Re: convergence problem

Some solutions (perhaps you have already try some of them!):

1- Add a trace statement to the unit which don't converge. You will see if there is no convergence, or a too slowly convergence, or if the input take alternatively two values (for example 1.3, then 1.4, then 1.3, then 1.4, ....)

2- you can try to increase the convergence tolerance (control cards)

3- you can try to increase the number of iterations before having an error message (control cards, "before 'WARNINGS' " and/or "before 'ERRORS', see more))

4- you can try to introduce an under-relaxation formulation in your equation card

I hope it can help you (and you understand my poor english!)

Best regards

Laurent

----- Original Message -----

From: Marco Sangiorgi

To: Trnsys

Sent: Thursday, September 25, 2003 1:07 PM

Subject: convergence problem

This process is for cooling a multi-zone building, and it is referred to one zone (repeated once for each zone with different parameters).

1.      I have a dry bulb temperature T and a Humidity Ratio HR for each zone (type 56)

2.      I mix a certain amount of air of the building with outside air (Type 9)

3.      I cool the air mix with an Heat Exchanger (Type 55)

4.      I blow the cooled air mix to the building

q       The psychrometrics and the Equations are just for certain calculations of condensations.

q       I set the FLOW of cool water that flows througth the Heat Exchanger (Type 55) to keep the temperature T of the building within a certain range

This seems easy and sometimes it works fine, but most of times I get errors (I’d reather say “horrors”) like THE INPUTS TO UNITS xx xx xx etc WERE NOT CONVERGED WITHIN THE SPECIFIED TOLERANCE AT TIME =   xx

I don’t know how to deal with it. I tried with both solvers, with the NOCHECK statement, with the backsolving technique, but as I’m not an expert I can’t achieve the convergence.

Referred to the Heat exchanger (Type 55), If I try to control the outlet temperature of the hot side (the air I need to cool) changing the flow of the cold water (cold side inlet flow), I never get a convergence. See picture.

Could somebody please give an advice? Now that the simulation (a hudge simulation) works, would be a pity to leave it because of this error.

Thank you