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.
time step
Posted 2023年3月18日 GMT-4 10:07 Studies & Solvers 8 Replies
Please login with a confirmed email address before reporting spam
What is the reason why the convergence curve is peak shaped (up and down), and after a few hours, the calculation progress is only 4%? The time step is like this:range(0,0.001,0.01),range(0.01,0.03,0.1),range(0.1,0.3,1),range(1,1,20)s.The transient process I studied is approximately 20 seconds.Thank you.
Please login with a confirmed email address before reporting spam
Based on the information you are providing I can assume that there must be a problem.
-------------------Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Please login with a confirmed email address before reporting spam
In case it isn't obvious, Edgar is suggesting that without looking at your model only the wildest guess is possible.
If your model contains proprietary elements, you may want to consider making a simpler model that exhibits the same sort of behavior. Indeed, putting together a model with simpler geometry and the same physics is often very helpful.
We need a "sticky" addressing "how do I ask a question that is likely to result in a helpful response?"
Please login with a confirmed email address before reporting spam
We need a "sticky" addressing "how do I ask a question that is likely to result in a helpful response?"
Shouldn't that be something one learns at school and university in the first place? Not sure if any sticky would help. Well, I am really surprised how often complex questions are based on a null information, not only in this forum.
-------------------Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Please login with a confirmed email address before reporting spam
Hello Dave and Edgar,
As it were, for a while I have thought of writing a document that would suggest to users of this Discussion Forum ways of improving their chance of getting useful responses from other users.
One of the reasons I have not done so already is that I don't think I can do much better than this fantastic page which I remember reading back when I was a wee grad student. While it's not written specifically with COMSOL in mind, much of it is applicable to this forum anyways, and it's a fun read. I'll just point out that its author uses the term "hacker" to mean something other than what most people think of today when they hear that word.
Best,
Jeff
-------------------Jeff Hiller
Please login with a confirmed email address before reporting spam
Jeff
Well if you DID write something include the following
- No acronyms
- clearly explain the problem- well enough for those not in the immediate area
- Post your model if at all possible !!!
- Ask an interesting question
- Be sure to browse the application gallery if trying to get started
Interesting link, I will have a look at it.
I remember many occasions when grad students would ask for advice and want a short answer...and I would spend time working back to find out what the real problem was.
Please login with a confirmed email address before reporting spam
The link that Jeff posted does contain very useful advice. However rather wordy and some parts not really applicable to Comsol questions.
Apparently the people who respond here are not as testy as those who haunt Linix forums.
Please login with a confirmed email address before reporting spam
Dave and Edgar,
It's a small step, but on the form that is used to submit posts to the Discussion Forum, we've now added language encouraging users to attach their mph files in addition to simply screenshots. We'll see if that helps.
Best,
Jeff
-------------------Jeff Hiller
Please login with a confirmed email address before reporting spam
Even better- have a checkbox that needs to be checked before submitting that says
"If I have not attached an .mph file I understand that the chances of a useful response are greatly reduced"
Anyway- since lots of posters don't come back...maybe they are just venting?
Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.
Suggested Content
- KNOWLEDGE BASE Solving Wave-Type Problems with Step Changes in the Loads
- KNOWLEDGE BASE Resolving Time-Dependent Waves
- BLOG How to Simulate the Carrier Dynamics in Semiconductor Devices
- BLOG Automatic Time Step and Order Selection in Time-Dependent Problems
- KNOWLEDGE BASE Solving time dependent models with inconsistent initial values