The Fortran 90 version of Cantera 1.2 will be released in limited
form this weekend, probably Sunday. This initial libraries released
will be for Visual Fortran under Windows. The unix/linux version will
probably come out some time next week, time permitting.

You can get a preview of what's coming by looking at the latest
draft of the in-progress Cantera Fortran manual (117 pages so far).
It's at http://blue.caltech.edu/cantera12.pdf.

Cantera 1.2 is a major upgrade from 1.1b. The two big-ticket items
are the multicomponent transport package, and the Fortran 90
interface, but there are many other improvements and additions as
well. As always, I welcome your comments and suggestions for
improvements.

Dr. Goodwin,

How do I get the 'dot' program?  What is the command?  Is it "dot
rxnpaths.dot"?

Jianhui


--- In cantera@y..., "Dave Goodwin" <dgoodwin@c...> wrote:
>
> If you generate a postscript plot using the 'dot' program, you
should be
> able to scale it to any size...
>
>
> -----Original Message-----
> From: Jianhui Hong [mailto:jianhuihong@y...]
> Sent: Monday, November 26, 2001 2:39 PM
> To: cantera@y...
> Subject: [cantera] path diagram too small to see
>
>
> Dr. Goodwin,
>
> I ran Cantera and generated a reaction path diagram.
> But the diagram is so small and the image is so hard
> to see.  Can you or somebody shed light on this issue?
>  Does anyone have the same problem?
>
> Jianhui Hong
>
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! GeoCities - quick and easy web site hosting, just
$8.95/month.
> http://geocities.yahoo.com/ps/info1
>
>
> To unsubscribe from this group, send an email to:
> cantera-unsubscribe@egroups.com
>
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/

If you generate a postscript plot using the 'dot' program, you should be
able to scale it to any size...


-----Original Message-----
From: Jianhui Hong [mailto:jianhuihong@...]
Sent: Monday, November 26, 2001 2:39 PM
To: cantera@yahoogroups.com
Subject: [cantera] path diagram too small to see


Dr. Goodwin,

I ran Cantera and generated a reaction path diagram.
But the diagram is so small and the image is so hard
to see.  Can you or somebody shed light on this issue?
  Does anyone have the same problem?

Jianhui Hong


__________________________________________________
Do You Yahoo!?
Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1


To unsubscribe from this group, send an email to:
cantera-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Dr. Goodwin,

I ran Cantera and generated a reaction path diagram.
But the diagram is so small and the image is so hard
to see.  Can you or somebody shed light on this issue?
  Does anyone have the same problem?

Jianhui Hong


__________________________________________________
Do You Yahoo!?
Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

Hasan,

One way to implement it in the Python script is to use one inlet with

mdot = mdot_1 + mdot_2
mdot Y_k  = (mdot_1 * Y_k,1 + mdot_2 * Y_k,2)
mdot h = (mdot_1 * h_1  + mdot_2 * h_2)




>
> Dave :
>
> I was thinking of running a two-inlet PSR. Is there a way, I can
> change the input file that you provided to accomplish this or do I
> have to write a script for it
>
> thanks
>
> hasan

Dave :

I was thinking of running a two-inlet PSR. Is there a way, I can
change the input file that you provided to accomplish this or do I
have to write a script for it

thanks

hasan

I wanted to try out the python cantera package. I downloaded the linux
tar file and extracted the files.

The web site says there should be a readme to get me started. When I
download, I get no readme file?

Thanks
Brian-

If you downloaded the new zero-dimensional kinetics program from
the 'Files' section, you may want to download it again. There was a
small bug that caused it to use the wrong name for the thermo
database.

It also has been modified to read parameters from file 'zero.inp' if
it exists in the local directory. This allows you to simply click on
the icon for zero.exe in Windows to run it.

--------------------

It also contains the Fortran source, and you should be able to modify
it and rebuild it if you have Compaq Visual Fortran 6.0. It should
compile with standard default settings if you add all *.f files to
your project. To link it, just link to ctf.lib and use the
multithreaded DLL Fortran libraries.

I have just added to  the 'files' section a file 'zero.zip' that
contains a new zero-dimensional kinetics program. It contains the
program source and a Windows executable that should run 'out of the
box' on any Windows PC. It is written in Fortran, using the new soon-
to-be-released Cantera Fortran interface.

Dave,

I was at the Western States Section of CI.  I was impressed and am
now trying to switch from CK-II to Cantera. I have the same problem:
when I tried to read or print page 54 and 55 of the manual, my
internet explorer froze.

Can anyone in this group show me how to run a PSR reactor?  Maybe
post your driver_psr.py file or something?  Thanks.


Jianhui Hong
John Zink Company, LLC
Tulsa, OK

Try http://blue.caltech.edu/cantera.pdf, if you haven't tried that yet.

By the way, a new version *IS* coming very soon, with a new manual. It has
been delayed due to some travel and other pressing committments in the last
couple of weeks, but is almost ready for release. By the later part of the
week, I hope.

Hello ;

I am trying to use Cantera and  am having trouble downloading the
Cantera manual available at the site. Something at page 54 has frozen
my IE several times. Can someone please mail me a copy
at "hkarim@..."

I'll really appreciate receiving a copy of the manual and many thanks
for taking your time to send me the manual

Hasan

I had hoped to release the next version of Cantera before leaving for
the Western States Section (of the Combustion Institute) meeting, but
it is not going to happen.

But if you like you can look at what I will say about Cantera there
at http://blue.caltech.edu/cantera2.htm.

The big new features are the Fortran interface, multicomponent
transport, and reactor networks.

Well, maybe it will be ready by the end of the week with any luck.

Hope to see some of you at the meeting.

Dave

To all Cantera user's group members ---

Although I suggested in a previous message that you post here any
problems or suggestions, since they may be of general interest, you
are of course also welcome to contact me directly, if you prefer.

I have put a text file in the 'files' section with my contact
information (e-mail, phone, fax, ...).

Please feel free to contact me about Cantera if you need help or have
suggestions for improvements. Even if you are not having problems but
are using Cantera, I would be interested to know what your
application is and how well it is working for you.

Thanks for the patch. I'll apply it to the next version due out in a couple
of days.

Hi,

I've found some problems during compiling your library with Intel C++
compiler:
	 1) stream format flags should OR'ed not added;
	 2) result of container.find() should be compared for '!=' and
	  not "<" with container.end()


Here is the patch against your CVS:
Index: CKReader/src/CKParser.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/CKReader/src/CKParser.cpp,v
retrieving revision 1.10
diff -u -r1.10 CKParser.cpp
--- CKReader/src/CKParser.cpp   2001/07/25 02:14:24     1.10
+++ CKReader/src/CKParser.cpp   2001/09/21 13:59:47
@@ -25,7 +25,8 @@

  namespace ckr {

-    static int parseGroupString(string str, vector<string>& esyms,
vector<int>& result);
+    static int parseGroupString(string str, vector<string>& esyms,
+                               vector<int>& result);

      /// @internal
      static string dummy_string;
@@ -35,11 +36,11 @@
       *  1, 2, 3, or 4 in column 80 do not.
       */
      static void illegalThermoLine(ostream& f,
-        char n, int linenum = -1)
+                                 char, int linenum = -1)
      {
          throw CK_SyntaxError(f, "column 80 must "
              "contain an integer", linenum);
-    };
+    }


      /**
@@ -50,7 +51,7 @@
      {
          string msg = "illegal number: "+s;
          throw CK_SyntaxError(f, msg, linenum);
-    };
+    }


      extern void getDefaultAtomicWeights(map<string,double>& weights);
@@ -87,7 +88,7 @@
       *  @param log     log file output stream
       */
      static void addElement(string symbol, double atoms,
-        Species& sp, ostream& log) {
+                          Species& sp, ostream&) {

          if (atoms != 0.0) {
              Constituent e;
@@ -145,7 +146,7 @@
                      coeffs.push_back(m);
                  }
                  else
-                    coeffs.push_back(1.0);
+                    coeffs.push_back( 1.0 );
                  inplus = false;
              }
              else if (*p == '+')
@@ -330,17 +331,16 @@
          int ntok;

          elements.clear();
-        while (1 > 0) {
-
+        while (true) {
  next:
              if (advanceToKeyword("ELEM", "SPEC")) {
                  firsttok = 1;
-                while (1 > 0) {
+                while (true) {
                      do {
                          getCKLine(s, comment);
                          getTokens(s.begin(), s.size(), toks);
                          ntok = toks.size();
-                    } while (ntok == 0);
+                    } while (!ntok);

                      if (firsttok == 0 && isKeyword(toks[0])) {
                          putCKLine(s,comment);
@@ -364,7 +364,7 @@
                              if (el.atomicWeight > 0.0) el.valid = 1;
                              else el.valid = 0;
                              if (find(elements.begin(),
-                                elements.end(), el) < elements.end())
{
+                                elements.end(), el) !=
elements.end()) {
                                  *m_log << "warning... duplicate
element "
                                         << el.name << " (ignored)." <<
endl;
                              }
@@ -376,11 +376,10 @@
                  }
              }
              else {
-                if (elements.size() == 0) return false;
+                if (elements.empty()) return false;
                  else return valid(elements);
              }
          }
-        return false;
      }


@@ -399,17 +398,17 @@
          int ntok;
          int nsp = 0;

-        while (1 > 0) {
+        while (true) {

   next:
              if (advanceToKeyword("SPEC", "THER")) {
                  firsttok = 1;
-                while (1 > 0) {
+                while (true) {
                      do {
                          getCKLine(s, comment);
                          getTokens(s.begin(), s.size(), toks);
                          ntok = toks.size();
-                    } while (ntok == 0);
+                    } while (!ntok);

                      if (firsttok == 0 && isKeyword(toks[0])) {
                          putCKLine(s,comment);
@@ -421,7 +420,7 @@
                              Species sp;
                              sp.name = toks[i];
                              if (find(species.begin(), species.end(),
sp)
-                                < species.end()) {
+                                != species.end()) {
                                  *m_log << "warning... duplicate
species "
                                         << sp.name << " (ignored)." <<
endl;
                              }
@@ -505,7 +504,7 @@
              tmax = atof(s.substr(20,10).c_str());

              if (verbose) {
-                log.flags(ios::showpoint+ios::fixed);
+                log.flags(ios::showpoint|ios::fixed);
                  log.precision(2);
                  log << endl << " default Tlow, Tmid, Thigh: " << tmin
<< "  "
                      << tmid << "  " << tmax << endl;
@@ -521,15 +520,16 @@

          bool getAllSpecies = (nsp > 0 && match(names[0],"<ALL>"));

-        while (1 > 0) {
-            if (nsp == 0) break;
+        while (true) {
+            if (!nsp) break;

              Species spec;
              readThermoRecord(spec);

              if (spec.name == "<END>") break;

-            if (getAllSpecies || (find(names.begin(), names.end(),
spec.name) < names.end()))
+            if (getAllSpecies ||
+               (find(names.begin(), names.end(), spec.name) !=
names.end()))
              {
                  species[spec.name] = spec;
                  if (spec.tmid == 0.0) {
@@ -723,7 +723,7 @@
          Reaction rxn;
          bool ok = true;

-        while (1 > 0) {
+        while (true) {

              // skip blank or comment lines
              do {
@@ -778,12 +778,12 @@
                      }
                      int np = rxn.products.size();
                      int npatoms = 0;
-                    for (int ik = 0; ik < np; ik++) npatoms +=
int(rxn.products[ik].number);
+                    for (int ik = 0; ik < np; ++ik) npatoms +=
int(rxn.products[ik].number);
                      if (pg.size() != npatoms)
                          throw CK_SyntaxError(*m_log," groups not
specified for all products", m_line);
                      else if (ierp < 0)
                          throw CK_SyntaxError(*m_log," error in
product group specification", m_line);
-                    for (int ip = 0; ip < np; ip++) {
+                    for (int ip = 0; ip < np; ++ip) {
                          rxn.products[ip].groups = pg[ip];
                      }
                  }
@@ -927,7 +927,7 @@
                  bool hasAuxData;
                  string name, data;
                  map<string, int> kwindex;
-                while (1 > 0) {
+                while (true) {

                      hasAuxData = extractSlashData(s, name, data);
                      if (!hasAuxData && name == "") break;
Index: CKReader/src/CKReader.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/CKReader/src/CKReader.cpp,v
retrieving revision 1.7
diff -u -r1.7 CKReader.cpp
--- CKReader/src/CKReader.cpp   2001/07/25 02:14:24     1.7
+++ CKReader/src/CKReader.cpp   2001/09/21 13:59:47
@@ -251,7 +251,7 @@
      bool ok = true;
      //    int ns = species.size();
      int nrxns = reactions.size();
-    log.flags(0);
+    log.flags(ios::fmtflags( 0 ));
      log.precision(6);

      log << endl;
Index: CKReader/src/writelog.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/CKReader/src/writelog.cpp,v
retrieving revision 1.3
diff -u -r1.3 writelog.cpp
--- CKReader/src/writelog.cpp   2001/07/25 02:14:24     1.3
+++ CKReader/src/writelog.cpp   2001/09/21 13:59:47
@@ -200,12 +200,12 @@
          }
          log << ")";

-        log.flags(ios::showpoint+ios::fixed);
+        log.flags(ios::showpoint|ios::fixed);
          log.precision(2);
          log << endl << "   Tlow, Tmid, Thigh: (" << spec.tlow << ", "
<<
              spec.tmid << ", " << spec.thigh << ")" << endl << endl;
          log << "   coefficients (low, high):" << endl;
-        log.flags(ios::scientific+ios::uppercase+ios::internal);
+        log.flags(ios::scientific|ios::uppercase|ios::internal);
          log.precision(8);
          for (int j = 0; j < 7; j++) {
              log << "   a" << j + 1;
Index: Cantera/src/ReactionPath.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/Cantera/src/ReactionPath.cpp,v
retrieving revision 1.16
diff -u -r1.16 ReactionPath.cpp
--- Cantera/src/ReactionPath.cpp        2001/07/25 02:14:24     1.16
+++ Cantera/src/ReactionPath.cpp        2001/09/21 13:59:47
@@ -161,7 +161,7 @@
      {
          int i;
          doublereal flxratio, flmax = 0.0;
-        s.flags(ios::showpoint+ios::fixed);
+        s.flags(ios::showpoint|ios::fixed);
          s.precision(3);

          // a directed graph
@@ -332,7 +332,7 @@
                  }
              }
          }
-        s.flags(ios::showpoint+ios::scientific);
+        s.flags(ios::showpoint|ios::scientific);
          s.precision(2);
                      hasAuxData = extractSlashData(s, name, data);
                      if (!hasAuxData && name == "") break;
Index: CKReader/src/CKReader.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/CKReader/src/CKReader.cpp,v
retrieving revision 1.7
diff -u -r1.7 CKReader.cpp
--- CKReader/src/CKReader.cpp   2001/07/25 02:14:24     1.7
+++ CKReader/src/CKReader.cpp   2001/09/21 13:59:47
@@ -251,7 +251,7 @@
      bool ok = true;
      //    int ns = species.size();
      int nrxns = reactions.size();
-    log.flags(0);
+    log.flags(ios::fmtflags( 0 ));
      log.precision(6);

      log << endl;
Index: CKReader/src/writelog.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/CKReader/src/writelog.cpp,v
retrieving revision 1.3
diff -u -r1.3 writelog.cpp
--- CKReader/src/writelog.cpp   2001/07/25 02:14:24     1.3
+++ CKReader/src/writelog.cpp   2001/09/21 13:59:47
@@ -200,12 +200,12 @@
          }
          log << ")";

-        log.flags(ios::showpoint+ios::fixed);
+        log.flags(ios::showpoint|ios::fixed);
          log.precision(2);
          log << endl << "   Tlow, Tmid, Thigh: (" << spec.tlow << ", "
<<
              spec.tmid << ", " << spec.thigh << ")" << endl << endl;
          log << "   coefficients (low, high):" << endl;
-        log.flags(ios::scientific+ios::uppercase+ios::internal);
+        log.flags(ios::scientific|ios::uppercase|ios::internal);
          log.precision(8);
          for (int j = 0; j < 7; j++) {
              log << "   a" << j + 1;
Index: Cantera/src/ReactionPath.cpp
===================================================================
RCS file: /cvs/cantera-1.1b/Cantera/src/ReactionPath.cpp,v
retrieving revision 1.16
diff -u -r1.16 ReactionPath.cpp
--- Cantera/src/ReactionPath.cpp        2001/07/25 02:14:24     1.16
+++ Cantera/src/ReactionPath.cpp        2001/09/21 13:59:47
@@ -161,7 +161,7 @@
      {
          int i;
          doublereal flxratio, flmax = 0.0;
-        s.flags(ios::showpoint+ios::fixed);
+        s.flags(ios::showpoint|ios::fixed);
          s.precision(3);

          // a directed graph
@@ -332,7 +332,7 @@
                  }
              }
          }
-        s.flags(ios::showpoint+ios::scientific);
+        s.flags(ios::showpoint|ios::scientific);
          s.precision(2);
          double sat;
          map<int, SpeciesNode*>::const_iterator b = m_nodes.begin();

Yes, that version of the manual has printing problems.

I will soon be introducing an update to Cantera that should address all your
other questions too. It will hopefully be ready with new documentation in a
week or so. Among other things, it will include:

- a Fortran 90 API that allows creating and manipulating Cantera objects
from Fortran 90 (no knowledge of C++ required, but it does use Fortran 90
modules and derived types)

- a full Chemkin-compatible multicomponent transport package

- linkable stirred reactors with multiple inlets/outlets, closed-loop
control, etc.

I hope to get this all documented and released before classes start here (2
weeks), after which progress may be less predictable...

Dave



-----Original Message-----
From: Zhigang [mailto:zzwyxh@...]
Sent: Monday, September 10, 2001 3:17 PM
To: cantera@yahoogroups.com
Subject: [cantera] Re: misc


Hi, Dave,

I got the same problem, it seems that on page 51 and 52, three pics
showing the settings of working under Visual C++ is incomplete.

So, could you fix that?

And in the manual A.4.1 Building C++ application in Visual Studio,
cantera.lib, ckreader.lib and cvode.lib is specified, but in the
package I downloaded I can't find those files, as well as
the "Cantera.dsw". Could you look at that too?

Thanks for your work.

Zhiwei


--- In cantera@y..., Jay Jeffries <Jeffries@n...> wrote:
> Hello Dave,
> If I open this PDF I can read it from my netscape, but when I try
to print it
> my computer crashes.  This appears to be consistent behavior with
this file
> only.
> Jay
>
> At 06:16 AM 6/16/2001 +0000, you wrote:
>
> >There is now a pdf version of the manual. You can access it at
> >http://blue.caltech.edu/cantera.pdf. Like Cantera itself, the
manual
> >is a work in progress, and is subject to change...
> >
> >A number of you have e-mailed me with various questions, problems,
> >etc. I am happy to answer all e-mail correspondence, but I would
> >encourage you to post these messages here if the topic is
potentially
> >of interest to others. If you are having problems with something,
it
> >is likely someone else is too. I get an e-mail message from Yahoo
as
> >soon as any message is posted, so my reply will be no slower than
if
> >you e-mail me directly.
> >
> >I'm interested in hearing the comments of anyone who has tried to
> >download, install, and use Cantera. Have you tried the Python
> >interface? Or CtLib? How did they work? etc.
> >
> >A final note: much of the Cantera web site will be down tomorrow
> >morning due to a necessary power shutdown in the building housing
the
> >host machines. Should be up by noon.
> >
> >
> >
> >
> >
> >To unsubscribe from this group, send an email to:
> >cantera-unsubscribe@egroups.com
> >
> >
> >
> >Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
>
> ********************************************
> Jay B. Jeffries, Ph.D.
> Mechanical Engineering
> Thermosciences Division, Bldg. 520
> Stanford University
> Stanford, CA 94305-3032
> 650-736-0007 voice
> 650-723-1748 fax
> Jay.Jeffries@S...



To unsubscribe from this group, send an email to:
cantera-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Hi, Dave,

I got the same problem, it seems that on page 51 and 52, three pics
showing the settings of working under Visual C++ is incomplete.

So, could you fix that?

And in the manual A.4.1 Building C++ application in Visual Studio,
cantera.lib, ckreader.lib and cvode.lib is specified, but in the
package I downloaded I can't find those files, as well as
the "Cantera.dsw". Could you look at that too?

Thanks for your work.

Zhiwei


--- In cantera@y..., Jay Jeffries <Jeffries@n...> wrote:
> Hello Dave,
> If I open this PDF I can read it from my netscape, but when I try
to print it
> my computer crashes.  This appears to be consistent behavior with
this file
> only.
> Jay
>
> At 06:16 AM 6/16/2001 +0000, you wrote:
>
> >There is now a pdf version of the manual. You can access it at
> >http://blue.caltech.edu/cantera.pdf. Like Cantera itself, the
manual
> >is a work in progress, and is subject to change...
> >
> >A number of you have e-mailed me with various questions, problems,
> >etc. I am happy to answer all e-mail correspondence, but I would
> >encourage you to post these messages here if the topic is
potentially
> >of interest to others. If you are having problems with something,
it
> >is likely someone else is too. I get an e-mail message from Yahoo
as
> >soon as any message is posted, so my reply will be no slower than
if
> >you e-mail me directly.
> >
> >I'm interested in hearing the comments of anyone who has tried to
> >download, install, and use Cantera. Have you tried the Python
> >interface? Or CtLib? How did they work? etc.
> >
> >A final note: much of the Cantera web site will be down tomorrow
> >morning due to a necessary power shutdown in the building housing
the
> >host machines. Should be up by noon.
> >
> >
> >
> >
> >
> >To unsubscribe from this group, send an email to:
> >cantera-unsubscribe@egroups.com
> >
> >
> >
> >Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
>
> ********************************************
> Jay B. Jeffries, Ph.D.
> Mechanical Engineering
> Thermosciences Division, Bldg. 520
> Stanford University
> Stanford, CA 94305-3032
> 650-736-0007 voice
> 650-723-1748 fax
> Jay.Jeffries@S...

A beta version of the multicomponent transport library is now
available. The CVS repository has been updated, as has the on-line
C++ source code documentation. See classes Transport,
TransportFactory, and MultiTransport.

A Fortran interface (mtlib) has also been implemented that has an
interface very similar to the Chemkin TRANLIB library. The
combination of ctlib/mtlib can be used to port most Chemkin-based
reacting flow programs to Cantera.

A binary version of these libraries for Windows may be found in
the 'files' section of the user's group. Get file win_ctlib.zip,
unzip it, read README.txt, and follow the procedure described there.
The code is contained a DLL file, and linking to it should be
straightforward. You do not need a C++ compiler to use this, but will
need Compaq Visual Fortran (CVF) or another Fortran compiler
compatible with CVF. Fortran applications should build "out of the
box," with no special handling of mixed-language programming required
on your part. (This is handled by INTERFACE specifications in Fortran
90 module files ctlib.f and mtlib.f.)

Note also that a few things are missing, notably mixture-averaged
conductivity and diffusion coefficients. These will be added very
soon.

I will be out next week, but welcome your questions/comments if you
have a chance to try this out.

Dave Goodwin

In my last message I posted, I promised a long list of things that
would be available soon...

Well, as it has turned out things have taken longer that expected
(which is not unexpected), since I have been trying to fit this into
spare moments between the Things I Am Actually Paid To Do.

Anyway, here's the status.

A multicomponent transport package is now fully operational, but not
yet incorporated into the released version. The property values
computed agree well with those from TRANLIB (but see below), and it
runs on the systems I've tested about twice as fast as TRANLIB. The
biggest reason for the performance increase is that the L matrix
equation is solved iteratively using GMRES, rather than by LU
decomposition (although it can do it with LU decomposition too).

In the course of developing this package, I have discovered what may
be a minor inconsistency in the TRANFIT program used by the TRANLIB
package. Essentially, TRANFIT applies a certain second-order
correction to the binary diffusion coefficients that the TRANLIB
documentation says is not being applied, and theory indicates
probably shouldn't be applied when using these coefficients to
evaluate multicomponent properties. This correction affects some
binary diffusion coefficients by as much as 7% for the test cases I
ran, and affects the multicomponent diffusion coefficients by perhaps
3% or 4%. This is probably not a problem for most simulations. The
Cantera multicomponent trannsport package does not apply this
correction. If any of you are users of TRANFIT/TRANLIB and would like
to know more about this issue, send me an e-mail message.

If this correction is disabled in TRANFIT (which can be done in the
old Chemkin-II version if you have the source, but can't be in the
Chemkin-III version), then the properties computed agree to about
0.2%. The remaining discrepancy is apparently due to slightly
different ways of generating the polynomial fits to collision
integrals (which are only +/- 1% accurate anyway).

This transport package should be available in a few days.

I have also been cleaning up and slightly reorganizing some low-level
C++ code. If you are using Cantera in C++, let me know and I will try
to minimize disruption.

Although it may not seem like it, Cantera is moving along, and I plan
to release an updated version soon - probably next week. This message
is just to let you know what is coming....


1. Transport properties. As alluded to in the previous messages, a
multicomponent transport package will be added soon. The plan is to
eventually implement several interchangable transport modules, from
simple ones for quick work to multicomponent transport for accurate
calculations. The first one will be a multicomponent transport
package that nominally implements the expressions published in the
Sandia report (SAND86-8246) by Kee et al. which were implemented in
the Chemkin "TRANLIB" transport library. Of course, as always this is
being implemented in open-source C++. Actually, a portion of the
package is already integrated into Cantera. If you have the CVS
version of Cantera, update your working copy, and then the following
C++ fragment should work:


....
CKGas gas("chem.inp", "therm.dat");
gas.inititializeTransport("tran.dat");
....
cout << gas.viscosity();     /* print the viscosity in Pa-s */


The files are property databases in the usual Chemkin-compatible
format. So far only the viscosity implementation is finished, but the
rest will follow soon.


2. Linked stirred reactor models. Classes for stirred reactors are
working now and will be available next week. These simuulate full
transient behavior, are not restricted to ideal gases, and may be
linked to other stirred reactors through flow controllers (mass flow
controllers, pressure regulators, etc.).  By combining the
appropriate components, you can build everything from a simple
constant-pressure CSTR to an engine simulator. This will initially
only be available in C++; at some point it will be added to the
Python package too.


3. A Chemkin mechanism to C++ converter. Considerable performance
gains are possible by hard-coding a reaction mechanism in Fortran, C,
or C++. Cantera will soon have the capability to read in a Chemkin-
format file and write out C++ source code for a Cantera-compatible
C++ class that hardwires a particular mechanism. Due to its
widespread popularity, a fast version of GRI-Mech 3.0 will be built
into Cantera in this way.


4. An updated manual. The documentation still leaves much to be
desired, but is improving. An update to the manual will be available
soon that will hopefully not have the printing problems some have
experienced. Also, the on-line C++ code documentation is slowly being
fleshed out, and will be available also as a reference manual (PDF).


As always, feedback is welcome. I would be interested to know how you
are using or plan to use Cantera, etc. Also, if you are interested in
helping with Cantera development, contact me - there is always plenty
to do.

Dave Goodwin


[p.s. - if you don't want to receive these e-mails, you can elect to
receive a periodic digest, or to not receive any messages. Go to the
user's group page to change options.]

Kevin,

Yes, it is coming soon. Maybe another week or so.


Dave


-----Original Message-----
From: kevin_ellwood@... [mailto:kevin_ellwood@...]
Sent: Tuesday, July 03, 2001 4:53 AM
To: cantera@yahoogroups.com
Subject: [cantera] TRANLIB replacement


Hi Dave,

You had mentioned in an email exchange that there might be a TRANLIB
replacemnt coming soon.  I can't quite remember who you said was going
to contribute the library but has there been any success.

Thanks and keep up the great work.
Kev


To unsubscribe from this group, send an email to:
cantera-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Hi Dave,

You had mentioned in an email exchange that there might be a TRANLIB
replacemnt coming soon.  I can't quite remember who you said was going
to contribute the library but has there been any success.

Thanks and keep up the great work.
Kev

Hello Dave,
If I open this PDF I can read it from my netscape, but when I try to print it
my computer crashes.  This appears to be consistent behavior with this file
only.
Jay

At 06:16 AM 6/16/2001 +0000, you wrote:

>There is now a pdf version of the manual. You can access it at
>http://blue.caltech.edu/cantera.pdf. Like Cantera itself, the manual
>is a work in progress, and is subject to change...
>
>A number of you have e-mailed me with various questions, problems,
>etc. I am happy to answer all e-mail correspondence, but I would
>encourage you to post these messages here if the topic is potentially
>of interest to others. If you are having problems with something, it
>is likely someone else is too. I get an e-mail message from Yahoo as
>soon as any message is posted, so my reply will be no slower than if
>you e-mail me directly.
>
>I'm interested in hearing the comments of anyone who has tried to
>download, install, and use Cantera. Have you tried the Python
>interface? Or CtLib? How did they work? etc.
>
>A final note: much of the Cantera web site will be down tomorrow
>morning due to a necessary power shutdown in the building housing the
>host machines. Should be up by noon.
>
>
>
>
>
>To unsubscribe from this group, send an email to:
>cantera-unsubscribe@egroups.com
>
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

********************************************
Jay B. Jeffries, Ph.D.
Mechanical Engineering
Thermosciences Division, Bldg. 520
Stanford University
Stanford, CA 94305-3032
650-736-0007 voice
650-723-1748 fax
Jay.Jeffries@...

There is now a pdf version of the manual. You can access it at
http://blue.caltech.edu/cantera.pdf. Like Cantera itself, the manual
is a work in progress, and is subject to change...

A number of you have e-mailed me with various questions, problems,
etc. I am happy to answer all e-mail correspondence, but I would
encourage you to post these messages here if the topic is potentially
of interest to others. If you are having problems with something, it
is likely someone else is too. I get an e-mail message from Yahoo as
soon as any message is posted, so my reply will be no slower than if
you e-mail me directly.

I'm interested in hearing the comments of anyone who has tried to
download, install, and use Cantera. Have you tried the Python
interface? Or CtLib? How did they work? etc.

A final note: much of the Cantera web site will be down tomorrow
morning due to a necessary power shutdown in the building housing the
host machines. Should be up by noon.

I have put a first draft of a manual for Cantera in the "files"
section of the users group. I had some trouble trying to generate
pdf, so it is a zipped postscript file.

It documents the Python interface, as well as how to port Chemkin-
based Fortran programs to Cantera used CtLib.

The zip file of the Cantera source distribution may not be fully up-
to-date, but you can always get the latest version using CVS.

I'm heading out for a week's vacation at the beach... if you have a
chance to try it out in the meantime, I'd be interested to hear about
your experience with Cantera when I get back.

Dave G.

First of all, welcome to the new members.

The installation instructions in the Python package contained errors,
which have now been fixed. Other minor fixes:

1) reaction path diagram labels include '+ M', or '(+ M)' if
appropriate.

2) script 'driver.py' now prints both pressure and density to the CSV
file.

An interface to the Python scripting language is now available for
Windows PCs. You don't need to compile anything to use it. In fact,
you don't need to program at all, and you don't need to know the
Python language. A general-purpose driver script is included that
allows you to:

- carry out kinetics simulations in adiabatic or isothermal reactors
at constant pressure or volume

- compute chemical equilibrium

- generate reaction path diagrams (see www.cantera.org for an example)

Simply edit the driver script to set the options and parameter values
for your problem, and process it with python.

Of course, if you do know some Python, you can customize this script
to do other things too. Documentation will be available shortly. In
the meantime, take a look at the driver script (driver.py).

Go to www.cantera.org to download the package.
Let me know if you run into problems.

You can get it from http://www.cantera.org

Welcome to the Cantera users group. There is not much here yet, but
there will be soon. Quite a bit should be available by the end of
this month. For now, take a look at chemeq (under 'Files'), a simple
text-based driver for the Cantera chemical equilibrium solver.