Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
ntfluxInt - kinks in plot
Posted 2020年4月22日 GMT-4 19:57 Heat Transfer & Phase Change, Results & Visualization Version 5.3a 2 Replies
Please login with a confirmed email address before reporting spam
I want to simulate a block of aluminum that has a certain temperature and radiates to an empty space. So I set up a COMSOL simulation:
Physics: heat transfer in solids, time dependent
I created a simple block
added a material (Aluminum 6063-T83)
set the temperatures (delta is 10K)
turned on surface-to-surface radiation
all surfaces of the block are diffuse surface with a given epsilon of 0.5
mesh is physics controlled (coarser)
When I do so I get reasonable results (I verified them by hand), but I noticed some kinks in the temperature plot. I investigated that they result from even bigger kinks in the ntfluxInt global value (see graph). I already varied the mesh sizes, time steps, material, and several other parameters, but none of that removes the kinks.
So my question is: where do these kinks come from? Are they simulation errors? Is there any way to get rid of them?
Thanks for your help
Attachments:
Please login with a confirmed email address before reporting spam
Dear Laura,
such phenomena can occur if the time stepping is not appropriate set. Please control the time stepping in the time dependent solver as described here: https://www.comsol.com/support/knowledgebase/1254
Kind regards
Friedrich
Please login with a confirmed email address before reporting spam
Setting the relative tolerance to a smaller value worked perfectly. Thanks for your help!