[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TRNSYS-users] calling a private type from another private type




Dear Antoine,

The version 16 source coding standard as you provided does indeed work with version 17. You will notice that most of the TESS Component Libraries are still coded that way.

My adivce with the component that uses PSYCH is that you need to sometimes TRACE the component to determine what is going into and coming out of the component. This usually determines that something in the simulation that is going wrong.

Best regards,

Matt Duffy

----- Original Message -----
From: "Antoine Gautier" <antoine.gautier@solamen.fr>
Sent: Sat, July 16, 2011 1:24
Subject: Re: calling a private type from another private type

!
Dear Matt

 
Thank you very much for your answer.
I am afraid my problems were not of the kind of "Release VS Debug" building mode.

 
First problem = address error 
I found out that it was related to the subroutine first line statement which cannot accept the standard (TIME,XIN,OUT,T,DTDT,PAR,INFO,ICNTRL,*) in Trnsys version 17 despite what is stated in §7.3.2 of the user manual (quoting: Every Type begins with the same line: SUBROUTINE TYPEn(TIME,XIN,OUT,T,DTDT,PAR,INFO,ICNTRL,*).This is obviously aimed at version 16 and not compatible with version 17 from my experience (at least with the version 17 I was using: see below).

 
Second problem = storage error
I called a subroutine that had its own storage array and this was touchy: I have not found the correct strategy here so I resorted to including the whole storage management code in the main subroutine. This works fine but if you have any better advice on that point I will be grateful.

 
Third problem = invalid floating operation
This mistake comes from the patch version which has been provided to me to overcome the problem with PSYCHROMETRICS routine cf. https://mailman.cae.wisc.edu/pipermail/trnsys-users/2011/008138.html. No error occurs with the standard version. But the standard version has the pro! blem with PSYCHROMETRICS routine so I am kind of stuck here. What I do not know though is if the corrected version is more strict on certain rules that my component does not comply with or if it just erroneous. 

 
Best regards
--
Antoine Gautier
Solam
en
+33 6 98 78 56 46