Menu

Discussion Forum

New Discussion
Filtrer

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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 Machine , ACDC, rotating doesn't work the right way

Posted 20 juin 2011, 09:27 UTC−4 Low-Frequency Electromagnetics, Geometry Version 4.1 4 Replies

Norbert Bartscher
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Good day,

I'm having some issues modelling electrical machines in 4.1. The attached file is a synchronous generator with 4 poles. Remanent flux density should be radial and the prescribed rotation velocity is 3000/min.

Somehow the rotor rotates (spatial), but the remanent flux density doesn't rotate the right way. It just stays in the same place and gets more than just a radial component, instead of just rotating radial with the rotor.
I simulated 20ms to check 1 full turn starting at 10s.

For t=10,000s, remanent flux density looks great and right.
For t=10,005s, remanent flux density looks totally wrong. Should be radial an rotated for a 4th period.
For t=10,010s, remanent flux density looks great and right, but it should be multiplied with -1.
For t=10,015s, remanent flux density looks totally wrong again. Similar to t=10,005s
For t=10,020s, remanent flux density looks great and right. Just like t=10,000s

The magnetic vector potential is also not rotating. But I think that's a result of the not-rotating-remanent flux density.

I think the mistake might be something in my model.

Is anybody out there who can help me? : )

Thanks in advance
4 Replies Last Post 29 juin 2011, 10:57 UTC−4
Akmal
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21 juin 2011, 04:53 UTC−4
Hi Bartscher Norbert,

i would suggest you to try this:

Finalize the geometry as an assembly, because rotor should rotate separately from stator. Make all of non-rotate-parts as a union (stator, coper, air...) and all of rotate-parts also as union.

Define identity pair betwen boundary of non-rotate and rotate-parts. and Pair-Continuity-Node.

If you finish the step and you still have same problem, then we have same problem.

I have donwload your model but not simulate yet.

Try mesh instead of spatial, are the mesh rotate or not. On my model mesh arent rotate. Maybe thats why the result (spatial) wont rotate. I do also PM-Generator, the induced voltage are correct.

regards
akmal

Hi Bartscher Norbert, i would suggest you to try this: Finalize the geometry as an assembly, because rotor should rotate separately from stator. Make all of non-rotate-parts as a union (stator, coper, air...) and all of rotate-parts also as union. Define identity pair betwen boundary of non-rotate and rotate-parts. and Pair-Continuity-Node. If you finish the step and you still have same problem, then we have same problem. I have donwload your model but not simulate yet. Try mesh instead of spatial, are the mesh rotate or not. On my model mesh arent rotate. Maybe thats why the result (spatial) wont rotate. I do also PM-Generator, the induced voltage are correct. regards akmal
Norbert Bartscher
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21 juin 2011, 09:52 UTC−4
Hey,

thanks for the tip. I tried it (maybe wrong) but it didn't help.
What makes me wonder is that the vectors of the remanent flux density stay at the same place, but change direction with time, while the rotor rotates in spatial frame.

I defined it with:

\vec{Br} :

0.683[T]*cos(atan2(y,x)) = Bx
0.683[T]*sin(atan2(y,x)) = By
0 = Bz

So the components of Br depend on y and x and Br is radial. That's ok.

x and y change with time what should be the result of the time depending rotor position, due to the velocity.

But what about the position vectors? The position vectors of Br should rotate with the same speed as the rotor does.
Seems to me that they don't.
How can you change the position vectors?

Thanks in advance







Hey, thanks for the tip. I tried it (maybe wrong) but it didn't help. What makes me wonder is that the vectors of the remanent flux density stay at the same place, but change direction with time, while the rotor rotates in spatial frame. I defined it with: \vec{Br} : 0.683[T]*cos(atan2(y,x)) = Bx 0.683[T]*sin(atan2(y,x)) = By 0 = Bz So the components of Br depend on y and x and Br is radial. That's ok. x and y change with time what should be the result of the time depending rotor position, due to the velocity. But what about the position vectors? The position vectors of Br should rotate with the same speed as the rotor does. Seems to me that they don't. How can you change the position vectors? Thanks in advance
Norbert Bartscher
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 juin 2011, 12:42 UTC−4
Hello,

I decided to choose a way more simple geometry and follow advices.

The Geometry was finalized as assembly and is made out of 2 composed parts. I chose the identy pair between them and added it to the physics.

Sadly the results are still the same. In spatial frame, I see the rotor domains rotating without the domain characteristics. For example neither the relative permeability nor the remanent flux density rotate.
BUT something changes, because the magnetic vector potential changes, although it doesn't rotate...

I could really need some help here... Is anybody out there?

Thanks in advance!
Hello, I decided to choose a way more simple geometry and follow advices. The Geometry was finalized as assembly and is made out of 2 composed parts. I chose the identy pair between them and added it to the physics. Sadly the results are still the same. In spatial frame, I see the rotor domains rotating without the domain characteristics. For example neither the relative permeability nor the remanent flux density rotate. BUT something changes, because the magnetic vector potential changes, although it doesn't rotate... I could really need some help here... Is anybody out there? Thanks in advance!
Onur Ilkorur
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29 juin 2011, 10:57 UTC−4
Hi,

I have contacted to Comsol support with a similar question. They have forwarded me a 3.5a file called 2Dsolenoid.mph, saying that it should help. Maybe you can request a copy from them (22MB).

I hope it can help you solve your problem.
Onur
Hi, I have contacted to Comsol support with a similar question. They have forwarded me a 3.5a file called 2Dsolenoid.mph, saying that it should help. Maybe you can request a copy from them (22MB). I hope it can help you solve your problem. Onur

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