Problems with COUSCOUS 2.6.0.18

[nrgeng]

When I load previously solved flow sheets into COFE, I am experiencing problems with this revision: (1) CalcProp Exception in TEA & (2) COFE CAPE-OPEN Environment Stopped Working. I restarted my CPU after the first occurrence, but it occurred again on the clean reboot (not my CPU). Can you help?

[nrgeng]

My records show that the last download consisted of COUSCOUS 2.6.0.17 and COFE 2.6.0.18. I apologize for any confusion that my referral to COUSCOUS 2.6.0.18 may have caused.

I have sent an e-mail with a data dump and an image of the crash message received. Hope these can help.

[jasper]

Can you send me the actual fsd file again and send me the steps to reproduce the crash?

[nrgeng]

None of these fsd files have been sent to you. The steps to reproduce the crash for four (4) of the flows sheets is to load the flow sheet and to click [solve].

I have a suite of seven (7) flow sheets for a large project. The project is proprietary. Only one (1) of the seven (7) flow sheets solves since my last download of 01/31/14, which consisted of COUSCOUS 2.6.0.17 and COFE 2.6.0.18. Prior to the last download all seven (7) flow sheets solved. Each one solved before it was saved. Now four (4) flow sheets crash when [solve] is clicked and two (2) flow sheets will not solve now, but do not crash. I am very nervous!

The one (1) flow sheet that solves contains only water and air with no reactions. The other six (6), which do not solve, contain reactive compounds and multiple reactions.

The previous download was on 01/23/14 consisting of COFE 2.6.0.17 only. The download before that was on 01/19/14 consisting of COFE 2.6.0.16 only. Two (2) flow sheets that crash now solved and were saved as recently as 01/28/14. Hope you can figure this one out.

[jasper]

try COUSCOUS 2.6.0.18 (now available)?

[nrgeng]

Results of testing COUSCOUS 2.6.0.18: A. two crash (was four), B. four do not solve, but do not crash (was two), C. one solves (was one). This represents a slight improvement, but six flow sheets still no longer solve. Has this type of problem occurred before when revisions are downloaded? How to proceed from here? Will I have to create six flow sheets from scratch?

I am correct in thinking that if a flow sheet solves in COCO it should also solve using any future revisions to COCO without modifications?

[jasper]

Hopefully yes.

Crashes can of course be solved, provided I can reproduce them. I cannot, as all my test documents run fine. So you will need to help me out there, please contact me.

Regarding not solve: what do you mean? Convergence problems?

[nrgeng]

By solve, I mean converge with a Log message such as "converged in 2 iterations; Solve finished in 6s, 817ms." When the file is loaded, a Log message such as "Restoring flowsheet from file; Loaded Flowsheet V1.8.2.2 (normal).fsd" appears. Is not this normal?

The above Log messages are from one of the flow sheets that no longer solves (converges) when loaded. My opinion is that if a flow sheet solves, is saved, and again is loaded; then it should immediately solve again when [solve] is clicked. Am I correct? This does not occur for the six flow sheets described in a previous Post.

By contact you, do you mean on this Forum, by e-mail, or what? Thanks for any help you can provide.

[jasper]

Contact by mail is fine. I replied already.

Yes, flowsheets that are saved in a solved status should solve within 1 iteration if loaded, presuming that the flowsheet convergence tolerance is not set tighter than the accuracy of the thermo and each of the units.

Yes, the loading message in the log is normal.

[nrgeng]

Using COUSCOUS 2.6.0.21, four of the five remaining flow sheets solved today. The flow sheet that noes not solve does solve with some of the Gibbs Reactor Units changed to Fixed Reactor Units. I am in the process of changing some of the Gibbs Reactor Units to Fixed Reactor Units in the other flow sheets to increase the solving speed. Thanks for all of your help.