<html><head><style type='text/css'>p { margin: 0; }</style></head><body><div style='font-family: Times New Roman; font-size: 12pt; color: #000000'>Hi Frank,<br><br>Yes at the moment, the 'Gambit2Nek' will not handle an internal BC. And yes, the boundary condition inputs must be a single character. To get around this, you can call it a dummy character and then go back later and change it in the REA file. Markus is doing this for a symmetry boundary condition by calling it 's' and then going back and changing it to sym.<br><br>- Michael<br><br>----- Original Message -----<br>From: nek5000-users@lists.mcs.anl.gov<br>To: nek5000-users@lists.mcs.anl.gov<br>Sent: Monday, June 7, 2010 11:51:34 AM GMT -06:00 Guadalajara / Mexico City / Monterrey<br>Subject: Re: [Nek5000-users] to Michael--> Nek5000 & Gambit, blocked file<br><br>Hi Michael,<br><br>Thanks for looking at it. After removing all boundary conditions except<br>the inlet and outlet and letting Gambit default everything else to walls<br>it runs OK. The original flow problem is two phase with the two phases<br>seperated by a BC named "marangoni_interface". If that is left out,<br>then I assume the boundary is set to "E", i.e., element connectivity.<br>If that is left in, then "Gambit2Nek" gives:<br><br>??? Attempted to access FluidBC(0,1); index must be a positive integer<br>or logical.<br><br>Error in ==> getBC2 at 23<br> FluidBC(b,1) = Type(i);<br><br>Error in ==> rea_write at 78<br>FluidBC = getBC2(ElemConn,ElmF,FaceF,TypeF,Elem1,ElemInfo);<br><br><br>Perhaps this is because "Gambit2Nek" cannot deal with an internal BC,<br>i.e., one between what were two phases? <br><br><br>Also, it seems that it is not possible to give NEK bc names that have<br>more than 1 character? For example; "ms" and "shl".<br><br>MOMENTUM NEK BC = ms<br>??? Subscripted assignment dimension mismatch.<br><br>Error in ==> readingNEU2 at 85<br> type1(i) = input('MOMENTUM NEK BC = ','s');<br><br>Error in ==> rea_write at 17<br>[Npts,Nelm,Elem,ElemMidPts1,Pts,Elem1,ElmF,...<br><br><br>Cheers,<br>Frank<br><br><br><br>On Mon, 2010-06-07 at 10:39 -0500, nek5000-users@lists.mcs.anl.gov<br>wrote: <br>> Hi Frank,<br>> <br>> I think the problem is that you specified two different groups (maybe<br>> for a fluid & solid problem?)...at the moment, the matlab program only<br>> recognizes 1 fluid group from Gambit. The idea at the moment is to<br>> merge a fluid and solid REA using Prex. I don't know if you were<br>> trying to do this, but this is the first issue I see: The Number of<br>> groups should be 1.<br>> <br>> Also, For your Boundary Conditions, I see that you have marked an<br>> inlet and outlet, but the rest I'm not sure what they are. If they are<br>> simply Walls then you don't need to mark them and Gambit will label<br>> them as Walls by default.<br>> <br>> - Michael<br>> <br>> ----- Original Message -----<br>> From: nek5000-users@lists.mcs.anl.gov<br>> To: nek5000-users@lists.mcs.anl.gov<br>> Sent: Monday, June 7, 2010 12:30:48 AM GMT -06:00 Guadalajara / Mexico<br>> City / Monterrey<br>> Subject: Re: [Nek5000-users] to Michael--> Nek5000 & Gambit, blocked<br>> file<br>> <br>> Hi Michael,<br>> <br>> The mesh is at:<br>> http://tetra.fluid.tuwien.ac.at/fmuldoo/engineering-marangoni-flows/sim2/GambitToNek/simple.neu<br>> <br>> Cheers,<br>> Frank<br>> <br>> On Sun, 2010-06-06 at 08:58 -0500, nek5000-users@lists.mcs.anl.gov<br>> wrote:<br>> > Frank,<br>> > <br>> > It is trying to extract information about the boundary conditions<br>> from <br>> > the neu file, there is most likely an error here.<br>> > <br>> > Attach the Neu file and I'll have a look<br>> > <br>> > <br>> > - Michael<br>> > <br>> > On Jun 5, 2010, at 11:28 AM, nek5000-users@lists.mcs.anl.gov wrote:<br>> > <br>> > > Hi Michael,<br>> > ><br>> > > I have been running your code and have hit the error below.<br>> > > Unfortunately, I don't ever use Matlab.<br>> > ><br>> > > Cheers,<br>> > > Frank<br>> > ><br>> > ><br>> > > ELEM & PTS MATRIX RESTURCTURE: STARTED ****<br>> > > ??? Index exceeds matrix dimensions.<br>> > ><br>> > > Error in ==> readingNEU2 at 99<br>> > > ElmF(f+1:f+n1(i),1)=BC1m(f+1:f+n1(i),1);<br>> > ><br>> > > Error in ==> rea_write at 17<br>> > > [Npts,Nelm,Elem,ElemMidPts1,Pts,Elem1,ElmF,...<br>> > ><br>> > >>><br>> > ><br>> > ><br>> > ><br>> > > On Fri, 2010-06-04 at 18:42 -0500, nek5000-users@lists.mcs.anl.gov<br>> > > wrote:<br>> > >> Hi Frank,<br>> > >><br>> > >><br>> > >><br>> > >> This is achieved by the step 2 in the list I gave: You must set<br>> the<br>> > >> edge nodes from 2 to 3. Alternatively you could set the face from<br>> 4 <br>> > >> to<br>> > >> 8 or the volume from 8 to 20 nodes. The extra nodes that gambit <br>> > >> throws<br>> > >> because of this are the midpoints. For Nekton, the recent<br>> midside<br>> > >> node support feature which fits a parabola through the two corner<br>> > >> points and midpoint of the element side is where this comes into<br>> > >> play.<br>> > >><br>> > >><br>> > >><br>> > >> So all you need to do is make sure you have set the mesh nodes as<br>> > >> above, and the matlab code does the rest. It separates the<br>> element's<br>> > >> corner nodes from the midpoint nodes. The midpoint nodes are<br>> written<br>> > >> into the curved side section of the the REA, whereas the corner <br>> > >> points<br>> > >> are written into the Mesh data section. So...The matlab code<br>> only <br>> > >> uses<br>> > >> the flag 'm' for curved side data.<br>> > >><br>> > >><br>> > >><br>> > >> Also, in the inputs section you can specify a maximum radius for<br>> > >> consideration. The code will check each edge's corner and<br>> midpoints<br>> > >> and determine the radius curvature. If it is too large it<br>> considers <br>> > >> it<br>> > >> a straight edge and moves on.<br>> > >><br>> > >><br>> > >><br>> > >> Hope this helps,<br>> > >><br>> > >><br>> > >><br>> > >> Michael<br>> > >><br>> > >><br>> > >> ----- Original Message -----<br>> > >> From: nek5000-users@lists.mcs.anl.gov<br>> > >> To: nek5000-users@lists.mcs.anl.gov<br>> > >> Sent: Friday, June 4, 2010 6:24:06 PM GMT -06:00 US/Canada<br>> Central<br>> > >> Subject: Re: [Nek5000-users] to Michael--> Nek5000 & Gambit,<br>> blocked<br>> > >> file<br>> > >><br>> > >> Hi Michael,<br>> > >><br>> > >> Thanks very much for the code. One question concerning the<br>> curved<br>> > >> sides: if one had a 3D cylindrical pipe grid from Gambit, how<br>> would<br>> > >> the<br>> > >> curvature of the outside (boundary) faces of the grid be set in<br>> your<br>> > >> code?<br>> > >><br>> > >> Cheers,<br>> > >> Frank<br>> > >><br>> > >><br>> > >> On Thu, 2010-06-03 at 21:13 -0500,<br>> nek5000-users@lists.mcs.anl.gov<br>> > >> wrote:<br>> > >>> Hi Frank,<br>> > >>><br>> > >>><br>> > >>><br>> > >>> Try the attached.<br>> > >>><br>> > >>><br>> > >>><br>> > >>> - Michael<br>> > >>><br>> > >>><br>> > >>> ----- Original Message -----<br>> > >>> From: nek5000-users@lists.mcs.anl.gov<br>> > >>> To: nek5000-users@lists.mcs.anl.gov<br>> > >>> Sent: Thursday, June 3, 2010 9:01:14 PM GMT -06:00 US/Canada<br>> Central<br>> > >>> Subject: [Nek5000-users] to Michael--> Nek5000 & Gambit, blocked<br>> > >> file<br>> > >>><br>> > >>> Hello Michael,<br>> > >>><br>> > >>> It seems that you sent an email with a *.zip attachment, which<br>> due<br>> > >> to<br>> > >>> a<br>> > >>> spam filter, got blocked. If it is renamed to say *.zipp, then<br>> it<br>> > >>> should get through.<br>> > >>><br>> > >>> Cheers,<br>> > >>> Frank<br>> > >>><br>> > >>><br>> > >>> -- <br>> > >>> Frank Herbert Muldoon, Ph.D. Mechanical Engineering<br>> > >>> Technische Universität Wien (Technical University of Vienna)<br>> > >>> Inst. f. Strömungsmechanik und Wärmeübertragung (Institute of F <br>> > >>> luid<br>> > >>> Mechanics and Heat Transfer)<br>> > >>> Resselgasse 3<br>> > >>> 1040 Wien<br>> > >>> Tel: +4315880132232<br>> > >>> Fax: +4315880132299<br>> > >>> Cell:+436765203470<br>> > >>> fmuldoo (skype)<br>> > >>><br>> > >><br>> http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html<br>> > >>><br>> > >>> _______________________________________________<br>> > >>> Nek5000-users mailing list<br>> > >>> Nek5000-users@lists.mcs.anl.gov<br>> > >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> > >>><br>> > >>><br>> > >>><br>> > >>> _______________________________________________<br>> > >>> Nek5000-users mailing list<br>> > >>> Nek5000-users@lists.mcs.anl.gov<br>> > >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> > >> -- <br>> > >> Frank Herbert Muldoon, Ph.D. Mechanical Engineering<br>> > >> Technische Universität Wien (Technical University of Vienna)<br>> > >> Inst. f. Strömungsmechanik und Wärmeübertragung (Institute of Fl <br>> > >> uid<br>> > >> Mechanics and Heat Transfer)<br>> > >> Resselgasse 3<br>> > >> 1040 Wien<br>> > >> Tel: +4315880132232<br>> > >> Fax: +4315880132299<br>> > >> Cell:+436765203470<br>> > >> fmuldoo (skype)<br>> > >><br>> http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html<br>> > >><br>> > >> _______________________________________________<br>> > >> Nek5000-users mailing list<br>> > >> Nek5000-users@lists.mcs.anl.gov<br>> > >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> > >><br>> > >><br>> > >><br>> > >> _______________________________________________<br>> > >> Nek5000-users mailing list<br>> > >> Nek5000-users@lists.mcs.anl.gov<br>> > >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> > > -- <br>> > > Frank Herbert Muldoon, Ph.D. Mechanical Engineering<br>> > > Technische Universität Wien (Technical University of Vienna)<br>> > > Inst. f. Strömungsmechanik und Wärmeübertragung (Institute of<br>> Fluid<br>> > > Mechanics and Heat Transfer)<br>> > > Resselgasse 3<br>> > > 1040 Wien<br>> > > Tel: +4315880132232<br>> > > Fax: +4315880132299<br>> > > Cell:+436765203470<br>> > > fmuldoo (skype)<br>> > ><br>> http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html<br>> > ><br>> > > _______________________________________________<br>> > > Nek5000-users mailing list<br>> > > Nek5000-users@lists.mcs.anl.gov<br>> > > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> > _______________________________________________<br>> > Nek5000-users mailing list<br>> > Nek5000-users@lists.mcs.anl.gov<br>> > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> -- <br>> Frank Herbert Muldoon, Ph.D. Mechanical Engineering<br>> Technische Universität Wien (Technical University of Vienna)<br>> Inst. f. Strömungsmechanik und Wärmeübertragung (Institute of Fluid<br>> Mechanics and Heat Transfer)<br>> Resselgasse 3<br>> 1040 Wien<br>> Tel: +4315880132232<br>> Fax: +4315880132299 <br>> Cell:+436765203470<br>> fmuldoo (skype)<br>> http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html<br>> <br>> _______________________________________________<br>> Nek5000-users mailing list<br>> Nek5000-users@lists.mcs.anl.gov<br>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>> <br>> _______________________________________________<br>> Nek5000-users mailing list<br>> Nek5000-users@lists.mcs.anl.gov<br>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br>-- <br>Frank Herbert Muldoon, Ph.D. Mechanical Engineering<br>Technische Universität Wien (Technical University of Vienna)<br>Inst. f. Strömungsmechanik und Wärmeübertragung (Institute of Fluid<br>Mechanics and Heat Transfer)<br>Resselgasse 3<br>1040 Wien<br>Tel: +4315880132232<br>Fax: +4315880132299 <br>Cell:+436765203470<br>fmuldoo (skype)<br>http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html<br><br>_______________________________________________<br>Nek5000-users mailing list<br>Nek5000-users@lists.mcs.anl.gov<br>https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users<br></div></body></html>