Reported by michelpons, 15 years ago.

De: Michel Pons [technologyofficer@…] Envoyé: Sunday, April 26, 2009 10:26 PM À: 'Michael Halloran' Cc: 'Ahsan Akhtar' Objet: RE: GLCC UO In Aspen Hysys v7.1

Pièces jointes: COLTT09_011_Screenshots.doc

Michael,

Please find attached a Word file with screenshots obtained with the scenario of COLTT issue 09-011. I ignore the assertion failed error and then Aspen Hysys crashes.

Best regards

Michel ======================================================= De: Michael Halloran [michaeljhalloran@…] Envoyé: Saturday, April 25, 2009 11:35 PM À: Michel Pons Cc: Ahsan Akhtar Objet: Re: GLCC UO In Aspen Hysys v7.1 Michael,

Can you add information about the assertion? When you get an assertion do you try to ignore it and continue the run?

I don't understand the logs here. The 1.07 log should show the calls to get_ComponentName in the same places as the 1.08 log. The 1.07 logger is logging them and they must have been called or the name shown in the log wouldn't have changed to Check for the 1st parameter. So that looks like a 1.07 problem to me; the 1.08 log is what I would expect.

Regards Michael ========================================================== De: Michel Pons [technologyofficer@…] Envoyé: Thursday, April 23, 2009 10:15 PM À: 'Ahsan Akhtar' Cc: 'Michael Halloran' Objet: GLCC UO In Aspen Hysys v7.1

Pièces jointes: hysys_042309_215826v107.log; hysys_042309_215412v1082.log

Ahsan,

I have ran a combination involving a CAPE-OPEN UO using Thermo 1.0 and Aspen Hysys v7.1. The UO comes from TOTAL and has been made with the Visual Basic Wizard.

The scenario starts from an Unconnected case file that displays three streams, one to be used as inlet and two as outlets to the UO. I use “Add Operation” then Extensions the CAPE-OPEN 1.0 and select GLCC UO. Then I accept this choice. Things are differing from then on with/wothout COLTT and between COLTT versions.

Without COLTT enabled on the GLCC UO, I get message boxes from Aspen Hysys of the type “Error in ExposeCOParstoHysys(13)”. This has been documented to Aspentech Support and is in the process of being solved. Several of these message boxes have to be Okeyed before arriving at the point where connections between ports and material streams can be made. I stop the scenario here.

With COLTT v1.07 enabled on the GLCC UO, I get first the same error message as above and then I have a message box repeated quite a number of times “ActiveXtender Error setting cell color (SetFontCellColor).

With COLTT v1.08.02 (after changing the COLoggersST.dll file) I get, after accepting the UO, a debug assertion failed message and then Aspen Hysys crashes.

I have joined to this message the two log files obtained. I am surprised that the two logs do not show exactly the same sequence.

COLTT v1.07

Parameter Collection <Unnamed> : Call to Item requesting Item 1 Parameter Collection Item returns : 0x569a84 Parameter Collection <Unnamed> : Return from Item - No Error Parameter <Check> : Call to get_Specification

get_specification returns: 0x56e97c

Parameter <Check> : Return from get_Specification - No Error

COLTT v1.08.02 ParameterCollection <Anonymous> : Call to Item requesting Item 1 ParameterCollection Item returns : 0x5698ec ParameterCollection <Anonymous> : Return from Item - No Error Parameter <Anonymous> : Call to ICapeIdentification.get_ComponentName Parameter <Anonymous> : Return from ICapeIdentification.get_ComponentName - No Error Parameter <Check> : Call to get_Specification

get_specification returns: 0x56ccfc

Parameter <Check> : Return from get_Specification - No Error

I have logged this issue as 09-011 in the meantime.

Best regards

Michel

Reported by michelpons, 15 years ago.

De: Michael Halloran [michaeljhalloran@…] Envoyé: Saturday, April 25, 2009 7:40 PM À: Michel Pons Cc: Ahsan Akhtar Objet: Re: Simulis Thermodynamics in Aspen Hysys 2004 Michel,

You'll notice that the calls which use anonymous as the name are all on the ICapeUtilities Interface which is a separate logger in 1.08 after the refactoring. The problem is that this logger hasn't been told that the name of the unit has changed.

Ahsan, there needs to be a mechanism to pass the name of the owning PMC to all aggregated loggers when the name is changed. It should be similar to SetLoggedComponent but it only needs to pass name. It needs to be called from put_PMCName. Could you investigate?

Regards Michael =========================================================== De: Michel Pons [technologyofficer@…] Envoyé: Wednesday, April 15, 2009 10:28 AM À: 'Ahsan Akhtar' Cc: 'michaeljhalloran@…' Objet: FW: Simulis Thermodynamics in Aspen Hysys 2004

Pièces jointes: hysys_041009_170955.log

Ahsan,

I was not able to pinpoint which of the « old » issues this was related to. So please consider this as issue 09-007 till one knows better.

Best regards

Michel =============================================================== De: Michel Pons [technologyofficer@…] Envoyé: Friday, April 10, 2009 5:54 PM À: 'Michael Halloran' Cc: 'Ahsan Akhtar' Objet: RE: Simulis Thermodynamics in Aspen Hysys 2004

Pièces jointes: hysys_041009_170955.log

Michael,

Please find attached a log of a MatLabUO in Aspen Hysys v7.1 as logged with COLTT v1.08.02.

I don’t see a difference between having MatLabUO logged and not logged. I get the same kind of errors displayed by Aspen Hysys.

Still I am a bit bewildered by the fact that once Aspen Hysys has put CO-100 as component name,

Unit <Anonymous> : Call to put_ComponentName Unit put_ComponentName puts: CO-100 Unit <CO-100> : Return from put_ComponentName - No Error

I still get now and then <Anonymous> to be used as name for the UO:

Unit <Anonymous> : Call to get_parameters

Unit <Anonymous> : Return from get_parameters - No Error

Unit <Anonymous> : Call to Edit Unit <Anonymous> : Return from Edit - No Error

Amidst numerous entries using <CO-100> as the Unit name. I think this is an old issue though.

Best regards

Michel

Reported by michelpons, 15 years ago.

De: Michel Pons [technologyofficer@…] Envoyé: Tuesday, May 19, 2009 9:13 AM À: 'Ahsan Akhtar' Cc: 'michaeljhalloran@…' Objet: RE: Friday Session

Ahsan,

Sorry for the delay but I was tied up by other tasks.

I have transferred the DLL you sent to the COLTT 09-006 VM and I have run the scenario. I can only report that it does not work any better. I can’t display the list of parameters for the UO and the live log screen in COLTT controller remains empty.

Best regards

Michel ========================================================= De: Ahsan Akhtar [ahsan.akhtar@…] Envoyé: Friday, May 15, 2009 8:10 AM À: 'Michel Pons' Objet: FW: Friday Session

Michel,

I am writing this email just to check whether you get a chance to test the issue 09-006 with the updated DLL?

Best Regards

Ahsan ======================================================== De: Ahsan Akhtar [ahsan.akhtar@…] Envoyé: Wednesday, May 13, 2009 12:47 PM À: 'Michel Pons' Objet: Friday Session

Pièces jointes: COLoggersST.dll

Michel,

May I request you to test the case 09-006 with the attached DLL on the VM on which I had a session on last Friday. If you remember, on Friday session, COLTT was causing a problem in displaying parameters and was showing blank window, due to which I stopped debugging the case.

Best Regards

Ahsan =========================================================== De: Michel Pons [technologyofficer@…] Envoyé: Thursday, May 07, 2009 7:48 PM À: 'sidathyder5@…' Objet: RE: PCNow Session

Ahsan,

Go to sleep. I will find the number. Have a good night.

Best regards

Michel ======================================================= De: Ahsan Akhtar [sidathyder5@…] Envoyé: Thursday, May 07, 2009 7:46 PM À: Michel Pons Objet: Re: PCNow Session I think u can copy the source code from VM prepared for issue of FTN mixer in aspen hysys

I don't remember the issue no. Actually I am going to sleep as it is late night here at Pakistan. But if u need no I can open my note book?

Best regards

Ahsan ========================================================= De: Michel Pons [technologyofficer@…] Envoyé: Thursday, May 07, 2009 6:48 PM À: 'Ahsan Akhtar' Cc: 'Michael Halloran' Objet: RE: PCNow Session

Ahsan,

OK for a session tomorrow morning, May 8.

I have a virtual machine set up for COLTT issue 09-006 with VS2008 Express edition installed as well as TOTAL HDPEReactor and Aspen Plus 7.1. However I don’t know where to get the proper version of the 1.08.02 COLTT source code. Could you send me a link from which I could download the source code before the session takes place?

I have also virtual machines for issues 09-008 and 09-009.

Best regards

Michel =========================================================== De: Ahsan Akhtar [ahsan.akhtar@…] Envoyé: Thursday, May 07, 2009 10:59 AM À: 'Michel Pons' Objet: PCNow Session

Michel,

Is it possible for you to have a session tomorrow morning at 8.00 am Paris time?

I want to debug and fix the issue 09-006. I will need TOTAL HDPEReactor and Aspen Plus v7.1 components installed on the machine. I will also need VS2008 and COLTT source code on that machine.

I will also run 09-008 and 09-009 issues to check if these are fixed.

Best Regards

Ahsan ========================================================= De: Michael Halloran [michaeljhalloran@…] Envoyé: Saturday, April 25, 2009 7:28 PM À: Michel Pons Cc: Ahsan Akhtar Objet: Re: TOTAL HDPEReactor in Aspen Plus v7.1 Michel, Ahsan,

I haven't seen a response to this yet so I took a look at this problem.

The final crash seems to occur in GetComponentConstant in MaterialObjectSt.cpp this code has not changed between 1.07 and 1.08.2 comparing my 1.08 copy and the 1.07 copy on sourceforge.net. As Michel notes what has changed is that the log in 1.08.2 is larger. That's because more information is being generated by the newer version, for example calls to unsupported interface are all now logged and there are a lot of them everytime a parameter is accessed.

Since the code hasn't changed I think this one has to be debugged on the Co-Lan laptop; it could be some sort of corruption caused by the code generating the new output.

Regards Michael ================================================================ De: Michel Pons [technologyofficer@…] Envoyé: Tuesday, April 14, 2009 4:56 PM À: 'Ahsan Akhtar' Cc: 'michaeljhalloran@…' Objet: TOTAL HDPEReactor in Aspen Plus v7.1

Pièces jointes: apmain_041409_162341v107.zip; Scenario.doc; apmain_041409_163539v10802.zip; Debugassertionfailed.doc

Ahsan,

I have encountered a major difference in using COLTT v1.07 and COLTT v1.08.02 when enabled on a Unit Operation provided by TOTAL. The Unit Operation is working with CAPE-OPEN Thermo 1.0 in Aspen Plus. Since the simulation ends up with errors, I have used COLTT in order to figure out what goes on. The scenario is quite simple (see Scenario.doc), starting from an Aspen Plus document that contains the necessary feed streams and setting up the connections. The unusual thing is that Edit does not work on such a UO and one needs to right-click on the block and select “input” to proceed with changing a parameter. Odd behaviour still not fixed in Aspen Plus v7.1.

Using COLTT v.1.07 seems fine, providing the same behaviour with and without COLTT enabled on the UO and providing a log file (see apmain_041409_162341v107 in the zipped file attached).

Now with COLTT v1.08.02, I get a much larger log (apmain_041409_163539v10802 in the zipped file attached) but a debug assertion error message when selecting the UO and subsequently Aspen Plus crashes (see Debugassertionfailed.doc attached).

So this shows some kind of regression from v1.07 to v1.08.02 that should be looked upon.

I am giving number 09-006 to this issue since there is also a non-numbered issue on the blank screen in COLTT controller that should be numbered 09-005 (sent on April 9). I am setting up a VM with the necessary software for further tests on 09-006.

Best regards

Michel