[Dr Youssef Hafez]

I checked CSU website for IP policy which could be found in the following link:

https://policy.csuci.edu/aa/01/aa-01-002.htm


Below is the part about student creation mentioned in CSU IP website.

Student Creations.
Students will normally own the copyright to the scholarly and creative publications they develop, including works fulfilling course requirements (term papers and projects), Capstone Projects, and Masters Theses/Projects. Students retain copyright ownership as long as they are not paid for the work that results in the creation and do not receive extraordinary University resources in support of the work. Nonetheless, by enrolling at the University, the student grants the University a nonexclusive, royalty-free license to mark on, modify, publicize and retain the work as may be required by the faculty, department, or the University. The University is not entitled to an equity share in any ownership profits, except in the circumstances covered below.
When the student is employed by the University and the creation falls within the scope of that employment, either the University or the faculty member (when the student is hired specifically to work on a faculty project) owns the copyright according to the same standards that apply to staff creations, under sections II.A.3 above, or faculty creations under Section II.A.2.

[LuckyTran]

Gents, CSU is not a ghost entity. We don't need to go all Sherlock Holmes on this. Just go contact the university and sort it out. The part of the policy being quoted is regarding copyright (which is not all of IP) of trivial things such as student homework. We're clearly talking about something else! See the sections of the policy regarding inventions and software. Note especially the section regarding responsibility to file disclosure statements to the University in order to assist in assigning such rights. Even when the student themselves are not employed blah blah blah, other people are (the 5 professors for example) that participated. A university policy is also not a substitute for local and international law, although they will certainly help sorting it out much faster and CSU policy strongly favors that the author benefits. You still need all parties involved to assign rights (or an explicit waiver of rights by owners not vested in this IP). Again, I am not saying it is impossible to publish, you just have more work to do. So just do it if you really want to put this discussion to rest.



My statements about IP ownership are really only relevant if we are actually talking about 20 year old source code. You can easily get around it by reverse-engineering it and just writing new code which will be infinitely more useful anyway since I doubt 20 year old code works well in a modern parallelized environment and considering the changes in other tools used nowadays: changes in CAD software, meshing software, plotting software, etc.

[andy_]

Originally Posted by Dr Youssef Hafez

I checked CSU website for IP policy which could be found in the following link:

https://policy.csuci.edu/aa/01/aa-01-002.htm


Below is the part about student creation mentioned in CSU IP website.

Student Creations.
Students will normally own the copyright to the scholarly and creative publications they develop, including works fulfilling course requirements (term papers and projects), Capstone Projects, and Masters Theses/Projects. Students retain copyright ownership as long as they are not paid for the work that results in the creation and do not receive extraordinary University resources in support of the work. Nonetheless, by enrolling at the University, the student grants the University a nonexclusive, royalty-free license to mark on, modify, publicize and retain the work as may be required by the faculty, department, or the University. The University is not entitled to an equity share in any ownership profits, except in the circumstances covered below.
When the student is employed by the University and the creation falls within the scope of that employment, either the University or the faculty member (when the student is hired specifically to work on a faculty project) owns the copyright according to the same standards that apply to staff creations, under sections II.A.3 above, or faculty creations under Section II.A.2.

Yes that looks pretty standard and similar to the UK. However, were you working on a project setup between your government and the university? If so how have the IP issues been handled. Were you paid during your time in the US by some institute in your country or the US? If so there will be IP issues which may or may not be in your favour. Did you subsequently work for an institute where CFD was part of your normal duties? If so, the institute that paid your wages will almost certainly own your revisions even if you performed them at home and not at work. In this case you will need an agreement with your employer transferring the IP back to you. Some employers will do this if there is no real conflict, some will refuse to give away anything because that is easiest for them and for some it will be impossible to find anyone prepared to address the issue. It really is something that is best sorted out before starting employment.

[Dr Youssef Hafez]

Diversity and multiplicity are both a big advantage and essential part in life and this is true in CFD as will be explained later. Imagine what life would be if there is only one single fruit or one food type that we have to eat, one single juice to drink, or one single colour of our cloths, one singe car model we have to drive…etc.
In the field of numerical methods several methods exist such as Finite Difference, Finite Element and Finite Volume. Each method can be the best under certain conditions. Even for each particular method there are several variations and subtypes.
In the field of turbulence models there are many types, k-l, k-e, k-w…etc.; linear and nonlinear types. Each turbulence model has also several variations, subtypes and different constants.
As far as code building and programming styles there are no limits and every code bears the personal style and personality of the code author(s) even for the same model structure, and the same numerical method.

Some things that appeared initially did not get any attention at all, however, years later its value was discovered that it can serve a very useful purpose. So what might not be with no or less value now might be of great value in the immediate or far future.

Therefore go ahead and feel free to explain and publicize about your CFD codes and I wish you all the good luck.

[Dr Youssef Hafez]

Dr. Paolo Lampitella thankfully provided an excellent opinion about the issue of IP which I enclose again here in case that it might have not been noticed. I hope focus should turn to the quality of the CFD code as can be seen in the enquiry by Dr. Denaro and others. I thank those who where concerned about the IP issue for their advises. I hope that every body should not worry and leave this matter to me or to any potential buyer to handle.

Dr. Paolo Lampitella said:

[I think the legal matter here is not so obvious or relevant.

In fact, what may seem obvious from a common law perspective, would instead spectacularly fall apart on civil law grounds (and vice-versa, obviously). If anything, this is a matter of international law and/or specific agreements between two institutions that existed at a very specific point in time. Assuming that a common law model was arbitrarily in place doesn't seem correct to me.

Also, in such cases one typically assumes a guilty frame that, honestly, is always far from proven. Can the university prove that the code was actually made in the Ph.D. context? Can the university actually define what the Ph.D. context is? Is there a daily time window or place where someone's work is accountable for the Ph.D. and some other where it isn't? Is all the IP produced by an individual during the Ph.D. years owned by the institution granting the title? Can the university prove that such code wasn't instead already there before the Ph.D.? Has the university actually checked that the said work wasn't, indeed, already IP of someone else? What if student and university have conflicting versions and none of them can prove their own?

But even assuming a 100% guilty use of the code, let's be honest here, is the university reasonably willing to chase a case, again far from clear and possibly linked to international law, without at least few tens or hundreds thousands being involved? I don't think so.

Finally, if it ever becomes a trouble, such codes can be rewritten in few weeks in such a manner that any claim becomes unreasonable.

What I think the matter is, here, is how to determine if a code has actual value and how to determine it. As I stated on RG, I think this is only a matter that you can resolve.

Any code for a book is certainly added value, but I would not think of it like actual value, at least no more than a dedicated chapter on a specific subject that few other treat.

So, besides the book example, the code value is what someone else actually pays for it. At the moment it is 0. It is the combined effort of making the code and selling it that, like the Schrodinger's famous experiment, makes the value collapse on its actual state.

And you are the only one who can actually come up with a rationale to buy your code, in whatever form. We can only point you at what the alternatives out there are and how they work.

Note that the effort in selling the code involves finding possible use cases as well as possible clients and convincing arguments for them, besides all the bells and whistles that make the code actual selling material.

I have an impression of how, under certain conditions, your 2D code, properly rearranged for unstructured grids, could be of some value as part of a general 3D code. But this is me, and you may disagree, and there would be no point in discussing it.

It is not really different from the sell of a house (or any other good). House owners tend to consider their house to have an intrinsic value, while it only has it for them.]

end of Dr. Paolo Lampitella statements.

[sbaffini]

Yes, let me summarize it with the fact that legal matters are different between common law and civil law systems and agreements might have existed back then that now don't anymore. So, while I also suggest to take legal advice, for the exact same reason I personally, strongly disagree on approaching certain legal matters starting from the passive/guilty position. In a civil law system, if the matter isn't specifically covered by a norm, it is debatable and the fact that a private agreement existed, like the one between the university and the student, really means nothing. What if human sacrifices were asked in the agreement? What if you were not in your full capacity when accepting that agreement? And so on. In the end, if it is your code and has already changed, you can finally make it 100% different with some additional effort. So, what would be their final IP in it, how the SIMPLE algorithm is written? If you are the sole author of a piece of code, it is likely that you will consistently rewrite it in the same way in different moments in time because it depends from your understanding of the algorithm and how to efficiently implement it, not from some random number initialization depending from where you were when writing it.

Legal, in my opinion, should never become a problem if you actually are the author of the code, because you should be able to write it again and again, because that is what you do and you will alsways be able to prove that code is indeed yours. My current 28k lines code comes from 490k lines changed in commits over time, and I am very lazy. Anyone reasonably doing this as a job should not be afraid of this. Typically those who have issues are the ones who can't actually rewrite something in the first place.

So, again, check the legal matter, but focus on the actual problem, which is much harder. Let me also say that there is no real point in debating the legal matter here. You may agree or not with the comments of others but, if ever, it is not this jury that you need to convince. So the legal matter, in my opinion, can be peacefully archived for whenever you will need to face it. Just don't neglect it because you think you're right as well.

[andy_]

Originally Posted by sbaffini

Let me also say that there is no real point in debating the legal matter here. You may agree or not with the comments of others but, if ever, it is not this jury that you need to convince. So the legal matter, in my opinion, can be peacefully archived for whenever you will need to face it. Just don't neglect it because you think you're right as well.

There is every point in debating it because most people reading this forum likely don't understand how IP works for the CFD code they write.

It seems rather unlikely the OP wholly owns "his" CFD code. Not because the US university has a claim (though it might) but because of being employed when he wrote some of it. His silence on the subject is probably wise. He also appears to be employed to work on CFD at present which only emphasises how little he seems to understand. Likewise advertising the issue on the web before sorting out the IP/ownership.

If some other bodies have a claim then the only way the legal status of a CFD code can be changed (e.g. transfer of ownership) is with the explicit written consent of everyone with a share no matter how small. Retrospectively getting this is likely to be very difficult once large bureaucratic institutions like universities and government institutions are involved. Getting it before starting employment is much less problematic.

In the rather unlikely event the OP finds a prospective buyer for "his" Fortran code they will obviously want to see proof he owns it which he currently likely won't be able to provide. In order to get it he will need his relevant past employers (i.e. those that were paying him to do FEM and CFD for them when he was writing some of "his" FEM and CFD) to produce a document that they have given up any claims to "his" software. He might get them but first he would have to understand and accept why they are needed which I suspect he doesn't at present.

[Dr Youssef Hafez]

First of all I hope whoever is speaking that he/she should know what he/she is talking about and should speak based on sound facts not speaking based on his/her imaginary assumptions or own hypothesis derived from his own local personal experience in another country, and under different circumstances and boundary conditions.

I feel the speaker is trying every possible way to object my ownership to my own code(s). It seems that he felt the university IP issue might not work, so surprisingly he turns to my previous employment trying to dig hard hoping to find anything against my claim without knowing anything about work environment at home. You say the IP issue is so important so why you Sir make a special post by you away from my post. Now let me give you Sir some facts that refute every claim you falsely and unknowingly made:

1- First of all, for my previous employer at the time of scholarship, only two things were required from me. The first was to get a Ph.D. degree from CSU in any subject related to water and river engineering (hydraulics mostly and maybe fluid mechanics). I was not required at all to do any specific topic or CFD or develop any code. I myself chose to study distribution of the boundary shear stress which later I found myself willing to investigate it via numerical modelling of RANS because I love applied mathematics a lot.

2- The second requirement was to spend a couple of years at home at my employer place after finishing the Ph.D. which I did. In fact just before departing to the USA for the scholarship I had to sign an agreement with one of my close family co-signing to guarantee my return. There was never any IP agreement with either the university or my employer, and this subject of IP has never been brought or discussed all the years during my Ph.D. (1991-1995).

3- I returned soon after I finished my Ph.D. in 1995 and fulfilled exactly these two requirements. Later, I resigned ten years ago (@ 2012) while obtaining “CLEARANCE” from my employer regarding everything. I hope you Sir to understand the meaning of the word “CLEARANCE”, it simply means I don’t owe my previous employer at scholarship time anything at all and I am cleared from any claim or responsibility.

4- I was never at any work, anywhere and anytime required to develop any code for anything. All was required at home was to solve the real-life Nile River problems using any method. To solve some problems I was voluntarily developing codes amid my love to CFD and my education background. For other problems I developed analytical models and theories (please see my RG page).

5- I never up to this date used the Ph.D. code in anything because it was dealing with cross sectional geometry and was never required by anyone; believe it or not this year only was the first time I looked into it. I felt that my Ph.D. was just an academic exercise and what I learned was how to model and solve any differential equation. Since I returned, I turned into hydrodynamic modelling and to analytical models related to river engineering (you can check my publications on my RG page). The Ph.D. code is a piece of cake for me and I can re-write it just sitting at STARBUCKS having a cup of coffee. It is not a big deal for me at all, so I don’t know why are you making a big issue of it, I can write 10 different codes solving the same problem with each code is different from the other. Real modellers know these facts so I am surprised of your attitude, again is there a conflict of interest here? Do you think someone with over 30 years of numerical modelling would be stuck to a Ph.D. code developed about 30 years ago? I have developed more sophisticated codes which are really unsteady and fully three-dimensional with even large scale 2D flood models and river flow, scour and deposition.

6- The speaker mentioned falsely that I am now working and I am quoting him saying “He also appears to be employed to work on CFD at present”. Sir, sorry you are totally wrong. I retired from a second employer (which has nothing to do at all with any modelling or codes or any research) on March 2020 so almost three years ago. I have been since then a free-lancer.

7- Throughout my Ph.D. study or at any place that I worked I was never and ever required to do any CFD coding or any modelling at all, so I don’t owe anybody anything as far as modelling or coding. I have been developing models amid my love to the subject and thanks to the high training and education that I got, after all isn’t that what a university should prepare its students to do. Unless the university thought that everything is right from my part, it would not have given me the letter of completion and my Ph.D. certificate. I would not be graduated unless everything which is required has to be done.
Examples of some codes (almost all are non-CFD) I developed voluntarily without being commissioned by anyone are:
1- Water Surface Profiles under Unsteady Motion with applications to
(Sea Channels, Wetland Hydrodynamics, River Flooding and Overland flow).
2- A Two Dimensional Flow Model in Lakes, Wetlands, and River Reaches.
3- Modelling of Water Pipe Distribution Networks.
4- Rainfall-Runoff overland flow modelling (1D network model).
5- Rainfall-Runoff overland flow modelling (two- dimensional model).
6- Open channel network modelling (steady, 1D network model).
7- Open channel network modelling (unsteady, 1D network model).
8- Agriculture water requirements for plants.
9- Water quality (1D modelling) for salinity, DO and BOD in channels.
10- Water quality (2D modelling), channels, lakes and wetlands.
11- Water quality (3D modelling), channels, lakes and wetlands.
12- Water surface profiles (1D self-calibrating model) for river and channels.
13- Hydrodynamic model (2D), lakes, channels and wetlands.
14- Hydrodynamic model (3D), lakes, channels and wetlands.
15- Dam-break model (1D).
16- Temperature model (1D, 2D and 3D) for environmental studies.
17- Sediment modelling (1D, 2D and 3D) for environmental studies.
18- Pipe network modelling (unsteady).
19- Groundwater modelling (2D and 3D).
20- Nile basin network model.
21- Modelling nonlinear Pendulum motion.
22- Modelling Radon gas distribution in soils.


8- I have been silent just because I am busy and because the issue was covered herein in great details by several Respected Gentlemen who spoke very clearly. There is not any feeling of any guilt as I have not done anything wrong. I live far away from the USA so I don’t have the resources to consult a paid IP attorney nor having the time now to make any contact with the university. I strongly believe that all the codes I have developed are mine, as I was never hired or required to write or develop any code. And CSU if you are listening or anybody else who thinks he/she has rights to my Ph.D. code please contact me and let us sort this matter. If anybody can contact CSU to resolve this issue that would be appreciated. I don’t know why the speaker assumes a guilty verdict. I put this post amid good faith and feeling confident that I have every right to ask the value of codes that I developed. Amid good faith I mentioned the Ph.D. code because it was used and existed results of several case studies but not on the grounds that it is the corner stone of my codes. I have been using my real name and my true identity unlike those who are speaking and hiding behind fake names and unknown identity. I hope next time if anyone to speak please introduce yourself clearly with your identity and let us know who you are so we can appreciate your background and value your opinions.

9- Finally I think enough is enough, if any one wants to discuss any issues related to IP please make your own post and I appreciate you not referring to me at all. I came here to discuss CFD issues so please let us keep this way; at least in my thread.

[LuckyTran]

Btw, the policy quoted is from CSUCI. The Colorado State Policy states that you must assign rights to the University with only one exception defined: Academic Material (i.e. lecture notes). There is also contact information for CSU if you really can't find it. I hope it is clearer that you should talk to the people involved to sort things out if you plan to commercialize your old source cold. I would offer to sort it out for you but I'm not a vested party. You don't actually need to hire a lawyer, just write an email to CSU and they have the lawyers to take care of it if it is really needed. In most cases it costs you no $ and less time than you can finish a Starbucks drink. No one is calling you guilty except to tell you to do the right thing. The more you resist the more it degrades your act as being "in good faith." For example, you could have just responded originally with "okay, I will look into it" and the conversation would have been over ages ago.

I (and likely the others as well) respond here not to argue with you specifically but for the sake of the someone that googles this thread 10 years from now, so that they don't receive the false impression that they are free to openly sell things that they might not wholly own without executing the proper procedures.


I see in a newer thread that you are looking for funding and partners to continue this work. Maybe you should contact CSU as a partner? Perhaps?

[andy_]

Originally Posted by Dr Youssef Hafez

9- Finally I think enough is enough, if any one wants to discuss any issues related to IP please make your own post and I appreciate you not referring to me at all. I came here to discuss CFD issues so please let us keep this way; at least in my thread.

It's your thread. A shame though because establishing what the standing of your CFD code actually is and what might need to be done to move to a position where it could be sold is likely to be an interesting example to a fair few here. Particularly those had thought they obviously owned "their" code and now might be wondering whether that is true or not.

[sbaffini]

Originally Posted by andy_

There is every point in debating it because most people reading this forum likely don't understand how IP works for the CFD code they write.

It seems rather unlikely the OP wholly owns "his" CFD code. Not because the US university has a claim (though it might) but because of being employed when he wrote some of it. His silence on the subject is probably wise. He also appears to be employed to work on CFD at present which only emphasises how little he seems to understand. Likewise advertising the issue on the web before sorting out the IP/ownership.

If some other bodies have a claim then the only way the legal status of a CFD code can be changed (e.g. transfer of ownership) is with the explicit written consent of everyone with a share no matter how small. Retrospectively getting this is likely to be very difficult once large bureaucratic institutions like universities and government institutions are involved. Getting it before starting employment is much less problematic.

In the rather unlikely event the OP finds a prospective buyer for "his" Fortran code they will obviously want to see proof he owns it which he currently likely won't be able to provide. In order to get it he will need his relevant past employers (i.e. those that were paying him to do FEM and CFD for them when he was writing some of "his" FEM and CFD) to produce a document that they have given up any claims to "his" software. He might get them but first he would have to understand and accept why they are needed which I suspect he doesn't at present.

I totally agree with you. What I meant is that while with CFD we can actually give what I'm proud to define the most educated help anyone can get on the Internet, for legal matters we admittedly can't, we can only give guesses based on personal experience, which can vary and are not, in most cases, fully educated.

As LuckyTran suggests, I would have expected the matter to end with "okay, I will look into it" as well, but it didn't. Now, my opinion is that that side of the discussion is slipping toward something resembling an actual debate on the matter that, as I stated above, we can't really sustain here in my opinion.

Of course, anyone is free to write whatever he wants within respect for others, but in my opinion there is much more to this discussion than the legal aspect which, as a developer, I consider relevant only if you actually want to use someone esle code or rewriting is out of question.

[FMDenaro]

Turning back to the original question, immagine I am interested in your code.
Before going to legal issues I first need to evaluate the quality of the solutions you can provide for some standard test-cases.
Just few of them:
- Taylor solution (accuracy analysis)
- lid driven cavity
- backward facing-step
- flow around a cylinder

Then, there are further test for turbulence, multi-phase flows and so on.
Without a clear example of the perfomances for such problems I wouldn’t go ahead on any kind of evaluation of the code.

[Dr Youssef Hafez]

Thank you Mr. LuckyTran for sending the CSU link, it really proved that my claim is absolutely genuine and right.
First, in the page of CSU policy: Intellectual property-copyrights and patents, it is stated that this policy is effective starting from Original Effective Date: 6/14/2000 while I completed my Ph.D. in 1995. Since you cite material from this link, why didn’t you see this crucial and very important part? or your eyes are trying only to locate things you think against my case?. I advise you to give yourself a break and take rest and try to find something useful to do such as discussing useful CFD issues.

[Dr Youssef Hafez]

However, for the sake of argument assuming just hypothetically that this policy applies. In the section of Academic material computer programs (like my developed computer program) are included there and it is stated that these Academic material may be copyrighted, patented, and/or trademarked. The policy states “the University will continue the tradition of not claiming ownership or a share of the proceeds from scholarly works such as Academic Materials, textbooks or their equivalent not supported through the use of University Resources”. In the section of university resources it is stated that
“the University has concluded that University Resources shall not include the Member’s time, use of the library, or use of personal office including any office computer or data processor located therein”. These are exactly the things I have been using I am O.K. again.
By the way if you argue that it is a case of software, you should know that Software by definition must have documentation, and my Ph.D. code was never documented at all.
So even for a policy that does not apply, my claim is genuine and true.

[LuckyTran]

Dude, read the room. Stop talking about it. The more you argue about it instead of actually contacting CSU and addressing it, the more it paints you as a bad actor and the less you seem genuine. You were not faculty at CSU, you can't generate academic material. Academic material is scholarly, not a commercial work. You are grasping at straws to prove your case to random people in the internet. I wish you would put just a tiny bit of this effort towards resolving the case with the parties affected. The policy being published online in 2000 doesn't mean that there weren't any legal policies affecting you in 1995 and even more are a reason you need to contact the parties to figure out what the policy was in 1995. Tell them! Not me!

[Dr Youssef Hafez]

Originally Posted by FMDenaro

My suggestion is to focus on the methodology inside the code, you developed it in the FEM framework and you should address the advantages of your structure compared to other FEM codes.
It is well known that some issues are:
- large memory required
- well assessed theory for linear problems but less for non linear ones.
- problems with highly convective flows.
- time integration method outside the FEM framework.


Thus,
prepare some standard test-cases, well accepted in the CFD community, show the resulting accuracy order and computational performances. The case of the square channel is not relevant.

Is the code able to read grid generated by other softwares?



Then,
could be your code relevant for didactic aims ? Why?

I thank you Dr. Denaro for your valuable issues that you raised and I will try here to answer them in a sufficient manner in as much as possible.
- As far as memory storage it was explained in a previous post that in my codes using SKYLINE storage method, the storage requirement might reach 3% of the size of the theoretical coefficient matrix, i.e. saving up to 97% is achieved.

- It dealing with any problems that involve corners, surface roughness irregularities, flow separation, and curvature, I think a non-linear stress versus strain-rate or velocity gradients should be always used. So, no matter that you are using k-omega SST, or k-epsilon (standard or realizable) that would not make much difference as long as the stress versus strain-rate relation is linear. It was demonstrated in my Ph.D. that the non-linear algebraic stress versus strain-rate by Speziale (1987) was successful to predict the three-dimensional velocity and boundary shear field in duct-cross section under steady and fully developed conditions. The 8-vortex structure representing the secondary motion in a square duct was obtained with high accuracy. Using Reynolds stress transport differential equation models would be very expensive computationally speaking and uncertainty still exist in such types of models. I am surprised so many researchers are trying to improve the prediction of the turbulent viscosity using different types of turbulence models (k-w) while forgetting the stress-velocity gradients contribution.


- For convective turbulent flows especially with high Peclet number, Pe, (Pe is defined here as Pe= u dx/effective viscosity), where u is the maximum convective velocity, dx is the maximum step size and effective viscosity is the sum of the molecular and turbulent viscosity. Pe represents ratio between advection to diffusion which for stability should be less than unity. I believe this condition should always be met without resorting to upwinding schemes that introduces artificial diffusion or some other sorts of artificial mechanisms. With my due respect to the vast research on high convective flows with unwinding schemes where researchers try to prove its validity against some test cases that to me looks like cancelling the number 6 in the numerator and denominator when dividing 64 by 16 to obtain 4 (correct answer but wrong method!!!). So the step size or element length should be chosen to meet that Pe is less than 1. Forcing larger step lengths therough upwinding or any other means would make the original source equations Navier-Stokes invalid in the first place. So why some researchers start with invalid equations and try to force a solution???

- Time integration in the FEM framework is done using finite difference schemes (central, explicit, and implicit) which is a traditional approach. The code has these three options while the default is the central difference (Crank-Nikolson method). If funding could be made available to me, I will investigate a new treatment of the time domain in which no difference methods are used.


- Some standard test cases will be presented later after some preparation work is completed. However, I should point out that non-square or complex geometry problems should not be the only or most relevant criteria to demonstrate model performance and handling of tough problems. The type of boundary conditions, the type of nonlinear stress-strain, and the type of boundary roughness would make even flow in square channel very challenging. Some well-known commercial software could not predict flow in square duct successfully while when one sees the so many impressive graphs of complex 3D, multi-phase, and various types of problems that one would think they reached perfection but this is not the case yet.

- The code, of course generally speaking, can be made to read grids generated by other softwares. Seeing exactly the type of mesh file I will be able to confirm this point.


- The codes that I have, would be the best for didactic aims. The reasons are: (1) they cover all types of problems starting from simple 1D flow (complete second order differential equation where advection-diffusion flows could be investigated, heat flows, structural problems), (2) 2D Laplace, and Poisson differential equations with applications to heat, elastic, and fluid flow problems, (3) transient problems, and (4) CFD problems with various degrees of complexities (nonlinear 3D unsteady turbulence models), (5) even network problems or discrete FEM is modelled (nonlinear flow of polymer melts in processing units). That the source code could be made available and in addition the ease in which data are entered and implementation of different types of boundary conditions add to their value as a didactic aims. Commercial softwares are either closed-source or open-source but very complex and lengthy to understand (one software has one million lines of coding!!!). This is not the case in my codes which are as simpler as possible while they can tackle many practical real-life problems. Any one studying these problems will have an idea of knowing how to develop codes. Bitty to see focus by many now on just how to enter the data in commercial softwares without understanding how the fluid mechanics and numerical modelling components are working.

Finally; I am not claiming that my codes cover everything but to some degree they cover several important problems and their performance is excellent in the problems covered. These codes are good for someone who is willing to start a CFD company or add them as additional tools to his modelling tools. They could be also good for testing other models performance. Remember I am just an individual who developed these codes by himself so comparisons would be unfair to CFD companies that have thousands of employees. Have you ever seen a game of soccer where one player is playing against 11 players, even my case is worse than that!!!!.

[englandthii]

I wonder how can I had 5 faculty staff in my Ph.D. committee and no one of them even mention that I have to put the code in the dissertation, nor the university mention anything. It was 1995 so I don't know rules have been changed since then to what u r saying or not. If the code is not reported in the dissertation how can the university prove that this code they own. Plus any one can make an application without the source code being published. So I don't see any legal stands or grounds for the university in my case. They did not pay me, and it was the other way I was paying the university so you can not apply the same rules pertaining to students with scholarships from the university. It seems the gentlemen who spoke had scholarships so they talk based on scholarship rules.

vidmate.app stream videos

[Dr Youssef Hafez]

Originally Posted by englandthii

I wonder how can I had 5 faculty staff in my Ph.D. committee and no one of them even mention that I have to put the code in the dissertation, nor the university mention anything. It was 1995 so I don't know rules have been changed since then to what u r saying or not. If the code is not reported in the dissertation how can the university prove that this code they own. Plus any one can make an application without the source code being published. So I don't see any legal stands or grounds for the university in my case. They did not pay me, and it was the other way I was paying the university so you can not apply the same rules pertaining to students with scholarships from the university. It seems the gentlemen who spoke had scholarships so they talk based on scholarship rules.

Sir, I agree with every word you said as this is the voice of sanity, reason and common sense. Laws, rules and regulations are based on concepts of fairness and justice.

Would you believe that I won a case back then in the early 1990s against my university, CSU?. During spring break I parked my car at a university parking building to go to the computer lab. The parking meter said something like “coins are required except during semester breaks”, of course that was like 30 year ago so I don’t remember the exact statement on the meter. When I returned back to my car, I found a parking ticket. When I called the university to object I was told to pay first then appeal. After several months I was requested to come for the appeal. I entered a room with long table where at the head of the table a man was sitting; said he is the judge; to his right side was sitting the CSU university attorney or representative, REP, with a big volume book in front of him on the table. The judge asked me to sit on the other side. I was not expecting this at all or expecting what things would be.
Anyway, the CSU Rep. opened the big volume book (it looked like it is over 1000 pages) saying this is the CSU university code book of rules, and regulations. He said that on page hundred and something that according to CSU code book “spring break is not a semester break” arguing that the semester breaks are between the semesters not during it.
The judge asked me to reply or if I have anything to say. Without any preparation or idea about that situation and feeling I am by myself against someone who is experienced in the law but deep down inside me I felt that I am on the right side. My mind worked quickly and found a very good argument based on sanity, logic, reason and fairness. I asked CSU REP if the university distributed this code book to all students especially foreign or international students like me and informed them of this rule or if on the parking meter it was written that spring break is not a semester break and I confirmed that rules or regulations should be made known to all students without assuming they know them a priori. I even said that if I had the code book how can I read 1000 pages?, and remember all rules. The judge said that I made a good point and that the university must refund me the cost of the ticket and either to write the rules (clearly stating which break is a semester or not?) and find a way to inform the students of the most important rules.

So to those Gentlemen that spoke with some what a tone of superiority with some inferior attitude, I say don’t underestimate anyone who believes truly he is right and don’t underestimate someone who can develop a complicated CFD code by himself.

[Dr Youssef Hafez]

Originally Posted by FMDenaro

Turning back to the original question, immagine I am interested in your code.
Before going to legal issues I first need to evaluate the quality of the solutions you can provide for some standard test-cases.
Just few of them:
- Taylor solution (accuracy analysis)
- lid driven cavity
- backward facing-step
- flow around a cylinder

Then, there are further test for turbulence, multi-phase flows and so on.
Without a clear example of the perfomances for such problems I wouldn’t go ahead on any kind of evaluation of the code.

Dear Dr. Denaro,

Thanks for presenting these important questions or issues (in two posts by you) about my code(s). I will try to address one issue at a time since each issue requires significant and detailed explanations. However, some issues are overlapping and crossing such as accuracy and nonlinearity, so I will try to discuss both in as much as possible.

First, regarding ACCURACY ANALYSIS, in my views it depends mostly on the mathematical complexity of the problems at hand. The more complex is the system at hand, the harder is to obtain an accurate solution and sometimes to obtain any solution at all. The mathematical complexities could be, for example, due to complex geometry, complex boundary conditions, and complex nonlinear behaviour in the structure of the governing equations. Often, for linear problems the mathematics are relatively easy and convergence is easily obtained and accuracy is reached to a higher degree as long as the geometry and boundary conditions are not stiff or tough. Nonlinearity behaviour causes not only having inaccurate and multiple solutions but to get convergence problems that bellow out the solution. Two sources of nonlinearity exist; one is due to the existence of convection velocity terms in NS and RANS equations and the second is due to nonlinear stress-strain rate relationship which exists when using algebraic stress models for turbulent stresses in RANS. Direct equation solvers are used to solve linear problems while nonlinear problems require iterative equation solvers where quality and convergence of the solution depends on the initial guessed values and their closeness to the solution. For k-e model for example an added difficulty is to obtain positive values for k and e which cannot be by definition negative.


The case of fully developed turbulent flow in a square duct includes both types of nonlinearity; convection and stress-strain consecutive laws. Although you might think because of the simple square geometry it is an easy problem, however it has these two types of nonlinearity challenge that render them to the class of tough problems. Convergence might be obtained but accuracy in obtaining secondary motion or velocities and eddy structure might be difficult and not that easy. It is a case of axial turbulent flow in the longitudinal axis of the duct which generates 8 vortices structure in the cross sectional plane normal to the main flow direction due to turbulence an-isotropy.

The case for fully developed Turbulent Flows in ducts:

Before getting to the issue of accuracy, a brief description of the model is presented (please see my Ph.D. dissertation for more details, Hafez 1995). It is a finite element computer model which was developed to solve the Reynolds equations of motion and continuity for steady and fully developed mean turbulent flow. Turbulent stresses appearing in the Reynolds equations are modelled in terms of mean velocity gradients and turbulent viscosity. The non-linear algebraic stress model developed by Speziale (1987) is used in closed channels and ducts. A two equation turbulence model consisting of the turbulent kinetic energy (k) and its rate of dissipation (ε) evaluates the turbulent viscosity that appears in the algebraic stress model.

The system of governing equations is divided into two groups: one for the continuity and momentum equations (velocity phase), and the other for the turbulent kinetic energy (k) and its rate of dissipation ε equations (k- ε phase). The two groups of equations are solved sequentially rather than simultaneously, assuming one group is known while solving for the other, until convergence and compatibility are obtained. Ten iterations are taken for the velocity phase and also 10 iterations are taken for the k- ε phase. The 10 iterations were chosen after making several numerical trial runs and it was found that it gives sufficiently accurate velocity field for a given turbulent viscosity and vice-versa.

The four-node bilinear element is used to linearly approximate each of the velocities, k, and ε while a single point quadrature approximates the pressure as a constant over the element. The iterative penalty approach is used to eliminate the continuity equation and to replace the pressure in the equations of motion. The standard Galerkin finite element method together with the finite element approximations are used to transform the non-linear governing partial differential equations into non-linear integral equations.
These equations are casted in a form suitable for applying Newton's method which solves non-linear equations. The integral equations are integrated using a Gaussian quadrature (4 points) to reduce the equations to non-linear algebraic equations. The equations are stored in a non-symmetric skyline matrix which significantly reduces the matrix size (@ 95% reduction). The equations are linearized and solved using the Gauss forward elimination and back substitution technique. The relaxation factor is taken equal to unity and convergence is obtained in all the cases considered therein.

There is no unique way for defining convergence and accuracy especially for nonlinear systems. Traditionally, the maximum difference between the solution values at any node or grid point between two consecutive iterations (Maximum residual) is considered a measure of accuracy and convergence. The smaller is the value of the residual in as much as possible the higher the quality of obtaining converged and accurate solution.

For the velocity phase, if the sum of residuals for the u velocity (erru), the v velocity (errv) and the w velocity (errw) is less than 1.e-05 then resulting solution is considered accurate enough. For the k- ε phase, if the sum of the residuals in k (ERRK) and the sum of the residuals in ε (ERRE) is less than 1.0e-06, then convergence and accuracy of the turbulent viscosity phase is achieved. Satisfaction of the continuity constraint or equation should also be considered. The continuity equation is the divergence of the velocity vector that should be zero but in the context of approximate numerical solution a small value as much as possible is accepted. The maximum residual at any element in satisfying the divergence free condition is termed (EDIVVX) at element number KMAX and the minimum residual at any element in satisfying the divergence free condition is termed (EDIVVN) at element number KMIN.

A quarter of a square duct is considered with 1600 elements (40 by 40). A brief summary of the residuals (or errors) is in the attached file named “Summary of residuals in the last phases” while the residuals for all the phases are in the file “residuals”. At phase 17 in the k- ε phase at iteration 4 in that phase the error in k is @ 10^-7 while the error in ε is in the order of @ 10^-8, and their sum is less than 1.0e-06 which triggered stoppage of the k- ε phase and switching to the velocity phase. It took 10 iterations in the velocity phase to have the sum of errors (erru, errv, and errw) to be less than 1.0e-05. The errors in satisfying the continuity requirement is in the order of @ + or - 10^-13. If full duct is considered 8-symmetrical vortices are predicted as in the attached file.

To be continued later for other validation cases.

[LuckyTran]

Making alt accounts to comment on your own thread is cringe af

Your case is not extraordinary. "No one told me" is not an excuse especially not when you're literally being told (now). Not saying no, does not mean yes. There are thousands of of international PhD students and I write this mostly for them.

I also during my PhD was not under any scholarship, also paid tuition to the university, and also was not owner of IP that I generated over a weekend. The differences are: I was aware of it, dutily filed the invention disclosure to the University, and didn't play the fool. And if you want to know the outcome of it, which took only 2 meetings. The university did its independent study and decided it would not pursue patent protection, I told the university I wanted to anyway, and so the university assigned to me the rights to do so. I had to pay the fees to the patent lawyers to do the worker and so on. And while, effectively, the university gave me full rights to do whatever I wanted with the IP, it is still clear to me (to this day) that the university has rights reserved.