Menu

Discussion Forum

New Discussion
フィルター

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.

Rotating generator with bad inducted voltage

Posted 2012/07/02 19:14 GMT-4 Low-Frequency Electromagnetics, Modeling Tools & Definitions, Parameters, Variables, & Functions, Results & Visualization, Studies & Solvers 7 Replies

José Cardoso
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi everyone.

I am using an old version of Comsol, the 3.5 !

I am trying to simulate a generator with 32 poles and with permanent magnets of NdFeB. Because of this and because i want to simulate it with a specific stator sheet, i have already added this two materials to my Material Library.

I want to test if the voltage inducted at no load and at 127rpm is like the results of an article i have but as you can see my simulations are not giving a sinusoidal voltage (as it should). I tried to make almost every example of the AC/DC module but i am not getting any good feedback where my error could be. Anyone have an idea? I have attached the files.

Could the error be in the type of the system solver? I am using the default "Direct(UMFPACK). Is there other more apropriate for this "moving generator"? For the time stepping i am using "linspace(0,0.03,30)".

Thank you.

Best regards,
José Cardoso

Attachments:

7 Replies Last Post 2012/08/22 17:52 GMT-4
Edgar Kaiser
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/07/03 4:08 GMT-4
Hi
at 127 rpm and 32 poles it takes only 0.015s to rotate from 1 pole to the next. Is your time stepping too coarse and you are observing just an aliasing result?

Cheers
Edgar
Hi at 127 rpm and 32 poles it takes only 0.015s to rotate from 1 pole to the next. Is your time stepping too coarse and you are observing just an aliasing result? Cheers Edgar
José Cardoso
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/07/03 8:16 GMT-4
Hi,

As i said i am using for time stepping "linspace(0,0.03,30)". I have already increased it for linspace(0,0.03,100) but the bad induction presists. I thought that with more steps the problem should be gone but it didn't. Is this what you are suggesting?

Do you think that i should increase it even more?

Thank you for you reply.

Best regards,
José Cardoso
Hi, As i said i am using for time stepping "linspace(0,0.03,30)". I have already increased it for linspace(0,0.03,100) but the bad induction presists. I thought that with more steps the problem should be gone but it didn't. Is this what you are suggesting? Do you think that i should increase it even more? Thank you for you reply. Best regards, José Cardoso
Edgar Kaiser
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/07/03 8:26 GMT-4
Sorry, I got the syntax of linspace confused. The time stepping is probably ok. You might still check if you chose explicit timing in the solver.
Another issue might of course be the meshing.

Cheers
Edgar
Sorry, I got the syntax of linspace confused. The time stepping is probably ok. You might still check if you chose explicit timing in the solver. Another issue might of course be the meshing. Cheers Edgar
José Cardoso
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/07/03 10:04 GMT-4
Sorry to ask but what is the "explicit timing"?

I tried to attach the file of the comsol but it is too big. I will try to attach 1/8 of my generator so you could see what is wrong if you don't mind.

For the mesh i set it to "Finer" and then make for 2 in "Resolution of narrow regions" like it was in one of the models.

Thank you again for your support sir.

Sorry to ask but what is the "explicit timing"? I tried to attach the file of the comsol but it is too big. I will try to attach 1/8 of my generator so you could see what is wrong if you don't mind. For the mesh i set it to "Finer" and then make for 2 in "Resolution of narrow regions" like it was in one of the models. Thank you again for your support sir.
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/07/04 1:40 GMT-4
Hi

before upéloading, Edit clear the solutions and the mesh, and do a file reste model then save with a new name and upload the file becomes "small" again ;)
Sorry in 3.5 its somewhere else, AND you loose the mesh strategy, whxih can be a serious issue.
That is a good reason why so many are happy to switch to V4, when they can

--
Good luck
Ivar
Hi before upéloading, Edit clear the solutions and the mesh, and do a file reste model then save with a new name and upload the file becomes "small" again ;) Sorry in 3.5 its somewhere else, AND you loose the mesh strategy, whxih can be a serious issue. That is a good reason why so many are happy to switch to V4, when they can -- Good luck Ivar
José Cardoso
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/07/25 10:01 GMT-4
Hi,

I think what i had was indeed a aliasing result. I can get now a sin wave for my induction voltage.

Now i am trying to see what my losses are. I read that hysteresis losses are difficult to compute so i am focusing in eddy current losses. I have already read the exemples presented in the model library but i don't know if i am doing it right.

After i solve my generator for trasient mode, i am making a subdomain integration from the prostprocessing menu and selecting the "resistive heating" (Q_emqa) from predefined quantities.

I can get the results directly for each time step but the unit of my result comes in W/m when i was expecting in W/m^3.

I can't obtain the average resistive heating because i am not using the heat module.

What am i doing wrong? I don't know if i am making a confusion out of it.

Best regards
Hi, I think what i had was indeed a aliasing result. I can get now a sin wave for my induction voltage. Now i am trying to see what my losses are. I read that hysteresis losses are difficult to compute so i am focusing in eddy current losses. I have already read the exemples presented in the model library but i don't know if i am doing it right. After i solve my generator for trasient mode, i am making a subdomain integration from the prostprocessing menu and selecting the "resistive heating" (Q_emqa) from predefined quantities. I can get the results directly for each time step but the unit of my result comes in W/m when i was expecting in W/m^3. I can't obtain the average resistive heating because i am not using the heat module. What am i doing wrong? I don't know if i am making a confusion out of it. Best regards
Aruna Shah
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012/08/22 17:52 GMT-4
Hi ,

I am also getting the bad induced voltage in my generator.

Iam using a line space of (0,0.01,0.25)

Will you please let me know how you get the sine wave for the induced voltage?

Thank you.


Hi , I am also getting the bad induced voltage in my generator. Iam using a line space of (0,0.01,0.25) Will you please let me know how you get the sine wave for the induced voltage? Thank you.

Attachments:

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