CAPE-OPEN

I am working through the five reactor types provided in COUSCOUS. I have had success with the Fixed Conversion Reactor using several parallel reactions. From my reading of the Help system, I think that the Gibbs Minimization Reactor is the next (second) more complex reactor. The Help system under COUSCOUS does not list it. Is this an oversight? Only four reactors are shown in Contents: Reactor. However, it is listed in Contents: Gibbs.

I loaded Methanol_iecr49p6150.fsd to work with its provided reactor. It loads OK, and it solves on my computer in less than six seconds. The simulation uses a Fixed Conversion Reactor with two reactions specified. I switched the reactor to a Gibbs Minimization Reactor keeping the same property pack and reaction pack that accompanied the simulation. I made no Flowsheet Configuration changes. The simulation tried to solve with one reaction specified. The Last Run Report showed that the solution converged.

With this success, I then added the original second reaction. Therein lies the problem. The simulation crashed, dropping out of COFE after displaying an error dialog. The simulation proceed for a short time (5 sec) though before it crashed. Can you help with this problem?

My System Specifications: Intel CPU: 2 GHz, 2 GB RAM; 64-bit Operating System; Windows 7 SP1

nrgeng wrote:The Help system under COUSCOUS does not list it. Is this an oversight?

Ah - in the index you mean. Yes, oversight. Fixed.

nrgeng wrote:With this success, I then added the original second reaction. Therein lies the problem. The simulation crashed, dropping out of COFE after displaying an error dialog. The simulation proceed for a short time (5 sec) though before it crashed. Can you help with this problem?

Of course. Can you save the fsd file right before the crash and send me via e-mail or attach? And describe the exact steps to make it crash?

Thank you for reporting these issues.

Please notice the stack of reactors on the right side of the image. Working up from the bottom, calculate each unit in succession: (1) Reactor, (2) GibbsReactor_19 CO2, (3) GibbsReactor_19 CO, and then finally (4) GibbsReactor_19 CO & CO2. The jackpot is Item 4 – crash! The same thing happens if you click the Solve icon (but less dramatic). Thanks for your help.

Fixed. Thank you for reporting. Fix is available via CUP.

The problem was a crash in a 3rd party routine for an LU decomposition for a matrix containing a row of zeroes.

Thanks nrgeng for sharing your learnings. By sharing we all can learn and help improve this software.