[TRNSYS-users] Using an Output of a Component at One Time Step asan Input to another Component at a Following Time Step

[danenberg]

----- Original Message Follows -----
From: "Jeff Thornton" <thornton at tess-inc.com>
To: "Aaron Danenberg"
<danenberg at hartford.edu>,trnsys-users at cae.wisc.edu
Subject: Re: [TRNSYS-users] Using an Output of a Component
at One Time Step  asan Input to another Component at a
Following Time Step
Date: Thu, 28 Apr 2011 17:09:39 -0500 (CDT)

> <I am trying to figure out how use an output from one
> component as an input to another component in a following
> time step.  I also need to provide the initial input to
> the component, after which it reads the output the other
> component from a previous time step. I have tried to use
> Type 93 with no success.>
> 
> That is EXACTLY why Type 93 was written.  What seems to be
> the issue?
> 
> <I have also tried to write to a data file using type 65
> and type 25 printers.  When I try this the simulation
> stops before the first value is written to the file.>
> 
> Did you check the list file for the source of the error?
> 
> Jeff
> 
> 
> 
> Jeff Thornton
> President - TESS, LLC
> 
> 22 North Carroll Street - Suite 370
> Madison WI 53703 USA
> 
> Phone: 608-274-2577
> Fax: 608-278-1475
> E-mail: thornton at tess-inc.com
> Web: www.tess-inc.com
> 
When I tried to use a data reader, the error I recieved said
the file was empty (which it was, the simulation aborted
before an output could be written to a during the first time
step).

I have one component with an output going into the type 93,
and the type 93 outputs this variable to the other
component.  There is no error when I use this type.  However
it looks like TRNSYS is trying to solve for a value of this
variable that would satisfy equations in both components,
when it just needs to pass an output to a following time
step.  For the simulation I am running, it would not make
any sense to find a value that would satisfy both equations,
there would be an infinite number of possibilities.  I also
noticed I need to lower the tolerance in the control cards
in order to simulate a full year.

In the type 93, all I see is an option to define the number
inputs and time steps to store.   I am confused on how to
properly use type 93 just to pass outputs to inputs in
proceeding time steps, and how to set an initial value for
the first time step.  Thanks!

-Aaron