Area 51 » Science

followers

Follow It!

17 Example Questions (5 closed)

active newest votes

up vote 14 down vote

added by Kamil Kiełczewski Nov 11 '20 at 14:45

closed as not constructive by warren, Big Brother, ppwater, TheSimpliFire, niamulbengali 6 hours ago

This question is not a good fit to our Q&A format. We expect answers to generally involve facts, references, or specific expertise; this question will likely solicit opinion, debate, arguments, polling, or extended discussion.

	This will help you to create your own question: The "idea" behind this question is to ask about theoretical definitions, differences, details or any properties of methods for solving Navier-Stokes equations at any step, part or in general (as in this question). Other similar questions could be about other methods like spectral methods, finit element, influence-matrixes, SVV, etc. – Kamil Kiełczewski Nov 21 '20 at 10:05
	I would also add hybrid RANS/LES methods to the question since they are becoming very popular lately. – gmonvill Dec 28 '20 at 9:32
	@gmonvill I think hybird RANS/LES methods and differences can be separate question – Kamil Kiełczewski Dec 28 '20 at 9:38

up vote 10 down vote

added by Kamil Kiełczewski Nov 11 '20 at 16:27

This will help you to create your own question: The "idea" behind this question is about theoretical mathematical or physical details connected with Navier-Stokes equations (for given concrete flow or in general). Question here can be about eg. turbulence models, thermodynamical models (energy equation e.g. using Boussinesq approximation), NS equations forms in curvilinear coord systems, polar fluid NS-equations form etc. Here we can also ask about real experiments details and data for compare (or validate) simulation with experiment – Kamil Kiełczewski Nov 21 '20 at 10:09

Although this is a nice question, it is for me off-topic because it is about fluid dynamics, and not the modelling/computing aspect. – Bernhard Dec 29 '20 at 14:13

@Bernhard - in my opinion - form of equations have extremely impact in chose numerical approach - especially in nontrivial curvilinear coordinate systems. When you have Coroils force then you can use it to calculate Buoyancy force based on energy equation (for eg. ocean/atmosphere simulations...). If you write your solver (or part of solver) then also having explicit equations is needed – Kamil Kiełczewski Dec 30 '20 at 10:13

But when you're talking about writing the solver, it is about the discretization of the equation, rather than the derivation of the equation itself. If the question was "How to implement the Coriolis force in <whatever>" I would say it is on-topic. This question has a lot of upvotes, however, it also has three close votes. – Bernhard Dec 30 '20 at 16:24

up vote 10 down vote

added by Kamil Kiełczewski Nov 11 '20 at 16:42

closed as not constructive by warren, Big Brother, 10 Rep, ppwater, TheSimpliFire Jan 5 at 9:25

This will help you to create your own question: "idea": these question are mainly aimed at engineers who not create codes but rather try to find best tools for calculate concrete flow case (or compare them in general), find information of how to configure that tools (how to define geometry, grid, choose model etc.) to create simulation of concrete flow case, create grid, connect everything with visualisation layer (if it is not build-in) etc. – – Kamil Kiełczewski Nov 21 '20 at 10:10

I feel this question is way to broad, because the answer will be a list. I would rather taylor this question to a specific simulation goal. – Bernhard Dec 29 '20 at 13:22

@Bernhard - I update question – Kamil Kiełczewski Dec 29 '20 at 14:02

@Kamil That looks a lot better. I am expecting the body of such a question to contain all the details of the intended simulation. – Bernhard Dec 29 '20 at 14:06

up vote 9 down vote

added by Kamil Kiełczewski Nov 11 '20 at 23:32

This will help you to create your own question: The "idea" behind this question is to ask about finding best method (fast, precise etc.) for calculating part or whole given concrete simulation. In this type of question researches can also ask about finding best grid/mesh, numerical approach for simulation chosen flow and also ask about best numeric libraries and tools including also visualisation and hardware resources – Kamil Kiełczewski Nov 21 '20 at 10:11

Can people that voted to close explain why this question is fake? I hope as a community we can help improve the contents. – Bernhard Dec 29 '20 at 14:14

up vote 4 down vote

added by Ashish Kinge Dec 8 '20 at 4:25

@AshishKinge thank you for your question :) - can you also click on orange "follow" button to follow this proposal (and give some up-votes if you want to other questions)? This will help to make "CFD-Overflow alive" – Kamil Kiełczewski Dec 8 '20 at 9:50

up vote 2 down vote

added by Bassie-c Dec 10 '20 at 2:31

up vote 2 down vote

added by Dmitri Chubarov Dec 28 '20 at 14:58

Why the downvote? An approach on how to simulate a physical problem is perfectly suitable for what a CFD site is if you ask me? – Bernhard Dec 29 '20 at 14:15

I also don't understand why this question was downvoted - it is good question – Kamil Kiełczewski Dec 30 '20 at 18:20

@Bernhard: I suspect it's because it's far too open-ended. As it stands, it's unanswerable. – warren Jan 5 at 13:34

up vote 2 down vote

added by Bernhard Dec 30 '20 at 16:28

up vote 1 down vote

added by gmonvill Dec 28 '20 at 9:29

up vote 0 down vote

added by Nike Dattani Nov 24 '20 at 1:07

up vote -1 down vote

added by Nike Dattani Nov 24 '20 at 1:05

up vote -1 down vote

added by Nike Dattani Nov 24 '20 at 1:08

interesting question - I don't get why some people give vote down (without comments) – Kamil Kiełczewski Nov 25 '20 at 15:41

@Kamil - probably because it's a simple "yes/no" question...and not a likely good fit for an SE site – warren Nov 28 '20 at 3:21

@warren - in stack overflow (SO) such question also appears - but implicity everybody who give answers assume that answerer must give some kind of proof/explanation (like example code snippet or something) - if someone give answer in SO without proof/expalanation he will get minus votes for such answer. So in my point of view such questions are not a problem - eventually question author can change "Is it possible" to "Is it possible (and how)" - but this is only technical language detail... – Kamil Kiełczewski Nov 28 '20 at 10:48

@Kamil - just because these questions have appeared in the past does not make them good now. If you want to get the proposal to go anywhere, there need to be good sample questions. This is a lousy sample question. – warren Dec 8 '20 at 19:34

@warren - what I can do? I already ask 5 questions and reach the limit. Maybe you will add some your good questions? – Kamil Kiełczewski Dec 8 '20 at 19:44

show 9 more improvement suggestions

up vote -2 down vote

added by Kamil Kiełczewski Nov 26 '20 at 10:48

	This will help you to create your own question: The "idea" behind this question is to ask about: results validation, statistical processing and visualisation which is usually non trivial for CFD. You can ask about validation of any part of computation or whole simulation, statistics, flow structures - because used formulas, code, parameters etc. can cause errors with simple but non-trivial solutions. You can also ask here about literature related to your case. – Kamil Kiełczewski Dec 7 '20 at 9:33
	The answer to this question is probably a Bachelor thesis. I don't think it is really answerable in SO format. Although I didn't downvote, because that's not constructive, I do understand the downvotes. – Bernhard Dec 30 '20 at 16:26
	Actually I know answer to this question (and put it here only as example): in my case the normalisation of NS equations cause different values of Q criterion (this is way I write that I put formulas). The above question not have details because this is only "example" and it looks like Area51 alows to put only question titles – Kamil Kiełczewski Dec 30 '20 at 17:59

up vote -3 down vote

added by Nike Dattani Nov 24 '20 at 1:05

closed as duplicate of Is it possible for a visco-elastic fluid to have a memory kernel that is not convex? by warren, Big Brother, ppwater, niamulbengali, Nij Jan 4 at 12:44

This question covers exactly the same ground as another question in this proposal.

convex is not equal to monotonic - this is why this two questions are not the same - everybody who have knowledge please click on "reopen" – Kamil Kiełczewski Jan 4 at 18:07

@Kamil: they're exactly the same type of question - that's probably why it got VtC'd – warren Jan 5 at 13:36

@warren but answers can be totally different - so it doesn't mean that questions are equal. It is not bad nor forbidden to ask similar but not equal questions – Kamil Kiełczewski Jan 5 at 14:29

@Kamil: you missed the point: propose types of questions that should appear. You can ask far more "sample questions" when you stick to the "pattern" method. Eg, "Why is {php} preferred to {python} when writing a {simple website}?" That pattern / question type encompasses 100s of real questions. "Is it possible for a visco-elastic fluid to have a memory kernel that is non-monotonic?" is *exactly the same type* of question as "Is it possible for a visco-elastic fluid to have a memory kernel that is not convex?" And both are *bad* because they're yes/no questions. – warren Jan 5 at 14:38

@warren I already told you in another question: in stack overflow (SO) yes/no questions also appears - but implicity everybody who give answers assume that answerer must give some kind of proof/explanation (like example code snippet or something) - if someone give answer in SO without proof/expalanation he will get minus votes for such answer. So in my point of view such questions are not a problem - eventually question author can change "Is it possible" to "Is it possible (and how)" - but this is only technical language detail... – Kamil Kiełczewski Jan 5 at 15:13

up vote -3 down vote

added by gmonvill Dec 28 '20 at 9:10

I think this question is downvoted because it is very broad – Bernhard Jan 4 at 16:15

up vote -4 down vote

added by Bassie-c, edited by Nike Dattani Dec 14 '20 at 16:19

closed as not a real question by Nij, Big Brother, warren, ppwater, damryfbfnetsi Jan 12 at 22:11

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form.

	Nice questions too (still I not understand why someone vote down) - there are no general test - but methodology is following: find real similar experiments and compare your results with it - if they valid - then you can extrapolate your simulation to get details which are (at this moment) not possible to get from experiment – Kamil Kiełczewski Dec 10 '20 at 8:11
	Hi, just as a reminder, you aren't meant to answer example questions. – Z9. Dec 24 '20 at 6:41
	I suggest to make the question specific. For example: How can I verify and validate my volume of fluid solver? – Bernhard Dec 30 '20 at 16:28