Issue with EMS Construction State Actuator and Finite Difference Nodes
Hello, I am trying to change the wall construction dynamically using EMS to simulate the shifting position of a PCM layer within the wall based on a condition statement. However, I am encountering an issue related to the number of finite difference nodes. The following error message appears, even though both constructions have the same thickness and number of layers:
** Severe ** InitEMSControlledConstructions: EMS Construction State Actuator not valid.
** ~~~ ** Construction named = WALLPCMEXTERNAL has number of finite difference nodes =10
** ~~~ ** While construction named = SISOPEN has number of finite difference nodes =0
** ~~~ ** This actuator is not allowed for surface name = SIS, and the simulation continues without the override
I have tried using HeatBalanceAlgorithm: ConductionFiniteDifference and adjusting the timestep settings to control the number of nodes, but the issue persists.
What could be causing this , and how can I resolve it?
Thanks in advance for any guidance!




@Alaa Salah the error message mentions three things:
Are you changing the construction state of the SIS surface between SISOPEN and WALLPCMEXTERNAL? If yes, then SISOPEN also requires finite different nodes (error message says it has 0 nodes while WALLPCMEXTERNAL has 10 nodes).
Can you upload your model to Google Drive, Dropbox, etc. and then share a URL for others to download? That would help others provide better solutions.
@Aaron Boranian you have described what I am trying to model. https://drive.google.com/drive/folder... This link contains 2 idf files of 2 trials to implement the concept in the sketch
@Alaa Salah I am getting something very similar with one of my models. Did you find a solution to this yet?
Interestingly, I have made the two construction states identical and it gives me the same error! This means it is throwing the "0" node error when there are clearly identical number of nodes between the two construction states.
Which leads me to believe this is pointing to some broader error, either in CFD algorithm or just a bug of some sort....