to guaranty a good efficiency of oxygen transfer), or the addition of hydrogen peroxide, for example. Hot climates, large networks, and pipes under pressure that cannot be avoided due to difficult topographies are critical factors in conception of the systems and sulfide control.

When designing hydraulic systems in engineering practice, maximum conditions must be respected. Maximum flow velocity is normally limited in order to limit the head losses, problems associated with transitory regimens like water hammer in pipe flow, or mechanical and electrochemical erosion that can be associated to water characteristics, turbulence and, in some cases, to transportation of abrasive materials, like sand, for instance. In Urban Hydraulics, for example, these maximum limits of velocity are normally fixed in the national regulations and in manuals of practice of engineering. The requirement for a minimum average shear stress in the internal wall of the pipe, in open channel flow, instead of a minimum velocity for the guaranty of self-cleansing conditions, in sewer systems, has been studied, but the criterion fully predominant is of minimum velocity.

The energy-water cost ratio, exposes the widely different conditions that different locations may imply. The minimum transport velocity is conditioned by the solids deposition mechanism or by the laminar-turbulent transition

http://repositorio.uchile.cl/bitstream/handle/2250/126728/Identifying-the-relative-importance-of-energy-and-water-costs-in-hydraulic-transport-systems-through-a-combined-physics--and-cost-based-indicator.pdf?sequence=1

Thanks for your answer, Dear Krishnan, with excerpts of the paper from 2014 “Identifying the relative importance of energy and water costs in hydraulic transport systems through a combined physics- and cost-based indicator” by Christian F. Ihle, Aldo Tamburrino, and Santiago Montserrat. In fact, the object of Hydraulic Engineering and Civil Engineering are real systems (i.e., are not illusory windmills nor dragons).

@Antonio and Jeff: The topic is quite interesting. I learned that rouloux formation leads to increased particle size and thus decreased kinematic viscosity which in turn decreases the Kolmogorov scale provided that we talk about a developed turbulent flow. In sewer systems you always have a mix of simple solids more or less in suspension and shear-thickening as well as shear thinning components so that phase separations but also) phase transitions of the overall fluid motion and mixing characteristics become seemingly an extremely complex business. However, as long as you have taken the data for all the components in (more or less) isolation then you may today easily compute the set of phase transitions provided you use the most recent literature about the physics of turbulence. And actually it is these phase transitions which are of primary interest for the design engineer.