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.

Probe setting

Please login with a confirmed email address before reporting spam

Hello, dear reader, I want to set a probe in the magnetic field generated by a wire. The probe can detect the sum of the magnetic flux in the direction I specify, but I don't know how to set the expression of the probe.

I want to use T=T1cos β 1+T2cos β2+...+Tx cosβ x to set Probe, but how should the comsol corresponding to this variable Tx and variable β be set?

Or do you have a better way?!


5 Replies Last Post Feb 24, 2021, 10:47 a.m. EST
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 4 years ago Feb 22, 2021, 10:50 p.m. EST
Updated: 4 years ago Feb 22, 2021, 10:52 p.m. EST

First, I apologize for not knowing what your variables (T and beta) mean. Anyway, do you want to probe the magnetic field value(s) or the flux (field times area)? If you all you want is the field, use point probes wherever you want the values. But if you want to compute/probe the flux through some area, create that area (such as by a rectangle, circle, etc., in a workplane embedded in your model) and define a probe variable of type "integral" upon that area, to use to compute the flux through it.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
First, I apologize for not knowing what your variables (T and beta) mean. Anyway, do you want to probe the magnetic field value(s) or the flux (field times area)? If you all you want is the field, use point probes wherever you want the values. But if you want to compute/probe the flux through some area, create that area (such as by a rectangle, circle, etc., in a workplane embedded in your model) and define a probe variable of type "integral" upon that area, to use to compute the flux through it.

Please login with a confirmed email address before reporting spam

Posted: 4 years ago Feb 23, 2021, 5:35 a.m. EST

First, I apologize for not knowing what your variables (T and beta) mean. Anyway, do you want to probe the magnetic field value(s) or the flux (field times area)? If you all you want is the field, use point probes wherever you want the values. But if you want to compute/probe the flux through some area, create that area (such as by a rectangle, circle, etc., in a workplane embedded in your model) and define a probe variable of type "integral" upon that area, to use to compute the flux through it.

What I want to do is a simulation of measuring a copper wire with a Hall sensor. However, it is well known that only the magnetic field lines perpendicular to the plane of the Hall element are effective magnetic fluxes, which means that the β direction of the magnetic field from all directions Component (multiple magnetic fields in multiple directions in space).

That's what I said T= Tx*cosβx , The magnetic field lines passing through the Hall element come from x directions and have x values, so we only need the component in the vertical plane, which is represented by cos β here. β is the angle between the Hall plane and the corresponding magnetic field line.

So I now know how to measure all the magnetic flux of the Hall element with a probe, but how to select the effective magnetic flux from all the magnetic flux passing through the element, this is my current problem.

How to set up the probe so that I can get the magnetic flux perpendicular to the plane from the total magnetic flux passing through the Hall element plane.

>First, I apologize for not knowing what your variables (T and beta) mean. Anyway, do you want to probe the magnetic field value(s) or the flux (field times area)? If you all you want is the field, use point probes wherever you want the values. But if you want to compute/probe the flux through some area, create that area (such as by a rectangle, circle, etc., in a workplane embedded in your model) and define a probe variable of type "integral" upon that area, to use to compute the flux through it. What I want to do is a simulation of measuring a copper wire with a Hall sensor. However, it is well known that only the magnetic field lines perpendicular to the plane of the Hall element are effective magnetic fluxes, which means that the β direction of the magnetic field from all directions Component (multiple magnetic fields in multiple directions in space). That's what I said T= Tx*cosβx , The magnetic field lines passing through the Hall element come from x directions and have x values, so we only need the component in the vertical plane, which is represented by cos β here. β is the angle between the Hall plane and the corresponding magnetic field line. So I now know how to measure all the magnetic flux of the Hall element with a probe, but how to select the effective magnetic flux from all the magnetic flux passing through the element, this is my current problem. How to set up the probe so that I can get the magnetic flux perpendicular to the plane from the total magnetic flux passing through the Hall element plane.

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 4 years ago Feb 23, 2021, 7:56 p.m. EST
Updated: 4 years ago Feb 23, 2021, 8:00 p.m. EST

The three components of the magnetic induction B are already available to you separately, so just pick the one you are interested in and integrate it. For example, if you are using the AC/DC module with the magnetic fields (mf) configuration, the components of B available to you are named as follows: mf.Bx, mf.By, and mf.Bz . You can also set up components normal to a surface. For any given surface, the normal is usuall nx, ny, nz. So the normal component of B in the general case, for the surface in question, would be (in the example above):

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
The three components of the magnetic induction B are already available to you separately, so just pick the one you are interested in and integrate it. For example, if you are using the AC/DC module with the magnetic fields (mf) configuration, the components of B available to you are named as follows: mf.Bx, mf.By, and mf.Bz . You can also set up components normal to a surface. For any given surface, the normal is usuall nx, ny, nz. So the normal component of B in the general case, for the surface in question, would be (in the example above): nx*mf.Bx+ny*mf.By+nz*mf.Bz

Please login with a confirmed email address before reporting spam

Posted: 4 years ago Feb 24, 2021, 6:29 a.m. EST

The three components of the magnetic induction B are already available to you separately, so just pick the one you are interested in and integrate it. For example, if you are using the AC/DC module with the magnetic fields (mf) configuration, the components of B available to you are named as follows: mf.Bx, mf.By, and mf.Bz . You can also set up components normal to a surface. For any given surface, the normal is usuall nx, ny, nz. So the normal component of B in the general case, for the surface in question, would be (in the example above): nxmf.Bx+nymf.By+nz*mf.Bz

Thank you for your response. I still have some doubts. Do I need to add a function to the expression mf.Bx to express the normal of the plane you mentioned above? (For example, mf.Bx*cosβ, β is the angle between the normal and the plane, assuming that the plane is in any direction)

I am confused about the specific positional relationship between the normal and mf.Bx, mfBy, and mfBz

>The three components of the magnetic induction B are already available to you separately, so just pick the one you are interested in and integrate it. For example, if you are using the AC/DC module with the magnetic fields (mf) configuration, the components of B available to you are named as follows: mf.Bx, mf.By, and mf.Bz . You can also set up components normal to a surface. For any given surface, the normal is usuall nx, ny, nz. So the normal component of B in the general case, for the surface in question, would be (in the example above): >nx*mf.Bx+ny*mf.By+nz*mf.Bz Thank you for your response. I still have some doubts. Do I need to add a function to the expression mf.Bx to express the normal of the plane you mentioned above? (For example, mf.Bx*cosβ, β is the angle between the normal and the plane, assuming that the plane is in any direction) I am confused about the specific positional relationship between the normal and mf.Bx, mfBy, and mfBz

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 4 years ago Feb 24, 2021, 10:47 a.m. EST
Updated: 4 years ago Feb 24, 2021, 10:50 a.m. EST

The angle between the normal to the plane, and the plane, is 90 deg. When you define the probe, if you define it as on the plane in question, then (nx,ny,nz) will be the vector components of a unit normal to that plane. n dot B (as constructed by the expression I provided) yields the component of B that is normal to that plane. If you aren't convinced, then you should definitely test it for yourself with other simple models, where you can know the answer analytically in advance and can compare it to the numerical result.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
The angle between the normal to the plane, and the plane, is 90 deg. When you define the probe, if you define it as on the plane in question, then (nx,ny,nz) will be the vector components of a unit normal to that plane. n dot B (as constructed by the expression I provided) yields the component of B that is normal to that plane. If you aren't convinced, then you should definitely test it for yourself with other simple models, where you can know the answer analytically in advance and can compare it to the numerical result.

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.