Simulation CFD 2014 HELP - analysis stopped due to divergence

[Emma1706]

Hi

i am new to CFD and i really need some help i have made some models of homes in inventor fusion and i am trying to run them through simulation CFD so i can analysis the dynamic thermal properties of the buildings... But when i hit solve (im attempting to solve this over a 24hr day simulation) i get a error message saying "Analysis Stopped due to divergence" and i dont know how to fix it

this is for my major thesis at uni and i really need these models, any help or suggestions would be greatly appreciated.

Thank you

Emma

[OmkarJ]

Hi,

Are you able to share cfz file? 

There are number of factors that can lead to divergence. 

[Emma1706]

Hi!


Thanks for replying so quickly!! I am also having two other issues with the modelling other than the divergence.


1. Divergence which stops the analysis.


2. I am trying to simulate a fan/air-conditioner/no cooling environment inside a 6m x 6m wide unit for part of a university thesis throughout 24 different models (4 different construction types). I need the simulation to record results every hour for 24 hours, however with the intelligent solution control turned on the recorded time steps are completely incorrect and unless I run it hour by hour manually without the ICS switched on I can't get the results I need.


3. Applying a mass volume flow rate to the internal lower surface of the 'fan' cylinder does not produce any air movement. I have tried several different methods and followed the seemingly simple online tutorials on your website however I just cannot get it to work correctly.


I am using the solar heating function to simulate a single day, I have made sure to set all the details correctly - Hopefully these issues can be easily sorted as the thesis is due in two weeks and if I have to run each model individually hour by hour it may take some time! It may also be important to note that the purpose of these 24 models is to develop a trend between models rather than completely accurate results.



Thanks!


Emma?

[srhusain · Accepted solution]

As OmkarJ mentioned previously it would be very helpful if you can post the support file to the forum. This will provide useful information on your problem setup.

Without the support file (<desigh_study_name>.cfz), I can offer the following comments:

1. When modeling an internal fan, no boundary conditions of volume or mass flow rate should be applied to any of the surfaces- what should be applied is a flow rate versus pressure drop in the fan properties editor of the UI.
2. If you are simulating over a 24-hour duration, you can instruct the solver to output the results on an hourly basis, but numerical stability will still dictate the maximum time step size you can use, even if accuracy is not the main concern. So, as a rough guide, you might try the first 100 time steps with a step size of 60 seconds and then continue with another 100 steps with size of 120 seconds and finally run out the remainder of the 24-hours with a step size of 360 seconds. Also, set the number of inner iterations to 1.
3. If not already done so, switch to Adv5 for the advection scheme for better stability.

[Emma1706 · Accepted solution]

Thank you so much the suggestion with the time step and the iterations this compleatly stabilised my results!!

I have also done what you have suggested with the fan and i have been getting great temperature simulations, you really did resolve all of my issues!

Thank you

Emma