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.

halbach array 8 magnet

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Hej!
I made 3D model of Halbach array 8 magnet circular used NdFeB magnetic with 0.5 tesla magnetic strength.
Moreover, choose AC/DC module static magnetic field with no current. I am getting error when ever i simulate it. Moreover, i do not know how to define north and south pole of the magnet. I have attached pdf in which i explain almost everthing and also attached comsol file of halbach array of 8 magnetic. If some one can help me to sort this problem as soon as possible.

I want to use static magnetic field anaylsis to know magnetic field in the middle of the magnets

I will be thankful

I am looking forward to your mail.


Regards,

Mehdi Khan


13 Replies Last Post Jun 21, 2017, 8:25 p.m. EDT
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 30, 2011, 9:03 a.m. EDT
Hi

there are different ways to simuate a permananet magnet, for magnetisation along the axis, you can use surface current (coil equivalence) or remanent or magnetisation, if you choose mnfc the first is not possible, so probably the magnetisation M is the easiest. You need to define a Magnetic Flux Conservation per magnet in different orientations

Nice project yo have there, could be nice to hear how the simlations and measurements coincide ;)

--
Good luck
Ivar
Hi there are different ways to simuate a permananet magnet, for magnetisation along the axis, you can use surface current (coil equivalence) or remanent or magnetisation, if you choose mnfc the first is not possible, so probably the magnetisation M is the easiest. You need to define a Magnetic Flux Conservation per magnet in different orientations Nice project yo have there, could be nice to hear how the simlations and measurements coincide ;) -- Good luck Ivar

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 31, 2011, 11:42 a.m. EDT
Hello Ivar!!
Really thanx for helping me and liked my project :)... Well , we are designing NMR spectroscopy and i told my professor about Comsol that is best software ...

I use surface current and give value 5000 (Magnetic field strength) alone z axis for North and -5000 alone Z-axis for South ... As i told you half of the magnet is North pole and other half is south pole ... Also define multi surface current. am i right

I want to calcuate Magnetic field density .... I simulate it on comsol no error but do not know how to see magnetic field density alone x-axis and y axis

Moreover, how will i give orientataion ? ... by graphically,

i use Js=5000 or North and Js=-5000 for south pole am i right ? (Surface current) ...

I could not able to divide front face of both circular magnet half Norht and Half south Up and down side of magnet how can i do that. The magnetic field strength of each magnet is 0.5 Tesla

I have attached the file one with surface current. can you please look around my .mph file and solve my problem ... There is no error while simulation

Moreover, how can i define What is "swept" and inside "SWEPT" "Distribution" .. How can define them ?.... Also how can i define magnetic Insulation ?... I do not have any idea


I am looking forward ot your mail
Hello Ivar!! Really thanx for helping me and liked my project :)... Well , we are designing NMR spectroscopy and i told my professor about Comsol that is best software ... I use surface current and give value 5000 (Magnetic field strength) alone z axis for North and -5000 alone Z-axis for South ... As i told you half of the magnet is North pole and other half is south pole ... Also define multi surface current. am i right I want to calcuate Magnetic field density .... I simulate it on comsol no error but do not know how to see magnetic field density alone x-axis and y axis Moreover, how will i give orientataion ? ... by graphically, i use Js=5000 or North and Js=-5000 for south pole am i right ? (Surface current) ... I could not able to divide front face of both circular magnet half Norht and Half south Up and down side of magnet how can i do that. The magnetic field strength of each magnet is 0.5 Tesla I have attached the file one with surface current. can you please look around my .mph file and solve my problem ... There is no error while simulation Moreover, how can i define What is "swept" and inside "SWEPT" "Distribution" .. How can define them ?.... Also how can i define magnetic Insulation ?... I do not have any idea I am looking forward ot your mail


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 1, 2011, 3:56 a.m. EDT
Hi

first thing, if you have magnets, the field lines have to loop around. So if you do not define a "vacuum or air media" the fields lines cannot loop and the physics is not correctly defined. So you need to add a cube or a sphere around our magnets with air/vacuum, then you can solve correctly your case. perhaps about 2x the diameter of your magnet ring

Second with your magnets that are NOT axial magnets (N-S along the cylinder axis. You cannot use surface currents to model them, the right hand rule for current circulation and magnetic field vector cannot be defined lie that. Use rather the magnetisation M definition. But then you need to define the direction correctly, either via corrdinate frames, or via the vector development (rather easy as you hve 8 magnet, in pairs along X, Y, +/-45°XY)

to avoid making a hughe model (RAM gourmand) you could consider symmetry, and or the use of infinite elements, take a look at the doc

--
Good luck
Ivar
Hi first thing, if you have magnets, the field lines have to loop around. So if you do not define a "vacuum or air media" the fields lines cannot loop and the physics is not correctly defined. So you need to add a cube or a sphere around our magnets with air/vacuum, then you can solve correctly your case. perhaps about 2x the diameter of your magnet ring Second with your magnets that are NOT axial magnets (N-S along the cylinder axis. You cannot use surface currents to model them, the right hand rule for current circulation and magnetic field vector cannot be defined lie that. Use rather the magnetisation M definition. But then you need to define the direction correctly, either via corrdinate frames, or via the vector development (rather easy as you hve 8 magnet, in pairs along X, Y, +/-45°XY) to avoid making a hughe model (RAM gourmand) you could consider symmetry, and or the use of infinite elements, take a look at the doc -- Good luck Ivar

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 1, 2011, 1:25 p.m. EDT
Hej Ivar,
Thankx Ivar helping ... I made sphere almost 2 times the circular ring of magnet and made beautiful eye so that i can see the simulated result inside it.... When i simulate my result i could not see magnetism inside .. I do not know why ?

This time i used "Magnetic Field" instead of "Source current" last time...

And define Magnetic field Ho=5000 A/m for north and Ho= -5000 or South alone z -axis .... Did i did correct?

What do you mean by "magnetisation M". You mean Define in "Ampera Law" Magnestisation M? Well i use Ampera Law Magnestisation M= 5000 , Did i did correct ?

I am new in comsol and this is my second diagram simulating in comsol .... How can i define Coodinate frame for my magnets North and south ... Where i can find this option...

Moreover, how can i define symmetry ... I dont know this option.

I have attached my file ... If you could please look around and solve my problem ..

I will be thankful


Best Regards,
Mehdi
Hej Ivar, Thankx Ivar helping ... I made sphere almost 2 times the circular ring of magnet and made beautiful eye so that i can see the simulated result inside it.... When i simulate my result i could not see magnetism inside .. I do not know why ? This time i used "Magnetic Field" instead of "Source current" last time... And define Magnetic field Ho=5000 A/m for north and Ho= -5000 or South alone z -axis .... Did i did correct? What do you mean by "magnetisation M". You mean Define in "Ampera Law" Magnestisation M? Well i use Ampera Law Magnestisation M= 5000 , Did i did correct ? I am new in comsol and this is my second diagram simulating in comsol .... How can i define Coodinate frame for my magnets North and south ... Where i can find this option... Moreover, how can i define symmetry ... I dont know this option. I have attached my file ... If you could please look around and solve my problem .. I will be thankful Best Regards, Mehdi


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 1, 2011, 2:58 p.m. EDT
Hi

a few things:
- The material "air" should be defined on the domain "1" and not the boundary !!
- Your magnets are not really "iron", there are better choices in the mat library, bt you could also use "user defined" then you kearn wha you need and what you really are doing ;)
- your magnetif field physics MF should also inlcude the "air" = domain "1"
- If you define a magnetisation for the air of Mz=5000 you are also making the air an magnet, not realistic, use "relative permeability"
- Your magnetic field N, S etc are boundary conditions, they should be domain defined,
- You need to define an Ampere law per magnet, or magnet pair, and then define the correct Mz, that I though should be along the X-Y plane and NOT the Z direction ?

Try out a few of the model library magnetic field examples and study them in detail, there are many things to set correct, and it's easy to get wrong and your model is already medium complex

--
Good luck
Ivar
Hi a few things: - The material "air" should be defined on the domain "1" and not the boundary !! - Your magnets are not really "iron", there are better choices in the mat library, bt you could also use "user defined" then you kearn wha you need and what you really are doing ;) - your magnetif field physics MF should also inlcude the "air" = domain "1" - If you define a magnetisation for the air of Mz=5000 you are also making the air an magnet, not realistic, use "relative permeability" - Your magnetic field N, S etc are boundary conditions, they should be domain defined, - You need to define an Ampere law per magnet, or magnet pair, and then define the correct Mz, that I though should be along the X-Y plane and NOT the Z direction ? Try out a few of the model library magnetic field examples and study them in detail, there are many things to set correct, and it's easy to get wrong and your model is already medium complex -- Good luck Ivar

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 21, 2011, 9:57 a.m. EST
Hello,

I could not fix the problem halbach array of 8 magnet.. i change the domain air= 1 .... did some changes but cannot see mangetic field around the magnet and in the center.

Please help me.. I have attached the comsol file and also .pdf to give u idea what i want to do

I am looking forward to your mail




I will be thankful


Best Regards,
Mehdi Khan

Hello, I could not fix the problem halbach array of 8 magnet.. i change the domain air= 1 .... did some changes but cannot see mangetic field around the magnet and in the center. Please help me.. I have attached the comsol file and also .pdf to give u idea what i want to do I am looking forward to your mail I will be thankful Best Regards, Mehdi Khan


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 22, 2011, 3:59 a.m. EST
Hi

You should define one "Ampere Laws" for the air, and one specifically for all the magnets that are oriented in the same direction (two in your case ?). Then define their field by the Magnetisation (possibly with different coordinate frames if you have complex geometries).

And that should do, you do not need the "Magnetic surface fields" you have selected the continuity should take care of that

--
Good luck
Ivar
Hi You should define one "Ampere Laws" for the air, and one specifically for all the magnets that are oriented in the same direction (two in your case ?). Then define their field by the Magnetisation (possibly with different coordinate frames if you have complex geometries). And that should do, you do not need the "Magnetic surface fields" you have selected the continuity should take care of that -- Good luck Ivar

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 23, 2011, 4:23 a.m. EST
Hej Ivar,
Now i feel i am doing correct ... I define ampera law alone x -y axis magnaization +35000 for north and -35000 for south...

My simulation are working but i cannot see magnetic field in the middle I want to se B=magnetic flux density in the middle.. like in arrow 3-d . I want to see field arround magnetc

I feel i did almost correct. Maybe i did minor mistakes

Can you please check and remove the error ... Or tell me how to remove it

Best Regards,
Mehdi Khan
Hej Ivar, Now i feel i am doing correct ... I define ampera law alone x -y axis magnaization +35000 for north and -35000 for south... My simulation are working but i cannot see magnetic field in the middle I want to se B=magnetic flux density in the middle.. like in arrow 3-d . I want to see field arround magnetc I feel i did almost correct. Maybe i did minor mistakes Can you please check and remove the error ... Or tell me how to remove it Best Regards, Mehdi Khan


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 23, 2011, 5:59 a.m. EST
Hi

I believe COMSOL has a issue with their site (disk full ?) when I try to download your file I get a 0 byte copy ;)
Try to upload it again later

--
Good luck
Ivar
Hi I believe COMSOL has a issue with their site (disk full ?) when I try to download your file I get a 0 byte copy ;) Try to upload it again later -- Good luck Ivar

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 23, 2011, 6:06 a.m. EST
Hello Ivar

There was some problem uploading file .. well i fix the problem ... you can download on last forum i have writen... Marry chrismas
Hello Ivar There was some problem uploading file .. well i fix the problem ... you can download on last forum i have writen... Marry chrismas

Dr. Mehdi Khan Total Artificial heart

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 23, 2011, 11:44 p.m. EST
Hej Ivar,
I check the attachment ... Its working ...
Well,

Now i feel i am doing correct ... I define ampera law alone x -y axis magnaization +35000 for north and -35000 for south... I want to know each magnet of mine is 0.5 T magnetic field strength (What value should i put in magnaization alone x-y direction)

Moreover, my simulation are working but i cannot see magnetic field in the middle of each 8 magnet I want to see B=magnetic flux density in the middle.. like in arrow 3-d . I want to see field arround magnetc


Also want to plot graph alone x y axis of magnetic flux density in the middle

I feel i did almost correct. Maybe i did minor mistakes

Can you please check and remove the error ... Or tell me how to remove it

Best Regards,
Mehdi Khan
Hej Ivar, I check the attachment ... Its working ... Well, Now i feel i am doing correct ... I define ampera law alone x -y axis magnaization +35000 for north and -35000 for south... I want to know each magnet of mine is 0.5 T magnetic field strength (What value should i put in magnaization alone x-y direction) Moreover, my simulation are working but i cannot see magnetic field in the middle of each 8 magnet I want to see B=magnetic flux density in the middle.. like in arrow 3-d . I want to see field arround magnetc Also want to plot graph alone x y axis of magnetic flux density in the middle I feel i did almost correct. Maybe i did minor mistakes Can you please check and remove the error ... Or tell me how to remove it Best Regards, Mehdi Khan


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 24, 2011, 3:26 a.m. EST
Hi

A few comments
1) you should use a manual mesh with a finer settings for the magnets and use the coarser for the sphere
2) I can advise you to centre the sphere in the middle of your magnet array (now its sidewise) minor effect, but essential if you use infinite elements later on
3) another trick is to have two spheres, one about 2x the magnet size to allow for better finer/intermediate mesh transition and the outer shell for the coarser mesh
3) for the Streamline plots you must define the area where the lines should start, for your case typically the magnet cylinder top and bottom.

To learn all thee plot techniques it's easier to start with a simple N-S pole magnet in the centre of a sphere, then test out how to define BCs, and plot the desired variables, then you can go for a more complex magnetic set up as you have

--
Good luck
Ivar
Hi A few comments 1) you should use a manual mesh with a finer settings for the magnets and use the coarser for the sphere 2) I can advise you to centre the sphere in the middle of your magnet array (now its sidewise) minor effect, but essential if you use infinite elements later on 3) another trick is to have two spheres, one about 2x the magnet size to allow for better finer/intermediate mesh transition and the outer shell for the coarser mesh 3) for the Streamline plots you must define the area where the lines should start, for your case typically the magnet cylinder top and bottom. To learn all thee plot techniques it's easier to start with a simple N-S pole magnet in the centre of a sphere, then test out how to define BCs, and plot the desired variables, then you can go for a more complex magnetic set up as you have -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 7 years ago Jun 21, 2017, 8:25 p.m. EDT

Hej Ivar,
Thankx Ivar helping ... I made sphere almost 2 times the circular ring of magnet and made beautiful eye so that i can see the simulated result inside it.... When i simulate my result i could not see magnetism inside .. I do not know why ?

This time i used "Magnetic Field" instead of "Source current" last time...

And define Magnetic field Ho=5000 A/m for north and Ho= -5000 or South alone z -axis .... Did i did correct?

What do you mean by "magnetisation M". You mean Define in "Ampera Law" Magnestisation M? Well i use Ampera Law Magnestisation M= 5000 , Did i did correct ?

I am new in comsol and this is my second diagram simulating in comsol .... How can i define Coodinate frame for my magnets North and south ... Where i can find this option...

Moreover, how can i define symmetry ... I dont know this option.

I have attached my file ... If you could please look around and solve my problem ..

I will be thankful


Best Regards,
Mehdi


Dear Sir,
Based on indicating south and north poles through magnetization constitutive relation, I still have some challenges. I am simulating 2 rectangular magnets facing each other, with intention of having the same pole. Hence I indicate positive for one magnet and negative for the other. The result I have for magnetic force is that of positive force, whereas it should be negative. Could you please help me how to indicate North and South pole of a rectangular magnet? Thanks. Akin
[QUOTE] Hej Ivar, Thankx Ivar helping ... I made sphere almost 2 times the circular ring of magnet and made beautiful eye so that i can see the simulated result inside it.... When i simulate my result i could not see magnetism inside .. I do not know why ? This time i used "Magnetic Field" instead of "Source current" last time... And define Magnetic field Ho=5000 A/m for north and Ho= -5000 or South alone z -axis .... Did i did correct? What do you mean by "magnetisation M". You mean Define in "Ampera Law" Magnestisation M? Well i use Ampera Law Magnestisation M= 5000 , Did i did correct ? I am new in comsol and this is my second diagram simulating in comsol .... How can i define Coodinate frame for my magnets North and south ... Where i can find this option... Moreover, how can i define symmetry ... I dont know this option. I have attached my file ... If you could please look around and solve my problem .. I will be thankful Best Regards, Mehdi [/QUOTE] Dear Sir, Based on indicating south and north poles through magnetization constitutive relation, I still have some challenges. I am simulating 2 rectangular magnets facing each other, with intention of having the same pole. Hence I indicate positive for one magnet and negative for the other. The result I have for magnetic force is that of positive force, whereas it should be negative. Could you please help me how to indicate North and South pole of a rectangular magnet? Thanks. Akin

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.